RF Network Planning and Optimization Service V200R001

RF Network Planning and Optimization Service
V200R001
Technical Guide to Cell Parameter Planning During Network Design (PCI, PRACH,
Neighboring Cell, and TA&TAL)
Issue 01
Date 2016-05-30
HUAWEI TECHNOLOGIES CO., LTD.

Copyright © Huawei Technologies Co., Ltd. 2016. All rights reserved.
No part of this document may be reproduced or transmitted in any form or by any means without prior
written consent of Huawei Technologies Co., Ltd.
Trademarks and Permissions
and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd.
All other trademarks and trade names mentioned in this document are the property of their respective
holders.
Notice
The purchased products, services and features are stipulated by the contract made between Huawei and
the customer. All or part of the products, services and features described in this document may not be
within the purchase scope or the usage scope. Unless otherwise specified in the contract, all statements,
information, and recommendations in this document are provided "AS IS" without warranties, guarantees
or representations of any kind, either express or implied.
The information in this document is subject to change without notice. Every effort has been made in the
preparation of this document to ensure accuracy of the contents, but all statements, information, and
recommendations in this document do not constitute a warranty of any kind, express or implied.
Huawei Technologies Co., Ltd.

Address: Huawei Industrial Base
Bantian, Longgang
Shenzhen 518129
People's Republic of China
Website: http://www.huawei.com
Email: [email protected]
Contents
Contents
Contents...............................................................................................................................................ii
1 LTE PCI Planning............................................................................................................................5
1.1 Overview...........................................................................................................................................................................5
1.1.1 Background and Benefits...............................................................................................................................................5
1.1.2 Introduction....................................................................................................................................................................5
1.1.3 Application Restriction...................................................................................................................................................6
1.1.4 Maturity..........................................................................................................................................................................6
1.2 Principles...........................................................................................................................................................................7
1.2.1 Technical Principles of PCI Planning.............................................................................................................................9
1.2.2 Technical Principles of PCI Check...............................................................................................................................11
1.2.3 Technical Principles of PCI Re-allocation....................................................................................................................11
1.3 Scenario Description.......................................................................................................................................................12
1.4 Application Guide...........................................................................................................................................................12
1.4.1 U-Net PCI Planning.....................................................................................................................................................12
1.4.2 SmartRNO PCI Planning.............................................................................................................................................34
2 PRACH ZC Root Sequence Planning........................................................................................66

2.1 Overview.........................................................................................................................................................................66
2.1.1 Background and Benefits.............................................................................................................................................66
2.1.2 Introduction..................................................................................................................................................................67
2.1.3 Restrictions...................................................................................................................................................................67
2.1.4 Maturity........................................................................................................................................................................68
2.2 Principles.........................................................................................................................................................................68
2.2.1 Introduction..................................................................................................................................................................68
2.2.2 PRACH ZC Root Sequence Planning Principles.........................................................................................................85
2.2.3 PRACH ZC Root Sequence Check Principles.............................................................................................................88
2.2.4 PRACH ZC Root Sequence Reallocation....................................................................................................................89
2.3 Application Scenarios......................................................................................................................................................89
2.4 Application Guide...........................................................................................................................................................90
2.4.1 U-Net PRACH ZC Root Sequence Planning...............................................................................................................90
2.4.2 SmarRNO ZC Root Sequence Planning.....................................................................................................................104
Issue () .
Contents
3 LTE Neighboring Cell Planning.....................................................................................................................134

3.1 Overview............................................................................................................................................................................134
3.1.1 Background and Benefits......................................................................................................................................................134
3.1.2 Introduction......................................................................................................................................................................134
3.1.3 Application Restriction..........................................................................................................................................................135
3.1.4 Maturity...........................................................................................................................................................................135
3.2 Principles............................................................................................................................................................................135
3.2.1 Intra-frequency Neighboring Cell Planning...............................................................................................................136
3.2.2 Inter-frequency Neighboring Cell Planning...................................................................................................................................137
3.2.3 Inter-RAT Neighboring Cell Planning..........................................................................................................................................137
3.3 Application Guidance..............................................................................................................................................................137
3.3.1 Neighboring Cell Planning Using U-Net......................................................................................................................................137
3.3.2 SmartRNO Neighboring Cell Planning........................................................................................................................154
4 TA&TAL Planning......................................................................................................................174
4.1 Overview.......................................................................................................................................................................174
4.2 Background and Benefits..............................................................................................................................................174
4.2.1 Introduction................................................................................................................................................................175
4.2.2 Requirements and Application Scenarios...................................................................................................................189
4.2.3 Maturity......................................................................................................................................................................189
4.3 Principles.......................................................................................................................................................................190
4.3.1 TA...............................................................................................................................................................................190
4.3.2 TAL.............................................................................................................................................................................190
4.3.3 LTE TAU and Paging Principles and Procedures.......................................................................................................193
4.3.4 TAC and TAL Configurations....................................................................................................................................194
4.3.5 Evaluating TA/TAL Baseline Specifications..............................................................................................................194
4.4 TA/TAL Planning..........................................................................................................................................................203
4.4.1 Planning Preparations and Outputs............................................................................................................................203
4.4.2 TA/TAL Planning Principles......................................................................................................................................204
4.4.3 Scenario-based TA/TAL Planning Solutions..............................................................................................................209
4.4.4 Suggestions for TA/TAL Planning Baselines.............................................................................................................214
4.4.5 Criteria for Evaluating the TAL Planning Solution....................................................................................................215
4.5 Checking and Evaluating TAs and TALs......................................................................................................................218
4.5.1 Application Scenarios.................................................................................................................................................218
4.5.2 Preparations................................................................................................................................................................218
4.5.3 Geographically Displaying and Checking TAs/TALs................................................................................................219
4.6 TA/TAL Replanning......................................................................................................................................................224
4.6.1 Scenarios....................................................................................................................................................................224
4.6.2 Principles....................................................................................................................................................................224
4.7 TA/TAL Optimization...................................................................................................................................................226
4.8 Application Guidance....................................................................................................................................................226
Issue () .
Contents
5 Application Cases........................................................................................................................228
5.1 PCI Planning.................................................................................................................................................................228
5.1.1 New Site Deployment................................................................................................................................................228
5.1.2 Capacity Expansion Scenario.....................................................................................................................................230
5.1.3 Replanning Scenario..................................................................................................................................................231
5.2 PCI Check.....................................................................................................................................................................233
5.2.1 Checking PCI Conflicts..............................................................................................................................................233
5.2.2 Evaluating PCI Planning............................................................................................................................................235
5.3 PRACH ZC Root Sequence Planning...........................................................................................................................240
5.3.1 New Site Deployment Scenario.................................................................................................................................240
5.3.2 Capacity Expansion Scenario.....................................................................................................................................243
5.3.3 Replanning Scenario..................................................................................................................................................244
5.3.4 Highway Speed Scenario...........................................................................................................................................247
5.4 PRACH ZC Root Sequence Check...............................................................................................................................255
5.4.1 Checking PRACH ZC Root Sequence Conflicts.......................................................................................................256
5.4.2 Evaluating the Reuse Isolation of PRACH ZC Root Sequences...............................................................................257
5.5 Configuration Delivery Checking.................................................................................................................................258
5.6 TAL Planning Case........................................................................................................................................................259
A Appendix......................................................................................................................................260
A.1 Impact of PCI Mod3 Staggering/Alignment on the Network Performance.................................................................260
A.2 Cell Searching and Downlink Synchronization...........................................................................................................261
A.2.1 Cell Searching-Why..................................................................................................................................................261
A.2.2 Cell Searching-What.................................................................................................................................................261
A.2.3 Cell Searching-When and How.................................................................................................................................262
A.3 Relationship Between the PCI and the Synchronization Code....................................................................................263
A.3.1 Parsing the PSS.........................................................................................................................................................263
A.3.2 Parsing the SSS.........................................................................................................................................................267
A.4 Relationship Between the PCI and the RS...................................................................................................................271
A.4.1 Time-Frequency Position of the CRS........................................................................................................................271
A.4.2 CRS Signal and Parsing............................................................................................................................................273
A.4.3 Functions of CRS Synchronization...........................................................................................................................274
A.5 Cyclic Shift Value.........................................................................................................................................................274
A.6 Number of ZC Root Sequences for Each Cell.............................................................................................................278
A.7 LTE TAU and Paging Principles and Procedures.........................................................................................................280
A.7.1 LTE TAU Principles and Procedure..........................................................................................................................280
A.7.2 LTE Paging Functions...............................................................................................................................................282
A.7.3 Paging Frames and POs.............................................................................................................................................282
A.7.4 LTE Paging Process...................................................................................................................................................283
A.7.5 Allocating Resources for LTE Paging Messages.......................................................................................................284
A.7.6 LTE Paging Mechanism............................................................................................................................................288
Issue () .
Contents
A.7.7 Criteria for Evaluating LTE Paging Performance.....................................................................................................289

A.8 Typical Traffic Models on the MME............................................................................................................................289
Issue () .
Technical Guide to Cell Parameter Planning During Network
Design (PCI, PRACH, Neighboring Cell, and TA&TAL) 1 LTE PCI Planning
1 LTE PCI Planning
1.1 Overview
This document describes the LTE physical cell identification (PCI) feature, including the
principles and applications of PCI planning, PCI check, and PCI re-allocation.
This document is delivered to NIS and colleagues engaged in professional services
1.1.1 Background and Benefits

The PCI feature includes the PCI planning, PCI check (evaluation), and PCI re-allocation.
The PCI is an important parameter for an E-UTRAN cell, affecting the downlink signal
synchronization, demodulation, and handover. Allocating proper PCIs to LTE cells, checking
improper PCIs, and solving problems of improper PCIs are vital for LTE radio network
construction and maintenance.
For newly-deployed networks or capacity expansion, a proper PCI needs to be allocated to a
new E-UTRAN cell to ensure that the E-UTRAN cell works properly.
When the network topology is modified, the planned PCIs are improper, or parameter settings
of live NEs are modified (such as the modification of neighbor relationship), the PCIs need to
be re-allocated for the E-UTRAN cells involved.
After PCI planning is complete, check, evaluate, and analyze the planning result, and slightly
adjust the PCI planning, and send the planning result to NEs.
1.1.2 Introduction
Figure 1.1.2.1.1.1.1.1 shows the PCI feature:
Figure 1.1.2.1.1.1.1.1 PCI feature
PCI feature
PCI planning PCI check PCI re-allocation

1.1.3 Application Restriction

Specification Restriction
N/A
Application Scenario Restriction

Table 1.1.3.1.1.1.1.1.1 lists the PCI application scenario restriction.
Table 1.1.3.1.1.1.1.1.1 PCI application scenarios
Scenario Restrictions Workaround
The latitude If the latitude and longitude are not Obtain the latitude and longitude
and longitude specified, the PCI planning, PCI of the NE.
of an NE are check, or PCI re-allocation cannot
not specified. be performed using the U-Net.
The PCIs need Currently, the U-Net does not The Wireless Parameter Auto-
to be planned support modeling for indoor configuration feature can be used
for indoor scenarios, including floor to complete the PCI planning
sites. information, room information, and online for micro base stations.
whether the indoor cells belong to
the same building.
Due to lack of key information, the
U-Net is not proper for PCI
planning for indoor cells.
1.1.4 Maturity
PCI auto-planning supports the following scenarios:
 New site deployment
 Capacity expansion: greenfield capacity expansion and mixed network capacity
expansion
 Re-allocation
PCI planning supports the following scenarios:
 Intra-frequency networking
 Inter-frequency networking: refarming, multiple carriers covering the same area (ensure
that inter-frequency cells served on the same site and covering the same areas have the
same PCIs), and inter-frequency cells covering different areas
 Multi-antenna, remote radio frequency, and single frequency network (SFN)
 Cells with compact bandwidths existing on the network
PCI planning principles are as follows:
 The PCI reuse tier and distance for LTE cells can be configured by users to set the
minimum isolation for PCI reuse.
 The impact on main synchronization code, downlink reference signal (RS) and uplink
RS is considered to ensure network KPIs.
The U-Net V300R010/SmartRNO V300R001 can check the following:
 PCI reuse distance and tier (the tier of neighboring cells)
 LTE cells with the same PCI
1.2 Principles
In an LTE system, 504 PCIs are available. These PCIs are categorized into 168 groups, three
in each group.
Table 1.2.1.1.1.1.1.1.1 PCI groups
Group Number PCIs in Each PCI Group
0 0, 1, and 2
1 3, 4, and 5
. .
. .
. .
167 501, 502, and 503
The PCI is an important parameter for an E-UTRAN cell, and each E-UTRAN cell is
allocated a PCI. The PCI affects the downlink synchronization and handover.
When a lot of E-UTRAN cells exist in an LTE system, some cells with the same frequency
must share one PCI. To avoid PCI conflicts, intra-frequency cells closely located cannot share
the same PCI.
Collision free: Neighboring intra-frequency cells cannot share the same PCI.
Confusion free: If two E-UTRAN cells are neighboring cells to an E-UTRAN cell and the two
cells use the same frequency, the two cells cannot use the same PCI.
PCI collision and PCI confusion are two PCI conflict cases. The LTE radio network must be
free of PCI conflicts.
PCI Collision
Figure 1.2.1.1.1.1.1.2 shows PCI collision.
Figure 1.2.1.1.1.1.1.2 PCI collision
When two adjacent cells have the same frequency and PCI, PCI collision occurs. In this
situation, if a UE is in the overlapped area of cell A and cell B shown in Figure 1.2.1.1.1.1.1.2,
the downlink synchronization and signal demodulation deteriorate. For example, the
synchronization success rate is affected, and the PCFICH and PHICH are prone to
interference.
PCI Confusion
Figure 1.2.1.1.1.1.1.3 shows PCI confusion.
Figure 1.2.1.1.1.1.1.3 PCI confusion
As shown in Figure 1.2.1.1.1.1.1.3, the serving cell has multiple intra-frequency cells using
the same PCI. In this situation, if a UE detects cell B, and handover requirements are met, the
UE reports the PCI of cell B to cell A. However, cell A cannot determine whether the target
cell for a handover is cell B or cell C. Therefore, the handover is not initiated.
If the UE supports cell global identification (CGI) measurement, and cell A detects PCI confusion shown
in Figure 1.2.1.1.1.1.1.3, the UE reads the CGI of cell B, the detected cell, and is handed over to cell B.
The main purpose of PCI planning is to avoid PCI conflicts. Then, consider the impact of
PCIs on the network and allocate proper PCIs to the cells.
This document describes the PCI feature based on the U-Net for technical service engineers.
PCI Collision Detection and Self-Optimization Feature Parameter Description is used to automatically
detect PCI conflicts and re-allocate PCIs. For details, see PCI Collision Detection and Self-Optimization
Feature Parameter Description.
The PCI auto-planning function of Huawei CME can be used to allocate PCIs for new sites, capacity
expansion, and PCI replanning scenarios. For details, see the online help of M2000.
1.1.1 Technical Principles of PCI Planning

The purpose of PCI planning is to allocate E-UTRAN cells to be planned with proper PCIs to
ensure PCI conflict free and enhance network performance.
Therefore, the following principles must be followed in PCI planning:
 Only PCIs available to users can be allocated to cells to be planned.
 When allocating PCIs to cells to be planned, ensure that the PCIs are conflict free. The
cells to be planned do not use the PCIs of intra-frequency cells within the reuse distance,
or the PCIs of intra-frequency cells within a certain tier of neighboring cells.
If two cells are geographically far from each other, the signal isolation of the two cells is large. If two
cells are geographically close to each other, the signal isolation of the two cells is small, and signals are
prone to overlapping or interference. To ensure normal synchronization and demodulation, cells that are
geographically close to each other use different PCIs. Therefore, the PCI reuse distance can be used to
define the set of E-UTRAN cells that must use different PCIs than those of the cells to be planned.
The neighboring cells configured in the parameter of an NE are the first-tier neighboring cells of the NE
(direct neighboring cell). If cell A and cell B are neighboring cells, and cell B and cell C are neighboring
cells, cell A and cell C are second-tier neighboring cells to each other. Theoretically, a higher tier
between two cells indicates a lower possibility of signal interference. Normally, bidirectional intra-
frequency cells in the first and second tiers must use different PCIs to avoid PCI confusion, because PCI
confusion leads to handover failures.
 If multiple PCIs available for cells to be planned are conflict free, choose the PCIs with
which the adjacent intra-frequency cells served on the same site have different main
synchronization codes.
In the LTE system, three main synchronization codes numbering 0 to 2 and 168 secondary
synchronization codes numbering 0 to 167 are defined. One PCI consists of one main synchronization
code and three secondary synchronization codes.
The main synchronization code has the length of 62 and consists of three values of the Zadoff-Chu
sequence, and the main synchronization code is of good orthogonality. The secondary synchronization
code is the different sequence in the two synchronization timeslots (0 and 10) in 10 ms (there are 168
combinations). The main synchronization code has a better orthogonality than the secondary
synchronization code does. There are 504 PCIs that consist of different main synchronization codes and
secondary synchronization codes. For details, see section C"Cell Searching and Downlink
Synchronization."
PCI Mod3 specifies the main synchronization code, which is vital in the signal synchronization.
Therefore, the values of PCI Mod3 for adjacent cells at the same site (with similar azimuths) must be
different.
 If multiple PCIs of the cells to be planned meet the requirements of 2 and 3, ensure that
the downlink RSs of cells to be planned and adjacent intra-frequency cells are stagger.
The position of downlink RS of E-UTRAN cells in the frequency domain is related to their PCIs. For
details, see section R"Relationship Between the PCI and the RS." In the initial phase of network
deployment, E-UTRAN cells are of light load, and the downlink RS of adjacent cells is the main source
of interference. In this situation, if the frequency-domain position of downlink RS of the serving cell is
the same as that of the adjacent cell, the downlink RS of the serving cell deteriorates severely, which
affects the channel estimation accuracy. As a result, the serving cell chooses a lower demodulation order,
and the downlink throughput drops.
When the E-UTRAN cell is served on a single antenna port and the PCI Mod 6 of the serving cell and
the adjacent cell is staggered, the downlink RSs of the serving cell and its adjacent cell are staggered in
the frequency domain. When the E-UTRAN cell is served by two or more antenna ports and the value of
PCI Mod3 of the serving cell is different from that of the adjacent cell, the downlink RSs of the serving
cell and its adjacent cell are staggered in the frequency domain. If the serving cell is served on a single
antenna port (or multiple antenna ports) and the adjacent cell is served on multiple antenna ports (or a
single antenna port), ensure that the value of PCI Mod3 of the serving cell is different from that of the
adjacent cell.
According to requirements in 3, the PCI planning ensures that the main synchronization codes in PCI
Mod3 are different among adjacent cells served on the same site, and therefore the downlink RSs of
these adjacent cells are staggered.
Compared with the number of E-UTRAN cells on a network, the resources of PCI Mod3 or PCI Mod 6
are insufficient. Therefore, ensure that the cell to be planned has different PCI Mod3 or PCI Mod 6 than
the intra-frequency adjacent cell with strong interference does.
 If multiple PCIs of cells to be planned meet the requirements of 2, 3, and 4, ensure that
the root sequence number of the uplink RS of the cell to be planned is different from that
of the intra-frequency adjacent cell.
The uplink RS uses the Zadoff-Chu sequence. The uplink RS in the cell is determined by the group
number u of the Zadoff-Chu sequence. Different group numbers generate pilot sequences of different
lengths, and the available sequences can be added by cyclic shifting of pilot sequences. If the value of u
differs in the two cells, the Zadoff-Chu sequences of the two cells are not strongly related. In this case, it
can be considered that the uplink RS interference is small between the two cells. If the value of u in one
cell is the same as that in the other one, the Zadoff-Chu sequences of the two cells are strongly related
and the uplink RS interference between the two cells is aggravated.
The value of u ranges from 0 to 29.
When neither of the group hopping or sequence hopping is applied, the value of u can be calculated by
using the following formulas:
PUSCH: u = (PCI + deltss) Mod30 (deltss ranges from 0 to 29)
PUCCH and SRS: u = (PCI) Mod30
When the group hopping is applied, the value of u can be calculated by using the following formulas:
PUSCH: u = (fgh + PCI + deltss) Mod30
fgh indicates the random pattern number in the range from 0 to 16, which must be consistent with floor
(PCI/30).
PUCCH and SRS: u = (fgh + PCI) Mod30
Currently, the value of u is calculated in the condition that neither of the group hopping
nor sequence hopping is applied and the value of deltss remains the same. In this case, if
the value of PCI Mod30 differs in two adjacent cells, the values of u in the two cells are
different, and the uplink RS quality and access success rate of edge users can be
improved.
 After the requirements of 2, 3, 4 and 5 are satisfied, there are extra available PCIs and
PCI reuse, ensure that the PCI reuse isolation is large based on the distance and number
of tiers between the cells to be planned and the planned cells.
The number of layers in the previous sentence indicates the number of base stations between two cells.
The number of layers indicates the geographical isolation, and can be calculated based on the latitude
and longitude of cells. The tier indicates the logical isolation, and can be calculated based on the
neighbor relationship.
The PCI reuse distance and reuse layers are calculated based on the latitude and longitude and can be
used when the remote radio unit (RRU) is used on the network.
 Other requirements:
− In the refarming scenario, if the cell to be planned and the planned cell use different
frequencies but overlap on the spectrum, the two cells are considered as intra-
frequency cells. Ensure that their PCIs are conflict free, uplink and downlink RSs are
staggered, and the reuse isolation is sufficient to avoid network replanning when the
network is evolved into an intra-frequency network.
− When multiple carriers cover the same area, ensure that the cells served on the same
site covering the same area are allocated with the same PCI to facilitate parameter
maintenance.
− In the SFN, multi-antenna, and RRU scenarios, consider the latitude and longitude,
and azimuth of each E-UTRAN cell to avoid PCI conflict, and ensure that the uplink
and downlink RSs are properly staggered.
− If the cells to be planned use the compact bandwidth, only PCIs in the frequency
band are allocated to the cells to avoid that the PCFICH and PHICH are established
on unavailable RBs.
− Ensure that the three intra-frequency adjacent cells served by the same site are
allocated with the PCIs in the same group to facilitate parameter maintenance.
− For new deployment, allocate proper PCIs to the new cells in the area to be planned.
For capacity expansion, allocate proper PCIs to new cells, and do not modify the
PCIs of existing cells. For re-allocation, re-allocation PCIs for the existing cells in the
area to be planned.
1.2.2 Technical Principles of PCI Check

The purpose of PCI check is to find PCI conflicts, including PCI collision and PCI confusion,
on the network.
 PCI collision
− If the distance between two E-UTRAN cells that use the same frequency and the
same PCI is less than the user-defined PCI reuse distance, PCI collision exists
between the two cells.
− If the PCIs of two cells that are intra-frequency neighboring cells are the same, PCI
collision exists between the two cells.
 PCI confusion
If two cells are second-tier intra-frequency neighboring cells to each other (for example,
the two cells are neighboring cells of a cell and the two cells use the same frequency)
and the PCIs of the two cells are the same, PCI confusion exists between the two cells.
Besides, PCI check can be used to evaluate the following on the network:
 PCI reuse isolation, including reuse distance and reuse tier
 Staggering of PCI Mod3 of adjacent cells (related to the downlink RS)
 Staggering of PCI Mod30 of adjacent cells (related to the uplink RS)
1.2.3 Technical Principles of PCI Re-allocation

The principles of PCI re-allocation are similar to those of PCI planning.
The purpose of PCI re-allocation is to allocate proper PCIs for cells with PCI conflicts and,
for example, to eliminate PCI conflicts.
If one of the following occasions occurs, re-allocate proper PCIs to cells with improper PCIs
based on the PCI planning principles:
 The PCI Mod3 of adjacent cells is not staggered.
 The PCI Mod30 of adjacent cells is not staggered.
 PCI reuse isolation is not sufficient. The reuse distance and reuse tiers are not up to the
expectation.
1.3 Scenario Description

Figure 1.3.1.1.1.1.1.1 Procedures related to the PCI feature
Check and evaluate Re-allocate PCIs to

Allocate PCIs to the the PCIs of E-UTRAN
new E-UTRAN cell. cells with improper
cells on the network. PCIs.
 In the new deployment and capacity expansion scenario, allocate proper PCIs to new E-
UTRAN cells to ensure that the E-UTRAN cells work properly.
 After PCI planning is complete, evaluate and analyze the planning result and then send
the result to the NEs. During network maintenance, the PCIs of E-UTRAN cells also
need to be checked and evaluated.
 E-UTRAN cells with improper PCIs can be found through evaluation and check. To
ensure that the network runs properly or enhance the network performance, re-allocate
proper PCIs to E-UTRAN cells with improper PCIs.
1.4 Application Guide

1.1.2 U-Net PCI Planning
PCI Planning
 Overview
− In the new deployment and capacity expansion scenarios, allocate PCIs to new E-
UTRAN cells.
− During PCI planning, collect the required data for the planning, prepare the
engineering parameter table, and then use a tool to plan PCIs.
 Tool
− Tool name: U-Net
− Tool version: V300R010
− R&D engineer: Yang Wenhua (employee ID: 00181336)
− Available at:
http://support.huawei.com/carrier/navi?lang=zh#col=software&path=PBI1-
7851894/PBI1-8132365/PBI1-8132371/PBI1-21395144/PBI1-13953
 Input
− Projection coordinate system

− (Optional) Electronic map
− Network engineering parameter information (For details, see Table 1.4.1.1.1.1.1.1.1,
Table 1.4.1.1.1.1.1.1.2, Table 1.4.1.1.1.1.1.1.3, Table 1.4.1.1.1.1.1.1.4, and Table
1.4.1.1.1.1.1.1.5.)
− PCI planning policy (For details, see Table 1.4.1.1.1.1.1.1.6.)
 If PCI Mod3 of adjacent cells needs to be staggered, calculate interference among adjacent cells
based on the topology or coverage prediction. Calculation based on the coverage prediction costs
more time and is controlled by more parameters, such as the propagation model, shadow fading, and
signal level threshold, than calculation based on the topology. Calculation based on the topology
costs less time and can be performed without setting parameters related to the coverage prediction
on NEs.
 If Planning with Exist PCI is selected, PCIs are allocated to cells without PCIs and cells with
planned PCIs use the existing PCIs. If Planning with Exist PCI is not selected, PCIs of all cells are
cleared and re-allocated. Therefore, select Planning with Exist PCI during PCI planning in a
capacity expansion scenario.
 If PCI Mod3 Matching Azimuth order is selected in PCI planning, both the PCI Mod3 and
azimuth of each cell are in ascending order. On some networks, the azimuths of base stations are
basically the same (for example, azimuths of all cells served by the base stations are 0, 120, and
240). In this situation, if PCI Mod3 of each cell is 0, 1, and 2 in ascending order, PCI Mod3 of
adjacent cells served by the adjacent based stations is staggered. If new sites are deployed for
capacity expansion and the azimuths of cells served by new sites are in the same patterns as those of
cells served by existing sites, manually allocate PCIs to new cells, the values of PCI Mod3 are 0, 1,
and 2 in ascending order. In this way, PCI Mod3 staggering of new cells and adjacent cells can be
ensured in a great possibility. If PCI Mod3 Matching Azimuth order is not selected, interference
among adjacent cells served by different sites is considered to ensure that PCI Mod3 is staggered.
However, the orders of PCI Mod3 and azimuth of each cell cannot be ensured. When the azimuth is
basically in a pattern, select PCI Mod3 Matching Azimuth order.
Table 1.4.1.1.1.1.1.1.1 Site information
Parameter Meaning Recommended Remarks

Value
Site Name Indicates the site N/A The U-Net uses the
name. site name as the
index keyword.
Longitude Indicates the site N/A N/A
longitude.
Latitude Indicates the site N/A N/A
latitude.
Altitude Indicates the site N/A This parameter is
altitude. used when
interference is
calculated based on
coverage prediction.
Table 1.4.1.1.1.1.1.1.2 Sector information

Value
Transceiver Indicates the sector The value is N/A

Name name. defined
manually.
Dx Indicates the N/A If the E-UTRAN cell is
horizontal offset of served by multiple antennas,
the antenna to the specify the horizontal offset
eNodeB. of each antenna to the base
station.
Dy Indicates the N/A If the E-UTRAN cell is
vertical offset of served by multiple antennas,
the antenna to the specify the vertical offset of
base station. each antenna to the base
station.
Antenna Indicates the N/A If the E-UTRAN cell is
antenna type. served by multiple antennas,
specify the type of each
antenna.
Height Indicates the N/A If the E-UTRAN cell is
antenna height. served by multiple antennas,
specify the height of each
antenna.
This parameter is used when
interference is calculated
based on coverage
prediction.
Azimuth Indicates the N/A If the E-UTRAN cell is
antenna azimuth. served by multiple antennas,
specify the azimuth of each
antenna.
Mechanical Indicates the N/A If the E-UTRAN cell is
Down tilt antenna mechanical served by multiple antennas,
downtilt. specify the mechanical
downtilt of each antenna.
based on coverage
prediction.

Value
Electrical Down Indicates the N/A If the E-UTRAN cell is

tilt antenna electrical served by multiple antennas,
downtilt. specify the electrical downtilt
of each antenna.
based on coverage
prediction.
Power Ratio Indicates the N/A If the E-UTRAN cell is
antenna power served by multiple antennas,
ratio. specify the power ratio of
each antenna.
based on coverage
prediction.
Number of Indicates the N/A N/A
Transmission number of antenna
Antenna Ports ports.
Table 1.4.1.1.1.1.1.1.3 Cell information

Value
Cell Name Indicates the N/A The U-Net uses the cell name as
cell name. the index keyword.
Frequency Indicates the N/A The frequency and bandwidth
Band attribute of the needs to be specified. If the
cell frequency actual bandwidth is less than the
band. standard bandwidth, the
bandwidth is considered as the
compact bandwidth.
RS Power Indicates the 15.2 dBm This parameter is used when
transmit power interference is calculated based
of the reference on coverage prediction.
signal (RS).
Main Indicates the N/A This parameter is used when
Propagation propagation interference is calculated based
Model model. on coverage prediction.

Value
Main Indicates the N/A This parameter is used when

Calculation calculation interference is calculated based
Radius radius of the on coverage prediction.
path loss
matrix.
PCI Indicates the N/A If this parameter is null, the cell
existing PCI of is a new cell.
an E-UTRAN
cell.
PCI Reuse Indicates the This parameter is On a live network, some LTE
Distance(Km) PCI reuse set according to the eNodeBs are deployed at the
distance. actual scenario. For same site where GSM BTSs and
example: UMTS NodeBs are deployed.
In common urban Therefore, you can set this
areas, this parameter by referring to the
parameter is set to reuse distance of PSC on the
4. UMTS network.
In suburban areas,
this parameter is set
to 10.
PCI Reuse Indicates the 2 N/A
Tier(Neighbor) PCI reuse tier.
Table 1.4.1.1.1.1.1.1.4 Intra-frequency neighboring cell information

Value
CellName Indicates the N/A The U-Net uses the cell name
source cell as the keyword for indexing an
name. E-UTRAN cell.
NCellName Indicates the N/A The U-Net uses the cell name
target cell name. as the keyword for indexing an
E-UTRAN cell.
Table 1.4.1.1.1.1.1.1.5 Inter-frequency neighboring cell information

Value
CellName Indicates the N/A The U-Net uses the cell name as the
source cell keyword for indexing an E-UTRAN
name. cell.

Value
NCellName Indicates the N/A The U-Net uses the cell name as the
target cell keyword for indexing an E-UTRAN
name. cell.
Table 1.4.1.1.1.1.1.1.6 Parameters involved in the PCI planning policy

Value
Available PCI Indicates the N/A This parameter is set according

available PCI to the actual scenario.
resources. Normally, indoor and outdoor
cells use different PCIs. For
example, in a Mexico office,
PCIs 0 to 440 are allocated to
outdoor sites, and PCIs 441 to
503 are allocated to indoor
sites.
In the boundary scenario, such
as the national boundary, cells
along the boundary belong to
different countries use different
PCIs. For example, European
countries must comply with a
certain protocol to ensure cells
belong to different neighboring
countries in the boundary area
must use different PCIs.
On some networks, PCIs are
reserved for follow-up capacity
expansion.
Planning With Indicates In the capacity N/A
Exist PCI whether to expansion scenario,
perform PCI re- if there are existing
allocation for cells in planned
planned cells. areas, select
Planning With
Exist PCI.

Value
PCI Mod3 Indicates that Deselect PCI On some networks, the

Matching both the PCI Mod3 Matching azimuths of base stations are
Azimuth order Mod3 and Azimuth order. basically the same (for
azimuth of the example, azimuths of all cells
cell are in served by the base stations are
ascending order. 0, 120, and 240). In this
situation, if PCI Mod3 of each
cell is 0, 1, and 2 in ascending
order, and the azimuths of new
sites are in the same pattern as
those as existing cells, manually
allocate PCIs to new cells.
When PCI Mod3 is 0, 1, and 2,
the PCI Mod3 staggering of the
new cell and adjacent cell can
be ensured in a great
possibility.
Iterative Indicates that Deselect Iterative If the PCI reuse distance is too
Generate PCI PCI reuse Generate PCI. large, resulting in PCI planning
distance iteration failures, select Iterative
is used. Generate PCI to reduce the
PCI reuse distance and ensure a
successful planning.
Distance Indicates the 20% This parameter becomes valid
Descending percentage of only when Iterative Generate
Percentage(%) decreased reuse PCI is selected.
distance due to
reuse distance
iteration.
Replan Indicates that the Deselect Replan During PCI planning, cells
adjacent cell PCIs of planned adjacent cell PCI. planned in the initial phase are
PCI cells' adjacent more likely to be allocated with
cells must be PCIs than cells planned in the
adjusted. later phase. In this situation,
change the PCI planning order
or do not allocate optimal PCIs
to cells planned in the initial
phase to enable cells planned in
the later phase to be allocated
with PCIs.
Inherit existed Indicates that Deselect Inherit This parameter becomes valid
PCI mod 3 PCI Mod3 existed PCI mod only when Replan adjacent
remains 3. cell PCI is selected.
unchanged after
re-allocation.

Value
Area Indicates the N/A If the user does not draw a

PCI planning polygon, the full map is for
area. planning.
If a user draws polygons on the
map, planning can be
performed in the areas in
polygons.
Topology Indicates Select Topology. N/A
interference
among adjacent
cells calculated
based on the
topology.
Prediction Indicates that Do not select Coverage prediction involves
interference Prediction. multiple parameters, such as the
among adjacent propagation model, shadow
cells is fading, and penetration loss.
calculated based Besides, the accurate values of
on coverage these parameters are difficult to
prediction. obtain. Different parameter
values result in different
planning results.
Resolution(m) Indicates the 50 m This parameter is used when
resolution of the interference is calculated based
predicted on coverage prediction.
coverage.
Min Signal Indicates the –120 dBm This parameter is used when
Level(dBm) RSRP threshold interference is calculated based
of an E-UTRAN on coverage prediction.
cell during
coverage
prediction.
Handover area Indicates the 6 dB This parameter is used when
threshold(dB) difference interference is calculated based
between on coverage prediction.
adjacent cell
signals and
serving cell
signals.
Shadowing Indicates that Do not select This parameter is used when
taken into shadow fading is Shadowing taken interference is calculated based
account considered. into account. on coverage prediction.

Value
Cell Edge Indicates the cell 75% This parameter is used when
Coverage edge coverage interference of adjacent cells is
Probability probability of calculated based on the
shadow fading. coverage prediction and
Shadowing taken into account
is selected.
The shadow fading variance of
the cluster must be specified.
For details, see U-Net
Operation Guide.
Indoor Indicates that the N/A This parameter is used when
Coverage penetration loss interference is calculated based
is taken into on coverage prediction.
consideration. The penetration loss of the
cluster needs to be specified.
For details, see U-Net
Operation Guide.
 Procedures
Step 1 Collect live network information, including NE engineering parameter information, cells to be
planned, and available PCIs for these cells.
Step 2 Prepare the engineering parameter table file according to the U-Net engineering parameter
template.
Step 3 Create a U-Net project.
Step 4 (Optional) Specify coordinates for the project and import the electronic map.
If the electronic map has a coordinate system, do not manually set the coordinate system.
If PCI planning is performed based on topology, the electronic map is not required.
If PCI planning is performed based on coverage prediction, import the electronic map to obtain a more
accurate planning result.
Step 5 Import engineering parameter files of NEs in the order of the site table, transceiver table, cell
table, and neighbor relationship.
If neighbor relationship is not imported, PCI conflicts among neighboring cells are not considered
during PCI planning.
Step 6 Set PCI auto-planning parameters, perform PCI auto-planning, and submit the planning result.
Step 7 Evaluate the planning result and manually and slightly adjust the planning result.
----End
 Output
Figure 1.4.1.1.1.7.1.1 shows the PCIs for the cells to be planned:
Figure 1.4.1.1.1.7.1.1 PCI planning result
Where,
Existing Code: indicates the existing PCI of the cell to be planned.
Suggest Code: indicates the exported PCI auto-planning result.
Confirm Code: indicates the PCI confirmed by the customer. After PCI planning is complete,
the user can manually modify the value of Confirm Code and submit the modified value as
the PCI to the cell table.
Min Reuse Distance: indicates the minimum PCI reuse distance of a cell to be
planned.
Min Reuse Tiers: indicates the minimum number of PCI reuse tiers of a cell to be planned.
 References and website
http://support.huawei.com/support/pages/kbcenter/view/product.do?
actionFlag=detailProductSimple&web_doc_id=SC0000764790&doc_type=ProductMan
ual&doc_type=ProductManual&saveBrowserLoged=true
PCI Check (Evaluation)

 Overview
− Application scenarios of the PCI check:
a. After PCI planning is complete, the PCI planning result must be checked and
evaluated.
b. During the network maintenance phase, PCIs on the live network must be
checked and evaluated.
− The following are checked and evaluated:
a. PCI conflicts (collision and confusion)
b. PCI reuse isolation, including reuse distance and reuse tier
c. Staggering of PCI Mod3 for neighboring cells
d. Staggering of PCI Mod30 for adjacent cells
− Evaluation methods:
a. Evaluation based on PCI result table: Used to summarize, filter, and record the
PCI reuse (of E-UTRAN cells with small PCI reuse isolation, such as the small
reuse distance and small number of reuse tiers) for filing and viewing.
b. Evaluation based on geographical presentation: Used to view the PCI reuse, and
staggering of PCI Mod3 and PCI Mod30.
c. Evaluation based on CDF statistics: Used to quantitatively evaluate PCI reuse
isolation in E-UTRAN cells. Especially in swap or PCI replanning scenarios,
this method can be used to quantitatively compare the PCI reuse before and after
swap or PCI replanning.
d. Evaluation based on coverage prediction (RS SINR when the neighboring cell is
not loaded): Used to quantitatively compare the staggering of PCI Mod3 before
and after swap or PCI replanning.
 Tool
− Available at:
7851894/PBI1-8132365/PBI1-8132371/PBI1-21395144/PBI1-13953
 Input
− Evaluation and check upon the PCI planning completion
The evaluation and check upon the PCI planning completion requires no parameter
settings. You can directly submit the planning result.
− Evaluation and check during network maintenance
Parameters of NEs listed in Table 1.4.1.1.1.7.1.1.1, Table 1.4.1.1.1.7.1.1.2, Error:
Reference source not found, Table 1.4.1.1.1.7.1.1.4, and Table 1.4.1.1.1.7.1.1.5 are
required.

Value
Site Name Indicates the site N/A The U-Net uses Site Name
name. as the index keyword.
longitude.
altitude.

Value
Transceiver Indicates the sector The value is N/A

Name name. defined
manually.
Dx Indicates the N/A If the E-UTRAN cell is served
horizontal offset of by multiple antennas, specify the
the antenna to the horizontal offset of each antenna
base station. to the base station.

Value
Dy Indicates the vertical N/A If the E-UTRAN cell is served

offset of the antenna by multiple antennas, specify the
to the base station. vertical offset of each antenna to
the base station.
Azimuth Indicates the N/A If the E-UTRAN cell is served
antenna azimuth. by multiple antennas, specify the
azimuth of each antenna.

Name Value
Cell Name Indicates the N/A The U-Net uses the cell name as the
cell name. keyword for indexing an E-UTRAN
cell.
Frequency Indicates the N/A The frequency and bandwidth needs
Band attribute of the to be specified. If the actual
cell frequency bandwidth is less than the standard
band. bandwidth, the bandwidth is
considered as the compact
bandwidth.
PCI Indicates the N/A N/A
existing PCI of
an E-UTRAN
cell.

Value
name. cell.
name. cell.
Table 1.4.1.1.1.7.1.1.5 Inter-frequency neighboring cell information

Value
name. cell.
name. cell.
 Procedures
Step 1 If the U-Net is used to perform the PCI auto-planning and the PCI planning needs to be
checked, submit the planning result. If the PCI check is used during network maintenance,
prepare engineering parameters, create a project, and import these engineering parameters to
the project.
Step 8 Check PCI conflicts, and evaluate the PCI reuse isolation, and staggering of PCI Mod3 and
PCI Mod30/PCI Mod6.
Step 9 Export the check (evaluation) report.
----End
 Checking PCI Conflicts

Evaluation criteria: whether PCI conflicts exist among E-UTRAN cells, including areas that
do not satisfy the minimum PCI reuse distance or areas within the second tier neighboring
cells.
− The following three methods are used to check PCI conflicts:
a. Export the result table and filter out E-UTRAN cells that do not meet the
requirement of PCI reuse isolation.
After PCI planning is complete, a PCI planning result table is displayed.
Alternatively, to import the engineering parameters of the live network, you can
right-click on the U-Net, choose PCI Planning, and click Open PCI Codes to open
the PCI table.
Export the result table, and use the filtering function of Excel to find E-UTRAN cells
that do not meet the PCI reuse isolation requirement, that is, E-UTRAN cells within
two reuse tiers or with a too small reuse distance. (For common urban areas, the reuse
distance is 4 km, and for suburban areas, the reuse distance is 10 km.)
Figure 1.4.1.1.1.9.1.1 Filtering E-UTRAN cells in the PCI result table

b. Spot E-UTRAN cells that do not meet the PCI reuse isolation requirement and
present the cells in a geographical figure.
Right-click on the PCI result table, and choose Filter to filter out E-UTRAN cells
that do not meet the PCI reuse requirement, that is, E-UTRAN cells within two reuse
tiers or with a too small reuse distance. (For common urban areas, the reuse distance
is 4 km, and for suburban areas, the reuse distance is 10 km.) Then present the cells in
a geographical figure. As shown in Figure 1.4.1.1.1.9.1.2, green cells are cells that do
not meet the PCI reuse isolation requirement, and yellow cells are normal cells.
Figure 1.4.1.1.1.9.1.2 Filtering out and presenting E-UTRAN cells that do not meet the PCI
reuse requirement
You can view an E-UTRAN cell that does not meet the PCI reuse requirement and its intra-
frequency cells using the same PCI, and the geographical distance between two cells
graphically. In Figure 1.4.1.1.1.9.1.3, the blue cell is the source cell, and the red cell is the
intra-frequency cell using the same PCI.
Figure 1.4.1.1.1.9.1.3 Intra-frequency E-UTRAN cells using the same PCI in a geographic figure
c. Filter out E-UTRAN cells that do not meet the PCI reuse isolation requirement
and export these cells in a file.
Right-click on the PCI result table, and choose Audit to export E-UTRAN cells that
do not meet the PCI reuse requirement, that is, E-UTRAN cells within two reuse tiers
or with a too small reuse distance. (For common urban areas, the reuse distance is 4
km, and for suburban areas, the reuse distance is 10 km.) Then view the E-UTRAN
cells with PCI conflicts in an Excel file.
Figure 1.4.1.1.1.9.1.4 Checking PCI conflicts using the audit function

 Evaluating PCI reuse isolation

− Application scenarios
a. In swap or PCI replanning scenarios, PCI reuse isolation is used to quantitatively
compare the results before and after planning.
b. In site deployment scenarios, PCI reuse isolation is used to quantitatively
analyze PCI planning on a network and determine whether the planning result
meets live network requirements (for example, PCI reuse distance of 98% E-
UTRAN cells exceeds 10 km).
− The following methods can be used to analyze PCI reuse isolation:
a. Minimum reuse distance of E-UTRAN cells on the network in CDF statistics
Export or copy the values of Min Reuse Distance of each cell in the PCI result
table to an Excel file, collect the CDF statistics of each reuse distance, and
generate a line graph.
Figure 1.4.1.1.1.9.1.5 PCI reuse distance in CDF statistics
Reuse distance in
initial configurations
Reuse distance
after planning
Cell PCI reuse distance (m)
b. Minimum number of E-UTRAN cell reuse tiers on the network in CDF statistics
Export or copy the values of Min Reuse Tier of each cell in the PCI result table
to an Excel file, collect the CDF statistics of reuse tiers, and generate a line
graph
Geographically presenting the PCI reuse information of each cell on the network
Use the Display Option in the U-Net to set the content and presentation form of
the PCI planning result. Click an E-UTRAN cell to view its intra-frequency E-
UTRAN cells using the same PCI.
Observe the E-UTRAN cell of the minimum PCI reuse distance and number of reuse
tiers in the PCI result table, to check the PCI reuse isolation in a geographic figure as
shown in Figure 1.4.1.1.1.9.1.3, and determine whether the PCI reuse isolation is
proper.
You can refer to the distance between intra-frequency E-UTRAN cells using the same
PCI and the number of eNodeBs between them. If the distance between the two cells
is large (for example, the two cells are over 4 km away from each other in the
common urban area or over 10 km away from each other in the suburban area), or not
neighboring cells (with multiple eNodeBs in between), the two cells can use the same
PCI.
 Evaluating PCI Mod3
PCI Mod3 is related to the main synchronization code and downlink reference signal
(RS). For details, see section 1.2"Principles."
− Evaluation criteria:
a. Intra-frequency cells served by the same site with adjacent azimuths use
different PCI Mod3.
b. Opposite cells served by adjacent sites (with no other sites in between) use
different PCI Mod3.
How close and opposite E-UTRAN cells are is determined based on geographically displayed figures
and experience.
c. In swap or replanning scenarios, ensure PCI Mod3 staggering (RS SINR when
the neighboring cell is not loaded) does not deteriorate.
− The PCI Mod3 can be evaluated by the following methods:
a. Evaluation based on coverage prediction (RS SINR when the cell is not loaded)
The PCI affects the downlink RS SINR when the cell is not loaded. In this case, if
PCI Mod3 is staggered, the downlink RS SINR is good
Therefore, you can use the coverage prediction function of the U-Net to calculate the
downlink RS SINR in each grill, export the calculated results, and collect the
statistics.
Coverage prediction involves multiple propagation model parameters, and
consequently accurate coverage prediction is difficult to achieve. However, you can
compare the RS SINR before and after network swap or PCI replanning to check
whether PCI planning is effective.
b. Analysis based on geographical presentation
You can analyze the following through geographical presentation:
Whether the PCI Mod3 of intra-frequency cells served by the same site with adjacent
azimuths is staggered.
 If the azimuths of cells A, B, and C served by a site are 0, 120, and 240, respectively, the three cells
have adjacent azimuths.
 If the azimuths of cells A, B, C, and D are 0º, 90º, 180º, and 270º, respectively, cell A has the
adjacent azimuth with cells B and D, cell B has the adjacent azimuth with cells A and C, cell C has
the adjacent azimuth with cells B and D, and cell D has the adjacent azimuth with cells A and C.
Whether PCI Mod3 of adjacent opposite cells served by adjacent sites is staggered.
How two cells are opposite to each other is related to the cell position and azimuths. If the azimuths of
the two cells are opposite to each other, the two cells are opposite to each other. If the azimuths of the
two cells face the other way, the two cells are not opposite to each other.
Perform the followings to view PCI Mod3 of each cell based on the geographical
presentation:
Step 1 Copy the PCI of each cell to an Excel file, and use Mod(PCI, 3) to calculate the PCI Mod3 of
each cell.
Step 10 Copy PCI Mod3 of each cell from the Excel file to the Comments column of the transceiver
table.
Ensure that the values in the Transceiver Name column in the transceiver table are consistent
with the names of transceivers of each cell.
Step 11 Geographically present PCI Mod3.

1. Right-click Transceiver, and choose Group By > Comments.
2. Right-click Transceiver, and choose Display Setting to configure the transceiver

display mode.
Specify different
colors for PCI Mod3.
Step 12 Analyze the PCI Mod3 staggering based on the geographical presentation.
Example:
PCI Mod3 is specified by different colors in different sectors. Red specifies that the PCI
Mod3 is 0, yellow specifies that the PCI Mod3 is 1, and blue specifies that the PCI Mod3 is 2.
On the network, the PCI Mod3 of intra-frequency cells with adjacent azimuths served by the
same site is staggered. Cells in red circles are opposite adjacent cells served by different sites
without PCI Mod3 staggering.
----End

PCI Mod30 is related to the uplink RS.
Adjacent intra-frequency cells (with no base stations in between) have different PCI
Mod30.
− Methods for evaluating and analyzing PCI Mod30 using the following methods:
a. Collect the PCI Mod30 reuse distance of each cell. If the PCI Mod30 reuse distance
is greater than the site spacing, the PCI Mod30 is properly staggered.
Step 1 Copy the PCI of each cell to an Excel file, and use Mod(PCI, 30) to calculate the PCI Mod30
of each cell.
Step 13 Copy the PCI Mod30 of each cell from the Excel table to the PCI column in the cell table.
Step 14 Ensure that the values in the Cell Name column in the transceiver table are consistent with
the cell names in the Excel file.
Step 15 Export the PCI planning result.

Step 16 Collect the information about minimum reuse distance of each cell using the Excel, and
display the distance in the CDF statistics.
A greater possibility that the PCI Mod30 reuse distance is greater than the site spacing
indicates a better staggering of PCI Mod30.
Figure 1.4.1.1.1.16.1.1 PCI Mod30 reuse distance

100
80
60
CDF
% 40
20
0
0 1 2 3 4 5 6 7 8
PCI Mod30 reuse distance (Km)
----End
b. Geographically present cells with the same PCI Mod30.

Step 1 Copy the PCI of each cell to an Excel file, and use Mod(PCI, 30) to calculate the PCI Mod30
of each cell.
Step 17 Copy the PCI Mod30 of each cell from the Excel table to the PCI column in the cell table.
Step 18 Choose Same PCI from the Display Option menu, and click an E-UTRAN cell to view other
cells with the same PCI Mod30.
Step 19 You can observe E-UTRAN cells with small PCI Mod30 reuse distance. For example, red
cells in the following figure use the same PCI Mod30.
----End
 Output
Provide the PCI evaluation report, including the following:
− Cells with PCI conflicts
− Cells with poor PCI Mod3 staggering
− Adjacent cells with the same PCI Mod30
− PCI reuse on the network
− PCI Mod3 staggering
− PCI Mod30 staggering
− In a network swap or replanning scenario, the PCI reuse, PCI Mod3 staggering, and
PCI Mod30 staggering before and after the network swap or replanning must be
provided.
PCI Replanning
 Overview
Use the U-Net to re-plan proper PCIs for cells that do not meet the requirement if any of
the following conditions occurs:
− PCI Mod3 of adjacent cells is poorly staggered.
− PCI Mod30 of neighboring cells is poorly staggered.
− PCI reuse isolation is insufficient.
− PCI conflicts exist.
Use the NPMaster-L (prototype tool) to optimize PCI Mod3 based on the data measured on the network.
Use the PCI auto-optimization function on the M2000 to re-allocate PCIs of cells with PCI conflicts to
reduce PCI conflicts.
 Tool
− Available at:
7851894/PBI1-8132365/PBI1-8132371/PBI1-21395144/PBI1-13953
 Input
− Engineering parameters described in "PCI Planning" of section 1.1.2"U-Net PCI
Planning.
− E-UTRAN cells requiring PCI replanning
 Procedure
planned, available PCIs for these cells, and cells requiring PCI replanning.
Step 20 Prepare the engineering parameter table based on the engineering parameter table of the U-
Net and set the PCIs of cells to be planned to null.
If PCI Mod3 of cells to be planned does not meet the requirement (for example, their PCI Mod3 is not
staggered with that of cells served by the same site or opposite adjacent cells), clear PCIs of the cell and
its intra-frequency cells served by the same site.

Step 23 Import engineering parameter files of NEs in the order of the site table, transceiver table, cell
table, and neighbor relationship.
Step 24 Set PCI auto-planning parameters (Planning with Exist PCI), perform PCI auto-planning, and
submit the planning result.
If the PCI planning of the cells to be planned fails or the reuse distance is insufficient and the PCI must
be re-planned, select Replan PCI for failed cell and one of its sub-items on the General tab page. For
details about sub-items, see Table 1.4.1.1.1.1.1.1.6. 1. When you choose Iteratively Generate PCI and
cells with planning failures exist or the reuse distance is is insufficient after replanning, PCI resources
are relatively insufficient or the number of reuse tiers is set to a too large value. If cells with planning
failures do not exist and the PCI reuse distance is too small, the PCI resource is relatively insufficient. If
the number of reuse tiers is set to a too large value, reduce the reuse tiers and then perform planning. If
the PCI resource is insufficient, release more PCI resources.
2. When you choose Replan adjacent cell PCI and cells with planning failures exist in the planning
area after the replanning, check whether the number of reuse tiers needs to be reduced or the reuse
distance must be iterated and reduced before performing the planning again. In this situation, if cells
with insufficient reuse isolation still exist, PCI resources are insufficient and more PCI resources must
be released.
Step 25 Evaluate the planning result and manually and slightly adjust the planning result.
----End
 Output
For details about output, see Figure 1.4.1.1.1.7.1.1.
 References
actionFlag=detailProductSimple&web_doc_id=SC0000764790&doc_type=ProductMan
ual&doc_type=ProductManual&saveBrowserLoged=true
1.4.2 SmartRNO PCI Planning

PCI Planning
 Overview
In site deployment or capacity expansion scenarios, provide PCI planning for new E-
UTRAN cells.
During PCI planning, collect required data for planning, prepare the engineering
parameter table, and then use a tool to perform PCI planning.
 Tool
− Tool name: SmartRNO
− R&D engineer: Wu Li (employee ID: 00228362)
− Available at: http://3ms.huawei.com/hi/group/2029711/thread_5270817.html?
mapId=6327757&for_statistic_from=all_group_forum
−
 Input
− NE engineering parameter information (For details, see Table 1.4.2.1.1.1.1.1.1.)
 If PCI Mod3 of neighboring cells served by adjacent sites must be staggered for PCI planning,
calculate interference among neighboring cells based on topology or coverage prediction.
Calculation based on coverage prediction costs more time than calculation based on topology and is
controlled by multiple parameters, such as the propagation model, shadow fading, and signal level
threshold. Calculation based on topology can be performed without setting NE parameters related to
 If Planning with Exist PCI is selected for PCI planning, PCIs are allocated to cells without PCIs
and cells with planned PCIs using existing PCIs. If Planning with Exist PCI is not selected, PCIs of
all cells are cleared for replanning. Therefore, select Planning with Exist PCI for PCI planning in a
 If PCI Mod3 Matching Azimuth order is selected for PCI planning, both the PCI Mod3 and
azimuth of each cell are in ascending order. On some networks, the azimuths of each site are
basically the same (for example, azimuths of all cells served by a site are 0, 120, and 240). In this
situation, if PCI Mod3 of each cell is 0, 1, and 2 in ascending order, PCI Mod3 of adjacent cells
served by different sites is staggered. If new sites are deployed for capacity expansion and the
azimuths of cells served by new sites are in the same patterns as those of cells served by existing
sites, manually allocate PCIs to new cells, the values of PCI Mod3 are 0, 1, and 2 in ascending order.
This way, PCI Mod3 staggering of new cells and adjacent cells served by existing sites can be
ensured. If PCI Mod3 Matching Azimuth order is not selected, interference among adjacent cells
served by different sites is considered to ensure that PCI Mod3 is staggered. However, the orders of
PCI Mod3 and azimuth of each cell cannot be ensured. When a site azimuth is basically in the same
pattern, select PCI Mod3 Matching Azimuth order.
Table 1.4.2.1.1.1.1.1.1 Engineering parameters
Parameter Name Meaning Mandatory or not

Site Name Indicates the site name and is Yes
the unique identifier for a
cell.
Site ID Indicates the site ID. No
Import this parameter and the

parameter can be set to any value
within the reasonable range during
PCI planning.
Transceiver Name Indicates the transmitter name No

and is the unique identifier
for a transmitter.
Cell Name Indicates the cell name. Yes
TAC Indicates the tracing area No
code.
Active Indicates whether to active No
the serving cell.
Frequency Band Indicates band information. Yes
Channel Index Indicates the index of Yes
available frequencies
corresponding to a band.
PCI Indicates the physical ID of a No
cell.
Longitude Indicates the site longitude. No
Latitude Indicates the site latitude. No
X Indicates the geodetic X No
coordinate of a site.
Y Indicates the geodetic Y No
Azimuth Indicates the antenna azimuth Yes
(degree). The value range is
from 0 to 360.
Antenna Indicates the antenna type. No
Height Indicates the antenna height No
(m).
Mechanical Downtilt Indicates the mechanical No
downtilt (degree) of an
antenna.
Electrical Downtilt Indicates the electrical No

antenna.
Total Electrical Downtilt Indicates the total electrical No
downtilt of an antenna.
PA Indicates the offset of the No
PDCCH RE transmit power
(dB) relative to the RS RE
transmit power. The value
range is from –15 to 15.
Max Power(dBm) Indicates the maximum No
transmit power of a
transmitter.
RS Power(dBm) Indicates the RS power No
(dBm) of a subcarrier.
Scene Indicates the cell scenario. No
Status Indicates the cell status. No
Site Comments Indicates site comments. No
Transceiver Comments Indicates transmitter No
comments.
Cell Comments Indicates cell comments. No
Dx Indicates the horizontal offset No
(m) of an antenna to the site.
Dy Indicates the vertical offset No

IsSFNCell Indicates whether a cell is on Mandatory for an
an SFN. If this parameter is SFN cell
left empty, or is set to 0, the
cell is regarded as a common
cell by default. However, the
cell will be regarded as an
SFN cell only when the value
of this parameter is 1.
SFNGroupId Indicates the SFN group ID. Mandatory for an
Cells with the same SFN SFN cell
group ID are regarded as an
SFN group.
Main Propagation Model Indicates the main No
propagation model.
Main Calculation Radius Indicates the calculation No
radius of the main
propagation model.
Main Resolution Indicates the calculation No
precision of the main
propagation model.
Site Equipment Indicates site equipment. No
TMA Indicates TMA equipment. No
Feeder Indicates feeder equipment. No
Feeder Length(DL) Indicates the downlink feeder No
length (m).
Miscellaneous Tx Loss(DL) Indicates the downlink No
miscellaneous loss (dB).
Jump Loss Ant-TMA(DL) Indicates the downlink No
jumper loss (dB) from the
TMA to the antenna port.
Jump Loss TMA-BS(DL) Indicates the downlink No
TMA to the cabinet top.
Input Total Loss Indicates the method of No
calculating the initial total
loss.
 Selected: Indicates that the
user enters the total loss.
 Deselected: Indicates that the
program calculates the total
loss.
Total Loss(DL) Indicates the total loss, No

including TMA, feeder,
jumper, and miscellaneous

loss.
Jump Loss Ant-BS(DL) Indicates the downlink No
antenna to the cabinet top.
Number of Transmission Antenna Indicates the number of Yes
Ports transmission antenna ports.
Number of Transmission Indicates the number of Yes
Antennas transmission antennas of an
eNodeB.
CI Indicates the cell ID. Yes
 If this parameter is left empty,

the tool will automatically
combines the values of
eNodeB ID and Local Cell ID
to generate the value of this
parameter.
 If the value of this parameter is
set, the tool will directly
import the value.
Transmission Mode Indicates the transmission No

mode.
Is MOCN Indicates whether a cell is on No
an MOCN.
Operator Indicates the operator name. No
TTI Bundling Indicates whether to perform Yes
TTI bundling. This function
is disabled by default.
Inter-site CA Indicates whether to enable Yes
inter-site CA.
Load Balance Indicates whether to enable No
load balancing in simulation
capacity. This function is
disabled by default.
VMIMO Indicates whether to use No
VMIMO. This function is
PSIC Indicates whether parallel No
signal interference
cancellation must be
considered. PSIC is suitable
for the VMIMO scenario of
two RX antennas. Use PSIC
to bring gains by eliminating
interference between UEs.

This function is disabled by
default.
PUCCH IRC Indicates whether to enable No
the IRC function of the
PUCCH. When colored
interference is strong, the
PUCCH IRC function can be
enabled to suppress
interference power of
combined signals and
therefore improve uplink
gains. This function is
PRACH IRC Indicates whether to enable No
the PRACH IRC function.
When colored interference is
strong, the PRACH IRC
function can be enabled to
suppress interference power
of combined signals and
PUSCH IRC Indicates whether to enable No
the PUSCH IRC function.
strong, the PUSCH IRC
COMP Indicates whether to enable No
intra-site macro diversity.
Enable uplink CoMP function
on the eNodeB side of the
live network to improve cell
edge capacity and cell
average throughput. This
function is disabled by
default.
Channel Relativity Indicates whether channel No
relativity is considered. This
default.
PDSCH Actual Load(DL) Indicates the actual downlink No
load. The value range is from
0 to 1.
Actual Load(UL) Indicates the actual uplink No

0 to 1.
MCC Indicates the MCC that No
specifies the country where a
mobile subscriber belongs to,
for example, the MCC of
China is 460. The value is a
3-bit string of 0 to 9 digits.
MNC Indicates the MNC that No
specifies the public land
mobile network (PLMN) to
which a mobile subscriber
belongs. The value is a 2- or
DlEarfcn Indicates the downlink Yes
frequency of a cell.
UlEarfcn Indicates the uplink frequency No
of a cell.
Bandwidth(MHz) Indicates the bandwidth. No
Automatically match
frequency bands based on
frequencies.
Local Cell ID Indicates the internal code of No
a cell for differentiating the
cell from other cells under the
same eNodeB.
Cell Speed Indicates the cell speed. No
 LowSpeed
 HighSpeed
 HighwaySpeed
Radius(m) Indicates the cell radius. No

Min Root Sequence Index Indicates the minimum ZC No
sequence of a cell.
Preamble Format Indicates preamble sequence No
that is used for a user to
randomly access a cell.
Configuration Index Indicates the time frame No
index for a cell.
CA(DL)  Indicates whether the No
downlink CA is supported.
If the value is set to Off,
downlink CA is not
supported.
 If the value is set to CA2CC,
downlink 2CC CA is
supported.
downlink 3CC CA is
supported.
Is Double Deck Indicates a double deck cell No

ID.
 If the value is set to True,
the cell is a double deck
cell.
 If the value is set to False,
the cell is not a double deck
cell.
Cell Double Deck Bandwidth Indicates non-standard No

bandwidths (MHz) for a
double deck cell.
2.6
4
5.7
7.3
7.6
10.8
12.2
12.5
17.1
17.4
27.6
28.5
GSM Interference Frequencies Indicates GSM interference No
frequencies. Only configure
this parameter to LTE FDD.
Multi-user Beamforming Indicates whether to enable No
the multi-user beamforming
function for downlink
scheduling. Only configure
this parameter to LTE TDD.
Prach Density Per 10ms Indicates the PRACH density No
of a cell. Only configure this
parameter to LTE TDD.
Offset Frequency Domain Indicates frequency domain No
offset. Only configure this
DwPTS-GP-UpPTS Indicates special subframe No
configuration proportion.
Only configure this parameter
to LTE TDD.
Frame Configuration Indicates uplink and No
downlink subframe
to LTE TDD.
It includes uplink, downlink,
and special subframe
If the value is set to
DSUUUDSUUU, D is for
downlink subframe, S is for
special subframe, and U is for
uplink subframe.
Table 1.4.2.1.1.1.1.1.2 Parameters involved in the PCI planning policy

Name Value
Available PCI Indicates N/A This parameter is set based on actual
available scenarios. Normally, indoor and
PCI outdoor cells use different PCIs. For
resources. example, in a Mexico office, PCIs 0 to
440 are allocated to outdoor sites, and
PCIs 441 to 503 are allocated to
indoor sites.
On boundaries, such as the national
boundary, cells along the boundary
belonging to different countries use
different PCIs. For example, European
countries must comply with a certain
protocol to ensure cells along the
boundary belonging to different
neighboring countries use different
PCIs.
On some networks, PCIs are reserved
for follow-up capacity expansion.
Planning With Indicates In a capacity N/A
Exist PCI whether to expansion
re-allocate scenario, if there
PCIs for are existing cells
planned in planned areas,
cells. select Planning
With Exist PCI.
PCI Mod3 Indicates Do not select On some networks, the azimuths of
Matching that both the PCI Mod3 each site are basically the same (for
Azimuth order PCI Mod3 Matching example, azimuths of all cells served
and azimuth Azimuth order. by a site are 0, 120, and 240). In this
of the cell situation, if PCI Mod3 of each cell is

are in 0, 1, and 2 in ascending order, and the
ascending azimuths of new sites are in the same
order. pattern as those of existing cells,
manually allocate PCIs to new cells.
When PCI Mod3 is 0, 1, and 2, the
PCI Mod3 staggering of a new cell
and an adjacent cell served by an
existing site can be ensured in a great
possibility.
Iterative Indicates Do not select If the PCI reuse distance is too large,
Generate PCI that PCI Iterative resulting in PCI planning failures,
reuse Generate PCI. select Iterative Generate PCI to
distance reduce the PCI reuse distance and
iteration is ensure a successful planning.
used.
Distance Indicates the 20% This parameter becomes valid only
Descending percentage when Iterative Generate PCI is
Percentage(% of decreased selected.
) reuse
distance due
to reuse
distance
iteration.
Replan Indicates Do not select During PCI planning, cells planned in
adjacent cell that the PCIs Replan the initial phase are more likely to be
PCI of planned adjacent cell allocated with PCIs than cells planned
cells' PCI. in the later phase. In this situation,
adjacent change the PCI planning order or do
cells must be not allocate optimal PCIs to cells
adjusted. planned in the initial phase. This way,
cells planned in the later phase can be
allocated with PCIs.
Inherit existed Indicates Do not select This parameter becomes valid only
PCI mod 3 that PCI Inherit existed when Replan adjacent cell PCI is
Mod3 PCI mod 3. selected.
backup of
existing cells
is needed
and PCI
Mod3 needs
to remain
unchanged
after re-
allocation.
Area Indicates the N/A If a user does not draw a polygon, the
PCI full map is used for planning.
planning If a user draws polygons on the map,
area. planning can be performed in the areas
in polygons.
Topology Indicates Select N/A

that Topology.
interference
among
adjacent
cells is
calculated
based on
topology.
Prediction Indicates Do not select Coverage prediction involves multiple
that Prediction. parameters, such as the propagation
interference model, shadow fading, and penetration
among loss. Besides, the accurate values of
adjacent these parameters are difficult to
cells is obtain. Different parameter values
calculated result in different planning results.
based on
coverage
prediction.
Resolution(m) Indicates 50 m This parameter is used when
resolution of interference is calculated based on
predicted coverage prediction.
coverage.
Min Signal Indicates the –120 dBm This parameter is used when
Level(dBm) RSRP interference is calculated based on
threshold of coverage prediction.
an E-
UTRAN cell
during
coverage
prediction.
Handover area Indicates the 6 dB This parameter is used when
threshold(dB) difference interference is calculated based on
between coverage prediction.
adjacent cell
signals and
serving cell
signals.
Shadowing Indicates Do not select This parameter is used when
taken into that shadow Shadowing interference is calculated based on
account fading is taken into coverage prediction.
considered. account.
Cell Edge Indicates the 75% This parameter is used when
Coverage cell edge interference is calculated based on
Probability coverage coverage prediction and Shadowing
probability taken into account is selected.
of shadow The shadow fading variance of the
fading. cluster must be specified. For details,
see U-Net Operation Guide.
Indoor Indicates N/A This parameter is used when

Coverage that interference is calculated based on
penetration coverage prediction.
loss is The penetration loss of the cluster
considered. must be specified. For details, see U-
Net Operation Guide.
 Procedures
planned, and available PCIs for these cells.
Step 2 Prepare the engineering parameter table file based on the SmartRNO engineering parameter
template.
Step 3 Create a SmartRNO project.
If the electronic map has a coordinate system, you do not need to manually set the system.
Step 5 Import NE engineering parameter files.
Step 6 Set PCI auto-planning , perform PCI auto-planning, and submit the planning result.
Step 7 Evaluate the result and manually adjust the result slightly.
----End
 Output
Figure 1.4.2.1.1.7.1.1 shows the PCIs for cells to be planned.
Figure 1.4.2.1.1.7.1.1 PCI planning result output
Where,
Existing Code: indicates the original PCI of a cell to be planned.
Suggest Code: indicates the PCI auto-planning output.
Confirm Code: indicates the PCI confirmed by a user. After PCI planning is complete, the
user can manually modify the value of Confirm Code and submit the modified value as the
PCI to the cell table.
Min Reuse Distance: indicates the minimum PCI reuse distance of a cell to be planned.
Min Reuse Tiers: indicates the minimum number of PCI reuse tiers of a cell to be planned.
 References and website: to be provided
PCI Check (Evaluation)

 Overview
− Application scenarios of the PCI check:
a. After PCI planning is complete, the PCI planning result must be checked and
evaluated.
b. During the network maintenance phase, PCIs on the live network must be
checked and evaluated.
− The following items are checked and evaluated:
a. PCI conflicts (collision and confusion)
b. PCI reuse isolation, including reuse distance and reuse tier
c. Staggering of PCI Mod3 for neighboring cells
d. Staggering of PCI Mod6 for neighboring cells
e. Staggering of PCI Mod30 for adjacent cells
a. Evaluation based on the PCI result table: is used to summarize, filter, and record
PCI reuse (of E-UTRAN cells with small PCI reuse isolation, such as the small
reuse distance and the small number of reuse tiers) for filing and viewing the
result.
b. Evaluation based on geographical presentation: is used to view the PCI reuse,
and the staggering of PCI Mod3, PCI Mod6, and PCI Mod30.
c. Evaluation based on CDF statistics: Used to quantitatively evaluate PCI reuse
isolation in E-UTRAN cells. Especially in swap or PCI replanning scenarios,
this method can be used to quantitatively compare the PCI reuse before and after
swap or PCI replanning.
d. Evaluation based on coverage prediction (RS SINR when the neighboring cell is
not loaded): Used to quantitatively compare the staggering of PCI Mod3 before
and after swap or PCI replanning.
 Tool
−
 Input
− Evaluation and check after PCI planning is complete
Evaluation and check after PCI planning is complete require no parameter settings.
You can directly submit the planning result.
− Evaluation and check during network maintenance
NE parameters listed in Table 1.4.2.1.1.7.1.1.1 are required.
Table 1.4.2.1.1.7.1.1.1 Engineering parameters

Site Name Indicates the site name and is Yes
the unique identifier for a
cell.
Import this parameter and the

parameter can be set to any value
within the reasonable range during
PCI planning.
Transceiver Name Indicates the transmitter name No

and is the unique identifier
for a transmitter.
TAC Indicates the TAC. No

the serving cell.
PCI Indicates the physical ID of a No
cell.
Azimuth Indicates the antenna azimuth Yes
(degree). The value range is
from 0 to 360.
(m).
antenna.

antenna.
Total Electrical Downtilt Indicates the total electrical No
downtilt of an antenna.
PDCCH RE transmit power
(dB) relative to the RS RE
transmit power. The value
range is from –15 to 15.
transmit power of a
transmitter.

comments.
Dx Indicates the horizontal offset No
IsSFNCell Indicates whether a cell is on Mandatory for an
an SFN. If this parameter is SFN cell
left empty, or is set to 0, the
cell is regarded as a common
cell by default. However, the
SFN cell only when the value
of this parameter is 1.
SFNGroupId Indicates the SFN group ID. Mandatory for an
Cells with the same SFN SFN cell
group ID are regarded as an
SFN group.
propagation model.
radius of the main
propagation model.
propagation model.
TMA Indicates TMA equipment. No
Feeder Length(DL) Indicates the downlink feeder No
length (m).
loss.
 Selected: Indicates that the
user enters the total loss.
 Deselected: Indicates that the
program calculates the total
loss.

loss.
Number of Transmission Antenna Indicates the number of Yes
Ports transmission antenna ports.
Antennas transmission antennas of an
eNodeB.
 If this parameter is left empty,

the tool will automatically
combines the values of
eNodeB ID and Local Cell ID
to generate the value of this
parameter.
 If the value of this parameter is
set, the tool will directly
import the value.

mode.
Is MOCN Indicates whether a cell is on No
an MOCN.
Operator Indicates the operator name. No
TTI Bundling Indicates whether to perform Yes
TTI bundling. This function
inter-site CA.
load balancing in simulation
capacity. This function is
VMIMO. This function is

signal interference
considered. PSIC is suitable
for the VMIMO scenario of
two RX antennas. Use PSIC
to bring gains by eliminating
default.
the PUCCH IRC function.
strong, the PUCCH IRC
strong, the PRACH IRC
strong, the PUSCH IRC
Enable uplink CoMP function
on the eNodeB side of the
live network to improve cell
default.

relativity is considered. This
default.
PDSCH Actual Load(DL) Indicates the actual downlink No
0 to 1.
0 to 1.
specifies the country where a
mobile subscriber belongs to,
China is 460. The value is a
specifies the PLMN to which
a mobile subscriber belongs.
The value is a 2- or 3-bit
string of 0 to 9 digits.
UlEarfcn Indicates the uplink frequency No
of a cell.
Automatically match
frequencies.
Local Cell ID Indicates the internal code of No
a cell for differentiating the
cell from other cells under the
same eNodeB.
 LowSpeed
 HighSpeed
 HighwaySpeed

Min Root Sequence Index Indicates the minimum ZC No
sequence of a cell.
Preamble Format Indicates preamble sequence No
that is used for a user to
randomly access a cell.
index for a cell.

downlink CA is supported.
If the value is set to Off,
downlink CA is not
supported.
downlink 2CC CA is
supported.
downlink 3CC CA is
supported.

ID.
 If the value is set to True,
the cell is a double deck
cell.
 If the value is set to False,
the cell is not a double deck
cell.

double deck cell.
2.6
4
5.7
7.3
7.6
10.8
12.2
12.5
17.1
17.4
27.6
28.5
GSM Interference Frequencies Indicates GSM interference No
frequencies. Only configure
this parameter to LTE FDD.
Multi-user Beamforming Indicates whether to enable No
the multi-user beamforming
this parameter to LTE TDD.
Prach Density Per 10ms Indicates the PRACH density No
of a cell. Only configure this

Offset Frequency Domain Indicates frequency domain No
offset. Only configure this
to LTE TDD.
downlink subframe
to LTE TDD.
It includes uplink, downlink,
and special subframe
special subframe, and U is for
uplink subframe.
Step 1 If the SmartRNO is used to perform PCI auto-planning and PCI planning must be checked,
submit the planning result. If the PCI check is used during network maintenance, prepare
engineering parameters, create a project, and import these engineering parameters to the
project.
Step 8 Check PCI conflicts, evaluate PCI reuse isolation, and staggering of PCI Mod3, PCI Mod6,
and PCI Mod30.
----End
 Checking PCI conflicts

− Evaluation criteria
Check whether PCI conflicts exist among E-UTRAN cells, including areas that do
not satisfy the minimum PCI reuse distance or among the second-tier neighboring
cells.
− Three methods are used to check PCI conflicts:
a. Export the result table and filter out E-UTRAN cells that do not meet the
requirement of PCI reuse isolation.
After PCI planning is complete, a PCI planning result table is displayed.
Alternatively, you can import the engineering parameters of the live network, right-
click on PCI Planning, and click Open PCI Codes to open the PCI table.
Export the result table, and use the filtering function of Excel to identify E-UTRAN
cells that do not meet the PCI reuse isolation requirement, that is, the number of tiers
for neighboring E-UTRAN cells is less than or equal to two or the reuse distance is
smaller. (For common urban areas, the reuse distance is 4 km. For suburban areas, the
reuse distance is 10 km.)
Figure 1.4.2.1.1.9.1.1 Filtering out E-UTRAN cells that do not meet the PCI reuse isolation
requirement listed in the PCI result table
b. Spot E-UTRAN cells that do not meet the PCI reuse isolation requirement and
geographically present the cells.
Right-click on the PCI result table, and choose Filter to filter out E-UTRAN cells
that do not meet the PCI reuse requirement, that is, the number of tiers for
neighboring E-UTRAN cells is less than or equal to two or the reuse distance is
smaller. (For common urban areas, the reuse distance is 4 km, and for suburban areas,
the reuse distance is 10 km.) Then geographically present the cells. Green cells do not
meet the PCI reuse isolation requirement, and yellow cells are normal cells, as shown
in Figure 1.4.2.1.1.9.1.2.
Figure 1.4.2.1.1.9.1.2 Filtering out and presenting E-UTRAN cells do not meet the PCI reuse rule
You can view an E-UTRAN cell that does not meet the PCI reuse requirement and its intra-
frequency cells using the same PCI, and the geographical distance between two cells
graphically. In Figure 1.4.2.1.1.9.1.3, the blue cell is the source cell, and red cells are intra-
frequency cells using the same PCI.
Figure 1.4.2.1.1.9.1.3 Geographically presenting intra-frequency E-UTRAN cells using the same
PCI
c. Filter out E-UTRAN cells that do not meet the PCI reuse isolation requirement
and export these cells in a file.
Right-click on the PCI result table, and choose Audit to export E-UTRAN cells that
do not meet the PCI reuse requirement, that is, the number of tiers for neighboring E-
UTRAN cells is less than or equal to two or the reuse distance is smaller. (For
common urban areas, the reuse distance is 4 km. For suburban areas, the reuse
distance is 10 km.) Then view E-UTRAN cells with PCI conflicts in the exported
Excel file.
Figure 1.4.2.1.1.9.1.4 Checking PCI conflicts using the audit function

 Evaluating PCI reuse isolation

a. In swap or PCI replanning scenarios, PCI reuse isolation is used to quantitatively
compare the results before and after planning.
b. In site deployment scenarios, PCI reuse isolation is used to quantitatively
analyze PCI planning on a network and determine whether the planning result
meets live network requirements (for example, PCI reuse distance of 98% E-
UTRAN cells exceeds 10 km).
− The following methods can be used to analyze PCI reuse isolation:
a. Minimum reuse distance of E-UTRAN cells on the network in CDF statistics
Export or copy the values of Min Reuse Distance of each cell in the PCI result table
to an Excel file, collect the CDF statistics of each reuse distance, and generate a line
graph.
Figure 1.4.2.1.1.9.1.5 PCI reuse distance in CDF statistics
Initial
configuration
After PRACH
ZC root
sequence
planning
PRACH ZC root sequence reuse distance for E-UTRAN cells

b. Minimum number of E-UTRAN cell reuse tiers on the network in CDF statistic
Export or copy the values of Min Reuse Tier of each cell in the PCI result table to an
Excel file, collect the CDF statistics of reuse tiers, and generate a line graph as shown
in Figure 1.4.1.1.1.9.1.5.
c. Geographically presenting the PCI reuse of each cell on the network
Use Display Option in the SmartRNO to set the geographical presentation content
and form of the PCI planning result. Click an E-UTRAN cell to view its intra-
frequency E-UTRAN cells using the same PCI.
Observe the E-UTRAN cell with the minimum PCI reuse distance and number of
reuse tiers in the PCI result table, check PCI reuse isolation in a geographic figure,
and determine whether PCI reuse isolation is reasonable.
You can refer to the distance between intra-frequency E-UTRAN cells using the same
PCI and the number of eNodeBs between them. If the distance between two cells is
large (for example, these cells are over 4 km away from each other in urban areas or
over 10 km away from each other in suburban areas), or two cells are not neighboring
cells (with multiple eNodeBs in between), the two cells can use the same PCI.
PCI Mod3 is related to the main synchronization code and downlink RS.
a. Intra-frequency cells served by the same site with adjacent azimuths use
different PCI Mod3.
b. Opposite cells served by adjacent sites (with no other sites in between) use
different PCI Mod3.
Determining how close and opposite E-UTRAN cells are based on geographical presentation and
experience.
c. In swap or replanning scenarios, ensure PCI Mod3 staggering (RS SINR when
the neighboring cell is not loaded) does not deteriorate.
− Evaluating PCI Mod3 using the following methods:
a. Evaluation based on coverage prediction (RS SINR when the cell is not loaded)
The PCI affects the downlink RS SINR when the cell is not loaded. In this case, if
PCI Mod3 is staggered, the downlink RS SINR is relatively good.
Therefore, you can use the coverage prediction function of the U-Net to calculate the
downlink RS SINR in each grid when the cell is not loaded, export calculated results,
and collect statistics.
Coverage prediction involves multiple propagation model parameters, and
consequently accurate coverage prediction is difficult to achieve. However, you can
compare the RS SINR before and after network swap or PCI replanning to check
whether PCI planning is effective.
b. Geographical presentation and analysis
You can analyze the followings through geographical presentation:

a) Whether the PCI Mod3 of intra-frequency cells served by the same site with
adjacent azimuths is staggered.
 If the azimuths of cells A, B, and C served by the same site are 0º, 120º, and 240º, respectively.
Then, cell A and cell B are neighboring cells, cell A and cell C are neighboring cells, and cell B and
cell C are neighboring cells.
 If the azimuths of cells A, B, C, and D are 0º, 90º, 180º, and 270º, respectively, cell A has the
adjacent azimuth with cells B and D, cell B has the adjacent azimuth with cells A and C, cell C has
the adjacent azimuth with cells B and D, and cell D has the adjacent azimuth with cells A and C.
b) Whether PCI Mod3 of adjacent opposite cells served by adjacent sites is
staggered.
 How two cells are opposite to each other is related to the cell position and azimuths: If the azimuths
of the two cells are opposite to each other, the two cells are opposite to each other.
 If the azimuths of the two cells face the other way, the two cells are not opposite to each other.
− Methods:
Click LTE PCI Planning, select LTE-FDD, and then right-click on Display Option
to set the PCI Mod3 display mode.
Take Mod3 display mode as an example. PCI Mod3 is specified by different colors
in different sectors. Red indicates that the PCI Mod3 is 0, yellow indicates that the
PCI Mod3 is 1, and blue indicates that the PCI Mod3 is 2, as shown in the following
figure. On the network, the PCI Mod3 of intra-frequency cells with adjacent azimuths
served by the same site is staggered. Cells in red circles are opposite adjacent cells
served by different sites without PCI Mod3 staggering.
 Evaluating PCI Mod30 or PCI Mod6

PCI Mod30 or PCI Mod6 is related to the uplink RS.
Adjacent intra-frequency cells (with no sites in between) have different PCI Mod30
or PCI Mod6.
− Evaluate and analyze PCI Mod30 or PCI Mod6 using the following methods:
a. Collect the PCI Mod30 or PCI Mod6 reuse distance of each cell. If the PCI
Mod30 or PCI Mod6 reuse distance is greater than the inter-site distance (ISD),
the PCI Mod30 or PCI Mod6 is properly staggered.
Step 1 Copy the PCI of each cell to an Excel file, and use Mod (PCI, 30) or Mod (PCI, 6) to calculate
the PCI Mod30 or PCI Mod6 of each cell.
Step 2 Copy the PCI Mod30 or PCI Mod6 of each cell from the Excel table to the PCI column in the
cell table. Ensure that values in the Cell Name column in the cell table are consistent with
those of cell names in the Excel file.
Step 10 Export the PCI planning result table.

Step 11 Collect information about Min Reuse Distance of each cell using the Excel, and display the
distance in CDF statistics.
A greater possibility that the PCI Mod30 or PCI Mod6 reuse distance is greater than ISD
indicates a better staggering of PCI Mod30 or PCI Mod6. Figure 1.4.2.1.1.11.1.1 shows
an example of PCI Mod30 reuse distance.
Figure 1.4.2.1.1.11.1.1 PCI Mod30 reuse distance
100
80
60
CDF % 40
20
0
0 1 2 3 4 5 6 7 8
PCI Mod 30 reuse distance (km)
----End
b. Geographically present cells with the same PCI Mod30 or PCI Mod6.
Methods:
Click LTE PCI Planning, select LTE-FDD, and then right-click on Display Option to set
the PCI Mod30 or PCI Mod6 display mode.
Click OK to present other cells using the same PCI Mod30 or PCI Mod6 with the evaluated
cell. You can observe E-UTRAN cells with small PCI Mod30 or PCI Mod6 reuse distance.
Red cells shown in Figure 1.4.2.1.1.11.1.2 use the same PCI Mod30.
Figure 1.4.2.1.1.11.1.2 Collecting PCI Mod30 reuse distance
----End
 Output
The PCI evaluation report includes:
− Cells with PCI conflicts
− Cells with poor PCI Mod3 staggering
− Adjacent cells with the same PCI Mod30
− PCI reuse on the network
− PCI Mod3 staggering on the network
− In network swap or replanning scenarios, PCI reuse, PCI Mod3 staggering, PCI
Mod30 staggering, and PCI Mod6 staggering before and after network swap or
replanning must be provided.
PCI Replanning
 Overview
Use the SmartRNO to re-allocate proper PCIs for cells that do not meet the requirement
if any of the following conditions occurs:
− PCI Mod3 of neighboring cells is poorly staggered.
− PCI reuse isolation is insufficient.
− PCI conflicts exist.
Use the PCI optimization function of the OMStar to optimize PCI Mod3 based on the drive test (DT)
data, traffic statistics, and MR data measured on the live network.
 Tool
−
 Input
− NE engineering parameters
If PCI Mod3 of neighboring cells served by adjacent sites must be staggered for PCI planning, calculate
interference among neighboring cells based on topology or coverage prediction. Calculation based on
coverage prediction costs more time than calculation based on topology and is controlled by multiple
parameters, such as the propagation model, shadow fading, and signal level threshold. Calculation based
on topology can be performed without setting NE parameters related to coverage prediction.
If Planning with Exist PCI is selected for PCI planning, PCIs are allocated to cells without PCIs and
cells with planned PCIs using existing PCIs. If Planning with Exist PCI is not selected, PCIs of all
cells are cleared for replanning. Therefore, select Planning with Exist PCI for PCI planning in a
If PCI Mod3 Matching Azimuth order is selected for PCI planning, both the PCI Mod3 and azimuth
of each cell are in ascending order. On some networks, the azimuths of each site are basically the same
(for example, azimuths of all cells served by a site are 0, 120, and 240). In this situation, if PCI Mod3 of
each cell is 0, 1, and 2 in ascending order, PCI Mod3 of adjacent cells served by different sites is
staggered. If new sites are deployed for capacity expansion and the azimuths of cells served by new sites
are in the same patterns as those of cells served by existing sites, manually allocate PCIs to new cells,
the values of PCI Mod3 are 0, 1, and 2 in ascending order. This way, PCI Mod3 staggering of new cells
and adjacent cells served by existing sites can be ensured. If PCI Mod3 Matching Azimuth order is
not selected, interference among adjacent cells served by different sites is considered to ensure that PCI
Mod3 is staggered. However, the orders of PCI Mod3 and azimuth of each cell cannot be ensured. When
a site azimuth is basically in the same pattern, select PCI Mod3 Matching Azimuth order.
 Procedures
planned, available PCIs for these cells, and cells requiring PCI replanning.
Step 12 Prepare the engineering parameter table file based on the engineering parameter table
template of the SmartRNO and set the PCIs of cells to be planned to null.
If PCI Mod3 of cells to be planned does not meet the requirement (for example, their PCI Mod3 is not
staggered with that of cells served by the same site or opposite adjacent cells), clear PCIs of the cell and
its intra-frequency cells served by the same site.

Step 16 Set PCI auto-planning parameters (Planning with Exist PCI), perform PCI auto-planning,
and submit the planning result.
If PCI planning of cells to be planned fails or the reuse distance is insufficient, PCI must be re-planned.
During PCI replanning, select Replan PCI for failed cell and one of its sub-items on the General tab.
For details about sub-items, see Table 1.4.2.1.1.1.1.1.2.
 When you choose Iteratively Generate PCI and cells with planning failures exist or the reuse
distance is insufficient after replanning, PCI resources are relatively insufficient or the number of
reuse tiers is set to a too large value. If cells with planning failures do not exist but the PCI reuse
distance of some cells is too small, PCI resources are relatively insufficient. If the number of reuse
tiers is set to a too large value, reduce the reuse tiers and then perform planning. If PCI resources are
insufficient, release more PCI resources.
 When you choose Replan adjacent cell PCI and cells with planning failures exist in the planning
area after replanning, check whether the number of reuse tiers must be reduced or the reuse distance
must be iterated and reduced before replanning. In this situation, if cells with insufficient reuse
isolation still exist, PCI resources are insufficient and more PCI resources must be released.
Step 17 Evaluate the planning result and manually adjust the result slightly.
----End
 Output
 References: http://3ms.huawei.com/hi/group/2029711/thread_5270817.html?
Technical Guide to Cell Parameter Planning During Network 2 PRACH ZC Root Sequence
Design (PCI, PRACH, Neighboring Cell, and TA&TAL) Planning
2 PRACH ZC Root Sequence

Planning
2.1 Overview
This document describes the PRACH Zadoff-Chu (ZC) root sequence feature,
which covers PRACH ZC root sequence planning, PRACH ZC root sequence
checking, and PRACH ZC root sequence reallocation.
This document is intended for NIS personnel and personnel dedicated for
professional services.

Random access (RA) is critical in LTE system, specifically for initial
connection setup, handovers, connection re-establishment, and uplink
synchronization restoring. The PRACH ZC root sequence is an important cell-
level parameter for the LTE system and determines the preamble sequence
during RA.
If adjacent cells are allocated the same ZC root sequence, PRACH false
alarms may be generated. See Figure 2.2.1.1.1.1.1.3. Therefore, it is critical to
allocate appropriate PRACH ZC root sequences to E-UTRAN cells and check
and eliminate potential PRACH ZC root sequence conflicts on the LTE
network.
The PRACH ZC root sequence features involve PRACH ZC root sequence
planning, PRACH ZC root sequence checking (evaluation), and PRACH ZC
root sequence reallocation.
At the phase of network establishment and capacity expansion, a ZC root
index must be allocated to each established E-UTRAN cell to ensure
successful RA.
The NCS and PRACH ZC root sequences must be reallocated for each E-
UTRAN cell under any of the following conditions:
Issue 01 (2016-05-30) Huawei Proprietary and Confidential 68

Copyright © Huawei Technologies Co.,
Ltd..
 The network layout changes.

 The original PRACH ZC root sequences are incorrectly allocated.
 The NE parameter settings change on the live network. For example, the
cell radius changes due to RF adjustment.
After PRACH ZC root sequence is completed, check, evaluate, and analyze
the planning results. If necessary, conduct fine adjustment to the planned
PRACH ZC root sequences and then deliver the planning results to
corresponding NEs.
2.1.2 Introduction
Figure 2.1.2.1.1.1.1.1 PRACH ZC root sequence features
Prach ZC root feature
Prach ZC root sequence PRACH ZC root PRACH ZC root

planning sequence checking sequence reallocation
2.1.3 Restrictions
Specifications
N/A
Applicable Scenarios
Table 2.1.3.1.1.1.1.1.1 Restrictions on applicable scenarios for PRACH ZC root

sequence features
Scenario Restrictions Workarounds
The latitude If the latitude and longitude are not Obtain the latitude
and longitude specified, the U-Net cannot be used and longitude of
of an NE are to perform PRACH ZC root the NE.
not specified. sequence planning, PRACH ZC root
sequence checking, or PRACH ZC
root sequence reallocation.
PRACH ZC Currently, the U-Net does not The Wireless
root sequence support modeling for indoor Parameter Auto-
planning on scenarios, including floor configuration
indoor information, room information, and feature can be used
eNodeBs whether the indoor cell belongs to to complete the
the same building. PRACH ZC root

Ltd..
Due to lack of key information, the sequence planning

U-Net is not suitable for PRACH online for small-
ZC root sequence planning for cell base stations.
indoor cells.
2.1.4 Maturity
The PRACH ZC root sequence auto-planning supports the following
scenarios:
 Capacity expansion for greenfields and mixed networking
 Reallocation
The PRACH ZC root sequence planning supports the following scenarios:
 Inter-frequency networking
 Multi-antenna, remote radio frequency, and single frequency network
(SFN)
The PRACH ZC root sequence planning principles are as follows:
The PRACH ZC root sequence reuse tier for E-UTRAN cells can be
configured by users to set the minimum isolation for PRACH ZC root
sequence reuse.
PRACH ZC root sequence checking involves the following items:
 PRACH ZC root sequence reuse distance
 PRACH ZC root sequence reuse tier
2.2 Principles
2.2.1 Introduction
RA may be triggered by paging from the network, UEs' service requests, or
RRC connection setup requests.
RA can be classified into contention-based RA and non-contention-based RA.
Figure 2.2.1.1.1.1.1.1 shows the contention-based RA procedure.

Ltd..
Figure 2.2.1.1.1.1.1.1 Contention-based RA
In the contention-based RA procedure, the RA preamble is generated by each

UE based on the broadcast messages. Different UEs served by a cell may
generate the same preamble. In this case, the eNodeB initiates a contention-
based RA on all UEs to allow RA.
Table 2.2.1.1.1.1.1.3.2 shows the non-contention-based RA procedure.
Figure 2.2.1.1.1.1.1.2 Non-contention-based RA

Ltd..
In the non-contention-based RA procedure, the RA preambles are dedicated

preambles and allocated to UEs by the eNodeB to implement services such as
handovers. Therefore, preamble conflicts do not occur on UEs served by a
cell.
Each cell is allocated a maximum of 64 preambles for RA. Due to
specifications restrictions on some products, the number of available
preambles for a cell may be less than 64. The RA preamble is generated with
ZC root sequences through column cyclic shifting. To generate 64 preambles,
multiple ZC root sequences must be allocated to each cell. Logically, these ZC
root sequences are consecutively numbered, composing a ZC root sequence
group. The cyclic shift value is determined by the parameters
ZeroCorrelationZoneConfig, HighSpeedFlag, and PreambleFormat. For
details, see 3GPP TS 36.211.
If a cell and its neighboring cells use the same preamble, the preamble sent
from UEs under the cell may be detected by neighboring E-UTRAN cells.
Consequently, a false alarm is generated and an incorrect RACH optimization
decision may be reached. If the eNodeB cannot correctly identify the UE that
is allowed to access, the access delay will increase and access success rate
decreases for the UE. If multiple false alarms are generated, the CPU usage
increases. As a result, flow control is performed, thereby affecting network
capacity.
Figure 2.2.1.1.1.1.1.3 provides an example. In this example, Cell A and Cell B
are configured with the same ZC root sequence and a UE camps on Cell A.
When the UE initiates a random access to Cell A, Cell A and Cell B can detect
the preamble sent by the UE and therefore both cells send a RACH Response
(RAR) to the UE. However, the UE can only detect the RAR sent by Cell A
and therefore send Msg3 to Cell A. Cell B reports a false alarm because this
cell does not receive Msg3 from the UE. Consequently, an incorrect RACH
optimization decision is reached and Cell outage detection and compensation
(CODC) resets the eNodeB.
Figure 2.2.1.1.1.1.1.3 PRACH false alarm
CellA
CellB
2,Detect Preamble 2,Detect Preamble 2,Detect Preamble
4,Detect RAR from cellA,and

neglect RAR from CellB
The PRACH ZC root sequence planning based on U-Net aims to:

Ltd..
 Determine the NCS or ZeroCorrelationZoneConfig for the cells to be

planned.
 Determine the ZC root sequence index for the cells to be planned. The
ZC root sequence index is the first logical ZC root sequence number.
PRACH ZC root sequence checking based on U-Net aims to identify the cells
that are close to one another and are allocated the same ZC root sequence.
PRACH ZC root sequence reallocation based on U-Net aims to reallocate ZC
root sequences to E-UTRAN cells that do not meet PRACH ZC root sequence
evaluation requirements.
Preamble Generation
In the contention-based RA, a UE generates a corresponding ZC sequence
based on the values of the information elements (IEs) PreambleFormat,
HighSpeedFlag, zeroCorrelationZoneConfigConfig, and rootSequenceIndex
in the SIB message for RA. SIB is short for system information block.
In the non-contention-based RA, the eNodeB generates preambles based on
the settings of the parameters PreambleFormat, HighSpeedFlag,
zeroCorrelationZoneConfigConfig, and rootSequenceIndex and then
selects one preamble for a UE to initiate RA.
Where,
 The preamble sequence is generated with ZC root sequences through
cyclic shifting.
 PreambleFormat determines the length of a preamble sequence as well
as the number of preambles that are generated with one ZC root
sequence. For preamble formats 0 to 3, the preamble length is 839. For
preamble format 4, the preamble length is 139.
 HighSpeedFlag determines the method for calculating the cyclic shift
value, the NCS value corresponding to
zeroCorrelationZoneConfigConfig, and the number of preambles that
are generated with one ZC root sequence.
 zeroCorrelationZoneConfigConfig determines the zero correlation
window (ZCW) for preamble detection and the NCS value and indirectly
determines the cyclic shift value as well as the number of preambles that
are generated with one ZC root sequence. The NCS value is related to the
cell radius.
 rootSequenceIndex determines the set of the ZC root sequences for
generating preambles. rootSequenceIndex specifies the ZC root
sequence with the minimum logical ZC root sequence number in the ZC
root sequence set. Preambles for RA are generated through cyclic
shifting on each ZC root sequence in the ZC root sequence set.
In preamble formats 0 to 3, there are 838 available ZC root sequences. In
preamble format 4, there are 138 available ZC root sequence. In different
preamble formats, the same logical root sequence number corresponds to
different physical root sequence numbers.

Ltd..
ZC Root Sequence Index

RootSequenceIndex indicates the logical number of the first ZC root
sequence in the E-UTRAN cell ZC root sequence set. The eNodeB sends the
E-UTRAN cell ZC root sequence index to UEs through the broadcast
message. The UEs generate a total of 64 preambles through cyclic shifting
based on the ZC root sequence indicated by rootSequenceIndex and
subsequent ZC root sequences. The UEs initiate RA using one of the
generated preambles.
Table 2.2.1.1.1.1.1.3.1 describes the mapping between the logical root
sequences and physical root sequences for preamble formats 0 to 3.
Table 2.2.1.1.1.1.1.3.1 ZC root sequence order for preamble formats 0 – 3
Logical Physical Root Sequence Number u

Root (in increasing order of the corresponding logical sequence
Sequence number)
Number
0–23 129, 710, 140, 699, 120, 719, 210, 629, 168, 671, 84, 755,
105, 734, 93, 746, 70, 769, 60, 779
2, 837, 1, 838
24–29 56, 783, 112, 727, 148, 691
30–35 80, 759, 42, 797, 40, 799
36–41 35, 804, 73, 766, 146, 693
42–51 31, 808, 28, 811, 30, 809, 27, 812, 29, 810
52–63 24, 815, 48, 791, 68, 771, 74, 765, 178, 661, 136, 703
64–75 86, 753, 78, 761, 43, 796, 39, 800, 20, 819, 21, 818
76–89 95, 744, 202, 637, 190, 649, 181, 658, 137, 702, 125, 714,
151, 688
90–115 217, 622, 128, 711, 142, 697, 122, 717, 203, 636, 118, 721,
110, 729, 89, 750, 103, 736, 61, 778, 55, 784, 15, 824, 14,
825
116–135 12, 827, 23, 816, 34, 805, 37, 802, 46, 793, 207, 632, 179,
660, 145, 694, 130, 709, 223, 616
136–167 228, 611, 227, 612, 132, 707, 133, 706, 143, 696, 135, 704,
161, 678, 201, 638, 173, 666, 106, 733, 83, 756, 91, 748, 66,
773, 53, 786, 10, 829, 9, 830
168–203 7, 832, 8, 831, 16, 823, 47, 792, 64, 775, 57, 782, 104, 735,
101, 738, 108, 731, 208, 631, 184, 655, 197, 642, 191, 648,
121, 718, 141, 698, 149, 690, 216, 623, 218, 621
204–263 152, 687, 144, 695, 134, 705, 138, 701, 199, 640, 162, 677,
176, 663, 119, 720, 158, 681, 164, 675, 174, 665, 171, 668,

Ltd..
170, 669, 87, 752, 169, 670, 88, 751, 107, 732, 81, 758, 82,
757, 100, 739, 98, 741, 71, 768, 59, 780, 65, 774, 50, 789,
49, 790, 26, 813, 17, 822, 13, 826, 6, 833
264–327 5, 834, 33, 806, 51, 788, 75, 764, 99, 740, 96, 743, 97, 742,
166, 673, 172, 667, 175, 664, 187, 652, 163, 676, 185, 654,
200, 639, 114, 725, 189, 650, 115, 724, 194, 645, 195, 644,
192, 647, 182, 657, 157, 682, 156, 683, 211, 628, 154, 685,
123, 716, 139, 700, 212, 627, 153, 686, 213, 626, 215, 624,
150, 689
328–383 225, 614, 224, 615, 221, 618, 220, 619, 127, 712, 147, 692,
124, 715, 193, 646, 205, 634, 206, 633, 116, 723, 160, 679,
186, 653, 167, 672, 79, 760, 85, 754, 77, 762, 92, 747, 58,
781, 62, 777, 69, 770, 54, 785, 36, 803, 32, 807, 25, 814, 18,
821, 11, 828, 4, 835
384–455 3, 836, 19, 820, 22, 817, 41, 798, 38, 801, 44, 795, 52, 787,
45, 794, 63, 776, 67, 772, 72
767, 76, 763, 94, 745, 102, 737, 90, 749, 109, 730, 165, 674,
111, 728, 209, 630, 204, 635, 117, 722, 188, 651, 159, 680,
198, 641, 113, 726, 183, 656, 180, 659, 177, 662, 196, 643,
155, 684, 214, 625, 126, 713, 131, 708, 219, 620, 222, 617,
226, 613
456–513 230, 609, 232, 607, 262, 577, 252, 587, 418, 421, 416, 423,
413, 426, 411, 428, 376, 463, 395, 444, 283, 556, 285, 554,
379, 460, 390, 449, 363, 476, 384, 455, 388, 451, 386, 453,
361, 478, 387, 452, 360, 479, 310, 529, 354, 485, 328, 511,
315, 524, 337, 502, 349, 490, 335, 504, 324, 515
514–561 323, 516, 320, 519, 334, 505, 359, 480, 295, 544, 385, 454,
292, 547, 291, 548, 381, 458, 399, 440, 380, 459, 397, 442,
369, 470, 377, 462, 410, 429, 407, 432, 281, 558, 414, 425,
247, 592, 277, 562, 271, 568, 272, 567, 264, 575, 259, 580
562–629 237, 602, 239, 600, 244, 595, 243, 596, 275, 564, 278, 561,
250, 589, 246, 593, 417, 422, 248, 591, 394, 445, 393, 446,
370, 469, 365, 474, 300, 539, 299, 540, 364, 475, 362, 477,
298, 541, 312, 527, 313, 526, 314, 525, 353, 486, 352, 487,
343, 496, 327, 512, 350, 489, 326, 513, 319, 520, 332, 507,
333, 506, 348, 491, 347, 492, 322, 517
630–659 330, 509, 338, 501, 341, 498, 340, 499, 342, 497, 301, 538,
366, 473, 401, 438, 371, 468, 408, 431, 375, 464, 249, 590,
269, 570, 238, 601, 234, 605
660–707 257, 582, 273, 566, 255, 584, 254, 585, 245, 594, 251, 588,
412, 427, 372, 467, 282, 557, 403, 436, 396, 443, 392, 447,
391, 448, 382, 457, 389, 450, 294, 545, 297, 542, 311, 528,
344, 495, 345, 494, 318, 521, 331, 508, 325, 514, 321, 518
708–729 346, 493, 339, 500, 351, 488, 306, 533, 289, 550, 400, 439,
378, 461, 374, 465, 415, 424, 270, 569, 241, 598

Ltd..
730–751 231, 608, 260, 579, 268, 571, 276, 563, 409, 430, 398, 441,
290, 549, 304, 535, 308, 531, 358, 481, 316, 523
752–765 293, 546, 288, 551, 284, 555, 368, 471, 253, 586, 256, 583,
263, 576
766–777 242, 597, 274, 565, 402, 437, 383, 456, 357, 482, 329, 510
778–789 317, 522, 307, 532, 286, 553, 287, 552, 266, 573, 261, 578
790–795 236, 603, 303, 536, 356, 483
796–803 355, 484, 405, 434, 404, 435, 406, 433
804–809 235, 604, 267, 572, 302, 537
810–815 309, 530, 265, 574, 233, 606
816–819 367, 472, 296, 543
820–837 336, 503, 305, 534, 373, 466, 280, 559, 279, 560, 419, 420,
240, 599, 258, 581, 229, 610
Table 2.2.1.1.1.1.1.3.2 describes the mapping between the logical root

sequences and physical root sequences for preamble format 4.
Table 2.2.1.1.1.1.1.3.2 ZC root sequence order for preamble format 4
Logical Physical Root Sequence Number u (in increasing order of the corresponding
Root logical sequence number)
Sequence
Number
0–19 1 138 2 137 3 136 4 135 5 134 6 133 7 132 8 131 9 130 10 129
20–39 11 128 12 127 13 126 14 125 15 124 16 123 17 122 18 121 19 120 20 119
40–59 21 118 22 117 23 116 24 115 25 114 26 113 27 112 28 111 29 110 30 109
60–79 31 108 32 107 33 106 34 105 35 104 36 103 37 102 38 101 39 100 40 99
80–99 41 98 42 97 43 96 44 95 45 94 46 93 47 92 48 91 49 90 50 89
100–119 51 88 52 87 53 86 54 85 55 84 56 83 57 82 58 81 59 80 60 79
120–137 61 78 62 77 63 76 64 75 65 74 66 73 67 72 68 71 69 70 - -
138–837 N/A
During ZC root sequence planning, only logical root sequences are

considered. If the preamble format is specified, the physical root sequence
mapped to each logical root sequence is unique and determined.
 ZC sequence generation
Ltd..
The ZC sequence for number i logical ZC root sequence is generated through

the following steps:
Step 1 Obtain the physical root sequence u for the physical root sequence based on
number i root sequence index according to mappings described in Table
2.2.1.1.1.1.1.3.1 and Table 2.2.1.1.1.1.1.3.2.
Step 2 Generate a ZC sequence using the following formula. For each u, there is a
ZC sequence.
π un( n+1)
−j
N ZC
x u (n )=e , 0≤n≤N ZC −1
In the formula, Nzc indicates the length of the ZC sequence. For LTE
TDD/FDD preamble formats 0 to 3, the Nzc value is 839, indicating that each
ZC sequence has 839 symbols. For LTE TDD preamble format 4, the Nzc
value is 139, indicating that each ZC sequence has 139 symbols.
----End
 Cyclic shift
After a ZC sequence is generated, a preamble sequence can be generated with
the ZC sequence through cyclic shifting. The cyclic shift value is calculated
using the following formula:
Cv=¿{vNcs v=0,1,. ⌊Nzc/Ncs⌋−1,Ncs≠0 Low−spe dcel ¿{0 Ncs=0 Low−spe dcel ¿ ¿

For details about NCS calculation, see "NCS" in section 2.2.1"Introduction."
For high-speed cells, the cyclic shift value Cv is obtained through complex
computation. For details, see section X"Cyclic Shift Value."
NCS
The NCS is generated based on the cell radius.
The NCS value is related to the cyclic shift value of the preamble sequence to
the ZC root sequence. In addition, the NCS determines the preamble detection
window.
The preamble sequence for an E-UTRAN cell can be obtained through cyclic
shifting of the ZC root sequence based on the NCS, cell type, and preamble
length.
A UE detects for the RAR message within the detection window specified by
ZeroCorrelationZoneConfig. If the UE does not receive the RAR message
within the detection window, it retransmits the preamble.

Ltd..
The following formula shows the relationships between the NCS and the cell
radius and delay spread:
NCS x Ts ≥ RTD + SpreadDelay + error
Where
 Ts refers to the preamble sampling interval.
 RTD refers to the round trip delay between the eNodeB and the cell edge.
 SpreadDelay refers to the multi-path delay spread.
 Error refers to the product error.
The greater the cell radius, the greater the NCS value increases if the cell
radius becomes greater. If the estimated NCS value is relatively small, the
eNodeB cannot detect the preambles sent by cell edge UEs.
The E-UTRAN cell sends UEs a message that contains
ZeroCorrelationZoneConfig, HighSpeedFlag, and logical ZC root sequence
index over the air interface. The UEs can obtain the NCS value specified by
ZeroCorrelationZoneConfig and the ZC root sequence by analyzing the
obtained information and then generates preambles with the ZC root sequence
through cyclic shifting.
Table 2.2.1.1.1.1.1.3.3
N CS or preamble generation (Preamble Formats 0-3).
zeroCorrelationZone N CS Value
Config
Low-speed Cell High-speed Cell
(Unrestricted Set) (Restricted Set)
0 0 15
1 13 18
2 15 22
3 18 26
4 22 32
5 26 38
6 32 46
7 38 55
8 46 68
9 59 82
10 76 100
11 93 128
12 119 158

Ltd..
zeroCorrelationZone N CS Value
Config
13 167 202
14 279 237
15 419 -
Table 2.2.1.1.1.1.1.3.4
N CS for preamble generation (Preamble Format 4).
zeroCorrelationZoneConfigConfig N CS Value
0 2
1 4
2 6
3 8
4 10
5 12
6 15
7 N/A
8 N/A
9 N/A
10 N/A
11 N/A
12 N/A
13 N/A
14 N/A
15 N/A
Preamble Sequence Generation

This section describes how to generate a preamble sequence for a low-speed
cell in preamble format 0, 1, 2, or 3.

Ltd..
The cyclic shift value for low-speed cells is NCS and for high-speed cells is
Cv that is obtained through complex calculation. The principles for generating
preamble sequences with ZC sequences through cyclic shifting are the same
for high-speed and low-speed cells. The following figure shows the procedure
of generating preamble sequences for a low-speed cell. For the procedure of
generating preamble sequences for a high-speed cell, replace Cv with NCS.
Preamble 0
839
Step 1. Divide the ZC sequence with a length of 839 into
Ncs Ncs Ncs Ncs
Insufficie  839 / Ncs  - 1 groups. The length of each group is Ncs. If the
nt Ncs number of symbols of the last group is less than the Ncs
value, the group cannot be a preamble sequence. The initial
ZC sequence is preamble sequence 0.
...
Preamble 1
839
Insufficie
Ncs Ncs Ncs Ncs
nt Ncs Step 2. Perform cyclic shifting on the initial ZC sequence with
a cyclic shift value of 1 Ncs to obtain preamble sequence 1.
...
Preamble 2
839
Insufficie
Ncs Ncs Ncs Ncs
nt Ncs
Step 2. Perform cyclic shifting on the initial ZC sequence with
a cyclic shift value of 1 Ncs to obtain preamble sequence 1
...
Preamble  839 / Ncs  - 1

839
Insufficie
Ncs Ncs Ncs Ncs Step n. Perform cyclic shifting on the initial ZC sequence with
nt Ncs
a cyclic shift value of (  839 / Ncs  - 1) Ncs to obtain preamble
sequence (  839 / Ncs  - 1 ).
...
After cyclic shifting is completed on all NCS groups for the initial ZC
sequence, all preamble sequences generated with the ZC sequence are
obtained.
Both the uplink SRS and PRACH preamble are allocated ZC sequences but do not
interfere with each other on ZC sequence allocation because they are separated in the
frequency domain by RBs.
Factors Affecting the Preamble Sequence

 Radius
The cell radius must be planned in advance and the planned cell radius
must be the same as that configured on the eNodeB. On the eNodeB side,
you can obtain the actual cell radius using the related MML command
(for example, Mo: Cell.CellRadius=XXm) and then import the actual cell
radius into the tool for planning.

Ltd..
It is recommended that the cell radius be planned greater than the actual
cell coverage radius. If the cell radius is planned less than the actual cell
coverage radius, cell edge users will fail to access the cell. However, if
the cell radius is planned to be much greater than the actual cell coverage
radius, the following adverse effects are generated:
− Isolation decreases due to restricted resources. The greater the NCS
value, the greater the number of ZC sequences occupied by a cell.
However, the ZC sequences are limited, and the number of cells using
the same ZC root sequence increases. Take low-speed cells for an
example. A cell with the cell radius of 5 km requires four ZC root
sequences to generate 64 preambles and a cell with the cell radius of
10 km requires eight ZC root sequences to generate 64 preambles.
− Interference increases. Two sequences obtained with one root
sequence through cyclic shifting are more orthogonal than the two
sequences obtained with two separate root sequences through cyclic
shifting. The greater the NCS value, the more the required root
sequences.
 The number of preambles generated with each ZC root sequence is variable for a
high-speed cell. Therefore, no mappings are provided.
 Each cell is allocated with 64 preambles according to the protocols. However, due
to specifications restrictions of products, fewer than 64 preambles may be
allocated to each cell.
 Due to differences in parameter settings of various vendors, the number of ZC

sequences used to generate 64 preambles for cells of different radii may be
different.
Table 2.2.1.1.1.1.1.3.5 The number of ZC root sequences occupied by cells with

different radii
Max R(Km)\Integer NumZC/Cell
59~100 64
[0~0.8) 1
1, 2 2
3, 4 3
5 4
6, 7 5
8, 9 6
10, 11, 12 8
13, 14, 15 10
16~22 13
23~38 22

Ltd..
39~58 32
To facilitate network management, the cell radius should be scenario

specific. The following two methods for cell radius planning are
recommended:
− Method 1: Use U-Net to predict the cell coverage area, thereby
obtaining the cell radius. Add 4/3 of the farthest point obtained using
U-Net to the obtained cell radius to obtain the planned cell radius.
− Method 2: Use 4/3 of the cell radius obtained through estimation on
the network planning promotion statement to obtain the planned cell
radius.
The network planning specifications are based on simulation and only for reference. If
test data on the live network is provided, adjust the network planning promotion
statement according.
Assume that the network planning promotion statement for LTE

FDD/TDD is as follows:
Table 2.2.1.1.1.1.1.3.6 Setting for site height in different scenarios
Scenario
Morphology eNodeB antenna height
DenseUrban 30 m
Urban 30 m
SubUrban 35 m
Rural (Open) 40 m
Table 2.2.1.1.1.1.1.3.7 Parameter setting for cell link budget
System Information
Feeder Loss 0.5 dB (tower-mounted RRU)
Margin Interference Margin Obtained using related formula
Slow fading Margin Refer to "Typical Parameter"
Penetration Loss/Std. Dev. Of Refer to "Typical Parameter"

Slow Fading(dB)
Propagation Model Huawei-modified Cost231-Hata

propagation model
UL/DL Configuration (TDD) #1 5 ms 2:2

Ltd..
S Subframe Configuration #7 10:2:2

(TDD)
Uplink MCS MCS6 -- QPSK 0.31
Uplink Edge Rate In the same scenario, different uplink edge

rates require different cell coverage radii.
The cell coverage radius decreases when
the uplink edge rate rises from 64 kbit/s to
128 kbit/s, then to 256 kbit/s, and then to
512 kbit/s.
 For LTE FDD eNodeBs, the MIMO mode is set to SFBC and antenna
configuration is 1T2R. for LTE TDD eNodeBs, the MIMO mode is set to
SFBC[[2T2R/4T4R] & BeamForming [8T8R] and the antenna configuration is
1T2R.
 The eNodeB is configured with three sectors.
 The cell radius depends on the uplink cell edge rate and the downlink rate is
calculated based on the uplink cell radius. The downlink rate in the table above
refers to the maximum downlink edge rate of the uplink cell edge. Rates for other
bandwidths can be obtained based on the proportion of the bandwidth to 20
Mbit/s.
 The eNodeBs in 2T2R mode and 4T4R mode have the same total power and 4-
way receive diversity provides higher gains than 4-way transmit diversity does.
Therefore, the downlink rate decreases after four antennas are used to enhance
uplink coverage.
 The cell coverage radius depends on multiple factors and cannot be included in
KPIs promised to customers. Therefore, do not promise the cell coverage radius to
customers under uncertain conditions.
 For 900 MHz, the Huawei-provided Okumura-Hata model is used. For 1800
MHz, the Huawei-provided Cost231-Hata model is used. On the 900 MHz
frequency, the antenna gain is 15 dBi. On the 1800 MHz frequency, the antenna
gain is 18 dBi. The edge coverage ratio is set to 90%.
The cell radii for LTE FDD and LTE TDD cells are described in the
following two figures.

Ltd..
LTE FDD Cell Radius (for Reference)
Total Power of Antenna Cell Radius Cell Radius Cell Radius DL Edge Cell Radius
Scenario eNodeB (W) Configuration Coverage at 1800 at 2300 at 2600 Rate 100RB at 2600 MHz
MHz (km) MHz (km) MHz (km) (Mbit/s) (km)
Indoor coverage
Outdoor coverage
Indoor coverage
Dense urban
area Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Urban area Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Suburban
area Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Rural area
Outdoor coverage
Indoor coverage
Outdoor coverage
Based on the preceding information, the recommended cell radius for the LTE
FDD or LTE TDD system is as follows:
FDD
Total Power of Antenna Cell Radius Cell Radius Cell Radius Cell Radius
Scenario Cell Type eNodeB (W) Configuration Coverage (900 MHz) (1800 MHz) (2100 MHz) (2600 MHz)
Indoor coverage
Dense urban Directional Outdoor coverage

area cell
Indoor coverage
Outdoor coverage
Indoor coverage
Urban area Outdoor coverage

Directional cell
Indoor coverage
Outdoor coverage
Indoor coverage
Outdoor coverage
Directional Indoor coverage
cell
Outdoor coverage
Suburban
area Indoor coverage
Outdoor coverage
Omnidirectional
Indoor coverage
cell
Outdoor coverage
Indoor coverage
Outdoor coverage
Directional Indoor coverage
cell
Outdoor coverage
Rural area
Indoor coverage
Outdoor coverage
Omnidirectional
cell Indoor coverage
Outdoor coverage
TDD

Ltd..
Scenario Total Power of Antenna Coverage 1800 MHz 2300 MHz 2600 MHz 3.5 G
eNodeB (W) Configuration
Indoor coverage
Outdoor coverage
Dense Indoor coverage

urban area
Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Urban area
Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Outdoor coverage
Suburban Indoor coverage

area
Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Outdoor coverage
Indoor coverage
Rural area
Outdoor coverage
Indoor coverage
Outdoor coverage
 The cell radius determines the NCS value and the NCS value is related to the
preambles for cells. Therefore, the cell radius configured on the eNodeB side must
be the same as that used during PRACH ZC root sequence planning. If the cell
radius on the eNodeB side is specified, you can import the cell radius information
into the tool and then perform PRACH ZC root sequence planning. If the cell
radius on the eNodeB side is not specified, you can set the cell radius based on the
reference value or calculate the cell radius using the tool. After PRACH ZC root
sequence planning is completed, the cell radius must be delivered to the eNodeB
together with the ZC root sequence no matter whether the cell radius is specified
or not.
 The cell radius in dense urban areas, suburban areas, or overlaps of urban
boundaries, you are advised to set the cell radius to a relatively large value.
For details, see LTE FDD Network Planning V2.2 and LTE TDD
Network Planning V2.2.
 PreambleFormat
The preamble format can serve as a case for encapsulating preamble
sequences. An RA preamble is a pulse, which consists of a Tcp and a
TSEQ in time domain. Here, Tcp denotes the time length of a Cyclic
Prefix (CP) and TSEQ denotes the time length of a preamble sequence.

Ltd..
CP Sequence
TCP TSEQ
The length of the RA preamble pulse can be determined based on the

preamble format configuration. The RA preamble has five formats:
format 0 through format 4. The cell radius varies with preamble formats.
The RA preamble format is sent to UEs through the broadcast message
and the UEs select corresponding RA preamble format to initiate RA.
Preamble Pulse TCP TSEQ LTE FDD LTE TDD
Format Cell Radius Cell Radius R
R
0 1000 103.1 µs 800 µs R ≤ 14 km 1.4 km < R ≤

µs 77 km
1 2000 684.4 µs 800 µs 29 km < R ≤ 29 km < R ≤

µs 77 km 77 km
2 2000 203.1 µs 1600 µs 14 km < R ≤ 14 km < R ≤

µs 29 km 29 km
3 3000 684.4 µs 1600 µs 77 km < R ≤ 77 km < R ≤

µs 100 km 100 km
4 167.9 14.58 µs 133.33 N/A R ≤ 1.4 km

µs µs
The length of the preamble in preamble formats 0 to 3 is 839 and each

preamble occupies 800 µs. The length of the preamble in preamble
format 4 is 139 and each preamble occupies 133.33 µs.
It is recommended that the preamble format be configured based on the
cell radius.
 HighSpeedFlag
HighSpeedFlag specifies the type of an E-UTRAN cell. It can be set to
low speed, high speed, or highway speed. The eNodeB sends the E-
UTRAN cell type to UEs through broadcast messages. Then, the UEs
identify the NCS value corresponding to
ZeroCorrelationZoneConfigConfig and the preamble sequence
generation mode according to the value of HighSpeedFlag. For details,
see 3GPP36.211.
Incorrect cell speed configurations reduce the RA success rate. You are
advised to configure the cell speed based on the coverage scenarios. For
example, configure the cells covering the high-speed railways as high-
speed cells.

Ltd..
Doppler spread occurs on high-speed moving UEs, which disables the

zero-autocorrelation feature on ZC sequences. Therefore, high-speed
UEs must use different ZC sequences, NCS, and preamble generation
modes from those for low-speed cells. Doppler spread depends on the
carrier and UE speed. Table 2.2.1.1.1.1.1.3.8 lists the speed division for
low-speed, high-speed, and highway-speed cells in common carriers.
Table 2.2.1.1.1.1.1.3.8 Speed division for low-speed, high-speed, and highway-speed

cells in common carriers
Carrier Speed Limit Between Low-speed Speed Limit Between High-
(GHz) and High-speed Cells (km/h) speed and Highway-speed
Cells (km/h)
0.8 317 527
0.9 276 462
1.8 111 204
2.1 87 167
2.6 60 125
3.5 31 79
 When the actual speed of a UE is close to the speed limit, the UE's performance
deteriorates to some extent. In actual conditions, there are some margins for the
speed limit. For example, if the speed of a cell is 110 on the 1.8 GHz carrier, the
cell can be configured as a high-speed cell.
 The preceding speed division takes effect only on PRACH ZC root sequence
planning.
 Cell speeds listed in the table are just simulation speeds. In actual scenarios,
configure the cell based on the simulation results and outfield verification results.
 Take the 800 MHz carrier for an example. If the cell speed is 310 km/h, the cell is
a low-speed cell. If the cell speed is 320 km/h, the cell is a high-speed cell. If the
cell speed is 520 km/h, the cell is a high-speed cell. If the cell speed is 535 km/h,
the cell is a highway-speed cell.
2.2.2 PRACH ZC Root Sequence Planning Principles

The following calculation operations are involved in PRACH ZC root
sequence planning:
 Calculate the NCS value for an E-UTRAN cell.
 Calculate the number of preambles that are generated with each available
ZC root sequence.

Ltd..
 Obtain available ZC root sequence groups. A ZC root sequence group

contains consecutively numbered ZC root sequences. Each ZC root
sequence group generates at least 64 preamble sequences.
 Select an optimal ZC root sequence group from these available ZC root
sequence groups and obtain the ZC root sequence index that is the
logical number of the ZC root sequence with the smallest number in the
optimal ZC root sequence group.
NCS Calculation
The NCS value determines the size of the zero correlation window for the
preamble sequence as well as the cyclic shift value of the ZC sequence. The
NCS value depends on the cell radius (CellRadious) and sampling interval
(Ts). Ts is related to the preamble sequence length (Nzc) and the occupied
time domain resources (Tseq). For details, see the following formula.
NCS ≥ (RTD + SpreadDelay + Error)/Ts
Where
 RTD = 2 x CellRadius/LightSpeed
 SpreadDelay refers to the delay spread.
 Error refers to the error during the calculation.
Ts = Tseq/Nzc
For preamble formats 0 to 3, the preamble length is 839 and 800 µs is
occupied. For preamble format 4, the preamble length is 139 and 133.33 µs is
occupied.
Number of Preambles Generated with Each ZC Root Sequence

 Low-speed cell
For low-speed cells, Nzc/ Ncs preamble sequences can be

⌊ ⌋
generated with each ZC root sequence.
 High-speed cell
For high-speed cells, the number of preamble sequences generated with a
ZC root sequence may differ because the number is related to the
physical number of the ZC root sequence (u), NCS value, and Nzc.
n RA
shift,
n RA
group , and n̄ RA
shift for each ZC root sequence are
calculated. Then, the number of preamble sequences generated with a ZC
RA RA RA
root sequence is (
n shift .
n group +
n̄ shift ).
Number of ZC Root Sequences for Each ZC Root Sequence Group

A ZC root sequence group consists of multiple consecutively numbered ZC
root sequences. Each ZC root sequence group generates at least 64 preamble
sequences.

Ltd..
Each E-UTRAN cell must be assigned sufficient ZC root sequences to

generate 64 preamble sequences.
 Low-speed cell
For a low-speed cell to generate 64 preamble sequences, the number of
required ZC root sequences can be calculated using the following
64
⌈ ⌉
formula: ⌊ Nzc/ Ncs ⌋
 High-speed cell
Calculate the number of preamble sequences generated with each of the
Nu logically numbered ZC root sequences. If the total number of
N u−1
∑ N v ≥64
i
generated preamble sequences is 64 or greater, that is i=0 ,
the Nu ZC root sequences can form a ZC root sequence group. The
number of ZC root sequences varies with ZC root sequence groups.
Nzc refers to the length of a ZC sequence and is related to the preamble format.
Number of Available ZC Root Sequence Groups for an E-UTRAN

Cell
The number of preamble sequences generated with a PRACH ZC root
sequence is calculated based on the available ZC root sequences for users. If
the number of preamble sequences generated with each of the available ZC
root sequences that are consecutive in logical numbers is greater than or equal
to 64, these ZC root sequences can be grouped into a ZC root sequence group.
Then, perform permutation and combination on all available ZC root
sequences based on the preceding principle to obtain available ZC root
sequence groups for an E-UTRAN cell.
Optimal ZC Root Sequence Group for an E-UTRAN Cell

During ZC root sequence allocation for E-UTRAN cells, the ZC root
sequences allocated to planned cells should be different from those for first-
and second-tier intra-frequency neighboring cells. If no such ZC root
sequence groups are available, perform tier reduction. That is, the ZC root
sequences allocated to planned cells can be the same as those for the second-
tier intra-frequency cells.
If multiple ZC root sequence groups meet the requirements, perform
operations accordingly. For example,
 If there are seldom used ZC sequence groups, allocate such ZC sequence
groups to E-UTRAN cells.
 If all available ZC sequence groups have been used, calculate the
distance between the planned cell and the cell to be planned and
topology tier and then reallocate the ZC root sequence group for the
planned cell with the maximum sum of the distance and topology tier

Ltd..
count to the cell to be planned. The ZC root sequence group is the

optimal ZC root sequence group.
 First-tier neighboring cells refer to the neighboring cells in the neighboring cell
list for an E-UTRAN cell.
 Second-tier neighboring cells refer to the first-tier neighboring cells of the first-
tier neighboring cells of an E-UTRAN cell.
 The topology tier between two cells is the number of eNodeBs in an ellipse that is
drawn with the distance between the two cells as the major axis.
If no ZC root sequence meets the requirements, perform tier reduction on the

cells to be planned by reducing the reuse tier of neighboring cells or making
the ZC root sequences allocated to neighboring cells of the highest tier be
available for cells to be planned. Then, check whether there are ZC root
sequences meeting the requirements. If no, perform tier reduction again.
Then, the start ZC root sequence of the optimal ZC root sequence group
serves as that for the E-UTRAN cell.
Procedures of calculating the optimal ZC root sequence group
Step 1 Obtain a group of available ZC root sequences and the set of cells that use the
group of ZC root sequence index. Then, calculate the value of CellScore using
the following formula:
CellScore = Topology tier between two cells x distance between two cells
GroupScore  Min( CellScore )

n
Step 2 Calculate the score of a group of available ZC root sequences for the target
cell using the following formula:
BestGroupS core  Max( GroupScore )

m
Step 3 Select the ZC root sequence group with the maximum reuse score as that for
the cell to be planned.

Ltd..
Root sequence index Group1 Root sequence index Group2
GroupScore=Min(CellScore)
Root sequence index Group3
CellScore=Tire*Distance
Object Cell
----End
2.2.3 PRACH ZC Root Sequence Check Principles

PRACH ZC root sequence checking aims to identify the E-UTRAN cell
with the lowest ZC root sequence reuse isolation.
The reuse isolation can be evaluated by the following two factors.
 Reuse distance of ZC root sequences
Based on the minimum reuse distance of ZC root sequences, identify the
intra-frequency E-UTRAN cells assigned with the same ZC root
sequence within the minimum reuse distance as the suspicious E-
UTRAN cells with ZC root sequence conflicts.
Based on live network conditions, determine the distance between two
cells without interference with each other. Use the distance as the
minimum distance between two intra-frequency cells assigned with the
same ZC root sequence, that is, the minimum reuse distance of ZC root
sequences.
In the event of proper RF resource planning, coverage balance between
the uplink and downlink can be achieved. Therefore, it is good practice
to refer to the reuse distance for PCI planning. For urban areas, the
minimum reuse distance is 4 km. For suburban areas, the minimum reuse
distance is 10 km.
 Reuse tier of ZC root sequences
The tier is related to neighboring E-UTRAN cells. First-tier neighboring
cells are the cells in the neighboring cell list for an E-UTRAN cell,
second-tier neighboring cells are the first-tier neighboring cells of the
first-tier neighboring cells for an E-UTRAN cell, and so on.
Generally, the source cell is adjacent to its first-tier neighboring cells,
and signal overlaps are introduced. Therefore, the ZC root sequences
assigned to the source cell and its first-tier neighboring cells are expected

Ltd..
to be different. If signal overlap occurs on two E-UTRAN cells that are

of multiple tiers, the ZC root sequences allocated to these two cells
should be different.
Based on the minimum reuse tier (1) of ZC root sequences, identify the
intra-frequency neighboring E-UTRAN cells assigned with the same ZC
root sequence within the same tier as the suspicious cells with ZC root
sequence conflicts. You are advised to set the tier count of neighboring
cells to be checked higher than that of the neighboring cells requested to
have different ZC root sequences from the cells to be replanned.
In swap or reallocation scenarios, you can also identify the ZC root sequence
reuse isolation by comparing the check results before and after reallocation.
Use a tool for performing PRAC ZC root sequence planning and checking only based
on theoretical analysis results. On live networks, it is possible that cell A has a long
ZC root sequence reuse distance and high reuse tier and UEs initiate multiple RA
attempts in cell A. In this case, ZC root sequence allocation on cell A needs to be
optimized. However, U-Net becomes unavailable in this scenario and frontline
personnel conduct optimization based on the false alarms in the cell.
2.2.4 PRACH ZC Root Sequence Reallocation

With ZC root sequence reallocation, ZC root sequences are reallocated for the
suspicious cells with ZC root sequence conflicts identified in section
2.2.3"PRACH ZC Root Sequence Check Principles."
2.3 Application Scenarios

Figure 2.3.1.1.1.1.1.1 illustrates the PRACH ZC root sequence planning
procedure.
Figure 2.3.1.1.1.1.1.1 PRACH ZC root sequence planning procedure
Check the ZC root Reallocate ZC root

Allocate ZC root sequences to cells
sequences allocated to
sequences to new E-
E-UTRAN cells on the with ZC root sequence
UTRAN cells. conflicts.
network.
In new site deployment and capacity expansion scenarios, allocate proper ZC

root sequences and NCS values to new E-UTRAN cells to ensure that the E-
UTRAN cells work properly.
After the PRACH ZC root sequence planning is complete, check the planning
result and then send the result to the NEs. During the network maintenance,
the PRAC ZC root sequences for E-UTRAN cells also need to be checked.
Check for suspicious E-UTRAN cells with improper ZC root sequences to
ensure normal network operation or improved network performance. Re-

Ltd..
allocate proper ZC root sequences to E-UTRAN cells with improper ZC root

sequences.
2.4 Application Guide

2.4.1 U-Net PRACH ZC Root Sequence Planning
PRACH ZC Root Sequence Planning
 Overview
In site deployment or capacity expansion scenarios, NCS values and the
PRACH ZC root sequence index must be planned for new E-UTRAN
cells.
During PRACH ZC root sequence planning, collect the required data for
the planning, prepare the engineering parameter table, and then use the
U-Net to plan PRACH ZC root sequences.
 Tool
− Available at:
http://support.huawei.com/carrier/navi?
lang=zh#col=software&path=PBI1-7851894/PBI1-8132365/PBI1-
8132371/PBI1-21395144/PBI1-13953
 Input
− NE engineering parameter information (For details, see Table
2.4.1.1.1.1.1.1.1, Table 2.4.1.1.1.1.1.1.2, Table 2.4.1.1.1.1.1.1.3, and
Table 2.4.1.1.1.1.1.1.4.)
− PRACH ZC root sequence planning policy (For details, see Table
2.4.1.1.1.1.1.1.5.)

Name Value
Site Name Indicates the N/A The U-Net uses the site
site name. name as the index
keyword.
Longitude Indicates the N/A N/A

site
longitude.
Ltd..
Latitude Indicates the N/A N/A

site altitude.
Altitude Indicates the N/A This parameter must be

site altitude. specified when the E-
UTRAN cell radius can
be calculated based on
cell coverage prediction.

Name Value
Transceiver Indicates the User-defined N/A

Name sector name.

horizontal served by multiple
offset of the antennas, specify the
antenna to horizontal offset of each
the eNodeB. antenna to the base
station.

vertical served by multiple
antenna to vertical offset of each
the eNodeB. antenna to the base
station.
Antenna Indicates the N/A If the E-UTRAN cell is

antenna served by multiple
type. antennas, specify the type
of each antenna.
This parameter must be
specified when the E-
Height Indicates the N/A If the E-UTRAN cell is

height. antennas, specify the
height of each antenna.

Ltd..

azimuth. antennas, specify the
Mechanical Indicates the N/A If the E-UTRAN cell is

Down tilt antenna served by multiple
mechanical antennas, specify the
downtilt. mechanical downtilt of
each antenna.
Electrical Indicates the N/A If the E-UTRAN cell is

Down tilt antenna served by multiple
electrical antennas, specify the
downtilt. electrical downtilt of
each antenna.
Power Ratio Indicates the N/A If the E-UTRAN cell is

power ratio. antennas, specify the
power ratio of each
antenna.

Name Value
Cell Name Indicates the N/A The U-Net uses the cell
cell name. name as the index

Ltd..

Name Value
keyword.
Frequency Indicates the N/A If necessary, the

Band attribute of frequency and bandwidth
the cell need to be specified.
frequency
band.
RS Power Indicates the 15.2 dBm This parameter must be

transmit specified when the E-
power of the UTRAN cell radius can be
reference calculated based on cell
signal (RS). coverage prediction.
Main Indicates the N/A This parameter must be

Propagation propagation specified when the E-
Model model. UTRAN cell radius can be
calculated based on cell
Main Indicates the N/A This parameter needs to

Calculation calculation be specified in any of the
Radius radius of the two conditions:
path loss  The E-UTRAN cell
matrix. radius can be
calculated based on
cell coverage
prediction.
 The E-UTRAN cell
radius can be
calculated based on the
propagation model
radius.
Min Root Indicates the N/A If this parameter is not

Sequence index of the specified, the cell is a new
Index minimum cell.
ZC root
sequence for
an E-
UTRAN
cell.
Radius Indicates the Set this parameter This parameter needs to

E-UTRAN according to the be specified when
cell radius. actual scenario. PRACH ZC root sequence
Unit: meter planning is performed
based on the set cell
radius.
Prach Reuse Indicates the 2 N/A

Ltd..

Name Value
Tier(Neighb PRACH ZC
or) root
sequence
reuse tier.

Name Value
CellName Indicates the N/A The U-Net uses the cell

source cell name as the index
name. keyword.
NCellName Indicates the N/A The U-Net uses the cell

target cell name as the index
name. keyword.
Table 2.4.1.1.1.1.1.1.5 Parameters involved in the PRACH ZC root sequence

planning policy
Name Value
Available Indicates the N/A Set this parameter

Root logical number according to the actual
Sequence for an scenario.
Index available Indoor and outdoor cells
PRACH ZC are allocated different
root sequence. PRACH ZC root
sequences. On the
boundaries, cells under
different areas use
different PRACH ZC
root sequences.
Planning Indicates N/A Select this option in the

Prach Based whether the capacity expansion
on existed PRACH ZC scenario. The PRACH
Prach root sequences ZC root sequences
allocated to the allocated to existing
planned cells cells are unchanged and
are changed. PRACH ZC root
sequence planning is
performed only on new
cells.

Ltd..

Name Value
Calculate Whether the E- N/A If this option is selected,

Cell Radius UTRAN cell the E-UTRAN cell
radius needs to radius is calculated. If
be calculated. no, the cell radius
specified in CellTable is
used.
Propagation Whether the E- N/A This option is selected

Radius UTRAN cell only when Calculate
radius can be Cell Radius is selected.
calculated
based on the
propagation
model radius.
Propagation Indicates the N/A If Calculate Cell

Radius factor for Radius and
Factor reference when Propagation Radius
the E-UTRAN are selected, the E-
cell radius can UTRAN cell radius is
be calculated the propagation model
based on the radius of the E-UTRAN
propagation cell multiplied by the
model radius. value of Radius Factor
during PRACH ZC root
sequence planning.
Coverage Whether the E- N/A This option is selected

Radius UTRAN cell only when Calculate
radius can be Cell Radius is selected.
calculated
based on cell
coverage.
Resolution( Indicates 50 m This parameter must be

m) coverage specified when the E-
prediction UTRAN cell radius can
resolution. be calculated based on
cell coverage
prediction.
Min Signal Indicates the –120 dBm This parameter must be

Level(dBm) RSRP specified when the E-
threshold of an UTRAN cell radius can
E-UTRAN cell be calculated based on
during cell coverage
prediction.
Shadowing Indicates that Do not select This parameter must be

taken into shadow fading Shadowing specified when the E-

Ltd..

Name Value
account is considered. taken into UTRAN cell radius can

account. be calculated based on
cell coverage
prediction.
Cell Edge Indicates the 75% This option is selected

Coverage cell edge when the E-UTRAN
Probability coverage cell radius can be
probability of calculated based on cell
shadow fading. coverage prediction and
Shadowing taken into
account is selected.
The shadow fading
variance of the cluster
must be specified. For
details, see U-Net
Operation Guide.
Indoor Indicates that N/A This parameter must be

Coverage penetration specified when the E-
loss is UTRAN cell radius can
considered. be calculated based on
cell coverage
prediction.
The penetration loss of
the cluster needs to be
specified. For details,
see U-Net Operation
Guide.
Area Indicates the N/A If the user does not

PRACH ZC draw a polygon, the full
root sequence map is for planning.
planning area. If a user draws
polygons on the map,
planning can be
performed in the areas
in polygons.
 Procedures
Step 1 Collect live network information, including NE engineering parameter
information, cells to be planned, and available PRACH ZC root sequences for
these cells.
Step 2 Prepare the engineering parameter table file according to the U-Net
engineering parameter template.

Ltd..
If the electronic map has a coordinate system, do not manually set the coordinate
system.
If the E-UTRAN cell radius can be calculated based on the coverage prediction,
import the electronic map to obtain a more accurate planning result.
Step 5 Import engineering parameter files of NEs in the order of the site table,
transceiver table, cell table, and neighbor relationship.
If the neighbor relationship is not imported, PRACH ZC root sequence conflicts

among neighboring cells are not considered during the PRACH ZC root sequence
planning.
Step 6 Set PRACH ZC root sequence self-planning parameters to start PRACH ZC

root sequence planning.
Step 7 Evaluate the planning result and manually adjust the planning result slightly.
----End
 Output
Table 2.4.1.1.1.1.1.1.6 PRACH planning result

Name Value
CellName Indicates the E- N/A N/A

UTRAN cell name.
Ncs Indicates the NCS N/A N/A

of an E-UTRAN
cell.
CellRadius Indicates the radius m During PRACH

of an E-UTRAN ZC root sequence
cell. planning, the E-
UTRAN cell
radius can be
calculated based
on coverage
prediction or
propagation model
radius.
StartRootSeq Indicates the start N/A N/A

uecnce logical ZC root
sequence number.
EndRootSeq Indicates the end N/A This parameter is

uence logical ZC root optional.

Ltd..

Name Value
sequence number.
Reuse Indicates the reuse m N/A

Distance distance of ZC root
sequences.
Reuse Indicates the reuse N/A The reuse tier of

Tier(Neighb tier of ZC root ZC root sequences
or) sequences. is the number of
tiers for
neighboring cells.
PRACH ZC Root Sequence Check

 Overview
a. Checking and evaluating PRACH ZC root sequence planning
results after PRACH ZC root sequence planning is completed
b. Checking and evaluating PRACH ZC root sequences allocated
to the live network at the network maintenance phase
− Information to be checked and evaluated
a. PRACH ZC root sequence reuse isolation (including the reuse
distance and reuse tier)
b. Cells with potential ZC root sequence conflicts
− Evaluation methods
a. Evaluation based on PRACH ZC root sequence planning result
table
Summarizes, filters, and records the PRACH ZC root sequence
reuse isolation (of E-UTRAN cells with low PRACH ZC root
sequence reuse isolation, such as the small reuse distance and
small number of reuse tiers) for filing and viewing.
b. Evaluation based on geographic display
Facilitates viewing PRACH ZC root sequence reuse isolation.
c. Evaluation based on CDF statistics
Evaluates the PRACH ZC root sequence reuse isolation of E-
UTRAN cells in a quantitative manner. Specifically, in swap or
PRACH ZC root sequence replanning scenarios, this method
facilitates comparing the PRACH ZC root sequence reuse
isolation before and after swap or PRACH ZC root sequence
replanning in a quantitative manner.
 Tool

Ltd..

− Available at:
8132371/PBI1-21395144/PBI1-13953
 Input
− Check and evaluation after PRACH ZC root sequence planning is
completed
No parameter settings are required. You can directly submit the
planning result.
− Check and evaluation during network maintenance
Parameters of NEs listed in Table 2.4.1.1.1.1.1.1.7, Table
2.4.1.1.1.1.1.1.8, Table 2.4.1.1.1.1.1.1.9, and Table 2.4.1.1.1.1.1.1.10
are required.

Name Value
Site Name Indicates the site name. N/A The U-Net uses
the site name as
the index
keyword.

longitude.

altitude.

Name Value
Transceiver Indicates the User-defined N/A

Name sector name.

horizontal served by multiple
antenna to the horizontal offset of each
eNodeB. antenna to the base
station.

Ltd..

Name Value
vertical offset served by multiple

of the antenna antennas, specify the
to the eNodeB. vertical offset of each
antenna to the base
station.

azimuth. antennas, specify the

Name Value
Cell Name Indicates the N/A The U-Net uses the cell
cell name. name as the index
keyword.
Frequency Indicates the N/A If necessary, the

Band attribute of the frequency and
cell frequency bandwidth need to be
band. specified. If the actual
bandwidth is lower than
the standard bandwidth,
the actual bandwidth is
compact bandwidth.
Min Root Indicates the N/A N/A

Sequence PRACH ZC
Index root sequence
index of an E-
UTRAN cell.

Name Value
CellName Indicates the N/A The U-Net uses the cell

source cell name as the index
name. keyword.

Ltd..
NCellName Indicates the N/A The U-Net uses the cell

target cell name as the index
name. keyword.
 Procedures
Step 1 If PRACH ZC root sequence self-planning is completed using the U-Net,
check the PRACH ZC root sequence based on the planning result. During
network maintenance, you need to prepare an engineering parameter file,
create a project, and import the engineering parameters before PRACH ZC
root sequence checking.
Step 2 Check the PRACH ZC root sequence reuse isolation (including the reuse
distance and reuse tier) to find the cells with potential ZC root sequence
conflicts.
----End
PRACH ZC Root Sequences Reuse Isolation Evaluation

 Overview
− Application Scenario
a. In swap and PRACH ZC root sequence replanning scenarios, the
PRACH ZC root sequence reuse isolation evaluation is used to
quantitatively compare the result before and after the replanning.
b. In site deployment scenarios, the PRACH ZC root sequence
reuse isolation evaluation is used to quantitatively analyze the
PRACH ZC root sequence planning on the network and
determine whether the planning result meets the actual
requirement (for example, the PRACH ZC root sequence reuse
distance of 98% E-UTRAN cell exceeds 10 km).
− Evaluation Methods
a. Minimum reuse distance of E-UTRAN cells on the network in
CDF statistics
Export or copy the values of Reuse Distance of each cell in the
PRACH ZC root sequence planning result table to an EXCEL
file, collect the CDF statistics of each reuse distance, and
generate a line graph.

Ltd..
Figure 2.4.1.1.1.1.1.2 PRACH ZC root sequence reuse distance
Initial
configuration
After PRACH
ZC root
sequence
planning
b. Minimum number of E-UTRAN cell reuse tiers on the network

in CDF statistics
Export or copy the values of Reuse Tier(Neighbor) of each cell
in the PRACH ZC root sequence planning result table to an
EXCEL file, collect the CDF statistics of reuse tiers, and
generate a line graph shown in Figure 2.4.1.1.1.1.1.2.
− Geographical presentation of the PRACH ZC root sequence reuse of
each cell on the network
Click an E-UTRAN cell on the network to view its intra-frequency E-
UTRAN cells with the same PRACH ZC root sequence. As shown in
Figure 2.4.1.1.1.1.1.3, cells in red are the intra-frequency cells with
the same PRACH ZC root sequence.

Ltd..
Figure 2.4.1.1.1.1.1.3 Cells with the same PRACH ZC root sequence
PRACH ZC Root Sequence Conflict Checking

PRACH ZC root sequence conflict checking aims to check whether there are
E-UTRAN cells allocated incorrect PRACH ZC root sequences on the
network and ensures that the PRACH ZC root sequence reuse distance and
reuse tier for all cells meet the requirements.
The PRACH ZC root sequence planning result table is exported after PRACH
ZC root sequence planning is complete. In addition, you can also view the
PRACH ZC root sequence planning result table by importing the network
engineering parameters, right-clicking Prach Planning, and then choosing
Open Prach Parameter in the displayed dialog box.
Figure 2.4.1.1.1.1.1.4 PRACH ZC root sequence planning result table
After the PRACH ZC root sequence planning result table is exported, find the
E-UTRAN cells whose PRACH ZC root sequence reuse isolation does not
meet the requirements. For example, the number of the neighboring cell tiers
is 1 (you can set the tier to a value greater than 1 based on actual conditions),
or the PRACH ZC root sequence reuse distance is small (4 km for common
urban areas and 10 km for suburban areas). The area encircled in red in Figure

Ltd..
2.4.1.1.1.1.1.5 displays the suspicious cells with PRACH ZC root sequence

conflicts, and the number of reuse tiers for these cells is 1.
Figure 2.4.1.1.1.1.1.5 Suspicious cells with PRACH ZC root sequence conflicts
PRACH ZC Root Sequence Reallocation

 Overview
Use the U-Net to re-allocate proper PRACH ZC root sequences for cells
that do not meet the requirement if any of the following conditions
occurs:
− Intra-frequency cells that are close to one another use the same
PRACH ZC root sequence
− Cells of the same tier use the same PRACH ZC root sequence
 Tool
− Available at:
8132371/PBI1-21395144/PBI1-13953
 Input
− Engineering parameter information described in "PCI Planning" in
section 1.1.2"U-Net PCI Planning."
− E-UTRAN cells requiring PRACH ZC root sequence replanning
 Procedures
information, cells to be planned, available PRACH ZC root sequences for
these cells, and cells requiring PRACH ZC root sequence replanning.
Step 2 Prepare the engineering parameter table based on the engineering parameter
table of the U-Net and set Min Root Sequence for cells to be planned to null.

Ltd..
system.
If the E-UTRAN cell radius can be calculated based on the coverage prediction,
import the electronic map to obtain a more accurate planning result.
transceiver table, cell table, and neighbor relationship.
If the neighbor relationship is not imported, PRACH ZC root sequence conflicts

among neighboring cells are not considered during the PRACH ZC root sequence
planning.
Step 6 Select Planning Prach Based on existed Prach, perform PRACH ZC root
sequence self-planning, and submit the planning result.
Step 7 Evaluate the planning result and manually and slightly adjust the planning
result.
Step 8 Export the planning result and prepare the delivery script to deliver the
planning result.
----End
 Output
For details about output, see Figure 2.4.1.1.1.1.1.5.
2.4.2 SmarRNO ZC Root Sequence Planning

PRACH ZC Root Sequence Planning
 Overview
In site deployment or capacity expansion scenarios, NCS values and the
PRACH ZC root sequence index must be planned for new E-UTRAN
cells.
During PRACH ZC root sequence planning, collect required data for
planning, prepare the engineering parameter table, and then use a tool to
perform planning.
 Tool
− Available at:
http://3ms.huawei.com/hi/group/2029711/thread_5270817.html?
 Input

Ltd..
− NE engineering parameter information (For details, see Table

2.4.2.1.1.1.1.1.1.)
− PRACH planning policy (For details, see Error: Reference source not
found.)
Table 2.4.2.1.1.1.1.1.1 Engineering parameter information
Site Name Indicates the site name Yes

and is the unique
identifier for a cell.
Import this parameter and

the parameter can be set to
any value within the
reasonable range during PCI
planning.
Transceiver Name Indicates the transmitter No

name and is the unique
identifier for a
transmitter.
Active Indicates whether to No

active the serving cell.
Frequency Band Indicates band Yes
information.

Ltd..
PCI Indicates the physical ID No
of a cell.
Longitude Indicates the site No
longitude.
Latitude Indicates the site No
latitude.
Azimuth Indicates the antenna Yes
azimuth (degree). The
value range is from 0 to
360.
Antenna Indicates the antenna No
type.
Height Indicates the antenna No
height (m).
antenna.

antenna.
Total Electrical Indicates the total No
Downtilt electrical downtilt of an
antenna.
PA Indicates the offset of No
the PDCCH RE transmit
power (dB) relative to
the RS RE transmit
power. The value range
is from –15 to 15.
transmit power of a
transmitter.

Ltd..
Scene Indicates the cell No
scenario.
comments.
Dx Indicates the horizontal No
offset (m) of an antenna
to the site.
Dy Indicates the vertical No
to the site.
IsSFNCell Indicates whether a cell Mandatory for an
is on an SFN. If this SFN cell
parameter is left empty,
or is set to 0, the cell is
regarded as a common
cell by default.
However, the cell will be
regarded as an SFN cell
only when the value of
this parameter is 1.
SFNGroupId Indicates the SFN group Mandatory for an
ID. Cells with the same SFN cell
SFN group ID are
regarded as an SFN
group.
Main Propagation Indicates the main No
Model propagation model.
Main Calculation Indicates the calculation No
Radius radius of the main
propagation model.
propagation model.
TMA Indicates TMA No

Ltd..
equipment.
Feeder Indicates feeder No
equipment.
Feeder Length(DL) Indicates the downlink No
feeder length (m).
Miscellaneous Tx Indicates the downlink No
Loss(DL) miscellaneous loss (dB).
Jump Loss Ant- Indicates the downlink No
TMA(DL) jumper loss (dB) from
the TMA to the antenna
port.
Jump Loss TMA- Indicates the downlink No
BS(DL) jumper loss (dB) from
the TMA to the cabinet
top.
calculating the initial
total loss.
 Selected: Indicates that
the user enters the
total loss.
 Deselected: Indicates
that the program
calculates the total
loss.

jumper, and
miscellaneous loss.
BS(DL) jumper loss (dB) from
the antenna to the
cabinet top.
Number of Indicates the number of Yes
Transmission Antenna transmission antenna
Ports ports.
Transmission Antennas transmission antennas of
an eNodeB.

Ltd..
 If this parameter is left

empty, the tool will
automatically combines
the values of eNodeB ID
and Local Cell ID to
generate the value of this
parameter.
 If the value of this
parameter is set, the tool
will directly import the
value.
Transmission Mode Indicates the No

transmission mode.
Is MOCN Indicates whether a cell No
is on an MOCN.
Operator Indicates the operator No
name.
TTI Bundling Indicates whether to Yes
perform TTI bundling.
This function is disabled
by default.
Inter-site CA Indicates whether to Yes
enable inter-site CA.
Load Balance Indicates whether to No
enable load balancing in
simulation capacity. This
default.
VMIMO. This function
PSIC Indicates whether No
parallel signal
interference cancellation
must be considered.
PSIC is suitable for the
VMIMO scenario of two
RX antennas. Use PSIC
to bring gains by
eliminating interference
between UEs. This
default.

Ltd..
PUCCH IRC Indicates whether to No

enable the PUCCH IRC
function. When colored
interference is strong,
the PUCCH IRC
function can be enabled
to suppress interference
power of combined
signals and therefore
improve uplink gains.
by default.
PRACH IRC Indicates whether to No
enable the PRACH IRC
the PRACH IRC
power of combined
by default.
PUSCH IRC Indicates whether to No
enable the PUSCH IRC
the PUSCH IRC
power of combined
by default.
COMP Indicates whether to No
enable intra-site macro
diversity. Enable uplink
CoMP function on the
eNodeB side of the live
network to improve cell
default.
Channel Relativity Indicates whether No

Ltd..
channel relativity is
considered. This
default.
PDSCH Actual Indicates the actual No
Load(DL) downlink load. The
1.
Actual Load(UL) Indicates the actual No
uplink load. The value
range is from 0 to 1.
specifies the country
where a mobile
subscriber belongs to,
China is 460. The value
is a 3-bit string of 0 to 9
digits.

specifies the PLMN to
which a mobile
subscriber belongs. The
value is a 2- or 3-bit
UlEarfcn Indicates the uplink No
Automatically match
frequency bands based
on frequencies.
Local Cell ID Indicates the internal No
code of a cell for
differentiating the cell
from other cells under
the same eNodeB.
 LowSpeed
 HighSpeed

Ltd..
 HighwaySpeed

Min Root Sequence Indicates the minimum No
Index ZC sequence of a cell.
Preamble Format Indicates preamble No
sequence that is used for
a user to randomly
access a cell.
index for a cell.
downlink CA is
supported. If the
value is set to Off,
downlink CA is not
supported.
 If the value is set to
CA2CC, downlink
2CC CA is supported.
CA3CC, downlink
Is Double Deck Indicates a double deck No

cell ID.
True, the cell is a
double deck cell.
False, the cell is not a
double deck cell.
Cell Double Deck Indicates non-standard No

Bandwidth bandwidths (MHz) for a
double deck cell.
2.6
4
5.7
7.3
7.6
10.8
12.2

Ltd..
12.5
17.1
17.4
27.6
28.5
GSM Interference Indicates GSM No
Frequencies interference frequencies.
Only configure this
parameter to LTE FDD.
Multi-user Indicates whether to No
Beamforming enable the multi-user
beamforming function
for downlink scheduling.
Only configure this
Prach Density Per Indicates the PRACH No
10ms density of a cell. Only
configure this parameter
to LTE TDD.
Offset Frequency Indicates frequency No
Domain domain offset. Only
to LTE TDD.
DwPTS-GP-UpPTS Indicates special No
subframe configuration
proportion. Only
to LTE TDD.
downlink subframe
configuration
proportion. Only
to LTE TDD.
It includes uplink,
downlink, and special
proportion.
DSUUUDSUUU, D is
for downlink subframe,
S is for special
subframe, and U is for

Ltd..
uplink subframe.
Table 2.4.2.1.1.1.1.1.2 Parameters involved in the PRACH planning policy

Name Value
Available Root Indicates the N/A This parameter is
Sequence logical number set based on actual
Index for an scenarios. Different
available PRACH ZC root
PRACH ZC sequences are
root sequence. allocated to indoor
and outdoor cells.
Cells under
different boundary
areas use different
PRACH ZC root
sequences.
Planning Indicates N/A If this parameter is
Prach Based whether the selected in a
on existed PRACH ZC capacity expansion
Prach root sequences scenario, the
allocated to PRACH ZC root
planned cells sequences allocated
remain the to existing cells
same. remain the same,
and PRACH ZC
root sequence
planning only is
performed for new
cells.
Calculate Cell Indicates N/A If this parameter is
Radious whether the E- selected, the E-
UTRAN cell UTRAN cell radius
radius must be is calculated. If no,
calculated. the cell radius
specified in
CellTable is used.
Propagation Indicates N/A This parameter is
Radious whether the E- selected only when
UTRAN cell Calculate Cell
radius can be Radius is selected.
calculated
based on the
propagation
Ltd..

Name Value
model radius.
Propagation Indicates N/A If Calculate Cell
Radious reference Radius and
Factor factors when Propagation
the E-UTRAN Radius are
cell radius can selected, the E-
be calculated UTRAN cell radius
based on the is the propagation
propagation model radius of the
model radius. E-UTRAN cell
multiplied by the
value of Radius
Factor during
PRACH ZC root
sequence planning.
Coverage Indicates N/A This parameter is
Radious whether the E- selected only when
UTRAN cell Calculate Cell
radius can be Radius is selected.
calculated
based on cell
coverage
prediction.
Resolution(m) Indicates 50 m This parameter
coverage must be specified
prediction when the E-
resolution. UTRAN cell radius
can be calculated
based on cell
coverage
prediction.
Min Signal Indicates the –120 dBm This parameter
Level(dBm) RSRP must be specified
threshold of an when the E-
E-UTRAN UTRAN cell radius
cell during can be calculated
coverage based on cell
prediction. coverage
prediction.
Shadowing Indicates that Do not select This parameter
taken into shadow fading Shadowing must be specified
account is considered. taken into when the E-
account. UTRAN cell radius
can be calculated
based on cell
coverage
Ltd..

Name Value
prediction.
Cell Edge Indicates the 75% This parameter
Coverage cell edge must be specified
Probability coverage when the E-
probability of UTRAN cell radius
shadow can be calculated
fading. based on coverage
prediction, and
Shadowing taken
into account is
selected.
The shadow fading
variance of the
cluster must be
specified. For
details, see U-Net
Operation Guide.
Indoor Indicates that N/A This parameter
Coverage penetration must be specified
loss is when the E-
considered. UTRAN cell radius
can be calculated
based on cell
coverage
prediction.
The penetration
loss of the cluster
must be specified.
For details, see U-
Net Operation
Guide.
Area Indicates the N/A If a user does not
PRACH ZC draw a polygon, the
root sequence full map is used for
planning area. planning.
If a user draws
polygons on the
map, planning can
be performed in the
areas in polygons.
 Procedures

Ltd..

information, cells to be planned, and available PRACH ZC root sequences for
these cells.
Step 2 Prepare the engineering parameter table file based on the SmartRNO
system.
If the E-UTRAN cell coverage radius is calculated based on coverage prediction,

import an electronic map to obtain a more accurate planning result.
If neighbor relationship is not imported, ZC sequence conflicts among neighboring

cells are not considered during PRACH ZC root sequence planning.
Step 6 Set PRACH ZC root sequence auto-planning parameters to perform PRACH

ZC root sequence planning.
Step 7 Evaluate the planning result and manually adjust the result slightly. Exported
results are shown in Figure 2.4.2.1.1.7.1.1.
----End
Figure 2.4.2.1.1.7.1.1 Checking PRACH ZC root sequence conflicts

Ltd..
Details about exported parameter are shown in Table 2.4.2.1.1.7.1.1.1.
Table 2.4.2.1.1.7.1.1.1 Parameter details
Parameter Name Meaning Unit Remarks

CellName Indicates the E- N/A N/A
UTRAN cell
name.
Ncs Indicates the N/A N/A
NCS of an E-
UTRAN cell.
CellRadious Indicates the E- m During
UTRAN cell PRACH ZC
radius. root sequence
planning, the
E-UTRAN cell
radius can be
calculated
based on
coverage
prediction or
propagation
model radius.
StartRootSequecnc Indicates the N/A N/A
e start logical
number of the
ZC root
sequence.
EndRootSequence Indicates the N/A Optional
end logical
number of the
ZC root
sequence.
Reuse Distance Indicates the m N/A
reuse distance
of ZC root
sequences.
Reuse Indicates the N/A The number of
Tier(Neighbor) number of reuse tiers of
reuse tiers of ZC root
ZC root sequences is
sequences. the number of
tiers for
neighboring
cells.

Ltd..
PRACH ZC Root Sequence Check

 PRACH ZC root sequence check
a. The planning result must be checked and evaluated after
PRACH ZC planning is complete.
b. PRACH ZC sequence on the live network must be checked and
evaluated during network maintenance.
− The following information must be checked and evaluated:
a. PRACH ZC root sequence reuse isolation (including the reuse
distance and reuse tier)
b. Cells with potential ZC root sequence conflicts
a. Evaluation based on PRACH ZC root sequence result table
Summarizes, filters, and records PRACH ZC root sequence
reuse information about E-UTRAN cells with small PCI reuse
isolation, such as the small reuse distance and the small number
of reuse tiers, facilitating result backup and query.
b. Evaluation based on geographic presentation
Directly views PRACH ZC root sequence reuse information.
c. Evaluation based on CDF statistics
Quantitatively evaluates PRACH ZC root sequence reuse
isolation in E-UTRAN cells, especially in swap or PRACH ZC
root sequence replanning scenarios, this method quantitatively
compares PRACH ZC root sequence reuse before and after swap
or PCI replanning.
− Tool
a. Tool name: SmartRNO
b. Tool version: V300R001
c. R&D engineer: Wu Li (employee ID: 00228362)
d. Available at:
− Input
a. Evaluation and check after PRACH ZC root sequence planning
is complete
The evaluation and check after PRACH ZC root sequence
planning is complete require no parameter settings. You can
directly submit the planning result.
b. Evaluation and check during network maintenance

Ltd..
Collect NE parameters on the network, as shown in Table

2.4.2.1.1.7.1.1.2.
Table 2.4.2.1.1.7.1.1.2 Parameters involved in the PRACH planning policy
Parameter Name Meaning Mandatory

or not
Site Name Indicates the site name Yes
and is the unique
identifier for a cell.
Import this parameter

and the parameter can be
set to any value within
the reasonable range
identifier for a
transmitter.
Active Indicates whether to No
active the serving cell.
Frequency Band Indicates band Yes
information.
PCI Indicates the physical ID No
of a cell.
Longitude Indicates the site No
longitude.

Ltd..

or not
360.
Antenna Indicates the antenna No
type.
Height Indicates the antenna No
height (m).
antenna.
antenna.
Total Electrical Downtilt Indicates the total No
electrical downtilt of an
antenna.
PDCCH RE transmit
power (dB) relative to
the RS RE transmit
power. The value range
is from –15 to 15.
transmit power of a
transmitter.
Scene Indicates the cell No
scenario.
comments.
to the site.
Dy Indicates the vertical No
to the site.

Ltd..

or not
IsSFNCell Indicates whether a cell Mandatory
is on an SFN. If this for an SFN
parameter is left empty, cell
or is set to 0, the cell is
regarded as a common
cell by default. However,
the cell will be regarded
as an SFN cell only when
the value of this
parameter is 1.
SFNGroupId Indicates the SFN group Mandatory
ID. Cells with the same for an SFN
SFN group ID are cell
regarded as an SFN
group.
propagation model.
radius of the main
propagation model.
propagation model.
TMA Indicates TMA No
equipment.
Feeder Indicates feeder No
equipment.
feeder length (m).
jumper loss (dB) from
the TMA to the antenna
port.
the TMA to the cabinet
top.

Ltd..

or not
calculating the initial
total loss.
the user enters the
total loss.
 Deselected: Indicates
that the program
calculates the total
loss.
jumper, and
miscellaneous loss.
the antenna to the cabinet
top.
Antenna Ports transmission antenna
ports.
Antennas transmission antennas of
an eNodeB.

automatically
combines the values
of eNodeB ID and
Local Cell ID to
generate the value of
this parameter.
 If the value of this
parameter is set, the
tool will directly
import the value.
Transmission Mode Indicates the No
transmission mode.
Is MOCN Indicates whether a cell No
is on an MOCN.

Ltd..

or not
name.
perform TTI bundling.
by default.
Inter-site CA Indicates whether to Yes
enable inter-site CA.
Load Balance Indicates whether to No
enable load balancing in
default.
VMIMO. This function
PSIC Indicates whether No
parallel signal
must be considered.
PSIC is suitable for the
RX antennas. Use PSIC
to bring gains by
eliminating interference
between UEs. This
default.
PUCCH IRC Indicates whether to No
enable the PUCCH IRC
PUCCH IRC function
can be enabled to
suppress interference
power of combined
by default.
PRACH IRC Indicates whether to No
enable the PRACH IRC
PRACH IRC function

Ltd..

or not
can be enabled to
power of combined
by default.
PUSCH IRC Indicates whether to No
enable the PUSCH IRC
PUSCH IRC function
can be enabled to
power of combined
by default.
COMP Indicates whether to No
enable intra-site macro
diversity. Enable uplink
CoMP function on the
default.
Channel Relativity Indicates whether No
channel relativity is
considered. This function
PDSCH Actual Load(DL) Indicates the actual No
downlink load. The value
Actual Load(UL) Indicates the actual No
uplink load. The value
specifies the country
where a mobile
subscriber belongs to, for
example, the MCC of
China is 460. The value

Ltd..

or not
digits.

specifies the PLMN to
which a mobile
subscriber belongs. The
value is a 2- or 3-bit
Automatically match
frequency bands based
on frequencies.
Local Cell ID Indicates the internal No
code of a cell for
differentiating the cell
from other cells under
the same eNodeB.
 LowSpeed
 HighSpeed
 HighwaySpeed
Min Root Sequence Index Indicates the minimum No
ZC sequence of a cell.
sequence that is used for
a user to randomly access
a cell.
index for a cell.
downlink CA is
supported. If the
value is set to Off,
downlink CA is not
supported.

Ltd..

or not
CA2CC, downlink
CA3CC, downlink
Is Double Deck Indicates a double deck No
cell ID.
True, the cell is a
double deck cell.
double deck cell.
double deck cell.
2.6
4
5.7
7.3
7.6
10.8
12.2
12.5
17.1
17.4
27.6
28.5
GSM Interference Indicates GSM No
Frequencies interference frequencies.
Only configure this
parameter to LTE FDD.
Multi-user Beamforming Indicates whether to No
enable the multi-user
beamforming function
for downlink scheduling.
Only configure this
Prach Density Per 10ms Indicates the PRACH No
density of a cell. Only
Ltd..

or not
to LTE TDD.
Offset Frequency Domain Indicates frequency No
domain offset. Only
to LTE TDD.
DwPTS-GP-UpPTS Indicates special No
proportion. Only
to LTE TDD.
downlink subframe
Only configure this
It includes uplink,
proportion.
DSUUUDSUUU, D is
for downlink subframe, S
is for special subframe,
and U is for uplink
subframe.
 Procedures
Step 1 If PRACH ZC root sequence self-planning based on the SmartRNO is
complete, check the PRACH ZC root sequence based on the planning result.
During network maintenance, you need to make an engineering parameter
file, create a project, and import the engineering parameters before checking
the PRACH ZC root sequence.
Step 8 Check the PRACH ZC root sequence reuse isolation (including the reuse
distance and reuse tier) to find the cells with potential PRACH ZC root
sequence conflicts.
----End

Ltd..
PRACH ZC Root Sequences Reuse Isolation Evaluation
Application Scenario
In swap and PRACH ZC root sequence replanning scenarios, the PRACH ZC
root sequence reuse isolation evaluation is used to quantitatively compare the
results before and after the replanning.
In site deployment scenarios, the PRACH ZC root sequence reuse isolation
evaluation is used to quantitatively analyze the PRACH ZC root sequence
planning on the network and determine whether the planning result meets the
actual requirement (for example, the PRACH ZC root sequence reuse
distances of 98% E-UTRAN cells exceeds 10 km).
Evaluation Methods
 Minimum reuse distance of E-UTRAN cells on the network in CDF
statistics
Export or copy the values of Reuse Distance of each cell in the PRACH ZC
root sequence planning result table to an EXCEL file, collect the CDF
statistics of each reuse distance, and generate a line graph. Line graph of
PRACH ZC root sequence reuse distances
Initial
configuration
After PRACH
ZC root
sequence
planning
 Minimum number of E-UTRAN cell reuse tiers on the network in CDF

statistics
Export or copy the values of Reuse Tier(Neighbor) of each cell in the
PRACH ZC root sequence planning result table to an EXCEL file, collect the
CDF statistics of reuse tiers, and generate a line graph shown in .
 Geographical display of the PRACH ZC root sequence reuse of each cell
on the network.

Ltd..
Click an E-UTRAN cell on the network to view its intra-frequency E-UTRAN

cells with the same PRACH ZC root sequence. As shown in Figure
2.4.2.1.1.9.1.1, cells in red are the intra-frequency cells with the same
PRACH ZC root sequence.
Figure 2.4.2.1.1.9.1.1 Cells with the same PRACH ZC root sequence
PRACH ZC Root Sequence Conflict Checking

PRACH ZC root sequence conflict checking aims to check whether there are
E-UTRAN cells allocated incorrect PRACH ZC root sequences on the
network and ensures that the PRACH ZC root sequence reuse distance and
reuse tier for all cells meet the requirements.
The PRACH ZC root sequence planning result table is exported after PRACH
ZC root sequence planning is complete. In addition, you can also import the
network engineering parameters, right-click Parch Planning, and then choose
Open Prach Parameter in the displayed dialog box to view the PRACH ZC
root sequence planning result table.
Figure 2.4.2.1.1.9.1.2 PRACH ZC root sequence planning result table

Ltd..
After the PRACH ZC root sequence planning result table is exported, find the
E-UTRAN cells whose PRACH ZC root sequence reuse isolation does not
meet the requirements. For example, the number of neighboring cell tiers is
less than or equal to 1 (you can set the tier to a value greater than 1 based on
actual conditions), or the PRACH ZC root sequence reuse distance is small (4
km for common urban areas and 10 km for suburban areas). The area
encircled in red in Figure 2.4.2.1.1.9.1.3 displays the cells with potential
PRACH ZC root sequence conflicts, and the number of reuse tiers for these
cells is 1.
Figure 2.4.2.1.1.9.1.3 PRACH ZC Root Sequence Conflict Checking
Available at: to be provided
PRACH ZC Root Sequence Reallocation

 Overview
 Use the SmartRNO to re-allocate proper PRACH ZC root sequences for
cells that do not meet the requirement if any of the following conditions
occurs:
− Intra-frequency cells that are close to one another use the same
PRACH ZC root sequence
− Cells of the same tier use the same PRACH ZC root sequence
 Tool
− Available at:
 Input
− Engineering parameter information
− E-UTRAN cells requiring PRACH ZC root sequence replanning
 Procedures
information, cells to be planned, available PRACH ZC root sequences for
these cells, and cells requiring PRACH ZC root sequence replanning.
Step 2 Prepare the engineering parameter table based on the engineering parameter
table of the U-Net and set Min Root Sequence for cells to be planned to null.

Ltd..

If the electronic map has a coordinate system, skip this step.
If the E-UTRAN cell radius can be calculated based on the coverage prediction, it is
recommended that the electronic map be imported to obtain a more accurate planning
result.
Step 12 Import engineering parameter files of NEs.
If neighbor relationship is not imported, PRACH ZC root sequence conflicts among

neighboring cells are not considered during PRACH ZC root sequence planning.
Step 13 Select Planning Prach Based on existed Prach, perform PRACH ZC root
sequence self-planning, and submit the planning result.
Step 14 Evaluate the planning result and manually and slightly adjust the planning
result.
----End
Figure 2.4.2.1.1.14.1.1 PRACH ZC Root Sequence Conflict Checking
The following table provides the detailed descriptions of output items.

Ltd..
Table 2.4.2.1.1.14.1.1.1 Detailed descriptions of output items
Parameter Name Meaning Unit Remarks
CellName Indicates the E- N/A

UTRAN cell name.
Ncs Indicates the NCS of N/A

an E-UTRAN cell.
CellRadious Indicates the radius m During PRACH ZC root

of an E-UTRAN sequence planning, the E-
cell. UTRAN cell radius can be
calculated based on coverage
prediction or propagation model
radius.
StartRootSequecnc Indicates the start N/A

e logical ZC root
sequence number.
EndRootSequence Indicates the end N/A This parameter is optional.

logical ZC root
sequence number.
Reuse Distance Indicates the reuse m

distance of ZC root
sequences.
Reuse Indicates the reuse N/A The reuse tier of ZC

Tier(Neighbor) tier of ZC root root sequences is the
sequences. number of tiers for
neighboring cells.

Ltd..
Design (PCI, PRACH, Neighboring Cell, and TA&TAL) 4 TA&TAL Planning
3 LTE Neighboring Cell Planning
3.1 Overview
This chapter describes the LTE neighboring cell planning, including the initial
neighboring cell planning on new networks and the neighboring cells
replanning during the network capacity expansion.
This document is intended for Huawei NIS engineers and engineers dedicated
for professional services.

Planning neighboring cells is mandatory at the initial phase of network
deployment. The planning result determines network performance.
Traditionally, neighboring cells are manually planned and time consuming.
Currently, neighboring cells are automatically or semi-automatically planned,
which greatly improves work efficiency, reduces network construction cost,
and accelerates network construction.
The U-Net/SmartRNO provides the function of automatically planning
neighboring cells. It supports neighboring cell planning for special scenarios
where repeaters, remote RF units, or SFNs are deployed and provides reliable
planning results.
3.1.2 Introduction
This section provides four solutions to neighboring cells planning:
 Neighboring cell planning based on live network data
 Neighboring cell planning based on dedicated tools
 Neighboring cell planning based on preplanning software
 Automatic neighboring cell planning
Dedicated tools for neighboring cell planning are not involved in this
document temporarily.
3.1.3 Application Restriction

Specification Restriction
N/A
Application Scenario Restriction

Table 3.1.3.1.1.1.1.1.1 lists the application scenario restrictions for the
neighboring cell planning.
Table 3.1.3.1.1.1.1.1.1 Application scenario restrictions for neighboring cell planning
Scenario Restrictions Workaround
The latitude and longitude If the latitude and longitude are not Provide the
of an NE are not specified. specified, the neighboring cell planning, latitude and
check, or re-allocation cannot be longitude of the
performed. NE.
The neighboring cells need Currently, the U-Net/SmartRNO does The Wireless
to be planned for indoor not support modeling for indoor Parameter Auto-
sites. scenarios, including floor information, configuration
room information, and whether the feature can be
indoor cells belong to the same building. used to complete
Due to lack of key information, the U- the neighboring
Net/SmartRNO is not proper for planning online
neighboring cell planning for indoor for small cells.
cells.
3.1.4 Maturity
The neighboring cell planning supports the following scenarios:
 Capacity expansion for greenfields and mixed networking
The neighboring cell planning supports the following network modes:
 Inter-frequency networking
 Multi-antenna, remote radio frequency, and SFN
Net V300R010/SmartRNO V300R001 supports the preceding neighboring
cell planning contents.
3.2 Principles
To perform the ANR function, both the LTE network and the UE must support
this function. The effect of the ANR function is obvious when the traffic is
heavy. However, most UEs do not support ANR, and there are a few UEs at
the initial deployment stage. Therefore, the initial neighboring cell

configuration must be performed by field engineers. Compared with other
radio access technologies, LTE performs handover measurements based on
frequencies instead of neighboring cell lists. The UE locates the cell based on
the frequencies indicated in the configuration measurement. The upper UE
processes the measurement result and obtains the list of candidate handover
cells, and then sends the list to the network. The network selects the cell and
initiates the handover. A neighboring cell list provides information required by
handovers, such as CGI. Configure as many neighboring cells as possible for
an eNodeB because the number of neighboring cells does not affect
measurement time or accuracy. Specifically, obey the following basic
principles for LTE neighboring cell planning:
 Cells geographically adjacent to each other must be configured as
neighboring cells.
 Two cells are neighboring cells of each other. For example, if sector B is
the neighboring cell of A sector, sector A is also the neighboring cell of sector
B. However, unidirectional handovers are required in certain scenarios, such
as high-speed coverage scenarios. For example, if a UE expects to be handed
over from sector A to sector B, but not from sector B to sector A, add sector A
into the blacklist of sector B. In densely-populated urban areas and common
urban areas, the inter-site distance (ISD) is 0.3 km to 1.0 km. Therefore, plan
more neighboring cells for each cell. Currently, a maximum of 256 intra-
frequency neighboring cells and 256 inter-frequency neighboring cells can be
configured. (Note: Although a maximum 256 intra-frequency and 256 inter-
frequency neighboring cells can be configured, the total number of intra-
frequency and inter-frequency neighboring cells cannot be greater than 256.
That is, the intra-frequency and inter-frequency cells flexibly share 256
neighboring cells.)In terms of inter-RAT neighboring cells, a maximum of 128
UTRAN neighboring cells and a maximum of 128 GERAN neighboring cells
can be configured. Therefore, in the neighboring cell configuration, configure
only the cells neighboring to a cell to use the limited resource efficiently.
 In suburban areas where sites are far from each other, configure
geographically adjacent cells as neighboring cells to ensure timely handovers.
 The LTE neighboring cells are not in a sequential order, the neighboring
cell detection is independent from the number of neighboring cells
configured, and the measurement period is short. Therefore, all neighboring
cells must be included, and need not be arranged according to the signal
strength.
 Considering the waveguide effect, for a sector facing a lake, sea, or a
wide road, if its adjacent cell is also in front of the lake, sea, or a wide road,
the adjacent cell is configured as its neighboring cell.
3.2.1 Intra-frequency Neighboring Cell Planning

Intra-frequency neighboring cells are used for handover in the LTE system.
When an eNodeB initially sets up RB services, it preferentially includes intra-
frequency measurement configuration in the RRC Connection Reconfig
message and preferentially initiates an intra-frequency handover. In actual
planning, possible intra-frequency neighboring cells can be planned into the
neighboring cell list. After initialization, the system optimizes the neighboring
cell relationship based on the UE measurement.
3.2.2 Inter-frequency Neighboring Cell Planning

In an LTE system, different frequencies are used for capacity expansion.
Therefore, inter-frequency neighboring cells must be configured.
In hierarchical cells, if inter-frequency configurations are used, macro cells
must be configured as the inter-frequency neighboring cells of micro cells.
In multi-frequency networking, possible inter-frequency neighboring cells can
be planned into the neighboring cell list. After initialization, the system
optimizes the neighbor relationship based on the inter-frequency neighboring
cells detected by the UE.
Hierarchical cells: In traffic hotspot areas, micro cells or micro-micro

cells are added to macro cells to serve more subscribers by taking
advantage of geographical areas. UEs out of micro cells or micro-micro
cells can reselect larger cells.
3.2.3 Inter-RAT Neighboring Cell Planning

In the early stage of LTE network deployment or during LTE network
deployment, the WCDMA system may coexist with the GSM, CDMA2000, or
TD-SCDMA system. Therefore, the E-UTRAN cell must be configured with
inter-RAT neighboring cells. In general, inter-RAT cells adjacent to E-
UTRAN cells are configured as inter-RAT neighboring cells. If handovers to
inter-RAT neighboring cells are not allowed, the inter-RAT neighboring cells
can be added to the blacklist. It is recommended that these cells be configured
as inter-RAT neighboring cells. By configuring different handover parameters,
continuous coverage and flexible handovers are implemented.
If the indoor areas or subways are covered only by inter-RAT neighboring
cells in the early stage of LTE network deployment, these cells must be
configured as neighboring cells of corresponding E-UTRAN to ensure
continuous coverage.
3.3 Application Guidance

3.3.1 Neighboring Cell Planning Using U-Net
 Overview
In the site deployment and capacity expansion scenarios, plan
neighboring cells for new E-UTRAN cells.
During neighboring cell planning, collect the required data for the
planning, prepare the engineering parameter table, and then use the U-
Net to plan neighboring cells.
 Tool
− Available at:
8132371/PBI1-21395144/PBI1-13953
 Input
− Network engineering parameter information about NEs (For details,
see Table 3.3.1.1.1.1.1.1.1 to Table 3.3.1.1.1.1.1.1.3.)
Parameter Name Meaning Recommended Value Remarks
Site Name Indicates the site name. / The U-Net uses

the site name as
the search
keyword.
Longitude Indicates the site / /

longitude.
Latitude Indicates the site / /

latitude.
Altitude Indicates the site / This parameter is

altitude. used when
interference from
neighboring cells
is calculated
based on
coverage
prediction.
Transceiver Indicates the sector name. User-defined /

Name
Dx Indicates the / If the E-UTRAN cell is

horizontal offset of served by multiple
the antenna to the antennas, specify the
eNodeB. horizontal offset of
each antenna
corresponding to the
eNodeB.
Dy Indicates the vertical / If the E-

offset of the antenna UTRAN cell is
to the eNodeB. served by
multiple
antennas,
specify the
vertical offset
of each antenna
corresponding to the
eNodeB.
Antenna Indicates the antenna / If the E-UTRAN cell is

type. served by multiple
antennas, specify the
type of each antenna.
Height Indicates the antenna / If the E-UTRAN cell is

height. served by multiple
height of each
antenna.
This parameter
is used when
interference
from
neighboring
cells is
calculated
based on
coverage
prediction.
Azimuth Indicates the antenna / If the E-UTRAN cell is

azimuth. served by multiple
azimuth of each
antenna.
Mechanical Indicates the antenna / If the E-UTRAN cell is

Down tilt mechanical downtilt. served by multiple
mechanical downtilt
of each antenna.
This parameter
is used when
interference
from
neighboring
cells is
calculated
based on
coverage
prediction.
Electrical Indicates the antenna / If the E-UTRAN cell is

Down tilt electrical downtilt. served by multiple
electrical downtilt of
each antenna.
This parameter
is used when
interference
from
neighboring
cells is
calculated
based on
coverage
prediction.
Power Ratio Indicates the antenna / If the E-UTRAN cell is

power ratio. served by multiple
power ratio of each
antenna.
This parameter
is used when
interference
from
neighboring
cells is
calculated
based on
coverage
prediction.
Number of Indicates the number of / /

Transmissio transmission antenna
n Antenna ports.
Ports
Cell Name Indicates the / The U-Net uses the

cell name. cell name as the
search keyword.
Frequency Indicates the / If necessary, the

Band attribute of frequency and
the cell bandwidth need to
frequency be specified. If the
band. actual bandwidth is
lower than the
standard bandwidth,
the actual bandwidth
is compact
bandwidth.
RS Power Indicates the 15.2 dBm This parameter is

transmit used when
power of the interference from
RS. neighboring cells is
calculated based on
Main Indicates the / This parameter is used when

Propagation propagation interference from
Model model. neighboring cells is
calculated based on
Main Indicates the / This parameter is used when

Calculation calculation interference from
Radius radius of the neighboring cells is
path loss calculated based on
matrix. coverage prediction.
PCI Indicates the / If this parameter is null, the

existing PCI cell is a new cell.
of an E-
UTRAN cell.
PCI Reuse Indicates the This parameter is set On a live network,

Distance(Km) PCI reuse according to the actual some LTE eNodeBs
distance. scenario. For example: are deployed at the
In common urban areas, same site where
this parameter is set to 4. GSM BTSs and
UMTS NodeBs are
In suburban areas, this deployed. Therefore,
parameter is set to 10. you can set this
parameter based on
the reuse distance of
PSC on the UMTS
network.
PCI Reuse Indicates the 2 /

Tier(Neighbor) PCI reuse
tier.
 Procedures
Step 1 Collect live network information, including the engineering parameter
information and cells to be planned.
Step 2 Prepare the engineering parameter table file according to the U-Net

transceiver table, and cell table.
Step 6 Set neighboring cell planning parameters, perform neighboring cell planning,
----End
Parameter setting interface of neighboring cell planning provides three tag

pages, including General, Intra-Frequency, and Inter-Frequency tag pages.
Figure 3.3.1.1.1.6.1.1 to Figure 3.3.1.1.1.6.1.3 show these tab pages. Table
3.3.1.1.1.6.1.1.1 to Table 3.3.1.1.1.6.1.3.1 list descriptions of parameters
involved in these tab pages.
Figure 3.3.1.1.1.6.1.1 Neighboring cell planning setting interface – General tag

page
Table 3.3.1.1.1.6.1.1.1 Descriptions of parameters in General tab page
Parameter Name Description
Methods Select Indicates the networking planning method.

Topology: Plans neighboring cells based on

network topology.
Prediction: Plans neighboring cells based on
prediction results. This method is applicable
only to outdoor base stations.
Topology + Prediction: Plans neighboring
cells based on both the network topology and
the prediction results.
If repeaters are configured, neighboring

cell planning based on predictions is not
supported.
Max Neighbor Indicates the maximum distance between

Distance(km) neighboring cells. If the distance between two
neighboring cells exceeds the parameter value,
these two cells cannot be planned as
neighboring cells.
Planning Neighbor based Indicates whether to plan neighboring cells

on existed Neighbor based on existing neighbor relationships. If
this option is not selected, existing neighbor
relationships are deleted, and neighboring cells
are replanned.
Force Co-Site As Indicates whether to forcibly configure co-

Neighbor sited cells as bidirectional neighboring cells.
Co-Site Distance(m) Two cells are cofigured as bidirectional

neighboring cells when the distance between
the two cells is less than the value of this
parameter.
Reference Existed Indicates whether to reference the neighbor

Neighbors relationships of existing cells.
Reference Rules After clicking Reference Rules, users can

configure reference rules in the displayed
window.
Source Cell Frequency Band: Indicates the
frequency band of the cell to be planned.
Neighbor Cell Frequency Band: Indicates
the frequency band of the neighboring cell.
Referenced Cell Frequency Band: Indicates
the frequency band of the cell used for
reference.
Reference Neighbor Cell Frequency Band:
Indicates the frequency band of the
neighboring cell of the cell used for reference.
Frequency display mode:
Frequency band of a cell.
ARFCN of a cell. ARFCN of a UMTS

cell is displayed. In this case, users need
to configure the ARFCN attribute of the
cell.
DlEarfcn of a cell. DlEarfcn of an LTE

to configure the DlEarfcn attribute of the
cell.
Azimuth Difference(°) Indicates the azimuth difference between the

cell to be planned and the cell used for
reference.
Reference Site Indicates the distance difference between the

Distance(m) site of the cells to be planned and the site of
the cells used for reference.
Mapping Table Indicates the sector splitting mapping file.

Obtain a file template using Template and
import the mapping between cells before and
after sector splitting.
After this parameter is selected, you can click
Template to obtain the data template. You can
also click Browse to import a data file.
For details about the template, see GSM
Mapping Template, UMTS Mapping
Template, and LTE Mapping Template.
Consider Handover Indicates whether to consider the handover

Statistics data and the path for saving the handover data.
Handover Statistics Indicates the handover data.

For details about the template, see Handover
Statistics Template.
Planning Weight Indicates the weight of the planning result

upon neighboring cell ranking.
Handover Statistics Indicates the weight of the handover data upon

Weight neighboring cell ranking.
Auto Identification Indicates whether to automatically generate

Neighbor Planning Area cells to be planned.
The U-Net automatically selects new

cells and related cells as cells to be
planned.
Area Indicates a planning area.

You can select all cells in an area or click
Filter to select only the cells to be planned in

the area.
In the Filter dialog box, you can specify the
contents to be searched for and set the search
criteria such as search direction or whether the
uppercase and lowercase letters are
distinguished.
Figure 3.3.1.1.1.6.1.2 Neighboring cell planning setting interface – Intra-

Frequency tag page
Table 3.3.1.1.1.6.1.2.1 Descriptions of parameters in Intra-Frequency tab page
New Cell-Indoor Cell-Max Outdoor Indicates the maximum number

Neighbor Number of outdoor neighboring cells
planned for a newly deployed

indoor cell.
New Cell-Indoor Cell-Max Indoor Indicates the maximum number

Neighbor Number of indoor neighboring cells
indoor cell.
New Cell-Outdoor Cell-Max Outdoor Indicates the maximum number

outdoor cell.
New Cell-Outdoor Cell-Max Indoor Indicates the maximum number

outdoor cell.
Existed Cell-Indoor Cell-Max Outdoor Indicates the maximum number

planned for an existing indoor
cell.
Existed Cell-Indoor Cell-Max Indoor Indicates the maximum number

planned for an existing indoor
cell.
Existed Cell-Outdoor Cell-Max Indicates the maximum number

Outdoor Neighbor Number of outdoor neighboring cells
planned for an existing outdoor
cell.
Existed Cell-Outdoor Cell-Max Indoor Indicates the maximum number

planned for an existing outdoor
cell.
Figure 3.3.1.1.1.6.1.3 Neighboring cell planning setting interface – Inter-

Frequency tag page
Table 3.3.1.1.1.6.1.3.1 Descriptions of parameters in Inter-Frequency tab page
Total Max Number Indicates the maximum number of neighboring

cells.
Min Signal Level(dBm) Indicates the minimum signal receive level of
an inter-RAT neighboring cell.
Handover Indicates the handover area threshold.
Threshold(dB)
Frequency Band Indicates the frequency band to be handed over

when the RAT is switched to GSM.
ARFCN Indicates the ARFCN to be handed over when

the RAT is switched to UMTS.
 Output
Figure 3.3.1.1.1.6.1.4 shows neighboring cells of the cell to be planned.
Figure 3.3.1.1.1.6.1.4 Neighboring cell planning output
In this figure:
[Intra-Frequency]: Indicates intra-frequency neighboring cells.
[Inter-Frequency]: Indicates inter-frequency neighboring cells.
[Neighbor Name]: Indicates the name of a neighboring cell.
[Cause]: Indicates the reason for configuring a cell as the neighboring
cell of the serving cell.
"existed": Indicates that neighbor relationships exist on the network.
"planned": Indicates that neighbor relationships are planned.
"manual": Indicates that the neighbor relationship is manually
created by a user.
"inherited": Indicates that neighbor relationships are inherited.
[Confirm]: Indicates whether a cell is configured as the neighboring cell
of the serving cell.
If the option is selected, the cell is configured as the neighboring
Step 7 If the option is deselected, the cell is not configured as the neighboring cell of
the serving cell.
Step 8 Query the neighboring cell planning result.

After the neighboring cell planning, you can view the planning result on a
map.
---End
 Operation Procedure
Set the mode and color for displaying neighbor relationships in the map
window.
Figure 3.3.1.1.1.8.1.1 shows an operation example. Table 3.3.1.1.1.8.1.1.1 and
Table 3.3.1.1.1.8.1.1.2 list the parameter descriptions.
Figure 3.3.1.1.1.8.1.1 Setting the mode and colors for displaying neighbor
relationships
Display Links Identifies neighbor relationships by

using lines.
Display Cell Color Identifies neighbor relationships in

cell colors.
Fit Neighbor Cell Visible Displays the neighbor relationships

of a cell on the map after you select
the cell in the neighbor relationship
table.
That is, after you select the source
cell in the neighbor relationship
table, the neighbor relationships of a
cell are displayed on the map.
Display Deleted Existed Neighbor Displays the neighboring cells that

are available on the live network
but are currently deleted.
Display Uni-directional Neighbor Displays unidirectional neighbor
From Other Cell relationships from other cells to
this cell.
Select Neighbor type Sets the neighboring cell type to be
displayed.
Intra-Frequency Neighbors: Indicates
intra-frequency neighboring cells.
Inter-Frequency Neighbors: Indicates
inter-frequency neighboring cells.
Inter-RAT Neighbors: Indicates
inter-RAT neighboring cells.
Intra-Technology Neighbors:
Indicates intra-RAT neighboring
cells.
Table 3.3.1.1.1.8.1.1.2 Descriptions of parameters in Neighbor Display Color tab

page
Legend Description
Confirm Intra Frequency Indicates intra-frequency
unidirectional neighboring cells.
Confirm Inter Frequency Indicates inter-frequency

unidirectional neighboring cells.
Confirm Inter-RAT Indicates inter-RAT unidirectional

neighboring cells. This parameter
is available only for a multi-
mode network.
Confirm Intra Technology Indicates the intra-RAT neighbor
relationship.
Deleted Intra Frequency Indicates intra-frequency

neighboring cells that are deleted
from the live network.
Deleted Inter Frequency Indicates inter-frequency

neighboring cells that are deleted
Deleted Inter-RAT Indicates inter-RAT neighboring

cells that are deleted from the live
network. This parameter is
available only for a multi-mode
network.
Query the neighboring cell planning result.

----End
You need to select the RAT node. Otherwise, neighbor relationships

cannot be displayed in the map window.
Figure 3.3.1.1.1.8.1.2 shows an operation example. Table 3.3.1.1.1.8.1.2.1 and

Table 3.3.1.1.1.8.1.2.2 list the parameter descriptions.
Figure 3.3.1.1.1.8.1.2 Viewing the neighboring cell planning result
All cells on the network are displayed in the Cell Name area in the displayed window.
After the neighboring cell planning, if the planning result table window is
automatically displayed, cells used for planning calculation are displayed in the Cell
Name area.
Table 3.3.1.1.1.8.1.2.1 Tab page description
Intra-Frequency Indicates intra-frequency neighboring cells.
Inter-Frequency Indicates inter-frequency neighboring cells.
Inter-RAT Indicates inter-RAT neighboring cells. This tab

page is valid only for a multi-mode network.
Table 3.3.1.1.1.8.1.2.2 Parameter description
Neighbor Name Indicates the name of a neighboring cell.
Cause Indicates the reason for configuring a cell as the

neighboring cell of the serving cell.
"existed": Indicates that neighbor relationships
exist on the network.
"planned": Indicates that neighbor relationships
are planned.
"manual": Indicates that the neighbor relationship

is manually created by a user.
"inherited": Indicates that neighbor relationships
are inherited.
Confirm Indicates whether a cell is configured as the

If the option is selected, the cell is configured as
the neighboring cell of the serving cell.
If the option is deselected, the cell is not
configured as the neighboring cell of the serving
cell.
Available at:
actionFlag=detailProductSimple&web_doc_id=SC0000764790&doc_type=Pr
oductManual&doc_type=ProductManual&saveBrowserLoged=true
3.3.2 SmartRNO Neighboring Cell Planning

 Tool
− Available at:
 Input
− Engineering parameter information about NEs (For details, see Table
3.3.2.1.1.1.1.1.1.)
Table 3.3.2.1.1.1.1.1.1 Engineering parameter information
Parameter Name Description Mandatory or not
Site Name Indicates the site name and Yes

is the unique identifier for a
cell.
Import this parameter and

the parameter can be set to
any value within the
reasonable range during
PCI planning.

identifier for a transmitter.
the serving cell.
PCI Indicates the physical ID of No
a cell.
360.
(m).
antenna.
antenna.
Total Electrical Indicates the total electrical No
Downtilt downtilt of an antenna.
physical downlink control
channel (PDCCH) RE
transmit power (dB)
relative to the reference
signal (RS) RE transmit
power. The value range is
from –15 to 15.

transmit power of a
transmitter.
Transceiver Indicates transmitter No
Comments comments.
offset (m) of an antenna to
the site.
(m) of an antenna to the
site.
IsSFNCell Indicates whether a cell is Mandatory for an
on an SFN. If this SFN cell
parameter is left empty, or
is set to 0, the cell is
regarded as a common cell
by default. However, the
SFN cell only when the
value of this parameter is 1.
SFNGroupId Indicates the SFN group Mandatory for an
ID. Cells with the same SFN cell
SFN group ID are regarded
as an SFN group.
Main Propagation Indicates the main No
Model propagation model.
Main Calculation Indicates the calculation No
Radius radius of the main
propagation model.
propagation model.
TMA Indicates tower-mounted No
amplifier (TMA)
equipment.

feeder length (m).
Miscellaneous Tx Indicates the downlink No
Loss(DL) miscellaneous loss (dB).
TMA(DL) jumper loss (dB) from the
Jump Loss TMA- Indicates the downlink No
BS(DL) jumper loss (dB) from the
loss.
the user enters the total
loss.
 Deselected: Indicates that
the program calculates
the total loss.
loss.
BS(DL) jumper loss (dB) from the
Transmission Antenna transmission antenna ports.
Ports
Transmission transmission antennas of an
Antennas eNodeB.

automatically combines
the values of eNodeB
ID and Local Cell ID to
generate the value of
this parameter. If the
value of this parameter
is set, the tool will
directly import the
value.
mode.
Is MOCN Indicates whether a cell is No
on a multi-operator core
network (MOCN).
name.
perform transmission time
interval (TTI) bundling.
default.
inter-site carrier aggression
(CA).
load balancing in
default.
virtual multiple-input
multiple-output (VMIMO).
default.
signal interference
considered. Parallel soft
(PSIC) is suitable for the
receive antennas (RX
antennas). Use PSIC to
bring gains by eliminating
default.
interference rejection
combining (IRC) function
of physical uplink control
channel (PUCCH). When
colored interference is
strong, the PUCCH IRC
When colored interference
is strong, the PRACH IRC
When colored interference
is strong, the PUSCH IRC
Enable uplink coordinated
multipoint transmission
(CoMP) function on the
default.
relativity is considered.
default.
PDSCH Actual Indicates the actual No
Load(DL) downlink load. The value
load. The value range is
from 0 to 1.
MCC Indicates the mobile No
country code (MCC) that
specifies the country where
a mobile subscriber belongs
to, for example, the MCC
of China is 460. The value
digits.
MNC Indicates the mobile No

network code (MNC) that
specifies the public land
mobile network (PLMN) to
which a mobile subscriber
belongs. The value is a 2-
or 3-bit string of 0 to 9
digits.
Automatically match
frequencies.
Local Cell ID Indicates the internal code No
of a cell for differentiating
the cell from other cells
under the same eNodeB.
 LowSpeed
 HighSpeed
 HighwaySpeed
Min Root Sequence Indicates the minimum ZC No
Index sequence of a cell.
sequence that is used for a
user to randomly access a
cell.
index for a cell.
downlink CA is
supported.If the value is
set to Off, downlink CA
is not supported. If the
value is set to CA2CC,
downlink 2CC CA is
supported.
CA3CC, downlink 3CC
CA is supported.
ID.
True, the cell is a
double deck cell.
double deck cell.
Cell Double Deck Indicates non-standard No
Bandwidth bandwidths (MHz) for a
double deck cell.
2.6
4
5.7
7.3
7.6
10.8
12.2
12.5
17.1
17.4
27.6
28.5
GSM Interference Indicates GSM interference No
Frequencies frequencies. Only configure
this parameter to LTE
FDD.
Multi-user Indicates whether to enable No
Beamforming the multi-user beamforming
this parameter to LTE
TDD.
Prach Density Per Indicates the PRACH No
10ms density of a cell. Only
configure this parameter to
LTE TDD.
Offset Frequency Indicates frequency domain No
Domain offset. Only configure this
Only configure this
downlink subframe
Only configure this
It includes uplink,
proportion.
special subframe, and U is
for uplink subframe.
 Procedures
Step 1 Collect live network information, including the engineering parameter
information and cells to be planned.
Step 2 Prepare the engineering parameter table file based on the SmartRNO
Step 5 Import the engineering parameter file.

Step 6 Set neighboring cell planning parameters, perform neighboring cell planning,
----End
Parameter setting interface of neighboring cell planning provides three tag

pages, including General, Intra-Frequency, and Inter-Frequency tag pages.
Figure 3.3.2.1.1.6.1.1 to Error: Reference source not found show these tab
pages. Table 3.3.2.1.1.6.1.1.1 to Table 3.3.2.1.1.6.1.3.1 list descriptions of
parameters involved in these tab pages.

page
Methods Select Indicates the networking planning method.

Topology: Plans neighboring cells based on
network topology.
Prediction: Plans neighboring cells based on
prediction results. This method is applicable
only to outdoor base stations.
Topology + Prediction: Plans neighboring
cells based on both the network topology and
the prediction results.
If repeaters are configured, neighboring

cell planning based on predictions is not
supported.
Max Neighbor Indicates the maximum distance between

Distance(km) neighboring cells. If the distance between two
neighboring cells exceeds the parameter
value, these two cells cannot be planned as
neighboring cells.
Planning Neighbor based Indicates whether to plan neighboring cells

on existed Neighbor based on existing neighbor relationships. If
this option is not selected, existing neighbor
relationships are deleted, and neighboring
cells are replanned.
Force Co-Site As Indicates whether to forcibly configure co-

Neighbor sited cells as bidirectional neighboring cells.
Co-Site Distance(m) Two cells are cofigured as bidirectional

neighboring cells when the distance between
the two cells is less than the value of this
parameter.
Reference Existed Indicates whether to reference the

Neighbors neighbor relationships of existing cells.
Reference Rules After clicking Reference Rules, users can
configure reference rules in the displayed
window.
Source Cell Frequency Band: Indicates the
frequency band of the cell to be planned.
Neighbor Cell Frequency Band: Indicates the
frequency band of the neighboring cell.
Referenced Cell Frequency Band: Indicates
the frequency band of the cell used for
reference.
Reference Neighbor Cell Frequency Band:
Indicates the frequency band of the
neighboring cell of the cell used for reference.
Frequency display mode:
Frequency band of a cell.
ARFCN of a cell. ARFCN of a UMTS

to configure the ARFCN attribute of the
cell.
DlEarfcn of a cell. DlEarfcn of a cell is displayed

on a LTE network. In this case, users need to
configure the DlEarfcn attribute of the cell.
Azimuth Difference(°) Indicates the azimuth difference between

the cell to be planned and the cell used for
reference.
Reference Site Indicates the distance difference between
Distance(m) the site of the cells to be planned and the
site of the cells used for reference.
Mapping Table Indicates the sector splitting mapping file.
Obtain a file template using Template and
import the mapping between cells before
and after sector splitting.
After this parameter is selected, you can

click Template to obtain the data template.
You can also click Browse to import a data
file.
For details about the template, see GSM
Mapping Template, UMTS Mapping
Template, and LTE Mapping Template.
Consider Handover Indicates whether to consider the handover

Statistics data and the path for saving the handover
data.
Handover Statistics Indicates the handover data.

For details about the template, see Handover
Statistics Template.
Planning Weight Indicates the weight of the planning result

upon neighboring cell ranking.
Handover Statistics Indicates the weight of the handover data

Weight upon neighboring cell ranking.
Auto Identification Indicates whether to automatically

Neighbor Planning Area generate cells to be planned.
The U-Net automatically selects new cells

and related cells as cells to be planned.
Area Indicates a planning area.

You can select all cells in an area or click
Filter to select only the cells to be planned in
the area.
In the Filter dialog box, you can specify the
contents to be searched for and set the
search criteria such as search direction or
whether the uppercase and lowercase letters
are distinguished.
Figure 3.3.2.1.1.6.1.2 Neighboring cell planning setting interface – Intra-

Frequency tag page
Table 3.3.2.1.1.6.1.2.1 Descriptions of parameters in Intra-Frequency tab page
New Cell-Indoor Cell-Max Outdoor Indicates the maximum number of

Neighbor Number outdoor neighboring cells planned
for a newly deployed indoor cell.
New Cell-Indoor Cell-Max Indoor Indicates the maximum number of

Neighbor Number indoor neighboring cells planned
for a newly deployed indoor cell.
New Cell-Outdoor Cell-Max Outdoor Indicates the maximum number of

Neighbor Number outdoor neighboring cells planned
for a newly deployed outdoor cell.
New Cell-Outdoor Cell-Max Indoor Indicates the maximum number of

for a newly deployed outdoor cell.
Existed Cell-Indoor Cell-Max Indicates the maximum number of

Outdoor Neighbor Number outdoor neighboring cells planned
for an existing indoor cell.
Existed Cell-Indoor Cell-Max Indoor Indicates the maximum number of

for an existing indoor cell.
Existed Cell-Outdoor Cell-Max Indicates the maximum number of

Outdoor Neighbor Number outdoor neighboring cells planned
for an existing outdoor cell.
Existed Cell-Outdoor Cell-Max Indicates the maximum number of

Indoor Neighbor Number indoor neighboring cells planned
for an existing outdoor cell.
Figure 3.3.2.1.1.6.1.3 Neighboring cell planning setting interface – Inter-

Frequency tag page
Table 3.3.2.1.1.6.1.3.1 Descriptions of parameters in Inter-Frequency tab page
Total Max Number Indicates the maximum number of neighboring

cells.
Min Signal Level(dBm) Indicates the minimum signal receive level of an

inter-RAT neighboring cell.
Handover Indicates the handover area threshold.

Threshold(dB)
Frequency Band Indicates the frequency band to be handed over

when the RAT is switched to GSM.
ARFCN Indicates the ARFCN to be handed over when

the RAT is switched to UMTS.
 Output
Figure 3.3.2.1.1.6.1.4 shows neighboring cells of the cell to be planned.
Figure 3.3.2.1.1.6.1.4 Neighboring cell planning output
In this figure:
[Intra-Frequency]: Indicates intra-frequency neighboring cells.
[Inter-Frequency]: Indicates inter-frequency neighboring cells.
[Neighbor Name]: Indicates the name of a neighboring cell.
[Cause]: Indicates the reason for configuring a cell as the neighboring
"existed": Indicates that neighbor relationships exist on the network.
"planned": Indicates that neighbor relationships are planned.
"manual": Indicates that the neighbor relationship is manually created
by a user.
"inherited": Indicates that neighbor relationships are inherited.

[Confirm]: Indicates whether a cell is configured as the neighboring
If the option is selected, the cell is configured as the neighboring
If the option is deselected, the cell is not configured as the
After neighboring cells are successfully planned, the display condition of
current cells is displayed on the map.
Figure 3.3.2.1.1.6.1.5 Opening the neighboring cell planning interface
Click Neighbor Planning and choose LTE-TDD. Right-click on LTE-TDD

and choose Display Option. Set the mode and color for displaying neighbor
relationships in the map window.

page
Figure 3.3.2.1.1.6.1.7 Neighboring cell planning setting interface – Neighbor

Display Color tag page
:
Descriptions of parameters in General tab page:
Neighbor display mode: Indicates the display mode of neighboring cells.
[Display Links]: Identifies neighbor relationships by using lines.
[Display Cell Color] Identifies neighbor relationships in cell colors.
[Fit Neighbor Cell Visible] Displays the neighbor relationships of a cell on the map
after you select the cell in the neighbor relationship table. That is, after you select the
source cell in the neighbor relationship table, the neighbor relationships of a cell are
displayed on the map.
[Display Deleted Existed Neighbor]: Displays the neighboring cells that are available
on the live network but are currently deleted.
[Display Uni-directional Neighbor From Other Cell]: Displays unidirectional

neighbor relationships from other cells to this cell.
"Select Neighbor type": Sets the neighboring cell type to be displayed.
The types of neighboring cells include:
Intra Frequency Neighbors: Indicates intra-frequency neighboring cells.
Inter-Frequency Neighbors: Indicates inter-frequency neighboring cells.
Inter-RAT Neighbors: Indicates inter-RAT neighboring cells.

Descriptions of parameters in Neighbor Display Color tab page:
[Confirm Intra Frequency]: Indicates intra-frequency neighbor relationships.
[Confirm Inter Frequency]: Indicates inter-frequency neighbor relationships.
[Confirm Inter-RAT]: Indicates unidirectional inter-RAT neighboring cells. This

parameter is available only for a multi-mode network.
[Deleted Intra Frequency]: Indicates intra-frequency neighboring cells that are deleted
[Deleted Inter Frequency]: Indicates inter-frequency neighboring cells that are deleted
[Deleted Inter-RAT] Indicates inter-RAT neighboring cells that are deleted from the
live network. This parameter is available only for a multi-mode network.

4 TA&TAL Planning
4.1 Overview
This document:
 Describes how to plan, check, optimize LTE TA/TAL at different stages
of LTE network deployment and in different networking scenarios.
 Provides definitions of the TA/TAL, factors that affect TA/TAL planning,
TA/TAL planning principles, and detailed procedures for TA/TAL
planning.
 Lists methods for checking and evaluating the TA/TAL, optimization and
replanning principles, and application cases on live networks.
This document is intended for Huawei NIS engineers and engineers dedicated
for professional services.
4.2 Background and Benefits

The background and benefits of TA/TAL planning and optimization are as
follows:
 On a new LTE network, the TA/TAL, as important parameters, need to be
planned properly. Otherwise, inappropriate TA/TAL planning may cause
the following network performance problems.
− A small TA or TAL or an improper boundary causes UEs at the
boundary of a TA to initiate excessive TAUs.
− A large TA or TAL causes a heavy paging load. Consequently, a large
paging delay occurs or the paging messages are discarded.
− Frequent TAUs and paging procedures affect the call attempt per
second (CAPS) of the eNodeB and consequently subsequent user
access. The CAPS indicates CPU load.
 When planning the TA/TAL, provide an optimum TA/TAL planning
solution by considering the current network scale, user distribution,
subsequent network scale after capacity, and product-based paging
specifications. If CSFB services are required, align the TAL boundary of
the LTE network with the LAC boundary of the GSM and UMTS
networks to reduce the delay of CSFB services and configure the

mapping relationships between the TAL and the LAC on the EPC side.
 On an expanded LTE network, TACs need to be assigned to E-UTRAN
cells of expanded sites. In this case, mixed network capacity expansion
and large-area capacity expansion need to be distinguished. In mixed
network capacity expansion, some margins are reserved during initial
network planning and the TAC of the mixed site can be configured as the
TA nearest to the expanded site. In large-area capacity expansion, sites to
be expanded are planned based on initial network planning.
 Currently, the MOCN is used in multiple LTE offices. Independent
TA/TAL configurations of two operators are available surrounding the
MOCN sites. Generally, TA/TAL configurations of operators using the
MOCN need to be adjusted to ensure balance between the TAU and
paging load and that the TAL is applicable to the MOCN scenario.
 With the increase of users and expansion of commercial networks,
regularly monitor whether the TAU and paging load are abnormal,
evaluate whether the network TA/TAL are appropriate, and promptly
locate network performance problems caused by inappropriate TA/TAL
configurations to provide data input and case reference for subsequent
optimization.
This document provides scenario-based planning and optimization principles
according to product versions and capabilities and introduces related tools to
provide guidance for frontline personnel to quickly and efficiently implement
TA/TAL planning, checking, and optimization, thereby improving site
deployment efficiency and reducing operation and maintenance (O&M) costs.
4.2.1 Introduction
This section describes scenario-based TA/TAL planning procedures and key
technologies required.
Scenario Classification
Table 4.2.1.1.1.1.1.1.1 lists classification of scenarios during TA/TAL
planning. TA/TAL planning principles vary with scenarios. For details, see
related sections.
Table 4.2.1.1.1.1.1.1.1 Classification of scenarios during TA/TAL planning
New Networks Independent LTE site deployment
Network where eNodeBs are co-sited with

NodeBs and BTSs
Expanded Networks Mixed network capacity expansion
Large-area capacity expansion
Special Networks MOCN
Multiple frequency bands

Multiple vendors
Boundaries between countries and cities
Macro and micro hybrid networks
High-speed railways, subways, and highways
CSFB scenario
Indoor scenario
TA/TAL planning principles vary with scenarios. For details, see related sections.
Application Scenarios of TA/TAL Checking and Evaluation

TA/TAL checking and estimation are required in the following scenarios:
 Generally, TA/TAL planning is implemented by customers. To reduce
subsequent commercial acceptance risks and guarantee commercial
quality, it is good practice to check the planning solution made by
customers. Correct errors promptly if any.
 After the network is commercially used, regularly collect configuration
data and related counters on the live network, and check TA/TAL
configurations. If potential TA/TAL configuration errors are found, lead
customers to adjust TA/TAL configurations if necessary.
 During network capacity expansion, evaluate and check TA/TAL
configurations (especially the TAU and paging load) of the current
network to prevent signaling storms after the network is commercially
used after being expanded, which can result in TA/TAL configurations on
a large scale.
TA/TAL Optimization
TA/TAL optimization is based on checking and evaluation results. During
TA/TAL optimization, you are advised to observe existing principles, such as
adjusting the TAL first while ensuring that the TAL boundary is not the heavy-
traffic area or the area where UEs move frequently. In addition, the trend for
user increase and the network capacity proportion also need to be considered.
Generally, TA/TAL optimization is performed in scenarios where the TA or
TAL does not greatly change. For example, modifying TAC configurations at
a site or adjusting the TA/TAL in a small area.
TA/TAL Replanning
TA/TAL replanning is performed in the following scenarios:
 After LTE networks are expanded on a large scale or traffic volumes
increase, original TA/TAL configurations need to be greatly adjusted.
Compared with TA/TAL optimization, TA/TAL replanning involves

larger areas.
 In the CSFB scenario, changes in GSM and UMTS LAC configurations
result in LTE TA/TAL adjustment.
Planning Procedure and Key Technologies

This section describes scenario-based TA/TAL planning, key procedures and
actions.
New Networks
As listed in Table 4.2.1.1.1.1.1.1.1, new networks include the independent
LTE network and the network where eNodeBs are co-sited with NodeBs and
BTSs.
Figure 4.2.1.1.1.1.1.2 shows the TA/TAL planning procedure on a new LTE
network.
Figure 4.2.1.1.1.1.1.2 TA/TAL planning procedure on a new LTE network
Start
Estimate the site scale based on paging

capacity of NEs.
Estimate the number of eNodeBs in a TAL based on MME Estimate the number of eNodeBs in a TAL based on
paging capacity. eNodeB paging capacity
eNodeB paging capacity: CPU paging loads, PDCCH
MME paging capacity: board paging specifications and and PDSCH paging loads, and paging blocking rate
number of boards
Number of UEs: Number of UEs served by eNodeBs
Number of UEs: number of attached UEs at the same
time Traffic model: paging traffic model on the access network
Traffic model: paging traffic model on the MME
Number of eNodeBs contained in a TA/TAL
Planning TA/TAL boundaries with TAU

overheads considered
End
During TA/TAL planning, the following key technologies are involved:

 Evaluating the TAL scale: refers to evaluating the number of eNodeBs
contained by each TAL.
The following factors on the RAN side need to be considered:
− eNodeB paging specifications (paging times allowed per second)
− Number of UEs covered by each eNodeB

− Paging model on a LTE network
− Subsequent mixed capacity expansion (including deploying macro
base stations, micro base stations, and dual-carrier configurations)
The following factors on the EPC need to be considered:
− Paging specifications of boards
− Number of UEs that boards can accommodate
− Generally, capacity limitation on the EPC does not greatly affect the
TAL scale. For details, see related sections. When evaluating the TAL
scale, take only limitations on paging specifications of the eNodeB
into account.
 Determining the number of TAs in a TAL: The static TAL is used on the
current EPC. According to LTE protocols, each TAL can be configured
with a maximum of 16 TAs. To facilitate subsequent TAL splitting and
network capacity expansion, it is good practice to configure 3 to 10 TAs
in a TAL on a new network.
 On networks where eNodeBs are co-sited with NodeBs and BTSs, to
reduce the delay of CSFB services, it is recommended that the TAL
boundary of the LTE network align with the LAC boundary of
GSM/UMTS networks. (LAUs for CSFB services can be reduced by
configuring the mapping relationship between the TAL and LAC over the
EPS.) The LTE TAL scale (number of eNodeBs in a TAL) may be
inconsistent with the GSM/UMTS LAC scale because of differences
between the LTE paging model and the GSM/UMTS paging model. In
this case, it is good practice to enable multiple TALs to correspond to
one LAC, which ensures that the boundaries of multiple TALs and the
LAC boundary are consistent, reducing the impact on CSFB services.
Generally, if LTE networks are set up based on GSM/UMTS networks, the LAC
boundary has been optimized because the GSM/UMTS networks are mature. In this
case, the TAL boundary can be inherited from the GSM/UMTS LAC boundary.
If the GSM/UMTS network LAC scale is close to the LTE TAL scale, the TAL
boundary can be inherited from the LAC boundary. If the former is much greater that
the latter, divide sites within the LAC area into multiple TALs and divide the
corresponding TALs within the LAC.
Expanded Networks
As listed in Table 4.2.1.1.1.1.1.1.1, expanded networks include mixed
network capacity expansion and large-area capacity expansion.
 In mixed network capacity expansion, if the number of sites undergoing
mixed network capacity expansion is large, before assigning TACs to the
expanded sites, evaluate the traffic model on the live network and then
obtain the TAL scale. The method for evaluating the TAL scale is similar
to the method for evaluating the site scale on a new network. Generally,
the TAC of a site undergoing mixed capacity expansion is the TCP of the
nearest site planned. However, if the current TAL scale is large and
becomes unexpectedly large after capacity expansion by adding sites, the
TAL or TAL needs to be split.
Figure 4.2.1.1.1.1.1.3 TAC configurations in mixed network capacity expansion
TAC=05;
TAC=06;
TAC=05;
TAC=06;
TAC=06;
TAC=06
 In large-area capacity expansion, the method for planning sites in

expanded areas is the same as that on a new network.
Special Networks
MOCN
In the MOCN scenario, generally the TA/TAL configurations of the primary
operator (providing the MOCN site) do not need to be adjusted and only the
TA/TAL configurations of the secondary operator (using the MOCN site) need
to be planned or adjusted. Key actions are as follows:
 Obtain information about all sites (including the MOCN site) of the
secondary operator, cell engineering parameters, and TAC and TAL
configurations.
 According to the preceding engineering parameters, use the U-Net to
geographically display TA/TAL configurations of sites and cells and
mark the MOCN site. Only the TAC configuration of the MOCN site is
required.
 Evaluate the TAL scale by using the method for evaluating the TAL scale
on a new network and plan non-MOCN sites based on the procedure and
principles on a new network.
 According to the principle of proximity, divide the TA of the MOCN site
and assign sub-TAs to different TALs of current operators based on UE
distribution, terrains, and ground objects while ensuring that the TAL
scale is not exceeded and that the TAL boundary is not the heavy-traffic
area or the area where UEs move frequently.
Figure 4.2.1.1.1.1.1.4 TA planning in the MOCN scenario
Multi-frequency networking
In this scenario, it is good practice to plan sites using different frequency
bands in a uniform manner and frequency-band-based independent planning is
not recommended. Planning principles and procedures are similar to those on
new networks. If frequency bands are not put into commercial operation at the
same time, reserve certain margins when planning the TA/TAL for frequency
bands that need to be commercially used first. The method for TA/TAL
planning at other frequency bands is similar to that on expanded networks.
Some operators such as China Telecom provide both TDD and FDD network
topologies. UEs supporting both TDD and FDD may be available on the
network. In this case, to reduce TAUs caused by reselections and handovers
between TDD and FDD, the TDD network and the FDD network must be
deployed in different areas. It is recommended that the TA/TAL for TDD and
FDD sites are planned in a uniform manner.
Multi-vendor networking
In this scenario, independent scenarios are available. The method for planning
the TA/TAL is similar to that on new networks. If sites are covered by
different vendors, coverage overlap occurs. In this case, sites of different
vendors are recommended to be planned in a uniform manner by considering
differences of paging specifications between vendors and confirming the
paging capacity of vendors. Ensure that the TAL does not exceed the
specifications of Huawei equipment. Otherwise, paging problems occur.
Boundaries between countries and cities

In this scenario, ensure that TAIs in the TALs at the boundary are different.
For the boundary of the same operator, the TACs in the TALs at the boundary
should also be different because TAI is the result of PLMN ID plus TAC. If
different vendors exist, the TAC can be different because the PLMN ID varies
with vendors.
Mixed macro and micro networking

In this scenario, a macro network is set up first. Then small base stations are
deployed based on the scale of users and services. During macro base station
deployment, the principle for planning the TA/TAL is similar to that on new
networks. During micro base station deployment, the principle for planning
the TA/TAL is similar to that on expanded networks. Currently, Huawei small
base stations are integrated with the sniffer function. That is, Huawei small
base stations automatically implement TAC configurations by using the
sniffer function to scan TAC configurations of surrounding macro cells.
High-speed railways, subways, and highways

On high-speed railways and highways, the following principles for planning
the TA/TAL are as follows:
 If few or even no UEs are distributed along the high-speed railway or
highway, it is good practice to assign independent TAs and TALs to
eNodeBs along the high-speed railway and highway.
 If many UEs are distributed along the high-speed railway or highway,
you are advised to plan high-speed coverage sites along the high-speed
railway and non-high-speed coverage sites along the high-speed railway
in a uniform manner.
In high-speed railway coverage scenarios, to reduce loads at the air interface
of the network to maximum degree, the corresponding TAL boundary should
be located at the large platform (with a large passenger flow) along the high-
speed railway. Otherwise, a TAU storm probably occurs. In subway coverage
scenarios, if many adjacent TALs of sites along the subway are available, it is
good practice to independently plan the TAL of sites along the subway and
adjust TACs of sites near the subway site to prevent the TAL boundary from
being located at the large subway platform. If few adjacent TALs along the
subway are available, you are advised to plan the TAs and TALs of subway
sites and non-subway sites in a uniform manner.
Figure 4.2.1.1.1.1.1.5 shows distribution of GSM LAC areas at sites along
subway line 1 in a city. (No subway planning is provided at the initial stage of
network deployment.)
Figure 4.2.1.1.1.1.1.5 Distribution of GSM LAC areas at sites along subway line
1 in a city
As shown in the above figure, sites in different colors indicate different LAC
areas. Sites along the subway cross multiple LAC areas. In this case, sites
along the subway can be planned with an independent GSM LAC area that
corresponds to an independent LTE TAL.
CSFB scenario
For CSFB services, the Combined Attach function is supported according to
3GPP. That is, a UE can be registered in both the EPS domain on an LTE
network and the CS domain on a GSM/UMTS network. In this way,
information about this UE can be saved in both the PS domain on an LTE
network and the CS domain on a GSM/UMTS network. This facilitates an
LTE-to-GSM/UMTS CSFB. For details, see the following figure.
When the UE sends the Attach Request message on the LTE network,
ATTACH TYPE is Combined GPRS/IMSI ATTACH, indicating that the UE is
capable of implementing CSFB and SMS over the SGs interface. The MME
locates the VLR to which the UE belongs based on the IMSI and allocated
TAs and implements the LAU in the CS domain. If the attach procedure
succeeds and both the LAI and VLR TMSI are reallocated (that is, the LAI
and VLR TMSI are different from those reported in Attach Request), then the
LAI and VLR TMSI are carried in the Attach Accept message.
After Combined Attach is complete and when the UE triggers CSFB services:
 If the UE falls back to the LAC area of GSM/UMTS cells and the LAC
area with the Combined Attach function enabled (mapping between the
TA and the LAC is configured at the EPC side), the LAU is not required,
reducing the CSFB delay.
 Otherwise, the UE initiates the LAU. An LAC takes about 1s to 2s.
Generally, CSFB services are required in the scenarios where the LTE
network is co-sited with the GSM/UMTS network. To reduce unnecessary
LAUs, observe the following principles when planning the TA/TAL:
 The TAL boundary is consistent with the GSM/UMTS LAC boundary as
much as possible.
 The TAL coverage scope is not greater than the LA coverage scope.
 In most cases, the GSM/UMTS LA is within one MCS or MSC pool. If
the TAL boundary aligns with the LA boundary, the TAL is also with one
MCS or MCS pool. Otherwise, when planning the TAL, adjust the
corresponding LA to be within one MCS or MCS pool.
Assume that the LA to which the TAL corresponds is not within one MCS or MCS
pool:
When a CSFB UE falls back to the CS domain, the MSC server is not the one with
which the UE has been registered upon TA/LA updates. In this case, the MCS enables
mobile terminating roaming retry or mobile terminating roaming forwarding if it has,
which prolongs the CSFB handover delay; if the MSC does not support mobile
terminating roaming retry or mobile terminating roaming forwarding, call drops
probably occur.
 If a CSFB proxy is used, ensure that location areas of CSFB Proxy

configured on the MME are different from the broadcasting location
areas or the adjacent location areas configured on other end offices (EOs)
within the coverage scope, which guarantees accurate fallbacks. That is,
all TAs need to correspond to Common Fake LAs specified for CSFB
services to ensure that all CSFB UEs registered with the CSFB proxy
upon TA/LA updates.
 If two CSFB proxies are deployed and Common Fake LAIs on the two
proxies are the same, the MME needs to select the CSFB proxy based on
the IMSI hash. Assume that both GSM and UMTS networks are
available:
One TA/TAL on the MME corresponds to one LA. If the LA on the GSM
network is different from that on the UMTS network, consider whether
the TA/TAL is mapped to the GSM LA or UMTS LA based on network
coverage, coverage quality, and network load. If the preceding factors are
the same on the GSM and UMTS networks, it is good practice to
preferentially fall back the UE to the UMTS network, reducing the CSFB
delay. In addition, whether the UE falls back to a GSM or UMTS
network is subjective to the customer's CSFB policy. If UE and network
capabilities are limited, sometimes the UE may fall back to a GSM
network.
In the CSFB scenario where both GSM and UMTS networks are available, if the
MME supports an LAI depends on the TAI and IMSI, IMSI numbers can be planned
further to enable different IMSI numbers to correspond to different LAs. This meets
flexible requirements of customers. (Note: Huawei MME supports an LAI depends on
the TAI and IMSI in PS 10.0 or later.)In this case, a TA with different IMISs can
correspond to different LAs but a TA with the same IMSI can correspond to only one
LA. If users cannot be distinguished based on the IMSI on a network, users in the
same TA fail to correspond to different LAs. When planning the TA and LA, consider
IMSI number planning in the following scenarios:
 The IMSI numbers for CSFB users are independent.
 Multi-operator UEs share the same LTE network and eNodeB broadcasts only a
single PLMN.
The LTE network may have coverage areas different from the GSM/UMTS network,
for example, the eNodeB is not co-sited with NodeBs/BTSs. In this case, if the MCS
on the GSM/UMTS network does not support mobile terminating roaming retry or
mobile terminating roaming forwarding, the inter-MSC/MSC-pool TAL coverage
scope may occur, resulting in call drops. To prevent this situation, use the following
planning methods:
 Optimize the azimuths or downtilts of LTE sites at the MSC/MSC pool boundary
without compromising the predicted LTE coverage area. In this way, the coverage
scope of the LTE sites to can be within an LA and the UE does not fall back to the
LA to which another MSC or MSC pool corresponds after a CSFB is
implemented.
 Add BTSs/NodeBs at the place where eNodeBs are deployed while ensuring that
the eNodeB coverage scope is not greater than the BTS/eNodeB coverage scope.
Specify appropriate LAs to new BTSs/NodeBs.
 Adjust the LA boundary again to ensure that the coverage scope of eNodeBs not
co-sited with NodeBs/BTSs is within an LA coverage scope.
 In areas with many non-co-sited LTE sites, combine the MCSs/MCS pools to
which the corresponding LAs belong to form a larger MCC pool. In this way, the
inter-MCS/MCS-pool TALs are eliminated.
Indoor scenario
The principle for TA/TAL planning in indoor scenarios is the same as that for
outdoor scenarios.
Examples of New Network Planning
Typical Example of TA/TAL Planning in a City in China

This city, as a typical second-tier city, has 1002 eNodeBs contained in 21
TALs at the first stage used for covering first-ring and second-ring valuable
areas. TAL boundaries align with GSM LAC boundaries. The number of
eNodeBs and TAs in each TAL is listed in Table 4.2.1.1.1.1.1.5.1.
Table 4.2.1.1.1.1.1.5.1 Number of eNodeBs and TAs in each TAL
TAL Index eNodeB Quantity TA Quantity
1 117 2
2 49 2
3 222 8
5 42 1
6 230 6
7 80 2
8 15 3
9 19 3
10 29 2
11 182 6
13 4 1
17 8 1
18 2 1
TAL Index eNodeB Quantity TA Quantity
21 3 1
The number of eNodeBs in some TALs is small because only valuable areas are
covered at the first stage.
Figure 4.2.1.1.1.1.1.6 shows TAL distribution. Different TALs are marked in

different colors.
Figure 4.2.1.1.1.1.1.6 TAL distribution
Figure 4.2.1.1.1.1.1.7 shows distribution of TAs in TAL 6.

Figure 4.2.1.1.1.1.1.7 Distribution of TAs in TAL 6
Typical Example of TA/TAL Planning in a City Out of China

In this city, 1200 eNodeBs are available at the first stage used for covering
about 0.5 million UEs. TAL boundaries are planned based on the original
UMTS LAC boundaries. Figure 4.2.1.1.1.1.1.8 shows the TAL distributing
and scale.
Figure 4.2.1.1.1.1.1.8 TAL distributing and scale
Figure 4.2.1.1.1.1.1.9 shows distribution of TAs in TAL 411.

Figure 4.2.1.1.1.1.1.9 Distribution of TAs in TAL 411
TAC=41102
TAC=41101 TAC=41104
TAC=41103
TAC=41105
TAC=41106
TAC=41107 TAC=41108
TAC=41109
TAC=41110
TAC=41112
TAC=41111
Evaluating the Check Procedure and Key Technologies

Perform the following steps to check whether the TA/TAL is properly
configured:
Collect TA/TAL configurations on the entire network, engineering parameters
of site cells, number of current UEs on the network, the trend of UE quantity,
and networking information (such as MOCN).
Step 1 Evaluate the TAL scale on new networks.
Step 2 Import eNodeB information to the U-Net to geographically display TAL
distribution and then count the number of eNodeBs in each TAL.
Step 3 Obtain the number of UEs and the paging scale in each TAL on commercial
networks.
Step 4 Calculated the number of eNodeBs in each TAL based on the current number
of UEs and predicted UE increase trend.
Step 5 Pay special attention to the TAL containing more eNodeBs than expected and
provide suggestions for splitting the TAL.
Currently, on the EPC, paging is performed in the unit of TALs. The TA scale
does not affect paging but has impact on subsequent TAL splitting and
adjustment. In some extreme case, all eNodeBs in a TAL may be within the
same TA (that is, configured with the same TAC). In this case, it is labor-
intensive for splitting and adjusting the TAL because the TAC also needs to be
changed. Therefore, the TA scale is relatively small according to network
planning. For example, only 20 to 30 eNodeBs are in one TA in M1 and Yota.
Consequently, the TA scale is not the focus during TA/TAL configuration.
----End
Optimization Procedure and Key Technologies

Based on TAL check results, adjust and optimize the TALs that do not meet
related requirements on the scale by observing the following principles:
 Adjust TAL configurations preferentially For example, if the original

TAL scale is large, split the original TAL into two smaller TALs
 When splitting the TAL, ensure that the new primary TAL boundary is
not located at heavy-traffic areas or areas where UEs move frequently.
4.2.2 Requirements and Application Scenarios

Requirements
Product Paging Specifications

Detailed paging specifications are as follows:
 ECU (PS 10.0) for eRAN 7.0: 3300 paging times/s (paging 16 eNodeBs)
 ECU (PS 9.1/2) for eRAN 6.0/ eRAN 3.0: 14,000 paging times/s
 eRAN 6.0/ eRAN 3.0 macro base stations: 750/500 paging times/s
 eRAN 6.0/ eRAN 3.0 micro base stations: 500/500 paging times/s
PS 9.1/PS 9.2/PS10.0 supports accurate layer paging that is under license
control. The sequence of accurate paging is eNodeBs, neighboring eNodeBs,
TAs, and TALs.
Paging specifications of ECUs in PS 9.1/9.2 and PS 10.0 are different. The paging
specifications of PS 10.0 ECUs can be calculated by using the following formula:
 3300 x 16 = 52,800 paging/s.
According to tests results on the EPC, paging occupies about 34% of CPU when the
following requirements are met:
 Each TAL contains 256 eNodeBs.
 Each TA contains 16 eNodeBs.
 More than 100,000 paging messages are processed per second.
If the CPU occupied by paging is less than 20% if it is calculated based on the paging
specifications of 52,800 paging times/s.
Applicable Scenarios
N/A
4.2.3 Maturity
 No commercial tool is available for supporting the planning function in
the current version. Only planning principles are provided. Operations
are performed on the U-Net by observing the provided planning
principles.
 The RND tool calculates the TAL and TA scales, that is, provides the
number of eNodeBs in each TAL and TA.
 The U-Net can geographically display the TAs and TALs and supports
some check functions.
 At present, no optimization tools are available and only optimization

principles are provided.
4.3 Principles
4.3.1 TA
The TA is a concept introduced to the LTE/SAE system for location
management of UEs. The TA is similar to the RA for the UMTS/GSM/Edge
system where UEs do not need to update the VLR in the TA. In the TA, UEs
do not need to update serving MMEs. The size of a TA, that is the TAC
coverage area, is a key factor in the system.
To determine where UEs are located, the coverage area of an LTE network is
divided into multiple TAs based on TACs. The TA is a cell cluster containing
the same TAC. A cell can be within only one TA but cells in the same TA can
be covered by different eNodeBs under the same MME. The system sends
paging messages to all cells in the TA to located UEs in idle state. The paging
scope is all cells in all TAs in a TAL in the current TAL solution (also called
multi-registered-TA solution).
The TA performs location management of UEs containing paging
management and location update management. The network side establishes a
call connection with the UE and MME must record the UE location in real
time so that the UE can be paged when necessary. The UE initiates a static
TAU based on the network topology or initiates a dynamic TAU based on the
mobility mode of calls to update information in the UE location database,
which helps the system locate the UE more efficiently.
The location update management includes the TAU solution and the paging
solution.
For details about the TA function, see:
 3GPP TS36.300, Overall Description; Stage 2
 3GPP TS36.304, User Equipment (UE) Procedures in Idle Mode
4.3.2 TAL
Multi-registered-TA solution is an LTE location management solution
specified in 3GPP Release 8 and is combined by and summarized from
multiple TA solutions. In this solution, multiple TAs form a TAL and are
allocated to the same UE. TAUs are not required when the UE moves between
the TAs in the TAL. According to 3GPP 24.301, a TAL can contain a
maximum of 16 TAs.
Figure 4.3.2.1.1.1.1.1 TAL
According to 3GPP 24.301, the UE-level TAL is supported. This matches the
dynamic TAL solution, in which different TALs can be assigned to the same
UE based on the UE location. The dynamic TAL solution is difficult to be
performed due to the following aspects:
 Due to unpredictable UE mobility actions and direction, frequent TAL
updates are required, wasting signaling overheads in TAUs and paging.
 It is difficult to determine the TA scale. The TA scale is closely related to
the traffic model and UE mobility actions. In addition, the balance
between the TAU and paging needs to be considered. Therefore, an
appropriate recommended value is difficult to derive from stimulation or
theoretical analysis.
Huawei EPC supports only the static TAL solution. In this solution, when the
UE is on a network, the network determines which TAs in the same TAL are
assigned to the cell-level UE and the UE is registered with these TAs. When
the UE moves to a new TA that is not in the TAL with which the UE is
registered, a TAU is required. Then the network distributes another group of
TAs in the TAL associated with the cell to the UE. The new and original TALs
do not overlap.
In the protocol-defined UE-level TAL solution, the updated TAL is not associated
with the cell. The eNodeB sends the new TAI to the MME over the TAU request
message based on the UE mobility attribute in the cell. Then the MME updates the
UE-level TAL for the specified UE and the new TAL can also contain some TAs in the
original TAL.
The dynamic TAL solution helps the EPC flexibly control the areas where
UEs can be paged in the following aspects:
 The EPC can flexibly assign TAs to UEs based on initial planning and
TAs do not need to be deployed again.
 TAUs caused by inter-TA UE movement can be avoided and the ping-
pong impact can be prevented.
In a static TAL solution, a TAL contains multiple small TAs, which facilitates
subsequent network maintenance. If the original TAL scale becomes too large
because of sharp increase in the number of UEs on the network or network
capacity expansion, paging loads become heavy. In this case, the TAL needs
to be optimized. If a TA contains many eNodeBs during planning, change cell
TACs when splitting the TAL, which restarts the cells and interrupts ongoing
services. If a TAL contains multiple TAs and each TA contains a few eNodeBs
during initial planning, adjust the number of TAs in the TAL on the EPC side
when splitting the TAL, which does not interrupt ongoing services and
facilitates operations.
Huawei products support the layered paging mechanism. In the current
version, paging is first performed in TAs. If paging fails in the TAs, paging is
then initiated in the TAL. In this way, one TAL contains multiple TAs each of
which contains fewer eNodeBs, which reduces paging loads on the network.
In later versions, the paging is performed on eNodeBs, on neighboring eNodeBs

(accurate paging), in TAs, and then in TALs.
Figure 4.3.2.1.1.1.1.2 shows the TAs and TA planning in MME pools.
Figure 4.3.2.1.1.1.1.2 TAs and TA planning in MME pools

MME MME
pool 1 pool 2
TA configuration
(non overlapping)
possible TA-list
assignments to UEs
Figure 4.3.2.1.1.1.1.3 shows cell-level TA planning in MME pools.

Figure 4.3.2.1.1.1.1.3 Cell-level TA planning in MME pools
The MME pool area similar to the SGSN pool is introduced in LTE. The
MME pool area is defined as a radio access network where a group of MMEs
cover the same area. When a UE moves in an MME pool area, the MME
serving the UE keeps unchanged.
Each eNodeB can connect to multiple MMEs. Therefore, the eNodeB needs to
select an MME from those covering the UE. If a UE attaches to an MME and
the UE is within the area covered by signals in the pool area containing the
MME, the UE still connects to the MME to reduce signaling loads on the EPC
as much as possible. The pool area contains one or more TAs on the RAN
side. In the pool area, a group of MMEs provide services that can also serve
TAs out of the pool area. Such MMEs are called an MME pool.
In this case, if a UE is within the coverage scope of an MME in the MME
pool area, it is likely that the serving MME of the UE keeps unchanged,
reducing signaling overheads on the EPC. Different MME pool areas can
overlap.
 Multiple MMEs in the same MME pool area can connect to all eNodeBs in the
MME pool area. In this way, UEs in this area do not need to change its serving
MMEs. Otherwise, the serving MME of a UE may be changed even if the UE
moves in the same MME pool area.
 Different cells under the same eNodeB can be in different TAs and one TA can be
in different MME pool areas.
4.3.3 LTE TAU and Paging Principles and Procedures

For details, see section A"Appendix."
4.3.4 TAC and TAL Configurations

 According to the protocol, the TAC ranges from 0 to 65,535, of which
0x0000(0) and 0xFFFE(65534) are reserved and will not be configured
in later versions. Therefore, it is good practice not to set the TAC to 0 or

65,534.
 According to the protocol, the TAL ranges from 0 to 65534 and no fields
are reserved.
 If the operator's network scale is not large, you are advised to ensure that
the TAC number is unique on the same network. Otherwise, you must
ensure that TAC numbers are different at the boundaries of different
cities and that the TAC number is unique in the same city.
 The TAL number can be the same in different MMEs or MME pool areas
and the TAL and you are advised to ensure that the TAL is the same in
the same MME or MME pool area.
In actual activities, TAC and TAL numbers may be planned and configured
based on actual site scales by operators. When planning TACs and TALs,
consult operators on whether there are any mandatory numbering rules and
related limitations.
4.3.5 Evaluating TA/TAL Baseline Specifications

Factors Affecting Paging Capacity
The factors affecting paging capacity exist on both the MME side and the
eNodeB side.
On the MME Side
 Number of UEs registered with an MME
 Number of eNodeBs contained in the TAL associated with idle UEs
served by the MME
The number of eNodeBs in each TAL should not exceed the maximum
paging capacity of eNodeBs and MMEs. If the number of eNodeBs in
the TAL is small, the paging delay increases because the MME pages all
its supported TALs in serial mode. In addition, frequent TAUs occur,
reducing the paging success rate.
 Time distribution model for the network triggering services, that is, time
model for paging distribution
 Paging retransmission mechanism including the paging retransmission
timer, number of retransmissions, and discarding timer
On the eNodeB Side
 Paging cycle specified by defaultPagingCycle
 Number of paging occasions specified by nB
 Number of UEs contained in a paging group specified by
maxNoOfPagingRecords
 Time distribution model for the network triggering services, that is, time
model for paging distribution (time distribution for paging service
throughput)
 TAU load and cycle
 TA scale and TAL scale
 Number of UEs moving at a high speed
Analyzing Paging Capacity on the MME and eNodeB Sides

Evaluating MME Paging Performance
The maximum paging specifications for the ECU on the MME in PS 10.0 is
52,800 times/s (when about 20% of the CPU is occupied), which is calculated
by using the following formula:
52,800 times/s = 3300 x 16
If an ECU is deployed on the live network, the MME paging capacity must be
52,800 times/s or less.
In a typical Huawei traffic model (For details, see Appendix), the paging
traffic model value is 2.17/subscriber@BH and the paging times per second
per UE on the average is calculated using the following formula:
2.17
 0.0006027
3600 pages/(sub*s)
Assume that the ECU can accommodate a maximum of 0.5 million UEs. The
maximum number of eNodeBs that can be contained in a TAL is as follows:
52800
 175
500000 * 0.0006027
The message for paging a UE is sent to all eNodeBs in the TAL. The paging times per
second per UE is 0.0006027/(sub*s) and the MME in the TAL can accommodate a
maximum of 0.5 million UEs. In this case, the number of eNodeBs in a TAL is175
(obtained by using the above formula).
On live networks, services of 0.5 million UEs are carries on at least two
ECUs. In this way, each ECU carries 0.25 million UEs at most and the
number eNodeBs in the TAL is at least 350. The calculation formula is as
follows:
52800
 350
250000 * 0.0006027
According to preceding analysis, the following factors affect the number of

eNodeBs in the TAL on the MME:
 Paging capacity
 Traffic model
 Number of UEs
 The preceding formula is for reference only. On actual networks, it is impossible

that an ECU carries services of 0.5 million UEs. Generally, EPC paging
limitations do not need to be considered. If considered, it is good practice to add
ECUs.
 The preceding formula assumes that the MME uses common paging policies.
Huawei EPC supports accurate layered paging under license control on eNodeBs,
on neighboring eNodeBs, in TAs, and then in TALs. Based on the latest paging
performance on the EPC, the traffic model recommended by Huawei, and the
common paging model, it is estimated that the number of eNodeBs in a TAL can
be 500. In this case, the number of UEs on an ECU can be greater than 0.3
million, which meets the requirements in most scenarios.
 The paging traffic model value 2.17 /subscriber@BH is an empirical value on a

GSM/UMTS network. However, on an LTE network, a large number of smart
phones are introduced. The paging traffic model value on an LTE network differs
much from that on a GSM/UMTS network. For example, the paging traffic model
value on the M1 network in Singapore is 20.71/subscriber@BH after iPhones go
to the market.
Evaluating eNodeB Paging Performance

The eNodeB paging performance depends on CPU usage and resources that
can be consumed by paging. More paging resources indicate higher paging
performance.
Parameters affecting eNodeB paging performance are as follows:
 PDSCH load
 PDCCH load
 CPU load
 Paging blocking
Paging capacity and PDSCH load
The PDSCH paging load is affected by the following settings:
 Paging cycle (PC): Configured based on operators' demands. Huawei
paging cycle is 128 frames by default, that is, 1.28s.
 Paging subframe frequency (Psn) in a paging cycle: Number of paging
subframes in each paging cycle that is determined by the nB parameter.
The baseline value of Huawei nB is set to 1 by default.
 Pun: Number of UEs that can be paged in a paging subframe. The
maximum number of UEs that can be paged is 16 on a 20 MHz
bandwidth per TTI, that is, the eNodeB can page 16 UEs every 10 ms
and can page a maximum of 1600 UEs per second. An eNodeB can page
a maximum of 16 UEs per 2.5 ms, that is, 6400 UEs at most per second.
 COV: PDSCH coverage that is determined by the SCR and resource
allocation type.
If the UE ID is paged by the S-TMSI containing Ui bits, the Rb load
used by the PDSCH to send paging messages per second is calculated
（Ui * Pun  k）
Rb Pd schPaging  roundup（）* Psn / Pc
Modtyp * Re Pdsch * SCRpaging
Where,
The letter k indicates other information in the paging message, such as

systemInfoModification and etws-Indication.
According to 3GPP 36.331, the number of S-TMSI IE bits is 40. In
addition, when the 1 bit EPC domain and padding are included, the final
number of bits for paging messages is 48.
Padding and CRC are not considered in the following calculation.
Assume that input parameters are as follows:
SCR paging
− = 0.1;
− Modtyp = 2 (QPSK);
− Assume that the MCS for paging messages is MCS0 whose coding
rate is 0.1
− Pun = 16
− Pc = 1.28s
− Psn = 128
− k = 2 bits
− Ui = 40 bits
− Resource allocation type = LVRB
RePdschRb  12 subcarrie r * (14 - 3 PDCCH Overhead)symbol - RS
− = 12 * 11 - 12 = 120
Evaluation assumption: 2T2R, 3 PDCCH Symbols, then

（ 40 * 16  2）
RbPd schPaging  roundup（） * 128 / 1.28  2700
2 * 120 * 0.1
Table 4.3.5.1.1.1.1.1.1 Analysis on PDSCH paging load overheads
Channel Resource PDSCH RB/s
Resource Load Amount 2700
Resource Load Percentage (FDD) 10 MHz 5.4%
20 MHz 2.7%
Compared with other data services, paging services have a higher

priority. High-priority paging services may reduce the downlink service
capacity and effective bit rate. Therefore, it is necessary to control
PDSCH resources consumed by paging services.
C PDSCHload indicates the limitation on paging performance imposed by
PDSCH loads.
n SB , page
indicates the number of average scheduling blocks consumed by
n
the PDSCH to send a paging message. SB , page depends on the number
of paging records carried by the RRC paging message. Assume that a
n SB , page
maximum of 16 users can be paged on a 20 MHz bandwidth, is
calculated by using the following formula:
nSB , page  2.7
RB per subframe = 2700 RB per second
I page,s
− indicates the frequency at which paging messages reach the
eNodeB, that is, the number of paging messages received by the
eNodeB per second.
nSB, frame
− indicates the number of scheduling blocks contained in
each frame, as listed in the following table.
 Number of scheduling blocks contained in each frame

on different bandwidths
Bandwidth 5 MHZ 10 MHZ 15 MHZ 20 MHZ
nSB, frame 250 500 750 1000
n SB, page
indicates the number of scheduling blocks occupied by
processing a single paging message.
I
LPDSCH indicates the frame loads occupied when page, s paging
messages are processed and is calculated by using the following formula:
n SB , page * I page,s
LPDSCH 
100nSB , frame
LPDSCH ,max
indicates the maximum frame loads that can be received
LPDSCH ,max
by the eNodeB. is set by operators and usually does not
LPDSCH ,max I page,s
exceed 3%. determines paging messages that can
C
be processed by the PDSCH, that is PDSCHload (capacity for eNodeBs
C
processing paging messages). PDSCHload is calculated by using the
following formula:
100nSB , frame * LPDSCH ,max
C PDSCHload 
nSB , page
 PDSCH load on the eNodeB side

100 * n SB , frame * LPDSCH ,max 100 * 1000 * 0.03
C PDSCH load    1111
2.7 2.7
pages/s
Paging capacity and PDCCH load
The PDCCH paging load is affected by the following configurations and

demands:
 COV: indicates PDCCH coverage determined by the CCE aggregation
level (CRL).
The paging load for the PDCCH CCE per second is as follows:
CcePdcchPaging  CRL paging * Psn / Pc
Assume the input parameter is as follows:

CRL  4cces
Then the result is as follows:
CcePdcchPaging  4 * 128 / 1.28  400
CCEs per second
Table 4.3.5.1.1.1.1.1.2 Analysis on PDCCH paging load overheads
Channel Resource PDCCH CCE/s
Resource Load Amount 400
Resource Load Percentage (FDD) 10 MHz 0.98%
20 MHz 0.48%
In case of 3PDCCH symbols, when 2T2R is used, the number of CCEs is 41 at a 10

MHz bandwidth and the number of CCEs is 84 at a bandwidth of 20 MHz.
Compared with other data services, paging services have a higher

priority. High-priority paging services reduce signalings for downlink
scheduling distribution and uplink scheduling authorization during
PDCCH transmission. Therefore, it is good practice to control PDCCH
resources consumed by paging services to an appropriate range.
CPDCCHload indicates the limitation on paging performance imposed by
PDCCH loads.
Assume that the paging arrival rate matches Poisson distribution. The P SA
is calculated as follows:
 I pa ge,PO
PSA  1  e
According to the preceding analysis, four CCEs of each frame are

required on the average to transmit a paging scheduling assignment
message. The average number of PDCCH CCEs occupied per frame
(nCCE, frame) is calculated as follows:
 I p a ge,PO
nCCE , frame  4nPO , frame
（ 1 e ）
The PDCCH paging load is the proportion of the number of CCEs

distributed for transmission and paging to the number of CCEs per
frame. Figure 4.3.5.1.1.1.1.2 lists the number of CCEs in each frame in a
3GPP-defined 2T2R system.
Figure 4.3.5.1.1.1.1.2 Number of CCEs in each frame in a 3GPP-defined 2T2R

system
LPDCCH ,max
defined by customers is usually 1% or less. In this case,
I page, PO
can be obtained and the PDCCH paging load per frame can be
calculated.
 PDCCH paging capacity on the eNodeB side (nB is set to 1)
nCCE, frame,total * LPDCCH ,max 840 * 0.001

C PDCCHload  100nPO , frame * ln [ 1  ]  100 * 1* ln [ 1  ]  pages / s
4nPO , frame 4* 1
Paging capacity and CPU load

The eNodeB assigns CPU resources to paging service while ensuring other
services are properly processed. The assigned CPU resources determine
paging performance.
When planning paging areas, ensure that the frequency at which paging
messages reach the eNodeB does not exceed the CPU paging load.
CCPU indicates the limitation on paging performance imposed by CPU loads.
CCPU  500pages / s
The maximum CPU usage of Huawei eNodeB is 60%. According to the traffic model,
the number of paging messages that can be processed per second on an LTE network
is 500.
Paging capacity and paging blocking

The number of paging records that can be transmitted in a PO is limited by the
maxNoOfPagingRecords parameter. In case of a low blocking rate, paging
matches Boisson distribution and the paging capacity determined by the
blocking rate is calculated as follows:
R
R max Cblocking
 (R
Cblockin g,PO ,PO
Rmax  e * max R)
R!
Pblocking,max  1  R 0
C blocking,PO
Where,
Cblocking, PO
− : indicates the paging capacity affected by blocking
(paging messages in each PO).
Pblocking,max
− : indicates the paging blocking possibility defined by
operators and is not greater than 2% in most cases.
−
Rmax : indicates the maxNoOfPagingRecords parameter.
According to the preceding formula, on a 20 MHz bandwidth (with
Cblocking, PO Pblocking,max
maxNoOfPagingRecords set to 16), is 12 when
is 2%.
Paging block increases the paging delay and the duration for data connection.
Therefore, you are advised to limit the number of paging messages blocked
due to PDSCH limitations.
Cblocking
indicates the limitation on paging performance imposed by paging
blocking.
 Paging capacity of blocking on the eNodeB side (nB is set to 1)
nB
Cblocking  Cblocking,PO * 100n PO , frame  Cblocking,PO * 100  12 * 100 * 1  1200 pages / s
T
Planning Paging Areas Subject to Paging Capacity

eNodeB paging performance is the minimum among the preceding four
parameters.
C eNB  min(C CPU，C PDSCHload ，C blocking，C PDCCHload )  min(500,11 11,1200，)  500pages/s
In densely-populated urban areas, for example, Shanghai, 2000 sites cover 10

million UEs. According to network planning, the average penetration rate in
urban areas is 40%. In this case, the average number of UEs per site is 2000.
The average traffic model value per user is as follows:
2.17
 0.0006027
3600
(sub*s)
In this case, the number of eNodeBs in a TAL is limited due to eNodeB

paging performance. The calculation formula is as follows:
C eNB 500
n eNB, TAlist    414
N* 0.0006027 2000 * 0.0006027
The result of the preceding formula is obtained based on products' maximum

capacity. Generally, due to differences in traffic models on actual networks, it
is good practice to reserve 30% of paging margins. In this case, the number of
eNodeBs contained in a TAL is as follows:
n eNB, TAlist  414 * 0.7  289
The preceding description provides a method for calculating the number of eNodeBs
in a TAL and the result of the preceding formula is for reference only.
In suburban areas, the number of UEs is less than that in urban areas.
According to network planning in suburban and rural areas, the average UE
penetration rate is 20%. In this case, the number of UEs covered by a single
site is set to 1000. The number of eNodeBs in a TAL is as follows:
C eNB 500
n eNB, TAlist    829
N* 0.0006027 1000 * 0.0006027
Similarly, 30% of paging margins are reserved. The number of eNodeBs in a

TAL in suburban areas is as follows:
n eNB, TAlist  829 * 0.7  580
The preceding description provides a method for calculating the number of eNodeBs
in a TAL and the result of the preceding formula is for reference only.
According to preceding analysis, the number of eNodeBs in a TAL on the

eNodeB is subject to the following factors:
 Paging capacity
 Traffic model
 Number of UEs
The preceding values are obtained based on typical Huawei traffic models that
are greatly different from the number of UEs covered by a single site.
4.4 TA/TAL Planning

4.4.1 Planning Preparations and Outputs
Before TA/TAL planning, collect engineering parameters of LTE sites and
detailed digital maps.
 If GSM/UMTS LAC boundaries are inherited, collect engineering
parameters for GSM/UMTS sites and LAC planning information.
 Collect the predicted number of UEs covered on an LTE network and the
site scale after subsequent capacity expansion by cooperating with
customers and then roughly calculate the TAL scale.
 Predict the paging model on a LTE network, that is, number of paging
times for each UE during peak hours, to estimate the TAL scale.
 If the EPC is from other vendors, confirm with the customer whether the
TAL can be configured and the TA/TAL scale on the EPC.
 If multiple vendors deploy eNodeBs in an area and no obvious hard
boundaries are provided, collect TA/TAL configurations in other
scenarios. For details about planning principles, see "Planning Procedure
and Key Technologies."
After TA/TAL planning is complete based on the preceding information,
provide detailed TAC and TAL planning results as guidance for maintenance
personnel to configure eNodeBs and MMEs. Figure 4.4.1.1.1.1.1.1 shows
examples of TAC and TAL planning results.
Figure 4.4.1.1.1.1.1.1 Examples of TAC and TAL planning results
4.4.2 TA/TAL Planning Principles

Basic Principles for TAL Planning
Basic principles for TAL planning are as follows:
 In a static TAL solution, TAs in a TAL do not overlap.
 TAs are relatively small so that the TAs can be flexibly configured during
subsequent TAL optimization. In addition, GSM/UMTS LTC boundaries
can be inherited.
− A TA contains 30 to 70 eNodeBs and a TAL contains 3 to 10 TAs.
− At the initial stage of LTE deployment, the LTE TAL aligns with the
GSM/UMTS LA because LAC boundaries are optimized.
 If combined attach is used, LAUs can be reduced after the UE is handed
over to a UMTS network when CSFB services are required. In this case,
the TAL should align with the GSM/UMTS LAC boundary. Multiple
TALs can correspond to one LAC. One TA can correspond to one LAC.
Perform similar mapping configurations on the EPC and ensure that
inter-MSC or MSC-pool TALs do not exist.
 Configure TAs and TALs in specified areas to avoid unnecessary TAU
signaling overheads.
− Ensure that locations with frequent TAU signaling overheads are in a
light-traffic area. In this way, the eNodeB can have sufficient
resources to process extra TAU signaling overheads.
− Take UE movements into consideration in planning. (It is
recommended that TAL boundaries be not located in heavy-traffic
areas such as trunks and railways.)
 The TAL scope does not exceed the serving GW (SGW) area. Otherwise,
frequent ping-pong handovers may occur on the SGW.
 The TAL scope does not exceed the LA. Otherwise, extra LAUs may be
caused by CSFB calls, increasing the CSFB delay. For example, if a TAL
corresponds to multiple LACs, a 1s to 2s LAU delay may occur after a
CSFB is performed.
The TAL scale refers to the number of eNodeBs in a TAL. If a TAL contains a
small number of eNodeBs, TAUs are performed frequently, increasing MME
loads and UE power consumption. In addition, UEs fail to respond to paging
due to frequent TAUs, reducing the paging success rate. If a TAL contains
multiple eNodeBs, paging loads increase.
Therefore, the number of eNodeBs in a TAL depends on paging capacity of
the MME and eNodeB and considers balance between the TAU frequency and
the paging load.
Inter-MME or MME-pool TAs/TALs are not recommended. Otherwise,
location updates fail during UE movement between MMEs or MME pools,
causing location update failures.
TAL Scale
Number of eNodeBs in TAs in a TAL = Min (Number of eNodeBs in TAs in a
TAL with MME performance considered, Number of eNodeBs in TAs in a
TAL with eNodeB performance considered)
For details, see section "Analyzing Paging Capacity on the MME and eNodeB
Sides."
To estimate the TAL, the following parameters are required:
 MME paging specifications: number of boards in an MME and board
paging specifications
 eNodeB paging specifications: eNodeB CPU load for processing paging

messages, PDCCH and PDSCH load overheads allowed for processing
paging messages, and paging blocking rate defined by operators
 Number of UEs attached at the same time in the planned coverage scope
of MMEs
 Peak-hour paging arrival rate determined by the MME paging traffic
model
 UE types supported by eNodeBs and number of UEs under each eNodeB
 Peak-hour paging traffic models of UEs in various types on the access
network
Network traffic models vary with regions and time, paging specifications supported
by devices vary with versions, and the number of online UEs varies with time.
Solution for Planning TAL Boundaries

According to network planning experience, observe the following principles
during TAL planning:
 The TAL scale (maximum number of eNodeBs/cells in a TAL) is limited
by the capacity of paging channels and varies with traffic. The number of
eNodeBs in a TAL is limited due to products' paging specifications and
capacity. Considering subsequent capacity expansion, it is good practice
to reserve a 20% margin in initial network topology planning.
 At the initial stage of LTE network deployment, traffic is not heavy. Due
to many unknown factors, the network is to be expanded and adjusted
subsequently. It is normal to adjust the TAL. Therefore, it is necessary to
monitor traffic volumes and paging capacity in the paging area.
 In case of non-continuous coverage between urban and suburban areas,
an independent TAL is used in a suburban area.
Hotspot coverage is the focus at the initial stage of LTE network
deployment. In non-continuous coverage between suburban areas and
urban areas, the +E-UTRAN Deactivate ISR Timer may time out when
the UE implements the TAU periodically. In this case, the system regards
that the IMSI is detached implicitly (The MME delete context
information about the UE including the TAL, TAI of last TAU, and
GUTI). If UEs move to the urban area and the TAL is the same between
the urban area and the original suburban area, some UEs do not perform
location updates immediately, but they are not in the serving area and
have signals. For details, see GSM/UMTS network planning experience.
To solve this problem, an independent TAL is used in a suburban area. In
this case, TALs are distributed in a concentric manner. (In the OL urban
area, multiple TALs may exist due to paging capacity and these TALs
can be distributed in segments, in a concentric manner, or in a hybrid
mode.)In addition, this greatly improves the call connection rate and
paging success rate. Figure 4.4.2.1.1.1.1.1 shows TAL distribution in
case of non-continuous coverage between suburban areas and urban
areas.
Figure 4.4.2.1.1.1.1.1 TAL distribution in case of non-continuous coverage

between suburban areas and urban areas
County 1 County 2
Urban area
County 3
County 4
County n
 TAL planning is performed in a geographically continuous area and

mixed networking should be avoided if possible. TALs are divided based
on the distance, which ensures geographically continuous coverage. In
actual operations, TAL boundaries can be slightly adjusted based on
analysis on LAU traffic statistics of the original network by using
Google Earth.
 A TAL is within one MME coverage scope.
According to protocols, when several MMEs share one HSS, a TAL can
cross MMEs, that is, contain eNodeBs under different MMEs. However,
this situation is impossible in actual activities because in this case paging
is implemented over at least two routes on at least two MMEs. In most
cases, one MME is bound to one HSS. Therefore, the TAC or TAL
cannot cross MMEs, that is, only eNodeBs under the same MME can be
contained in the same TAC or TAL. Upon paging, the MME needs to
query the HHS to obtain the TAC of the UE and then paging packets can
be delivered to the corresponding TAL and TAC.
 Mountains and rivers are used as TAL boundaries, which reduces overlap
depth of cells in different TALs and minimizes costs for location updates
at the TAL edge.
Areas to be planned are greatly different and the GIS system cannot
identify an area based on clutters or terrains. In this case, TALs are
planned based on obvious mountains and rivers according to
administrative areas. Planning results need to be adjusted as required.
In heavy-traffic metropolises, mountains and rivers can be used as
boundaries of TALs, which reduces overlap depth of cells in different
TALs. If no mountains or rivers exist, TALs should not be separated by
streets and TAL boundaries should not be located at heavy-traffic areas
such as shopping malls. Generally, the TAL boundary is skewed with the
street. At boundaries between urban and suburban areas, the TAL
boundary is located at eNodeBs in suburban areas instead of at the
borders of urban and suburban areas. This avoids frequent location

updates for UEs at the borders of urban and suburban areas.
 During subsequent capacity expansion, if multiple frequency bands are
used on an LTE network, TALs can be planned based on frequency bands
and geographical positions.
If the 2.1 GHz frequency and 2.6 GHz frequency band use different
MMEs, TAs of the MMEs are also different. In this case, set related
parameters to enable UEs to camp on cells at the same frequency band,
reducing handovers and reselections between the two frequency bands.
In addition, consider loads on the system due to location updates when
designing signaling channels.
If the 2.1 GHz frequency and 2.6 GHz frequency band share one MME,
one TAL is recommended if system capacity permits. If two or more
TALs are required due to insufficient paging capacity, TALs can be
planned based on frequency bands and geographical positions.
Figure 4.4.2.1.1.1.1.2 shows TAL planning based on frequency bands.
Figure 4.4.2.1.1.1.1.2 TAL planning based on frequency bands
2.1 GHz cell 2.1 GHz cell 2.1 GHz cell
2.6 GHz cell 2.6 GHz cell 2.6 GHz cell 2.6 GHz cell
In this scenario, location updates are frequent due to handovers and

reselections between multiple frequency bands. In this case, set related
parameters to enable UEs to camp on cells at the same frequency band,
reducing handovers and reselections between the two frequency bands. In
addition, consider loads on the system due to location updates when designing
signaling channels.
Figure 4.4.2.1.1.1.1.3 shows TAL planning based on geographical positions.
Figure 4.4.2.1.1.1.1.3 TAL planning based on geographical positions
2.1 GHz cell 2.1 GHz cell 2.1 GHz cell 2.1 GHz cell
2.6 GHz cell 2.6 GHz cell 2.6 GHz cell 2.6 GHz cell 2.6 GHz cell 2.6 GHz cell
If TALs are planned based on geographical positions, frequent location

updates due to dual-frequency-band handovers and reselections can be
avoided. Data on the original network needs to be changed before capacity
expansion. In addition, location updates are caused by intra-frequency and
dual-frequency handovers and reselections at the TAL boundary with heavy
signaling traffic. Therefore, the TAL boundary needs to be properly designed.
4.4.3 Scenario-based TA/TAL Planning Solutions

High-speed UE Mobility in Dense Urban Areas
In dense urban areas, radio devices abound. The paging area should be as
small as possible to keep paging loads unchanged. However, in this scenario
some UEs move at a high speed at railways or bullet trains during peak hours
on working days, increasing signaling loads on the EPC. These UEs should be
paged in a larger area.
APs Installed in Home Environments – Forbidden Tracking Area

When APs are installed in home environments, the number of UEs is limited.
In this case, an independent TAL is required. In this way, some UEs that are
not allowed to access the APs are rejected during TAUs that are required when
these UEs are reselected to the APs from macro cells.
MOCN
In case of MOCN, that is, RNAs of multiple operators are connected to the
same EPC, hard boundaries should be planned between the mobility areas of
multi-operator RANs. However, if the mechanism of paging combination and
classification is available between different nodes of the EPC, hard boundaries
are not required. Hard boundaries are also required on inter-RAT networks
where the MME and the SGSN are provided by different vendors.
Classification of Paging Scenarios and TAL Planning Solutions

Model 1 Los Angles
In this model, mobility management functions in densely-populated areas
such as Islands connected over dedicated freeways. UEs in different areas
move among different areas every day. For example, commuters go to areas
A, B, and C over area D, as shown in Figure 4.4.3.1.1.1.1.1.
Figure 4.4.3.1.1.1.1.1 islands connected by freeways, with commuter routes
 In area A, residence communities prevail with many UEs sparsely

distributed that rarely move. In addition, fixed phones are popular and
home APs may be installed in some families in this area. Therefore,
potential paging capacity is small and an independent TAL is required. If
paging loads are heavy according to statistics, multiple TALs can be
planned based on population density in this area. TAU loads will not be
large because UEs rarely move in this area.
 In areas B and C, commercial districts abound with many UEs densely
distributed. UE mobility is of medium level and UE usage is high.
Potential paging capacity is large. Therefore, several TALs are required
and the TALs should be properly planned as required. This can decrease
paging loads but may increase TAU loads.
 In area D, there is a freeway connecting areas A, B, and C. Due to
geographical limitations, the number of UEs increases within a specified
period of time, UEs move frequently, and UE usage is of medium level.
Area D contains cells that must be passed when UEs move from area A
to area B or C. Therefore, you need to pay special attention when
planning TALs and prevent transient paging and peak TAU loads if
necessary.
According to preceding analysis, two TAL planning solutions are provided, as
shown in Figure 4.4.3.1.1.1.1.2.
Figure 4.4.3.1.1.1.1.2 Comparison of two TAL planning solutions in model 1 Los

Angles
Solution 1 Solution 2
Solution 1 is based on Qualcomm proposal. In this solution, each of the three

areas has a TA. In this case, when many UEs move between TA1 and TA2 or
TA3, TAU loads in the three middle cells at the boundary of the three areas
easily reach the transient peak if each UE is associated with a TA. If each UE
is associated with two TAs, TAU loads are reduced but paging loads may
reach the transient peak. Therefore, solution 1 is of great disadvantages.
Solution 2 is proposed to avoid disadvantages in solution 1. Multiple TAs are
planned in each area as required. The three middle cells at the boundary of the
three areas are in three independent TAs (TA 1, TA4, and TA8 in solution 2 in
Figure 4.4.3.1.1.1.1.2). UEs in area A move towards area D in different
directions. TA1 contains buffer cells in different directions. In this case, when
UEs enter TA1, TAU loads are distributed in different buffer cells. After UEs
enter TA1, UEs move at a high speed. To avoid frequent TAUs, associate TA1,
TA4, and TA8 in a TAL, which also obtains balance between paging loads and
TAU loads.
Model 2 London
In this model, areas are similar in geographical and UE distribution. Each area
contains residences and working places. Therefore, mobility management is
performed in each area.
Figure 4.4.3.1.1.1.1.3 London: equivalent areas
Two TA/TAL planning solutions are provided in this area, as shown in Figure
4.4.3.1.1.1.1.4.
Figure 4.4.3.1.1.1.1.4 Comparison of two TA/TAL planning solutions in model 2

London
Solution 1 Solution 2
Disadvantages in solution 1 in model 1 also lie in solution 1 in this model.

Similarly, solution 2 is provided to avoid disadvantages in solution 1, such as
frequent TAUs of UEs at the boundary of TAs, reducing UE battery
consumption and obtaining balance between paging loads and TAU loads. In
addition, solution 2 requires better TAU policies and higher processing
capabilities.
Model 3 Registration in Densely-populated Area – Red
In this scenario, densely-populated metropolises such as Tokyo in Japan are

the focus. In such areas, population density greatly differs during the daytime
and nighttime. In the morning, many UEs move to the central urban area on
trams. This high-speed mobility brings sudden registered communication
services near the central urban area (for example Tokyo city belt highways).
New registered communication services will occupy resources for CS or PS
services, greatly degrading the quality of user access services in this area.
The following are two examples:
 Eurostar TGV between Paris and Lyon, 393 m long, runs at a speed of
300 km/h and accommodates 784 passengers.
 Japan's Shinkansen, 480 m long, runs at a speed of 300 km/h and
accommodates 1300 passengers.
Figure 4.4.3.1.1.1.1.5 TAU storms in the RED scenario
TA2 TAU storm TA3 TAU storm TA4
TA1,TA2 TA2,TA3 TA3,TA4
As shown in Figure 4.4.3.1.1.1.1.5, all UEs in each TA are in the same TAL.
For example, UEs in TA2 are in TA1 and TA2 that are in the same TAL and
UEs in TA3 are in TA2 and TA3 that are in the same TAL. All UEs initiate
TAU requests during a short period of time at TA boundaries, resulting in peak
TAU loads on the MME and eNodeB. For example, when a Eurostar or
Shinkansen passes a TA boundary, a TAU request is sent every 6 ms or 4.4 ms,
respectively.
To address TAU storms, the UE-based TAL assignment policy can be used,
that is, the MME assigns different TALs to UEs in the same TA. As shown in
Figure 4.4.3.1.1.1.1.6, UEs are divided into two groups. UEs in different
groups are assigned with different TALs. In this case, only half of UEs at the
TA boundary need to initiate TAU requests, ensuring the service quality to a
certain degree. This solution is called a UE-level dynamic TAL solution,
which is still under research at present.
Figure 4.4.3.1.1.1.1.6 UE-based TAL assignment policy
TA2 TA3 TA4
TA0,TA1,TA2 TA2,TA3,TA4 TA2,TA3,TA4
TA1, TA2,TA3 TA1, TA2,TA3 TA3, TA4,TA5

Multi-frequency-band Scenarios
When multiple frequency bands (for example, 2.1 GHz and 2.6 GHz) are used
on an LTE network, TALs can be planned based on frequency bands and
geographical positions.
If 2.1 GHz and 2.6 GHz frequency bands use different MMEs, their TAs are
also different. In this case, set radio parameters to enable UEs to camp in cells
at the same frequency band if possible, which can reduce dual-frequency
handovers and reselections.
If 2.1 GHz and 2.6 GHz frequency bands share one MME, one TAL is
recommended if system capacity permits. If two or more TALs are required
due to insufficient paging capacity, TALs can be planned based on frequency
bands and geographical positions.
Figure 4.4.3.1.1.1.1.7 Planning TALs in inter-frequency cells based on frequency

bands and geographical positions
Based on Based on
frequency bands geographical positions
If TALs are planned based on frequency bands, frequent TAUs occur due to
handovers and reselections of multi-frequency cells. In this case, related
parameters need to be set properly so that UEs can camp in cells at the same
frequency band. If TALs are planned based on geographical positions,
frequency TAUs caused by dual-frequency handovers and reselections can be
avoided. However, radio parameters of the original network need to be
modified before capacity expansion. In addition, location updates are caused
by intra-frequency and dual-frequency handovers and reselections at TAL
boundaries and signaling loads are heavy at the TAL boundaries. Therefore,
TAL boundaries need to be properly designed.
4.4.4 Suggestions for TA/TAL Planning Baselines

Traffic and paging models on commercial LTE networks do not provide
uniform baseline reference values because UE and paging models greatly
differ in different LTE offices. Compared with eRAN3.0 eNodeBs,
eRAN6.0/7.0 eNodeBs improve by 40% of paging specifications. However, at
present, TA/TAL planning is based on suggestions in eRAN3.0. Currently,
because the number of eNodeBs in most offices is within 300, original
GSM/UMTS LAC boundaries can be inherited. For details, see Table
4.4.4.1.1.1.1.1.1.
Table 4.4.4.1.1.1.1.1.1 Suggestions for TA/TAL planning in different scenarios
Scenario TA Scale (Number of TAL Scale (Number of

eNodeBs in a TA) eNodeBs and TAs in a
A Maximum of 100 TAL)
eNodeB in a TA on the A Maximum of 16 TAs in a
USN Static TAL Solution on an
MME
Urban areas 30–50 150–300 eNodeBs/3–10
TAs
Suburban and 50–70 200–580 eNodeBs/3–12
rural areas TAs
 In dense urban or urban areas, it is good practice to configure about 200 (fewer
than 300) eNodeBs in a TAL.
 In suburban or rural areas, it is good practice to configure about 300 (fewer than
580) eNodeBs in a TAL.
 If an LTE network is deployed on a GSM/UMTS network, operators require that

GSM/UMTS location area planning be inherited, that is LTE TAL boundaries
align with GSM/UMTS LA boundaries. The number of TAs in a TAL can be
obtained by referring to Table 4.4.4.1.1.1.1.1.1.
 In actual network planning, the scales of TAs and TALs can be estimated by
entering different planning conditions based on operators' settings, the actual
number of UEs, and traffic models, which is supported by prototype tools.
 At the initial stage of LTE network deployment, the number of UEs is small. It is
good practice to plan 5 to 10 TAs in a TAL. (According to preceding description, a
TA contains a small number of cells and usually 30 to 50 eNodeBs are contained
in a TA.)As network loads increase, reduce the number of TAs in the TAL to
ensure that TA configurations are not affected and ongoing services are not
interrupted by TA configurations.
 If GSM LACs are inherited in which fewer than 100 eNodeBs are contained, the
TAL scale should also be small to be consistent with the LAC scale.
 If Huawei PS 9.1 EPC is used on the live network, a maximum of 128 eNdoeBs
can be configured in each TA and the recommended number of eNodeBs is 100.
The preceding limitations are canceled in PS 10.1.
4.4.5 Criteria for Evaluating the TAL Planning Solution

TAU Load (Registration Load)
During a TAU, resources are required for initiating an RACH request in the
uplink and providing a response in the downlink, resulting in UE power
consumption. Therefore, fewer TAUs initiated by a UE during a specified
period of time indicate smaller TAU loads. When UEs in a cell initiate many
TAUs, the cell capability of providing services for UEs will be limited due to
RACH loads. Therefore, registration loads should be similar in different cells
to prevent large load peaks.
Paging Load
The S1-MME port and air interface are required in paging capacity analysis.
The MME needs to transmit UE paging messages to all eNodeBs in the TALs
associated with the UEs over the S1-MME port. Consequently, if the number
of UEs to be paged is large in a specified period of time, the number of
eNodeBs in TALs associated with the UEs is large, resulting in heavy loads
over the S1-MME port. If a paging area covers many cells, many air interface
resources will be occupied by paging messages of the same UE. If the number
of UEs to be paged in a paging cycle is large, paging loads at the air interface
are heavy.
The size of a paging area is dynamically managed by an MME and the paging
mechanism is also designed by the MME. During a paging process, the eNodeB maps
paging messages delivered by the MME to specified paging occasions of UEs based
on certain rules to ensure that UEs read specified-occasion paging messages in a
paging cycle, which reduces UE power consumption.
Signaling Bandwidth
The number of TACs delivered by the eNodeB affects capacity and
performance of the eNodeB. Available radio resources on paging channels of
air interfaces vary with bandwidths.
In the TAL solution, only one TAC is delivered by the eNodeB, which does
not greatly affect eNodeB performance.
UE Ping-Pong Effect Causing High EPC Loads

The TAL solution reduces the impact of ping-pong TAUs as much as possible
to effectively control EPC loads and reduce UE power consumption.
In the TAL solution, paging areas of UEs can be configured flexibly by
updating TALs. No special attention is required for configurations of TA
boundaries.
UE and Network Complexity

After the UE obtains TACs broadcast by the eNodeB, the UE compares the
received TACs with the TACs in the TAL and then determines whether to
initiate a TAU request.
In the TAL solution, compare the number of TACs (x) assigned to a UE and
one TAC received by the UE in the corresponding cell. The value of x should
be controlled to reduce signaling overheads and UE design complexity. In this
condition, paging performance and the size of a paging area are also limited.
The MME manages the TAL of UEs to ensure paging performance, which
imposes high requirements on performance on the EPC. According to
protocols, the maximum number of TALs is 16.
Air interface signaling loads including broadcasting channel loads, air
interface paging loads, and RACH loads affect solution selection and
performance. Other factors are soft limitations. For example, paging loads at
the S1-MME port can be reduced by configuring wider bandwidths. However,
technical difficulties and cost increase should also be considered.
Performance Counters
When selecting the TA size and management policies of TALs, ensure balance
between paging loads and TAU loads.
The TAU is a procedure in the NAS signaling process. No counters are
provided on the RAN side to directly measure TAU loads. The TAU loads can
be deduced by using one of the following methods:
1. Compare the number of times for cells receiving RRC Connection
Request sent from the UE with a setup cause of
L.RRC.ConnReq.Att.MoSig and the number of times for cells
receiving RRC Connection Request sent from the UE with a setup cause
of L.RRC.ConnReq.Msg. If the absolute value of the former is high and
the former outnumbers the latter, check whether TAL boundaries exist.
2. When UEs initiate TAUs, RACH requests are sent in the uplink and
RAR responses are provided in the downlink, which occupies resources.
If many UEs implement TAUs due to inappropriate TAUs, preambles
including L.RA.GrpA.Att and L.RA.GrpB.Att will be received more
frequently.
3. A large TAL scale easily leads to high paging loads, which may exceed
eNodeB paging specifications. In this case, paging messages are
discarded or the paging delay prolongs. The following paging-related
counters are recorded on the current eNodeBs:
− L.Paging.S1.Rx: indicates the number of times for cells receiving
paging messages over the S1 interface.
− L.Paging.UU.Succ: indicates the number of times for cells receiving
responses to UE paging messages over the Uu interface.
Compare L.Paging.S1.Rx per hour and corresponding products'
specifications (for details, see section 2.2) and check whether paging
loads are high.
 You are advised to observe the preceding counters during peak hours. In addition,
because traffic fluctuates every day and greatly varies on working days and
weekends, hour-level traffic statistics is recommended. The average data during
peak hours per day or the maximum data per day can be used for reference. (Data
during peak hours per day is hour-level traffic statistics during peak-hours per
day.)
 Paging on the EPC is performed based on TAs or TALs. Therefore, the number of
paging times for eNodeBs in one TA is similar. The paging times of the eNodeB
that forwards the most paging messages in a TA/TAL are the maximum paging
times in the TA/TAL (L.Paging.S1.Rx).
 If frequent TAUs are caused by inappropriate TAL boundaries, it is good practice

to observe Top L.RRC.ConnReq.Att.MoSig or L.RA.GrpA.Att during peak hours.
4.5 Checking and Evaluating TAs and TALs

4.5.1 Application Scenarios
TAs and TALs are checked and evaluated in the following scenarios:
 For new networks, TAs and TALs are planned by customers. Before
delivering configuration data to NEs, check TA/TAL configurations to
prevent network performance problems due to incorrect TA/TAL
configurations.
 For expanded networks, check TA/TAL configurations on existing
networks and consider whether TA/TAL configurations need to be
adjusted after new eNodeBs are added.
 For commercial networks, check and evaluate network configuration
data on a regular basis to promptly discover incorrect TA/TAL
configurations, facilitating network optimization.
4.5.2 Preparations
Before checking and evaluating the TA/TAL, perform the following
operations:
 Obtain eNodeBs on the current network, engineering parameters of cells,
and TAC/TAL configurations of the current network. (TAL
configurations are obtained from the EPC or provided by customers).
 Ensure that U-Net V3R10 or later is available.
 On new networks, determine the predicted number of UEs covered by
networks, traffic models, and paging models with customers.
 Determine eNodeBs to be expanded and cell engineering parameters
before network capacity expansion.
 Customize traffic counters on the M2000 and EPC for commercial
networks.
Counters that need to be customized on the M2000 are as follows:
− L.Paging.S1.Rx: indicates the number of times for cells receiving
paging messages over the S1 interface.
− L.Traffic.User.Max: indicates the number of online UEs in cells.
Counters that need to be customized on the EPC are as follows:
− Number of UEs using the Attach function on the entire network
during peak hours
− Number of paging times on the entire network during peak hours
Based on preceding two counters, obtain the paging model by using the
following formula:
Paging model (paging/per User&HB) = Number of paging times on the
entire network during peak hours/Number of attached UEs on the entire
network during peak hours
− Number of UEs in ECM-idle mode during peak hours
− Number of UEs in ECM-Connected mode during peak hours
Based on the preceding two counters, obtain the proportion of the
number of UEs in idle mode to the number of UEs in connected mode by
Number of UEs in idle mode/Number of UEs in connected mode =
Number of UEs in ECM-idle mode/Number of UEs in ECM-Connected
mode
4.5.3 Geographically Displaying and Checking

TAs/TALs
Geographically Displaying TAs/TALs on the U-NetTo geographically display
TAs and TALs on the U-Net, perform the following steps:
Step 1 Start the U-Net and create an LTE project. Convert the formats of engineering
parameters of the live network into template formats on the site, transceiver,
and cell tab pages supported by the U-Net and import the engineering
parameters to the U-Net.
At present, the TAL field is unavailable in the NE model on the U-Net.
Therefore, you are advised to import TAL configurations of cells to the
Comment field in the Transceiver table.
Step 2 On the Project Explorer window, select Transceiver, right-click, and choose
Group By > Comment from the shortcut menu, as shown in Figure
4.5.3.1.1.2.1.1.
TAL configurations on the current network are geographically displayed, as
shown in Figure 4.5.3.1.1.2.1.2. Different TALs are marked in different
colors.
Figure 4.5.3.1.1.2.1.1 Operations of geographically displaying TALs

Figure 4.5.3.1.1.2.1.2 Geographically displayed TALs
Step 3 On the Project Explorer window, select Transceiver, right-click, and choose
Display Setting from the shortcut menu, as shown on the left side in Figure
4.5.3.1.1.3.1.1. On the Display Field dialog box, click Group Display tab
page. On the displayed tab page, select ValueIntervals under the Display
Type item and select LTE-FDDCell:TAC under the Field item, as shown on
the right side in Figure 4.5.3.1.1.3.1.1.
Figure 4.5.3.1.1.3.1.1 Operations of geographically displaying TAs

TAs are geographically displayed in different colors based on cell sizes and
shapes, as shown in Figure 4.5.3.1.1.3.1.2.
Figure 4.5.3.1.1.3.1.2 Geographically displayed TAs
TAC=xx TAC=xx TAC=xx

x01 x02
TAC=41 TAC=xx
x04
103
TAC=xx x05 TAC=xx
TAC=xx
x07 TAC=xx x06
TAC=xx
x08
TAC=xx x10 x09
TAC=xx
x12 x11
----End
Checking for Incorrect TAs/TALs

After TAs and TALs are geographically displayed, check for incorrect TACs.
 Incorrect TAL boundaries: TAL boundaries are discontinuous because of
incorrect TALs at a site. Figure 4.5.3.1.1.3.1.3 shows geographical
distribution of TALs at a site. Incorrect TACs are marked by black
circles. Because TAL boundaries correspond to TAU boundaries and the
paging scope. Abnormal TACs at these sites greatly affect network
performance and need to be manually confirmed and modified.
Figure 4.5.3.1.1.3.1.3 Checking for incorrect TACs by geographically displaying

TALs of an office
Incorrect TA boundaries: For example, if TACs in a cell of a site that are

not located at the boundary are inconsistent with TACs at adjacent sites,
they may be incorrectly configured by users. Generally, if a cell with
incorrect TACs and its neighboring cells are in the same TAL, network
performance is not affected because TAUs and paging are performed in
the unit of TAL. However, to facilitate maintenance, you are advised to

analyze causes of incorrect TAC configurations and it is good practice to
maintain continuous TAs. As shown in Figure 4.5.3.1.1.3.1.4, abnormal
TACs are marked by black circles but TACs ranging from XX01 to
XX12 are in the same TAL, thereby imposing no impact on network
performance.
Figure 4.5.3.1.1.3.1.4 Checking for incorrect TACs by geographically displaying

TAs of an office
Evaluating and Checking the TAL Scale

Evaluating the TAL Scale
Work out the allowed TAL scale by using the method for evaluating TA/TAL
baseline specifications described in section 4.2.3"Maturity."
2. For new networks, the number of UEs and paging models can be
obtained from the customers that own the networks. Paging models can
be obtained based on statistics of Huawei LTE networks. For details, see
the Appendix.
3. On expanded or commercial networks, paging models can be obtained by
using traffic statistics on live networks. For details, see section "Basic
Principles for TAL Planning."
Generally, when evaluating the TAL scale, reserve a 20% of margin for
capacity expansion. It is good practice to confirm the reserved margin with
customers.
Figure 4.5.3.1.1.3.3.1 shows how to evaluate the TAL scale at a site.
Figure 4.5.3.1.1.3.3.1 Example of evaluating the TAL scale at a site
Number of eNodeBs and UEs in Each TAL on the Current Network

The number of eNodeBs in a TAL can be obtained based on engineering
parameters and configuration data. The average/maximum number of UEs in
RRC connected mode per cell can be obtained based on traffic counters on the
M2000. Then the average/maximum number of UEs in RRC connected mode
per TAL can also be obtained.
Checking the TAL Scale
The TAL scale can be checked in two modes:
 Rough check: Compare the actual TAL scale and the evaluated TAL
scale. If the former is greater than the latter, provide warnings or prompt
users to split TALs and adjust TACs if necessary.
 Detailed check: Obtain the paging model, UE distribution, and UE
increase trend in each TAL. Provide warnings and adjustment
suggestions by using the TAL-based paging model, UE increase trend,
and predicted paging limitation time.
4.6 TA/TAL Replanning

4.6.1 Scenarios
TA/TAL replanning is performed in the following scenarios:
 During LTE TAL planning, LAC boundaries of GSM/UMTS networks
are inherited and the combined attach function is used on the EPC. In
this case, the mapping between TACs and LACs is configured on the
EPC. One TAC corresponds to one LAC. For example, TAC1 and TAC2
correspond to LAC1 and TAC3 and TAC4 correspond to LAC2. If
GSM/UMTS LACs are split or original LACs are changed due to
capacity expansion, original TAs/TALs are adjusted or replanned to
guarantee the Combined Attach function.
 At the initial stage of network construction, traffic increase and capacity
expansion are not thoroughly considered. As a result, as the number of
UEs increases, original TAs/TALs are inapplicable in most cases.
 New coverage scenarios are introduced along with the development of

cities. For example, if subways or elevated railways do not exist, it
would be difficult to consider the specific subway or elevated railway
lines at the initial stage of network construction. After subways or
elevated railways are set up, TAs/TALs along subway or elevated railway
stations may need to be replanned.
4.6.2 Principles
TA/TAL replanning principles are as follows:
1. If LTE TA/TAL replanning is triggered by LAC changes of GSM/UMTS
networks, principles are as follows:
− Ensure that TAL boundaries are aligned with LAC boundaries as
much as possible after adjustment and that one LAC can be mapped
to multiple TALs.
− If TAL boundaries cannot be aligned with LAC boundaries, adjust
configurations of TALs and TAs to reduce the number of sites whose
TAL boundaries are inconsistent with LAC boundaries and reduce
unnecessary LAUs as many as possible after a CSFB.
− Keep existing configurations including TAC configurations on the
EPC and eNodeB unchanged.
Example 1: Figure 4.6.2.1.1.1.1.1 shows TA/TAL adjustment due to
LAC changes. Customers adjust LAC1 and LAC2 configurations by
replanning some eNodeBs to LAC1 to reduce in the number of eNodeBs
in LAC2. Then assign a new TAC to these eNodeBs and add the new
TAC to TAL1. This ensures that TAL boundaries are aligned with LAC
boundaries.
Figure 4.6.2.1.1.1.1.1 TA/TAL adjustment due to LAC changes
The LAC1 scope increases and

the LAC2 scope decreases.
New TAC
Example 2: Figure 4.6.2.1.1.1.1.2 shows LTE TA/TAL replanning due to

LAC splitting. LAC1 is split into LAC1 and LAC3. Before LAC1 is
split, TAL1 aligns with LAC1. After LAC1 is split, TAL splitting and
replanning are required. TAL1 is split into TAL1 and TAL3,

corresponding to LAC1 and LAC3
Figure 4.6.2.1.1.1.1.2 LTE TA/TAL replanning due to LAC splitting
LAC1 is split into

LAC1 and LAC3.
TAL1 is split into TAL1 The TAL is split at

and TAL3, which the same time.
correspond to LAC1
and LAC3.
2. If TA/TAL replanning is caused by network capacity expansion or

increase in the number of UEs, principles are as follows:
− Collect traffic counters on the EPC and eRAN to obtain the traffic
model and UE increase trend and evaluate the TA/TAL replanning
scale based on the obtained information.
− If network-wide replanning is not mandatory, preferentially replan
such areas with the largest capacity expansion scale, the largest
number of UEs, or highest paging loads.
− During replanning, reserve some margins.
− The method for evaluating the TA/TAL scale and principles for
planning boundaries during replanning are the same as those during
initial planning.
3. If TA/TAL replanning is triggered by new coverage scenarios, principles
are as follows:
− eNodeBs along subways or railways are in an independent TAL.
− Refer to GSM/UMTS LAC planning along subway or elevated
railway stations and inherit GSM/UMTS LAC boundaries as many as
possible. Generally, if LTE TAs/TALs need to be adjusted,
GSM/UMTS LACs along stations also need to be adjusted.
If TA/TAL replanning is triggered by GSM/UMTS LAC adjustment, principles

described in "CSFB scenario" in "Special Networks" in section "Planning Procedure
and Key Technologies" should be observed.
4.7 TA/TAL Optimization

In eRAN6.0, TA/TAL optimization is implemented through a combination of
tool-assisted check and manual reconfiguration. When defining the actual
TA/TAL boundary, take factors such as terrain, ground objects, and hotspots
into consideration. Current solutions cannot achieve a balance between these
factors. The following are some basic optimization rules:
 If a TAL is too large, preferentially split the TAL, that is, reconfigure the
TAL.
 Ensure that one TA belongs to a unique TAL.
 Do not configure heavy-traffic areas and areas where UEs move
frequently around TAL boundaries.
 If CSFB services are required, align the split TAL boundary of the LTE
network with the LAC boundary of the GSM and UMTS network. The
boundaries of the union set of multiple TALs can be aligned with LAC
boundaries.
4.8 Application Guidance

In eRAN6.0, related TA/TAL planning tools are unavailable. This document
focuses on planning principles.
You are advised to use the Mapinfor or U-Net to help TA/TAL planning on the
live network. The geographical display and multi-layer control functions of
the Mapinfor or U-Net can be used. Figure 4.8.1.1.1.1.1.1 shows TALs
geographically displayed on the U-Net, in which different colors indicate
TALs. The TAL boundary and scale can be adjusted by using the geographical
display function, which facilitates planning and optimization.
Figure 4.8.1.1.1.1.1.1 TALs geographically displayed on the U-Net

For details about how to use the Mapinfor and U-Net, see related tool operation
guides.
Design (PCI, PRACH, Neighboring Cell, and TA&TAL) 5 Application Cases
5 Application Cases
5.1 PCI Planning

5.1.1 New Site Deployment
 Data: coordinate system (or electronic maps) and NE engineering
parameters
 Create a project, set the coordinate system or import the electronic map,
and then import the NE engineering parameters.
Geographical distribution of all cells on the network is displayed on the
project window of the U-Net. See the following figure.
Perform the PCI auto-planning.

Figure 5.1.1.1.1.1.1.1 Performing the PCI auto-planning
Figure 5.1.1.1.1.1.1.2 Specifying PCI planning parameters in the new site

deployment scenario
Figure 5.1.1.1.1.1.1.3 Specifying the PCI planning strategy

Figure 5.1.1.1.1.1.1.4 The planning result is displayed as follows:
Figure 5.1.1.1.1.1.1.5 Submit the planning result as shown in the following

figure:
5.1.2 Capacity Expansion Scenario

Capacity expansion involves mixed network capacity expansion and
greenfield capacity expansion.
Mixed network capacity expansion: adding E-UTRAN cells on planned sites
or adding E-UTRAN sites on the planned operating network.
Greenfield capacity expansion: adding E-UTRAN cells in sparely populated
areas, around which planned E-UTRAN cells exist.
Mixed network capacity expansion
a. Adding new sites b. Expanding capacity of cells served by

existing sites
Gray cells are new cells with no PCIs, and yellow cells are planned cells.
Figure 5.1.2.1.1.1.1.1 Greenfield capacity expansion
Green sites are new sites with no PCIs, and black sites are planned sites.
Operations for PCI planning in newly added cells using the U-Net in the
capacity expansion scenario are similar to those in the new site deployment
scenario. For details, see section 5.1.1"New Site Deployment." However, in
the capacity expansion scenario, Planning with exist PCI must be selected.
5.1.3 Replanning Scenario

Before the replanning, determine the cells to be re-planned and clear their
PCIs. Allocate new PCIs using the same method of PCI auto-planning in a
capacity expansion scenario and keep the PCIs of other cells unchanged.
NEs for Re-allocation Distributed in Continuous Areas

If a continuous area requires replanning (such as in the network swap
scenario), use a polygon to locate the area requiring PCI replanning as shown
in Figure 5.1.3.1.1.1.1.1 and perform PCI re-allocation in the area in the
polygon.
Figure 5.1.3.1.1.1.1.1 Replanning scenario (in a continuous area)

The E-UTRAN cells in the polygon require PCI replanning.

Use similar method of PCI planning for new site deployment: Deselect
Planning with exist PCI shown in Figure 5.1.1.1.1.1.1.2 to delete
corresponding PCIs in the cell table. In this way, you do not have to manually
clear PCIs for re-allocation. Set the planned area to a polygon, and perform
PCI planning. In this way, only PCIs of cells in the polygonal area are re-
allocated.
Figure 5.1.3.1.1.1.1.2 PCI replanning settings for a continuous area
NEs for Re-allocation Distributed in Non-continuous Areas

If the NEs requiring replanning are scattered on the network as shown in
Figure 5.1.3.1.1.1.1.3, clear the PCIs of cells to be planned, and re-allocate
PCIs using the PCI planning method for the capacity expansion scenario
described in section 5.1.2"Capacity Expansion Scenario." That is, allocate
PCIs only to cells to be planned and keep PCIs of other cells unchanged.
Figure 5.1.3.1.1.1.1.3 PCI replanning (NEs scattered on the network)

Gray cells require PCI re-allocation.
5.2 PCI Check

5.2.1 Checking PCI Conflicts
On a network, 4000 E-UTRAN cells are deployed and the PCIs are planned
by the third party. Our engineers use the U-Net to check whether PCI conflicts
exist among second tier neighboring cells.
As shown in Figure 5.2.1.1.1.1.1.1, the minimum number of PCI reuse tiers is
two. That is, PCI conflicts exist.
Figure 5.2.1.1.1.1.1.1 Checking PCI conflicts on the network

5.2.2 Evaluating PCI Planning

PCI Reuse Distance
On a live network, 9000 cells are deployed, and their PCIs are planned
manually in the initial phase. To increase the PCI reuse isolation, our
engineers use the U-Net to re-plan PCIs and compare the PCI reuse distance
before and after the PCI replanning. The reuse distance after the PCI
replanning increases, indicating that the PCI replanning takes effect.
As shown in Figure 5.2.2.1.1.1.1.1, collect statistics about PCI reuse distance
before and after the planning using the U-net, and display the PCI reuse in
CDF figures. PCI planning using the U-Net is better than manual PCI
planning. The reuse distance of U-Net-based planning is 2 km greater than
that of manual planning.
Figure 5.2.2.1.1.1.1.1 PCI reuse distance on the live network
PCI planned using the U-Net
PCI planned manually
PCI Mod3 Staggering

Import PCI Mod3 of a network into the U-Net, and observe whether PCI
Mod3 of opposite adjacent cells are staggered.
As shown in Figure 5.2.2.1.1.1.1.1, PCI Mod3 of red cells is 0, PCI Mod3 of
yellow cells is 1, and PCI Mod3 of blue cells is 2. PCI Mod3 of cells in red
circles is poorly staggered.
Manual analysis shows that some opposite adjacent cells use the same PCI
Mod3.
Re-plan the PCIs of E-UTRAN cells on the network using the U-net.
Conflicts among opposite adjacent cells decreases, and the RS SINR is good
when these cells are not loaded. That is, the U-Net-based planning result is
better than the PCI allocation before the replanning.
PCI Mod3 check on the live network
Before optimization After optimization
Figure 5.2.2.1.1.1.1.2 RS SINR before and after optimization when the

cells are not loaded
Before
optimization
After
optimization
RS SINR (dB) PDF (%)
Before After Optimization

Optimization
<=-5 4.331075 3.228378

(-5,0] 16.84602 15.61047
(0,5] 21.12395 20.6191
(5,10] 20.76525 21.64209
(10,15] 18.91856 19.74226
(15,20] 11.0934 11.70453
>=20 6.921748 7.453169
PCI Mod30 Staggering

Display E-UTRAN cells with the same PCI Mod30 geographically in the U-
Net as shown in Figure 5.2.2.1.1.1.1.3.
In the figure, red E-UTRAN cells have the same PCI Mod30.
In the figure, cells with the same PCI Mod30 are not adjacent cells. Then, it is
considered that the PCI Mod30 staggering meets the expectation.
Figure 5.2.2.1.1.1.1.3 PCI Mod30 on the live network
5.3 PRACH ZC Root Sequence Planning

5.3.1 New Site Deployment Scenario
 Data Preparation
− Coordinate system or electronic map
− NE engineering parameters
 Procedures
Step 1 Create a project, set the coordinate system or import the electronic map, and
then import the NE engineering parameters.
Geographical distribution of all cells on the network is displayed on the
project window of the U-Net. See the following figure.
Step 2 Perform PRACH ZC root sequence self-planning.
Step 3 Set the PRACH ZC root sequence planning strategy.

Available Root Sequence Index: indicates the available PRACH ZC root sequences.
Planning Prach based on existed Prach: indicates whether the ZC root sequence
allocated to the planned cell is unchanged.
Calculate Cell Radius: indicates whether the cell radius needs to be calculated.
Propagation Radius: indicates whether the cell radius is calculated based on the
propagation radius.
Propagation Radius Factor: indicates the proportion factor of the cell radius to the
propagation model radius.
Coverage Radius: indicates whether the cell radius is calculated based on cell
Resolution(m): indicates the resolution of coverage prediction.
Min Signal Level(dBm): indicates the minimum RX signal level for coverage
prediction.
Shadowing taken into account: indicates whether shadow fading is considered

during coverage prediction.
Cell Edge Coverage Probability(%): indicates the cell edge coverage probability for
shadow fading.
Indoor Coverage: indicates whether penetration loss is considered.
Area: indicates the PRACH ZC root sequence planning area.
The planning result is displayed as follows:

----End
5.3.2 Capacity Expansion Scenario

Capacity expansion involves mixed network capacity expansion and
greenfield capacity expansion.
Mixed network capacity expansion involves adding E-UTRAN cells on
planned sites or adding E-UTRAN sites on the planned operating network.
Greenfield capacity expansion involves adding E-UTRAN cells in sparely
populated areas, around which planned E-UTRAN cells exist.
Figure 5.3.2.1.1.1.1.1 and Figure 5.3.2.1.1.1.1.2 shows the two scenarios for
mixed network capacity expansion.
Figure 5.3.2.1.1.1.1.1 Adding new sites
Figure 5.3.2.1.1.1.1.2 Adding new cells to existing sites

Gray cells are new cells (Min root sequence idx is null), and yellow cells are planned
cells.
Figure 5.3.2.1.1.1.1.3 Greenfield capacity expansion
Green sites are new sites (Min root sequence idx is null), and black sites are planned
sites.
Operations for PRACH ZC root sequence planning in newly added cells using
U-Net in the capacity expansion scenario are similar to those in the new site
deployment scenario. For details, see section 5.3.1"New Site Deployment
Scenario." However, in the capacity expansion scenario, Planning Prach
Based on existed Prach must be selected.
5.3.3 Replanning Scenario

Before the replanning, spot the cells to be re-planned and clear the PRACH
ZC root sequences allocated to the cells to be planned. Allocate new PRACH
ZC root sequences using the same method of PCI self-planning in a capacity
expansion scenario and keep the PRACH ZC root sequences of other cells
unchanged.
NEs for Re-allocation Distributed in Continuous Areas

If some continuous area requires replanning (such as in the network swap
scenario), use a polygon to locate the area requiring PRACH ZC root
sequence replanning as shown in Figure 5.3.3.1.1.1.1.1 and perform PRACH
ZC root sequence reallocation in the area in the polygon.
Figure 5.3.3.1.1.1.1.1 Replanning scenario (in a continuous area)
Use a method similar to the method of ZC planning for new site deployment:
Do not select Planning Prach Based on existed Prach shown in Figure
5.3.3.1.1.1.1.2 to avoid manually clearing PRACH ZC root sequences for re-
allocation. Set Area to Polygon_Draw_1, and perform PRACH ZC root
sequence planning. In this way, PRACH ZC root sequences are reallocated
only to cells in the polygonal area.
Figure 5.3.3.1.1.1.1.2 PRACH ZC root sequence replanning parameter settings

for some continuous areas
NEs for Re-allocation Distributed in Non-continuous Areas

If the NEs requiring replanning are scattered on the network as shown in
Figure 5.3.3.1.1.1.1.3, clear the PRACH ZC root sequences of cells to be
planned, and re-allocate PRACH ZC root sequences using the PRACH ZC
root sequence planning method for the capacity expansion scenario described
in section 5.3.2"Capacity Expansion Scenario." That is, allocate PRACH ZC
root sequences only to cells to be re-planned and keep PRACH ZC root
sequences of other cells unchanged.
Figure 5.3.3.1.1.1.1.3 PRACH ZC root sequence replanning (NE scattered on the

network)
Gray cells require PRACH ZC root sequence reallocation.
5.3.4 Highway Speed Scenario

The U-Net V300R009 does not support ZC root sequence self-planning in
highway speed scenarios. Related parameters must be manually specified.
In highway speed scenarios, one ZC root sequence generates only one
preamble sequence. At present, a product in highway speed scenarios supports
a maximum of 20 preamble sequences. That is, one highway-speed cell
supports preamble sequences generated with only 20 ZC root sequences.
The number of ZC root sequences assigned to low-speed cells is fixed. See the
following figure.
Number of Preambles
Generated Per ZC
Root Sequence
Low-speed Cell Low-speed Cell Low-speed Cell
A cell can be configured as a low-speed cell using the U-Net. In this case, the
cell radius is greater than 22.8 km but not greater than 38.8 km and each cell
can be assigned with 22 ZC root sequences. (Note: A cell can be set as a high-
speed railway cell using the U-Net V300R009 so that ZC root sequence
planning in the cell is preferred. However, the number of allocated ZC root
sequences is the same as that in low-speed cells.)
In highway speed scenarios, the PRACH ZC root sequence planning
procedure is as follows:
Step 1 Set engineering parameters of highway-speed cells.
Step 2 Import the engineering parameter settings of common operating cells.
Step 3 (Optional) Import or generate neighboring cells for all cells.
Step 4 Obtain parameters for the simulation task.
Step 5 Set parameters for PRACH ZC root sequence planning.
Step 6 Check the planning result.
----End
Setting Engineering Parameters of Highway-speed Cells

Open the cell engineering parameter table in the U-Net and set High Speed
and Radius(m) to LowSpeed and 25000, respectively. In this case, each cell
occupies 22 root sequences. This is close to the configuration of 20 ZC root
sequences, which are supported by the product at most.
Importing the Engineering Parameter Settings of Common Operating

Cells
If intra-frequency cells have been configured on the network, obtain these
cells' engineering parameters related to root sequences. Engineering
parameters for highway-speed cells can be planned based on those for the
existing cells.
For example, Cell 21-1 is configured with the root sequences starting from 76,
as shown in the following figure.
(Optional) Importing or Generating Neighboring Cells for All Cells
If the actual neighboring cell relationship has been specified, import the actual
neighboring cell relationship for ZC root sequence planning.
Specify Prach Reuse Tier(Neighbor), which is set to 2 by default. During
the planning, exclude the ZC root sequence sets occupied by the neighboring
cells and their neighboring cells.
If the neighboring cell factor is not considered, do not import the neighboring
cell relationship or set the neighboring cell tier to 0. In this case, obtain the
highest score for reuse isolation when the product of the reuse distance and
topology tier is the same.
Calculation of the isolation distance and topology tier does not take the
azimuth and actual geographical propagation into consideration. Therefore,
the result is more accurate when the neighboring cell factor is considered.
For example, the distance between sites A and B is R. Draw an ellipse with R
as the major axis and K x R as the minor axis. The number of topology tiers
between sites A and B is the total number of eNodeBs in the ellipse excluding
the sites that have the same site address as sites A and B. The more the
number of topology tiers, the more the number of cells between sites A and B.
Therefore, sites A and B can use the same ZC root sequence.
Obtaining Parameters for the Simulation Task

Run a command (for example, Mo: Cell.CellRadius=XXm) to obtain the
actual cell coverage radius. Based on the obtained actual cell coverage radius,
query the range for the obtained cell radius in the following table.
Obtain the available ZC root sequences for the highway-speed cell based on
the cell radius by referring to the following table.
U-Net Simulation Data

Maximum Available ZC Root
Ultra-high- Radius of Sequences for
speed Cell Ultra-high- Ultra-high-speed
speed Cell Cell
Not supported
(recommended)
Not supported
(recommended)
Not supported
(recommended)
Not supported
For example, if the cell coverage radius of a highway-speed cell is 5 km, the
maximum cell radius is 5.52 according to query of the second column in the
table above, and the available ZC root sequence range is [76, 237][558, 777].
Setting Parameters for PRACH ZC Root Sequence Planning
 Available Root Sequence Index: Specify this parameter as 76-273, 558-

777 based on the available ZC root sequence range obtained in step 4.
 Planning Prach based on existed Prach: If there are operating intra-
frequency cells on the network before the ZC root sequence planning on the
high railway cells, obtain the engineering parameter configurations for the
root sequences of the operating cells and select this option.
 Radius: Do not select this option if the cell radius has been set in the cell
table.
 Area: Full Map is recommended.
Checking the Planning Result

The following result is obtained through theoretical analysis.
Assume that there are 99 cells, each cell uses 22 ZC root sequences, and the
available ZC root sequences are number 76 to 273 and number 558 to 777.
Then, there are 20 ZC root sequence groups.
The result is that each ZC root sequence is reused for four to five times.
For example,
As shown in the figure above, the cell with a circle at the arrow is cell 14_3.
Cells in red use the same ZC root sequence as cell 14_3.
If cell 14_3 must use a different ZC root sequence from its first-tier
neighboring cells (with red connections to cell 14_3 as shown in the following
figure), the cells encircled in blue with red connections to cell 14_3 as shown
in the following figure use the same ZC root sequence with cell 14_3.
PRACH ZC root sequence planning for highway-speed cells is performed

based on parameter settings for low-speed cells. Therefore, the values of
HighSpeed, Ncs, and CellRadius in the planning result are incorrect and
cannot be for reference.
According to the planning result, the values of Suggested Start

RootSequenceIndex and Suggested End RootSequenceIndex for each cell
meet the requirements and each highway-speed cell is allocated 20 ZC root
sequences. (In this example, 22 ZC root sequences are planned for each cell.)
Principles for PRACH ZC root sequence planning based on the planning
result are as follows:
 Site height ignored, the ZC root sequence with a great Reuse
Distance(m) value is preferred.
 The ZC root sequence with a great Reuse Tier(Topology) value is
preferred.
 The ZC root sequence with a high comprehensive score is preferred. The
comprehensive score is the product of Reuse Distance(m) and Reuse
Tier(Topology).
 After neighboring cells that do not meet the requirements, the ZC root
sequence with a high comprehensive score is preferred.
5.4 PRACH ZC Root Sequence Check

There are 587 E-UTRAN cells on the live network and these cells have been
allocated PRACH ZC root sequences 0 to 76. During the PRACH ZC root
sequence checking, if low PRACH ZC root sequence reuse isolation is
detected (for example, two or more first-tier neighboring cells are allocated
the same PRACH ZC root sequence or two or more intra-frequency E-
UTRAN cells within 4 km are allocated the same PRACH ZC root sequence),
PRACH ZC root sequence conflicts occur.
Use the U-Net to reallocate PRACH ZC root sequences to cells on the live
network and set the range of available PRACH ZC root sequences as 0 to 76.
Then, check the PRACH ZC root sequence reuse isolation before and after
reallocation.
5.4.1 Checking PRACH ZC Root Sequence Conflicts

Export the PRACH ZC root sequence table in the U-Net to an EXCEL file,
and use the sorting function of EXCEL to filter out the E-UTRAN cells whose
PRACH ZC root sequence reuse tier count is 1. These cells are allocated
incorrect PRACH ZC root sequences.
As shown in the following figure, the cells in red are intra-frequency E-

UTRAN cells allocated the same PRACH ZC root sequence. If two cells close
to each other share the same PRACH ZC root sequence, false alarms may be
generated.
5.4.2 Evaluating the Reuse Isolation of PRACH ZC Root

Sequences
Use the U-Net to reallocate PRACH ZC root sequences to cells on the live
network and set the range of available PRACH ZC root sequences as 0 to 76.
Copy data about the PRACH ZC root sequence reuse distance and reuse tier
count for each cell before and after PRACH ZC root sequence planning to an
EXCEL file to draw a CDF curve. View the CDF curve to check the PRACH
ZC root sequence reuse isolation before and after PRACH ZC root sequence
reallocation.
100
90
80
70
60
CDF 50
(%) 40 Before
After
30
20
10
0
0 2000 4000 6000 8000 10000 12000
PRACH ZC root sequence reuse distance (m)
100
90
80
70
60
CDF 50
(%) 40 Before
30 After
20
10
0
0 2 4 6 8 10 12 14 16 18 20
PRACH ZC root sequence reuse tier
According to the CDF curve, reallocation of PRACH ZC root sequences using

the U-Net improves the PRACH ZC root sequence reuse isolation, reduces the
number of cells with PRACH ZC root sequence conflicts, and increases the
PRACH ZC root sequence reuse distance. Therefore, the reallocation results
are recommended. If the PRACH ZC root sequence reuse distance is required
to be greater than 4 km, expand the range of available PRACH ZC root
sequences for PRACH ZC root sequence planning.
5.5 Configuration Delivery Checking

During the ZC root sequence planning, checking, and replanning using the
offline tool U-Net, ensure that parameter configurations must be consistent
between the U-Net and eNodeB. Otherwise, the planning result does not take
effect, resulting in ZC root sequence planning conflicts.
Check whether parameter settings for the cell are consistent with the planned.
Cell Parameter Description Mapped

Planned
Parameter
Cell.CellRadius Indicates the cell radius. Cell

Radius(m)
Cell.HighSpeedFlag Indicates whether a cell is a High Speed

high-speed cell.
Cell.PreambleFormat When this parameter is set to 4, -

ZZC root sequences numbered
0 to 137 are available.
Cell.RootSequenceInde Indicates the start ZC root Confirmed

x sequence number. Start Root
Sequence
Index
During PRACH ZC root sequence checking, manually export the four

parameters encircled in red for each cell in the figure above and check
whether the parameter settings are consistent with those in the U-Net.
5.6 TAL Planning Case

For details, see the following attachment.
M1 TA & TAL
Rationality Analysis.ppt
Design (PCI, PRACH, Neighboring Cell, and TA&TAL) A Appendix
A Appendix
B Impact of PCI Mod3
Staggering/Alignment on the Network

Performance
Test scenario: Samsung UE, 2T2R, LTE FDD 10 MHz, Rank1/Rank2 auto-
adaptation
When neighboring cells (not loaded) served by the same site are interfered by
each other in the ETU 30 km moderately related channel scenario, the
downlink throughput of Samsung UEs is better if the pilot frequencies are
staggered than that if the pilot frequencies are aligned. And the difference is
greater when interference is greater because the demodulation performance is
better and the MCS is higher when the pilot frequencies are staggered.
The ETU 30 km moderately related channel scenario refers to that the UE moves at
the speed of 30 km/h in the moderately related Extended Typical Urban channels.
Pilot staggered
Pilot aligned
Pilot staggered
Pilot aligned
Pilot staggered
Pilot aligned
C Cell Searching and Downlink
Synchronization
D Cell Searching-Why
When a UE, such as a cell phone, is turned on, it knows nothing about the
surrounding world before cell searching and synchronization. To access a cell,
the UE must learn the cell. The UE obtains the cell information, and then
perform synchronization with the cell and collect various information
provided by the cell until the UE obtains sufficient information.
Therefore, the base stations continuously broadcasts signals and the UE

continuously attempts to obtain cell information according to a certain
procedure, which includes cell searching and downlink synchronization.
E Cell Searching-What
Cell searching and downlink synchronization enable a UE to obtain:

 Time and frequency synchronization with cells
 Cell PCIs
 Information about the frame structure, such as the timeslot timing, frame
timing, frame sequence, and CP length
 Channel information for follow-up operations such as random access
Actually, cell searching is a process of synchronization between a UE
and cells. Sufficient information obtained in this process provides a good
foundation for the follow-up communication procedure.
F Cell Searching-When and How
Cell synchronization is performed on the downlink broadcast channel based

on the primary synchronization signal (PSS), secondary synchronization
signal (SSS), and cell-specific reference signal (CRS). Cell searching and
downlink synchronization are performed based on these signals. Figure
F.1.1.1.1.1.1.1 shows the detailed process:
Figure F.1.1.1.1.1.1.1 Cell searching process and related signals
Physical cell ID
and the CP length
System bandwidth
and number of
Resources on transmit antennas
the PHICH
Number of
occupied symbols
on the PDCCH
Frame ratio
PHICH period
adopted by the
and Ng value
system
After a UE is turned on, it must search for cells for radio access. After the UE
accesses a cell, it continues to search for the cell's neighboring cells and
evaluate its receive (Rx) signal quality to determine whether to hand over to
or reselect another cell.
A UE searches for cells by demodulating broadcast information from the cell.
After the UE is turned on, similar to a radio, the UE does not know when
information will be broadcast by a cell, and has only local time, which might
be inaccurate. In this situation, the UE must obtain correct information in
correct time to perform correct operations.
G Relationship Between the PCI and the
Synchronization Code
H Parsing the PSS
To obtain correct time, a UE must synchronize with a cell based on PSSs in

the cell's broadcast information. The PSSs are transmitted in a fixed frequency
and in a fixed position in the frequency domain, and therefore the UE can
initially synchronize time upon receiving the PSS.
I Time-Frequency Position of the PSS
The first step of cell searching is to parse the PSS.

In an LTE TDD system, the PPS is in the third orthogonal frequency division
multiplexing (OFDM) symbol of sub-frame 1 and sub-frame 6 in the time
domain, and is in the six radio bearers (RBs) in the middle of the frequency
domain. Seventy-two subcarriers (without DC subcarriers) are included, and
five subcarriers are distributed on each side of the PSS as the protection
frequency band to prevent PSS interference among cells.
Figure I.1.1.1.1.1.1.1 shows that the PSS in the third symbol of the special
subframe DwTPS (therefore, there must be three or more DwPTS symbols in
the special subframes).
Figure I.1.1.1.1.1.1.1 PSS in the frame structure (same as subcarrier

six)
The PSS is in the center of the frequency band and occupies six RBs in a fixed
frequency-domain position. Therefore, a UE needs to search only the six RBs
in the frequency band center to parse the PSS in different frequency
bandwidth configurations. In an LTE TDD system, even the frequency band is
configured as 1.4 MHz, which is the minimum configuration, time and
frequency resources of six RBs are ensured. Then, the UE can use the same
method to parse the PSS during cell searching.
J PSS Signal Composition and Parsing
The purpose of the PSS is to ensure that a UE can quickly and accurately
obtain a known signal in the broadcast information to obtain the time. The
working principle is the same as that in radio broadcasting. Before the PSS
synchronization, the UE knows a rough time-domain length of the OFDM
symbol according to the protocol, and performs sampling and demodulation

on the received signals.
Based on the previous requirements, the PSS uses a special sequence, the
Zadoff-Chu sequence, in the designing phase. The following defines the
Zadoff-Chu sequence:
Figure J.1.1.1.1.1.1.1 Formula for calculating the PSS
The previous two formulas provide two methods for calculating the PSS when
the coefficient n is an odd number or an even number. The mathematical
characteristics of the sequence enable the overlap signals of subcarriers to
meet the requirement of PSS. Therefore, during the designing phase, the PSS
uses the Zadoff-Chu sequence with the length of 62, and the PSS is
demodulated into 62 subcarriers in the frequency band center. With the ten
subcarriers, there are 72 subcarriers. Therefore, the PSS occupies six RBs.
Figure J.1.1.1.1.1.1.2 shows three root sequences of the Zadoff-Chu sequence.
Figure J.1.1.1.1.1.1.2 Three root sequences of the PSS
(2)
Root index u
N ID
0 25
1 29
2 34
The characteristics of the Zadoff-Chu sequence provide the following features

for the PSS:
 Frequency-domain features
− Constant amplitude: facilitates PSS channel evaluation and SSS
coherent demodulation.
− Good auto-correlation: The auto-correlation check function checks
obvious peak values.
− Good correlation: avoids PSS interference from other cells.
 Time-domain features
− The waveform is symmetric in the time domain. Two sequences with
the u being 29 and 34 are conjugate-symmetric.
− When the correlation feature is used, there are side peak values
increasing as the frequency-offset increases.
The purpose of the previous features is to facilitate PSS identification. When a
PSS is received by a UE, the UE performs the auto-correlation check by
comparing the waveforms of three root sequence modes to identify the

sequence mode of the PSS.
The UE measures the RSSI in the frequency range in which cells may exist.
Based on the RSSI changes, the UE locates a center frequency in which cells
may exist, and searches cells in the time domain. The UE determines a slide
window size based on one OFDM symbol and performs blind checks of the
three root sequences on each time point.
Figure J.1.1.1.1.1.1.3 PSS blink checks
If the check fails, the UE performs the check on the next sampling point until
the whole frame is checked.
When the UE is performing the checks, the duplicates of root sequences when
u is 25, 29, or 34 are stored in the UE. During the checks, the duplicates are
used to check the correlation.
Figure J.1.1.1.1.1.1.4 Auto-correlation results of PSS of different root

sequences
u=25 u=29 u=34
In Figure J.1.1.1.1.1.1.4, when the auto-correlation of the Zadoff-Chu

sequence with u being 34 is checked using the duplicates of the Zadoff-Chu
sequence with u being 25 or 39, the peak value is approximately 0. However,
if the duplicate of the Zadoff-Chu sequence with u being 34 is used, an
obvious peak value is displayed. Correct signal type is demodulated, and the
SSS demodulation is prepared.
After the PSS is parsed, the u value is obtained. Then the rough scheduled
synchronization, accurate scheduled synchronization, and the evaluation and
compensation of fine frequency. Through rough scheduled synchronization,
the PSS range is obtained. Through accurate scheduled synchronization, the
starting and ending positions of the OFDM symbol (without the CP header) of
the PSS are located. In addition, the evaluation and compensation of fine
frequency corrects the accuracy of center frequency detected through the
RSSI. These operations are fundamental to the accuracy of follow-up frame
structure parsing.
K Effects of PSS Synchronization
During the PSS parsing, the bit synchronization clock can be obtained from
the signals to achieve bit synchronization. Then, through PSS checks and
calculations of time difference of two PSSs, a 5 ms definite time is obtained.
Before this step, the UE still cannot demodulate the complete bit information
of an OFDM symbol and the UE just has a rough time. Through PSS parsing,
the UE obtains the channel estimation, which is fundamental to information
parsing, and the boundary definite time of 5 ms.
Besides, there are three root sequence modes for the PSS. The three sequence
modes map to the physical IDs of cells as shown in Figure K.1.1.1.1.1.1.1. In
an LTE TDD system, there are 504 PCIs available. These PCIs are categorized
into 168 groups with each, and in each group PCIs 0, 1, and 2 are defined.
Therefore, if the Zadoff-Chu sequence is obtained, the physical IDs of cells
searched by the UE are obtained.
The PSS can be used to calculate the rough CP length after the accurate time
synchronization in some scenarios according to different implementations of
the UE receiver.
Figure K.1.1.1.1.1.1.1 Estimating the CP length using the PSS
An OFDM symbol of
the PSS subcarrier
The CP part in Figure K.1.1.1.1.1.1.1 is one part of normal signal inserted in

the beginning of an OFDM symbol to maintain the orthogonality. Therefore,
this feature can be used to estimate the CP length and tune the value at the
beginning of the OFDM symbol at time d.
In the end, upon completing the PSS parsing, the UE has synchronized the
following with the cell:
 Rough time synchronization and frequency synchronization of the
OFDM symbol
 Boundary time of 5 ms
 Channel estimation used for SSS parsing
(2)
 Physical ID of the cell N ID
 (Optional) CP length
L Parsing the SSS
To obtain correct information, a UE must determine whether the received

information meets the requirement. When we listen to the radio, we determine
our desired language and frequency. Therefore, after an integral point
announcement, there is usually an announcement of current time and
frequency. This is the function of SSS, that is, obtaining the key information
from the cell and determine whether the receive mode meets the requirement.
M SSS Position in the Time Domain
After the PSS is parsed, the UE parses the SSS. The time-domain position of
the SSS differs in the LTE TDD system and LTE FDD system. In the LTE
TDD system, the SSS is in the last OFDM symbol of subframe 0 and
subframe 5.
Figure M.1.1.1.1.1.1.1 Time-domain position of the SSS (same as

subframe 5)
The SSS occupies similar time-domain resources as the PSS does. The SS
uses a 5 ms interval and occupies 73 x 1 RE resources (6 RBs and 1 dc
subcarrier).
N SSS Signal and Parsing
The SSS uses two M sequences with the length of 31 for interweaving
expansion. After the interweaving concatenation, the length of the SSS is 62.
Similar to the PSS, the SSS can be demodulated to the 62 subcarriers in the
frequency band center with five subcarriers on each side. Then 72 subcarriers
are occupied corresponding to the six RBs.
Figure N.1.1.1.1.1.1.1 Formulas for calculating the SSS
Subframe 0
Subframe 5
Subframe 0
Subframe 5
In Figure N.1.1.1.1.1.1.1, the SSS comprises two concatenation M sequences,

S(m0) and S(m1). C1(n) and C0(n) are related to the IDs parsed through the
PSS. On subframes 0 and 5, the sequences are disordered. Therefore, the SSS
sequences are disordered even they have the same two M sequences. The
following clarifies the principles:
O Sequences of the SSSs in different positions
Subframe 0 M Sequence Subframe 5 M Sequence
0 0 0 1
1 1 1 0
2 0 2 1
3 1 3 0
4 0 4 1
5 1 5 0
6 0 6 1
7 1 7 0
... ... ... ...
Subframe 0 M Sequence Subframe 5 M Sequence
62 0 62 1
The positions of SSSs differ in the LTE TDD and LTE FDD systems.
Therefore, the positions of SSSs must be detected through blind checks even
the PSS and SSS have a position relationship defined in the protocol.
Figure O.1.1.1.1.1.1.1 Synchronization signals in the LTE TDD and

LTE FDD systems
The SSS may be located at the third subframe before the PSS in the LTE TDD
system, or be located at the first subframe before the PSS in the LTE FDD
system. Due to different CP lengths, six subframes are in each timeslot (0.5
ms) in the expanded CP scenario, and seven subframes are in each timeslot in
the common CP scenario. Therefore, there are two types of SSS and PSS
offsets in the LTE TDD system.
During signal parsing, the UE attempts to parse SSS in the possible position
described above based on information of one frame parsed from the PSS. The
SSS differs in the format in subframes 0 and 5, and therefore, the parsed SSS
may be different and needs to be analyzed in different ways.
The information carried in the SSS is related to the physical group numbers of
cells. There are 504 PCIs available, which are categorized into 168 groups,
(2)
and three PCIs are in each group. The ID of each PCI in the group ( N ID ) is
parsed from the PSS, and there are 168 values of N (1)
ID used to generate the
'
m sequence with length of 31, m , and then generate M sequences of the
same content but in different formats, m0 and m1.
Figure O.1.1.1.1.1.1.2 Mapping between the physical group numbers of

cells and the M sequences
'
m0 =m mod 31
'
m1=( m 0 + ⌊ m /31 ⌋ +1 ) mod 31
(1 ) ' '
m
' ( 1)
=N ID + q(q +1)/2 , q= ⌊ N ID +q ( q + 1)/2
30 ⌋ ,
'
⌊ (1 )
q = N ID /30 ⌋
(m 0 )
s0 (n)=~
s ( (n+m 0 )mod 31 )
(m 1 )
s (n )=~
1 s ((n+m )mod 31 ) 1
c 0 ( n)=~
c ((n+N (2)
ID )mod 31)
( 2)
c 1 (n )=~
c ((n+N ID +3 )mod 31)
As shown in Figure N.1.1.1.1.1.1.1, C1(n) and C0(n) are related to the ID

parsed from the PSS, and Figure O.1.1.1.1.1.1.2 shows the calculation. P
describes the mapping between the 168 physical group IDs of cells and M
sequences based on the previous calculation:
P Mapping between the physical group IDs of cells and the M sequences
m0 m1 m0 m1 m0 m1 m0 m1 m0 m1
N (1)
ID N (1)
ID N (1)
ID N (1)
ID N (1)
ID
0 0 1 34 4 6 68 9 12 102 15 19 136 22 27
1 1 2 35 5 7 69 10 13 103 16 20 137 23 28
2 2 3 36 6 8 70 11 14 104 17 21 138 24 29
3 3 4 37 7 9 71 12 15 105 18 22 139 25 30
4 4 5 38 8 10 72 13 16 106 19 23 140 0 6
5 5 6 39 9 11 73 14 17 107 20 24 141 1 7
6 6 7 40 10 12 74 15 18 108 21 25 142 2 8
7 7 8 41 11 13 75 16 19 109 22 26 143 3 9
8 8 9 42 12 14 76 17 20 110 23 27 144 4 10
9 9 10 43 13 15 77 18 21 111 24 28 145 5 11
10 10 11 44 14 16 78 19 22 112 25 29 146 6 12
11 11 12 45 15 17 79 20 23 113 26 30 147 7 13
12 12 13 46 16 18 80 21 24 114 0 5 148 8 14
13 13 14 47 17 19 81 22 25 115 1 6 149 9 15
14 14 15 48 18 20 82 23 26 116 2 7 150 10 16
15 15 16 49 19 21 83 24 27 117 3 8 151 11 17
16 16 17 50 20 22 84 25 28 118 4 9 152 12 18
17 17 18 51 21 23 85 26 29 119 5 10 153 13 19
N (1) m0 m1 N (1) m0 m1 N (1) m0 m1 N (1) m0 m1 N (1) m0 m1

ID ID ID ID ID
18 18 19 52 22 24 86 27 30 120 6 11 154 14 20
19 19 20 53 23 25 87 0 4 121 7 12 155 15 21
20 20 21 54 24 26 88 1 5 122 8 13 156 16 22
21 21 22 55 25 27 89 2 6 123 9 14 157 17 23
22 22 23 56 26 28 90 3 7 124 10 15 158 18 24
23 23 24 57 27 29 91 4 8 125 11 16 159 19 25
24 24 25 58 28 30 92 5 9 126 12 17 160 20 26
25 25 26 59 0 3 93 6 10 127 13 18 161 21 27
26 26 27 60 1 4 94 7 11 128 14 19 162 22 28
27 27 28 61 2 5 95 8 12 129 15 20 163 23 29
28 28 29 62 3 6 96 9 13 130 16 21 164 24 30
29 29 30 63 4 7 97 10 14 131 17 22 165 0 7
30 0 2 64 5 8 98 11 15 132 18 23 166 1 8
31 1 3 65 6 9 99 12 16 133 19 24 167 2 9
32 2 4 66 7 10 100 13 17 134 20 25 - - -
33 3 5 67 8 11 101 14 18 135 21 26 - - -
Q Effects of SSS Synchronization
During the SSS parsing, whether the LTE TDD or LTE FDD mode is used by
the current cell is determined based on the SSS position in frames. Then, the
CP length is verified based on the distance between the SSS and the PSS on
the time domain to obtain an accurate subframe length and its internal
structure.
The M sequence of the SSS sequence is calculated, and therefore the value of
N (1)
ID can be obtained only through the SSS parsing. The SSSs carrying the
same information are of different formations in subframes 0 and 5. Therefore,
whether the SSS is in the first half frames or the second half frames can be
determined through one parsing. Then, the physical cell group IDs and 10 ms
frame timing are obtained.
In the end, upon completing the SSS parsing, the UE has synchronized the
following with the cell:
 Whether the system is an LTE TDD system or LTE FDD system
 Accurate CP length and subframe length

(2)
 Physical cell IDs (group ID N (1)
ID and cell ID in each group N ID )
 10 ms frame timing
R Relationship Between the PCI and the RS
In radio broadcasting, there are some auxiliary signals before the time
announcement to remind the users to receive the time announcement. These
auxiliary signals are reported once in each second, and therefore they can be
used to check whether your watch is slower or faster than the standard pace.
In a word, these signals are for reference, and they are reference signals (RSs).
The RS is a signal transmitted from the transmitter to the receiver for channel
estimation or detection. The RS is usually called the pilot. In the LTE system,
the RS occupies one RE in the time domain (one subframe in the frequency
domain or one OFDM symbol in the time domain).
The RSs can be classified into the uplink RSs and downlink RSs. The
downlink RSs include the cell RS (CRS), MBSFN RS, UE-specific RS, and
location RS. The uplink RSs include the uplink sounding RS (SRS) and
demodulation RS (DMRS)
During cell searching, the CRS is parsed. Similar to functions of other RSs,
the main functions of the CRS is to estimate the current channel, determine
the channel signal quality, and verify the obtained cell information and
frequency and time synchronization.
S Time-Frequency Position of the CRS
The CRS distribution is related to the number of antenna ports of the cell.
According to section 6.2.1 in 3GPP protocol TS36.211, the antenna port is
configured based on related RSs.
T
Antenna Ports
Mapping Between the CRS and the Number of
The antenna port is defined based on related RSs, not physical signals.
The antenna port is determined by the types of the RS. For example, p = {0},
p = {0,1}, and p = {0,1,2,3} indicate the ports for CRS, p=4 indicates the port
for MBSFN RS, and p = 5/7/8 indicates the ports for UE-specific RS.
By default, two antenna ports are configured for a base station. The number of
antenna ports affects the number of time-frequency positions for the RS, and
therefore has great impacts on the throughput.
U CRS Distribution
The CRS is used for downlink demodulation and measurement sent in all
downlink subframes. As described above, the CRS can use antenna ports 0 to
3. Each RS occupies one RE. In the protocol, the mapping of the RS resources
and antenna ports is shown in Figure U.1.1.1.1.1.1.1. In Figure
U.1.1.1.1.1.1.1, one large block indicates two RBs, that is, one subframe in
the time domain and 12 subcarriers in the frequency domain, which is a
normal CP.
Figure U.1.1.1.1.1.1.1 CRS distribution
R0 R0
One antenna port
R0 R0
R0 R0
R0 R0
l0 l6 l0 l 6
Resource element (k,l)
R0 R0 R1 R1
Two antenna ports
R0 R0 R1 R1
Not used for transmission on this antenna port
R0 R0 R1 R1
Reference symbols on this antenna port
R0 R0 R1 R1
l 0 l 6 l 0 l6 l0 l6 l0 l 6
R0 R0 R1 R1 R2 R3
Four antenna ports
R0 R0 R1 R1 R2 R3
R0 R0 R1 R1 R2 R3
R0 R0 R1 R1 R2 R3
l0 l6 l0 l6 l 0 l 6 l 0 l 6 l0 l6 l0 l6 l 0 l 6 l 0 l 6
even-numbered slots odd-numbered slots even-numbered slots odd-numbered slots even-numbered slots odd-numbered slots even-numbered slots odd-numbered slots
Antenna port 0 Antenna port 1 Antenna port 2 Antenna port 3
As shown in Figure U.1.1.1.1.1.1.1, the RS is distributed in different ways at

antenna ports 1, 2, or 4. In the time domain, the RS is in the first OFDM
symbol and the last third OFDM symbol at port 0 or 1 in each timeslot (0.5
ms), or in the second OFDM at port 2 or 3 in each timeslot. In the frequency
domain, one RS is in every six subcarriers and the RSs in the same symbol but
at different antenna ports are staggered. At each antenna port, time-domain

resources are reserved for RSs of other antenna ports (puncture). In the
reserved time-domain resources, no information of the local antenna is
transmitted to prevent interference of inter-antenna RSs. In a word, at all
antenna ports, the RSs are evenly distributed in the time domain and
frequency domain.
The RSs are distributed evenly in the whole time-frequency resources, but in
the center part (the six RBs in the center) the RSs remain unchanged.
Therefore, the UE can perform channel estimation based on the RSs in the
center of the frequency band in different frequency band configurations.
The CRS maps to different time-frequency positions based on different cell
information. The starting position of the RS is determined by the PCI. The
starting position is called the RS frequency offset. There are six types of
frequency offsets, and each type of offsets maps 84 PCIs. Therefore, there are
504 PCIs mapped to six different time-frequency positions. Different
frequency offsets enable cells to receive non-reference signals from
neighboring cells and prevent cells from receiving or interfering RSs of
neighboring cells. Therefore, the RS power can be increased to raise the
signal-to-noise ratio (SNR) of RSs.
In the same cell especially with the multiple-input multiple-output (MIMO)
technology is used, interference of inter-antenna RSs is prevented by good
channel estimation. Therefore, the puncture can be used to prevent
interference of inter-antenna RSs to obtain good antenna estimation.
When four antennas are used, the time-domain density of the third and fourth
antenna ports is decreased to limit the RSs, which may have some negative
impacts on the channel estimation in the fast channel fluctuation status.
However, the four antenna Space Division Multiplexing (SDM) is designed
for low mobility scenarios, and the impacts are small.
V CRS Signal and Parsing
The previous section has introduced the position and functions of the CRS,
and one important function of the CRS is downlink channel estimation.
Therefore, the information in the RS is known to the receiver.
During the CRS parsing, the CRS position must be determined. According to
the previous description, the CRS offset is determined by the PCI of
transmitted CRSs. In related protocols, the time-domain position l and
frequency-domain position of the CRS is determined by the following
formulas:
k =6 m+ ( v +v shift ) mod6
0 , N DL if p ∈ {0,1 }
l=
1 { symb−3
DL
if p ∈ {2,3 }
m=0,1 , .. . ,2⋅N RB−1
Where,
0 if p=0 and l=0
{
3 if p=0 and l≠0
3 if p=1 and l=0
v= 0 if p=1 and l≠0
3(n s mod 2) if p=2
3+3 (ns mod2 ) if p=3
cell
v shift =N ID mod 6
The following describes the coefficients in the above formulas:

k: indicates the frequency-domain position of the RS.
l: indicates the time-domain position of the RS.
p: indicates the antenna port. When only one antenna is configured, use port 0.
When two antennas are configured, use ports 0 and 1. When four antennas are
configured, use ports 0, 1, 2, and 3.
: indicates the number of OFDM symbols in a downlink timeslot (0.5

ms). Seven OFDM symbols are configured for the normal CP, and six OFDM
symbols are configured for the extended CP.
: indicates the downlink bandwidth configurations, which map the

number of RBs. For example, when the bandwidth is 20 MHz, 100 RBs are
used.
: indicates the slot number in 10 ms a radio frame starting from 0. In the
previous formulas, mod 2 is used to differentiate two timeslots in a subframe.
Based on the previous formulas, in the time domain, for antenna ports 1 and 2,
the CRSs are distributed on the first and the last third OFDM symbols of each
timeslot, and for antenna ports 3 and 4, the CRSs are distributed on the second
OFDM symbol of each timeslot. In the frequency domain, separated by six
OFDM symbols, CRSs have different offsets based on the PCIs and are
evenly distributed.
After the cell PCI is parsed from the PSS and SSS, the time-domain position
of the CRS is spotted, and then the CRS can be correctly parsed for downlink
channel estimation.
W Functions of CRS Synchronization
The CRS can be used to verify the PCI through the frequency-domain position
of the CRS. The CRS can also be used to obtain accurate time due to its good
time-frequency characteristics. During the channel parsing, the receive power
of the RS transmitted from the first antenna port is measured to obtain the
RSRP of the RS for follow-up handovers and reselection.
Therefore, the functions of the CRS synchronization are as follows:
 Obtaining the downlink channel estimation
 Verifying the PCI
 Obtaining more accurate time and frequency synchronization
 Obtaining the RSRP for handovers and reselection
X Cyclic Shift Value
A ZC root sequence can be obtained by using the following formula:

π un( n+1)
−j
N ZC
x u (n )=e , 0≤n≤N ZC −1
A preamble sequence can be obtained through cyclic shifting on a ZC

sequence with the physical number u. The cyclic shift value is calculated
Cv=¿{vNcs v=0,1,. .⌊Nzc/Ncs⌋−1,Ncs≠0 Low-spe d cel ¿{0 Ncs=0 Low-spe d cel ¿¿¿
If the cell is a low-speed cell and the NCS value is 119 (the NCS value is
calculated based on the cell radius in actual conditions), the cyclic shift value
is obtained through the following formula:
Cv = v x NCS = v x 119, v = 0, 1, 2, 3, 4, 5, 6
The ZC sequence can generate seven preamble sequences as follows:
x{u0,1u2x3. u83¿{xu19, 20xu1. u83,x0u1.xu18¿{u238,x9. u83,x0u1.x237¿{u57,x38. u38,x0. u356¿{xu476, 7.xu83,0.xu475¿{u95,x6. u83,x0. 594¿
 If the cell is a low-speed cell and the NCS value is 0, the generated
preamble sequence is as follows:
x u0 ,x u1 ,x u 2 ,x u 3 ......x u 838
 For a high-speed cell, the cyclic shift value Cv is calculated using the
following formula:
C v=d start ⌊ v /nRA RA RA RA RA

shift ⌋ +(v mod n shift ) N cs , v=0,1 , .. . .. , nshift n group + n̄shift −1
A variable du is defined to meet the following condition:
du=¿ { p 0≤p<N zc /2 ¿ ¿¿¿
In the formula, P is a nonnegative integer and meets the equation

(pu)modNzc = 1.
 If du is greater than or equal to NCS but less than Nzc/3, then
nRA
shift = ⌊ d u / N CS ⌋
d start =2 d u +nRA
shift N CS
RA
ngroup =⌊ N ZC /d start ⌋
n̄RA RA
shift =max ( ⌊ (N ZC−2d u −n group d start )/N CS ⌋ , 0 )
 If du is greater than or equal to Nzc/3 but less than or equal to (Nzc –

NCS)/2, then
nRA
shift = ⌊ (N ZC−2d u )/ N CS ⌋
d start =N ZC−2 d u +nRA
shift N CS
RA
ngroup =⌊ d u /d start ⌋
n̄RA ( RA RA
shift =min max (⌊ (d u −ngroup d start )/ N CS ⌋ , 0 ) , nshift )
For example, assume that the NCS value is 128, the cell is in preamble format
0, 1, 2, or 3, and the Nzc value is 839.
Obtain the logical root sequence number corresponding to NCS 128 with
reference to the following table.
Figure X.1.1.1.1.1.1.1 Relationship between the cell's NCS and ZC root

sequence index (only for high-speed cells)
Logical Root
CM Group Sub Group Number Ncs
Sequence Number
When the NCS value is 128, the available logical ZC root sequences are
numbered 204 to 263 and 630 to 659. Logical ZC root sequences are allocated
to the low CM group first and then to the high CM group. In the table above,
the NCS value belongs to the low CM group. Therefore, Logical ZC root
sequences 204 to 263 are allocated.
Based on the mapping between the logical root sequences and physical root
sequences for preamble formats 0 to 3 in Table 2.2.1.1.1.1.1.3.1, the u value is
152. Then,
p=138,
d u =138
Because du is greater than or equal to Nzc but less than Nzc/3, the following
values are obtained:
n RA
shift =⌊ 138/128 ⌋ =1
d start =2∗138+1∗128=404
n RA
group= ⌊ 839 /404 ⌋= 2
n̄ RA
shift =max ( ⌊ (839−2∗138−2∗404 )/128 ⌋ , 0)=0
Therefore,
Cv=404∗⌊ v /1 ⌋ +( v mod1 )∗128 , v =0,1
In this case, the ZC sequence can generate two preamble sequences as

follows:
{ xu0 ,xu1 ,xu2 ......xu838 ¿ ¿¿¿
Y Number of ZC Root Sequences for Each
Cell
To calculate the number of ZC root sequences for a cell, perform the
following steps:
Step 1 Calculate the NCS value.
Step 2 Calculate the number of preambles generated with each ZC root sequence.
Step 3 Calculate the number of ZC root sequences for each ZC root sequence group.
----End
For detailed operations, see the following sections.

The following sections are described based on the assumption that cell A is a
low-speed cell, the radius is 10 km, and preamble format 0, 1, 2, or 3 is used.
 Calculating the NCS value
The NCS value depends on the cell radius and maximum delay spread.
For details, see formula (6.2.1).
NCS x Ts ≥ RTD + SpreadDelay + error formula (6.2.1)
Where
− Ts indicates the length of the sampled ZC root sequence.
T S=¿ {800/839[μsec]PreambFmt=0~3 ¿ ¿¿¿

− RTD indicates the round trip delay of cell signals and is defined as
follows:
RTD=2 r/c=6 . 67 r[ μ sec ]
The cell radius r is in the unit of km, c is the light speed 3 x 108 m/s.
− SpeadDelay indicates the delay spread and is related to the preamble
format. Currently, SpreadDelay is set to 5 µsec.
− Error indicates the error. The downlink synchronization error is
about 2 µsec.
Step 1 After the preceding values are substituted into formula (6.2.1), the following
formula is obtained.
{NCS>1.04875⋅(6.67r+7)PreambFmt=0~3¿¿¿¿
Step 2 Based on the assumption, calculate the NCS value range for cell A. The result
is that the NCS value is greater than 77.29.
Step 3 The NCS value is fixed. Based on the following table, select a value greater
than 77.29 as the NCS value.
N CS Configuration
N CS Value

0 0 15
1 13 18
2 15 22
3 18 26
4 22 32
5 26 38
6 32 46
7 38 55
8 46 68
9 59 82
10 76 100
11 93 128
12 119 158
13 167 202
14 279 237
15 419 N/A
Cell A is a low-speed cell. Therefore, select a value greater than 77.29

from the second column, that is, 93.
----End
 Calculating the number of Preambles generated with each ZC root

sequence
Calculate the number of preambles generated with each ZC root
sequence through the following formula:
Num= ⌊ 839 /Ncs ⌋ =⌊ 839/93 ⌋ =9
 Calculating the number of ZC root sequences for each ZC root sequence

group
A ZC root sequence group consists of multiple consecutively numbered
ZC root sequences. Each ZC root sequence group generates at least 64
preamble sequences.
Each E-UTRAN cell must be assigned sufficient ZC root sequences to
generate 64 preamble sequences.
Calculate the number of ZC root sequences for each ZC root sequence
group through the following formula:
Nu=⌈64 / Num⌉=⌈64/9⌉=8
Z LTE TAU and Paging Principles and
Procedures
AA LTE TAU Principles and Procedure
TAUs contain static TAUs and dynamic TAUs. Static TAUs are triggered on
the network side when the periodic TAU timer expires and are independent
from UE actions. Dynamic TAUs are implemented based on UE calls and
mobility attributes, consuming more network resources. Dynamic TAUs are
triggered in the following scenarios:
 According to monitoring results, the UE has been in a new TA whose
TAI is unavailable in the TAI list of the network with which the UE is
registered.
 The UE in UTRAN PMM_Connected mode (for example, URA_PCH) is
reselected to the E-UTRAN.
 The UE in GPRS READY mode is reselected to the E-UTRAN.
 The GUTI update status is xxx, which can be changed by bearing
configurations on the GERAN/UTRAN.
 RRC Connection is released because TAUs are performed for load
balance.
 The UE capability of communicating with the EPC changes.
During a TAU, the MME sends a new TAL to the UE. At present, dynamic
TAL updates or UE-level TAL configurations are not supported. Therefore,
exercise caution when planning the TAL boundary. Otherwise, frequent TAU
signaling overheads occur at the TAL boundary after the network is
commercially used.
The TAU is initiated by the UE in ECM_IDLE or ECM_CONNECTED
mode. The SGW update is implemented by the MME during a TAU and is
independent from the preceding scenarios where dynamic TAUs are
performed.
Figure AA.1.1.1.1.1.1.1 Procedure for a tracking area update with the SGW
changed
UE eNodeB new MME old MME/ new Serving old Serving PDN GW HSS
old GW GW
SGSN
1. Trigger to start
TAU procedure
2. TAU Request
3. TAU Request
4. Context Request
5. Context Response
6. Authentication / Security
7. Context Acknowledge
8. Create Bearer Request
9. Update Bearer Request (A)
10. Update Bearer Response
11. Create Bearer Response
12. Update Location
13. Cancel Location
14. Cancel Location Ack
15. Update Location Ack
16. Delete Bearer Request
(B)
17. Delete Bearer Response
18. TAU Accept
19. TAU Complete
IEs contained in the TAU Request message are as follows:
 Old GUTI
 Last visited TAI
 Active flag
 EPS bearer status
 P-TMSI Signature
 Additional GUTI
 KSI
 NAS sequence number
 NAS-MAC
IEs contained in the TAU Accept messages are as follows:

 GUTI
 TAI list
 EPS bearer status
 KSI
 NAS sequence number
 NAS-MAC
During a TAU, uplink bandwidth resources at radio air interfaces and

communication resources on the EPC are occupied. Frequent TAUs increase
MME loads and UE battery consumption and reduce the paging success rate
of the MME (because the UE cannot respond to the paging message during a
TAU). Therefore, it is good practice to control TAU overheads within an
appropriate range.
MMEs in current versions support only static TALs. A maximum of 16 TAs can be
configured in a TAL. After UE-level dynamic TALs are achieved, UE-level TAL
configurations need to be saved on the MME and TALs also need to be saved in the
UE. When the UE detects that the new TAI is inconsistent with the TA in the TAL, the
UE will search for the new TAI in the TAL. If the new TAI is in the TAL, the UE will
not initiate a TAU request. Otherwise, the UE will initiate a TAU request and then the
MME updates the UE-level TAL and delivers the updated TAL to the UE. This greatly
reduces unexpected TAU storms in cells, reducing consumption of MME and UE
resources.
Products that can achieve dynamic TAL updates based on UE mobility directions and
the eNodeB topology are under research.
BB LTE Paging Functions
Paging provides the following functions on an LTE network:

 The MME sends a paging message to the UE in idle mode when another
UE calls the UE or downlink data reaches the EPC.
 The E-UTRAN sends update notification messages of eNodeBs to UEs
in idle or connected mode.
 The E-UTRAN sends ETWS notification messages to UEs in idle or
connected mode.
CC Paging Frames and POs
Paging messages sent to UEs are sent in specific paging frames (PFs) at a
specific paging occasion (PO), as shown in Figure CC.1.1.1.1.1.1.1.
Figure CC.1.1.1.1.1.1.1 Structure of PFs and POs with a 1280 ms DRX cycle
 The frequencies of the PF and PO are determined by the nB parameter.

 The default paging cycle determines the discontinuous reception (DRX)
cycle (that is, how often the UE monitors the PO). A small DRX cycle
shortens the paging time but increases UE battery consumption.
 One paging message can be used to page multiple UEs, that is, multiple
UEs can share one PO. The UE_ID ranges from 0 to 1023. The formula
for calculating the UE_ID is as follows:
UE_ID = (Last 10 bits of the IMSI) mod 1024.
If the last 10 bits of UEs' IMSIs are the same, the UEs are allocated in
the same paging group in a paging cycle.
DD LTE Paging Process
In idle mode, the MME traces UE locations by maintaining the TAL in several
cells, which reduces frequent TAUs. (The MME traces UEs in idle mode at a
TAL level.) When the UE enters from the idle mode to the connected mode
because downlink data reaches the EPC, the MME determines the cell where
the UE is located by paging (The MME traces UEs in connected mode at a
cell level).
The MME plays an important role in paging UEs. When downlink data
reaches the EPC, data packeting ends and the data is buffered on the SGW.
Meanwhile, the SGW sends a paging notification message to the MME and
the MME sends paging messages to all related eNodeBs in the TALs within
the paging area to instruct the eNodeBs to page UEs within their coverage
scope. In addition, downlink signaling also triggers the MME to page UEs to
set up a signaling connection between the UEs and the network.
S1-AP paging messages are delivered to related eNodeBs over the S1
interface to instruct the eNodeBs to send paging messages to UEs in all cells
within related TAs. After S1-AP paging messages reach the eNodeB, the
messages wait for the first valid PO in a queue. Then the paging messages are
transmitted through the PDCCH and PDSCH over the air interface. The
response to the paging message sent by the UE to the MME is generated at the
NAS layer.
The LTE paging procedure is as follows:
Figure DD.1.1.1.1.1.1.1 LTE paging procedure
 The paging message contains IDs of paged UEs and multiple UEs can be
paged in the same RRC paging message. (The number of paged UEs is
determined by maxNoOfPagingRecords.)
 UEs that do not detect their IDs in the paging message will enter the
DRX cycle. Otherwise, the UEs send the service request message to the
MME. If T3413 times out and the MME does not receive the service
request message, the MME resends the S1-AP Paging message. The
maximum number of times for the MME resending the paging message
is specified by N3413.
 After a paging message reaches the eNodeB, the RRC layer attempts to
send the message at the first valid PO. If this fails due to blocking, the
RRC layer resends the paging message at the next PO after the DRX
cycle.
 The RRC layer attempts to send the paging message within
PagingDiscardTimer. It is good practice to set PagingDiscardTimer to a
value greater than or equal to T3413 and less than daultPagingCycle so
that the paging message can be sent for at least twice. If
PagingDiscardTimer expires, the paging message will be discarded.
Paging occupies CCE resources on the PDCCH and scheduling block
resources on the PDSCH.
 The PDSCH paging load allowed by operators determines PDSCH

paging capacity and is calculated by using the following formula:
PDSCH paging load = (Average number of PDSCH scheduling blocks
occupied by the RRC paging message x Arrival rate of paging messages
sent by eNodeBs)/Number of available scheduling blocks for the radio
frame
Generally, the PDSCH paging load is 3% or less.
 The PDCCH paging load allowed by operators determine the PDCCH
paging capacity. The PDCCH paging load is calculated by using the
following formula:
PDCCH paging load = Number of CCEs scheduled and allocated for
transmitting paging messages/Number of available CCEs at the current
bandwidth of the eNodeB
Generally, the PDCCH paging load is 1% or less.
Planning paging areas helps control the eNodeB paging arrival rate
within an appropriate range and reduces PDSCH and PDCCH paging
loads without compromising transmission of downlink user data. In
addition, planning paging areas helps control paging delay.
Note: Paging messages have a higher priority than user data. In the
following scenarios, the eNodeB paging arrival will be high, affecting
user data transmission:
maxNoOfPagingRecords is set to a large value or is fully configured (all
PO bandwidths are occupied).
The paging area or TAL scope is small.
EE
Messages
Allocating Resources for LTE Paging
 Definitions of the PC, PF, and PO and Relationships Between the PC, the
PF, and the PO
The E-UTRAN sends paging messages to the corresponding UEs at the
PO in a PC. One paging message can be sent to multiple UEs, that is,
multiple UEs can share one PO (subframe). The UE in the subframe
monitors whether P-RNTI exists on the PDCCH to determine whether
there are paging messages on the PDSCH that are sent to itself. The radio
frame containing one or more POs is called a radio paging frame. One
paging cycle contains several paging frames. The paging cycle includes
defautpagingcycle and FFS.
The UE in idle mode is allocated with a unique PO (one or more TTIs)
within a paging cycle. To reduce battery consumption, DRX is used. In
this case, the PC for each UE is called the paging DRX cycle.
Figure EE.1.1.1.1.1.1.1 Scheduling paging messages
As shown in Figure EE.1.1.1.1.1.1.1, if the PF and PO are determined in

one paging DRX cycle, the UE can monitor the PDCCH in the
corresponding subframe and read corresponding paging messages by P-
RNTI on the PDCCH and corresponding frequency resources and
numbering formats. According to 3GPP 24.301, the PF and PO are
calculated by using the following formulas, respectively:
− SFN mod T= (T div N) x (UE_ID mod N)
− i_s = (UE_ID/N) mod Ns
Where,
T, length of the paging DRX cycle, can be set to 32, 64, 128, and 256
frames and is specified by using the defautpagingcycle parameter.
nB, number of POs in a paging DRX cycle, can be set to 4T, 2T, T, 1/2T,
1/4T, 1/8T, 1/16T, and 1/32T and is specified by the nB parameter in
SIB-2.
N = min(T, nB)
Ns = max (1, nB/T)
nIMSI mod 1024
UE_ID =
nIMSI is the last 10 bits of the IMSI.
If the UE does not have an IMSI, for example, the UE initiates an emergency call
without a USIM, the PO and PF are calculated based on the assumption that UE_ID is
0 by default.
According to preceding formulas, time resources of user paging

messages are determined by the IMSI, defautpagingcycle, and nB.
FF and GG list FDD and TDD paging subframe mapping models,
respectively. The paging subframe mapping models are the mapping
relationships between the paging message transmission subframe number
calculated by using the preceding formulas and the actual scheduling
subframe number. This ensures that the paging service does not conflict
with other services (for example, MBMS) and reduces the impact as
much as possible. For example when Ns is 1, the FFD allocates one

subframe to each radio frame for transmitting paging messages and
schedules paging messages in subframe 9 of each radio frame; when Ns
is 1, the TDD allocates one subframe to each radio frame for transmitting
paging messages and schedules paging messages in subframe 0 of each
radio frame.
FF Ns PO when i_s=0
FDD subframe mapping model
PO when i_s=1 PO when i_s=2 PO when i_s=3
1 9 N/A N/A N/A
2 4 9 N/A N/A
4 0, 4 5 9
GG Ns PO when i_s=0
TDD subframe mapping model (all UL/DL patterns)
PO when i_s=1 PO when i_s=2 PO when i_s=3
1 0 N/A N/A N/A
2 0 5 N/A N/A
4 0, 1 5 6
 Number of UEs That Can Be Paged at a PO

One paging record contains the RRC paging message for paging a
specific UE in a PO. maxNoOfPagingRecords specifies the maximum
number of paging records in a PO, that is, the maximum number of
paging messages in a PO at various bandwidths. According to 3GPP, the
upper threshold is 16.
To reduce the paging delay and accelerate the paging processing
procedure, it is good practice to configure maxNoOfPagingRecords
based on the relationships between bandwidths and paging messages, as
shown in Figure GG.1.1.1.1.1.1.1.
Figure GG.1.1.1.1.1.1.1 Suggestions for configuring

maxNoOfPagingRecords at various bandwidths
The paging message has a higher priority than user data. If

maxNoOfPagingRecords is set to a larger value than expected or fully
configured (maxNoOfPagingRecords is 7 at a bandwidth of 5 MHz and
paging records occupy the all bandwidth resources of the eNodeB for the
PO), user data transmission is affected. If maxNoOfPagingRecords is
greater than the preceding value, the paging message may fail to be
transmitted.
 nB Definition
In a broad sense, eNodeB paging performance depends on the number of
POs in a radio frame. The number of POs is specified by the nB
parameter sent by using the SIB2. nB can be set to one of the following
values:
− 1/32T
− 1/16T
− 1/8T
− 1/4T
− 1/2T
− T
− 2T
− 4T
T: indicates defaultPagingCycle.
nB  T : All radio frames are used for paging and each PF contains
nB/T POs.
nB  T : nB/T frames are used for paging and each PF contains one PO.
The number of POs contained in a radio frame is calculated by using the
following formula:
nB
nPO, frame 
T
nB determines distribution of PFs and POs, as shown in Figure

GG.1.1.1.1.1.1.2.
Figure GG.1.1.1.1.1.1.2 Distribution of PFs and POs determined by nB
− A small nB results in a small number of POs and each PO carries

more paging messages.
a. Advantage: Setting nB to a small value can reduce PDCCH
loads.
b. Disadvantage: More paging messages increase the risk that
paging may fail to pass the first PO, increasing paging delay.
− A large nB results in a large number of POs and each PO carries
fewer paging messages.
HH LTE Paging Mechanism
 Paging group
To avoid false alarms and ensure paging capacity, UEs are divided into
different groups in the paging mechanism in which a group of UEs
listens in the same TTI for possible paging messages. In this solution,
one paging group is supported at one PO. In this way, different paging
groups are allocated in different POs and thereby can use the same
control command. The number of POs can be configured, which offers
the following benefits:
− If the number of UEs to be paged is small at each PO, the possibility
of generating incorrect alarms for multiple users is low, that is, UE
battery consumption is low.
− The number of POs can be configured flexibly at different system
bandwidths. If the bandwidth is small, the size of paging messages is
limited, limiting the number of UEs to be paged. In this case, UEs are
divided into different POs, thereby reducing the possibility that too
many UEs are paged in one paging message.
 P-RNTI
In the 59th RAN2 conference, multiple companies support each PO uses
only one RNTI (called P-RNTI) and that only one RNTI is used on each
PCH. That is, all UEs use the same RNTI. The UEs check whether the P-
RNTI exists on the PDCCH and determine whether paging messages are
on the corresponding PDSCH. However, the UEs cannot determine
whether paging messages are sent to themselves. Therefore, paging

messages need to be demodulated.
 Paging DRX
The demands for UE Specific Paging DRX are discussed in 3GPP
24.301. The conclusion is that different UEs may access the E-UTRAN.
Some UEs are sensitive to battery consumption and some UEs need to
obtain paging messages quickly. Therefore, it is recommended that the E-
UTRAN support UE Specific Paging DRX. In addition, the
defaultpagingDRX parameter is required and related application
problems are discussed in 3GPP 24.301. The conclusion is that the
message update period for eNodeBs is defined based on the cell-level
defaultpagingDRX parameter. According to 3GPP 24.301, when UE
Specific Paging DRX exists, the smallest value of defaultpagingDRX
can be used as the paging cycle.
 Paging repetition
Paging messages are triggered by the MME and then is transparently
transmitted to UEs through eNdoeBs. The MME sends a paging message
to all eNodeBs in the TAL with which UEs are registered. After receiving
the paging message, the eNodeBs send the paging message to the UEs in
the TAL.
According to 3GPP 24.301, the MME is responsible for resending paging
messages. The MME sends a paging message to all eNodeBs in the TAL with
which UEs are registered. If the eNodeBs receive the paging response, the
MME stops resending paging messages and triggers RRC connection setup. If
the eNodeBs do not receive the paging response, the MME resends the paging
messages to the eNodeBs. If no response is received for many times, the
MME notifies the gateway of paging failures. The eNodeBs only transmit
paging messages because the UE can respond to the paging message only
under one eNodeB. Therefore, if the paging message is resent under other
eNodeBs, radio resources on radio channels will be greatly wasted.
II
Performance
Criteria for Evaluating LTE Paging
 Low power consumption

Battery consumption of UEs in idle mode is the top issue during paging
design. The DRX mechanism is used. With this DRX mechanism, the
paging cycle determines how many times the UEs can monitor the
paging channel during a specified period of time, thereby determining
battery consumption of UEs within a specified period of time.
 Low paging delay
The UE needs to read the paging message at specific time in a PC.
Therefore, the PC determines the paging delay. That is, a long PC may
bring a long delay.
 Low network loads
Paging capacity is determined by the model of services initiated by UEs

and is closely related to paging area planning. Attributes of the service
model vary with scenarios and requirements on paging area planning
also differ. Low network loads refer to reducing paging loads of a single
eNodeB and MME and ensuring low TAU loads in some scenarios.
JJ Typical Traffic Models on the MME
Parameter MME
Value
Average attached ratio of PS subscribers @ BH (%) 100%
CSFB ratio of PS subscribers @ BH (%) 50%
Default Bearer Activity Ratio of attached subscribers @ BH(%) 100%
Dedicated Bearer Activity Ratio of attached subscribers @ BH(%) 46%
ECM Connected Ratio of attached subscribers @ BH(%) 10%
EPS Attach per subscriber @ BH 0.8
EPS Detach per subscriber @ BH 0.8
Default bearer Context Activation times per attached subscriber @ BH 0.8
Default bearer Context Deactivation times per attached subscriber @ BH 0.8
Bearers numbers per attached subscriber @ BH 2
Duration per dedicated bearer context @ BH (minute) 43
Dedicated bearer Context Activation times per attached subscriber @ 0.64
BH
Dedicated bearer Context Deactivation times per attached subscriber @ 0.64
BH
Bearer Context Modification times per attached subscriber @ BH 0.14
Service Request Procedure per attached subscriber @ BH 9.67
Service Request data caused per attached subscriber @ BH 4.16
Service Request CSFB MO-Call caused per attached subscriber @ BH 2.00
Service Request CSFB MO-SMS caused per attached subscriber @ BH 0.60
S1 Release Procedure per attached subscriber @ BH 9.03
TAU per attached subscriber @ BH Intra MME 3
Inter MME 3.00
Periodic TA Updates per attached subscriber @ BH 1
Handover per ECM connected subscriber Intra MME 6.3

@ BH
Inter MME 0.01
Handover per attached subscriber @ BH Intra MME 0.63
Inter MME 0.001
Authentication procedures per attached subscriber @ BH 3.80
Paging CSFB MT-SMS caused per attached subscriber @ BH 0.80
Paging CSFB MT-Call caused per attached subscriber @ BH 2.00
Paging per attached subscriber @ BH 2.17

RF Network Planning and Optimization Service V200R001

Uploaded by

Copyright:

Available Formats

RF Network Planning and Optimization Service V200R001

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RF Network Planning and Optimization Service V200R001

Uploaded by

Copyright:

Available Formats

RF Network Planning and Optimization Service

HUAWEI TECHNOLOGIES CO., LTD.

Trademarks and Permissions

Huawei Technologies Co., Ltd.

2 PRACH ZC Root Sequence Planning........................................................................................66

3 LTE Neighboring Cell Planning.....................................................................................................................134

A.7.7 Criteria for Evaluating LTE Paging Performance.....................................................................................................289

1 LTE PCI Planning

1.1.1 Background and Benefits

Figure 1.1.2.1.1.1.1.1 PCI feature

PCI planning PCI check PCI re-allocation

1.1.3 Application Restriction

Application Scenario Restriction

Table 1.1.3.1.1.1.1.1.1 PCI application scenarios

Scenario Restrictions Workaround

Table 1.2.1.1.1.1.1.1.1 PCI groups

Group Number PCIs in Each PCI Group

Figure 1.2.1.1.1.1.1.2 PCI collision

Figure 1.2.1.1.1.1.1.3 PCI confusion

1.1.1 Technical Principles of PCI Planning

The value of u ranges from 0 to 29.

PUSCH: u = (PCI + deltss) Mod30 (deltss ranges from 0 to 29)

PUCCH and SRS: u = (PCI) Mod30

PUSCH: u = (fgh + PCI + deltss) Mod30

PUCCH and SRS: u = (fgh + PCI) Mod30

1.2.2 Technical Principles of PCI Check

1.2.3 Technical Principles of PCI Re-allocation

1.3 Scenario Description

Check and evaluate Re-allocate PCIs to

1.4 Application Guide

− Projection coordinate system

Table 1.4.1.1.1.1.1.1.1 Site information

Parameter Meaning Recommended Remarks

Table 1.4.1.1.1.1.1.1.2 Sector information

Parameter Meaning Recommended Remarks

Transceiver Indicates the sector The value is N/A

Parameter Meaning Recommended Remarks

Electrical Down Indicates the N/A If the E-UTRAN cell is

Table 1.4.1.1.1.1.1.1.3 Cell information

Parameter Meaning Recommended Remarks

Parameter Meaning Recommended Remarks

Main Indicates the N/A This parameter is used when

Table 1.4.1.1.1.1.1.1.4 Intra-frequency neighboring cell information

Parameter Meaning Recommended Remarks

Table 1.4.1.1.1.1.1.1.5 Inter-frequency neighboring cell information

Parameter Meaning Recommended Remarks

Parameter Meaning Recommended Remarks

Table 1.4.1.1.1.1.1.1.6 Parameters involved in the PCI planning policy

Parameter Meaning Recommended Remarks

Available PCI Indicates the N/A This parameter is set according

Parameter Meaning Recommended Remarks

PCI Mod3 Indicates that Deselect PCI On some networks, the

Parameter Meaning Recommended Remarks

Area Indicates the N/A If the user does not draw a

Parameter Meaning Recommended Remarks

Figure 1.4.1.1.1.7.1.1 PCI planning result

PCI Check (Evaluation)

Table 1.4.1.1.1.7.1.1.1 Site information

Parameter Meaning Recommended Remarks