GSM Um Signalling & Radio Ion Training Material
GSM Um Signalling & Radio Ion Training Material
Course Outline
GSM system overview Radio optimisation
Um interface
Section A
Topic Outline
System architecture
Downlink
Uplink
Downlink
1785 MHz
1710MHz
DCS-1800
DCS-1800 : 512-885
GSM radio optimisation . Figure 5
+95 MHz
Data
3 57
TSC
1 26 1
Data
57 3 8.25
4 4
BCS
40 4
114
114
114
114
57 57 57 57 57 57 57 57
Mapped on 4 TDMA TS
01234567 01234567
01234567
01234567
SACCH Channel
SACCH DOWNLINK Block Rsvd Ordered MS power (5 bits)
Typical message sequence on SACCH 3.270 DL RR System Info 5 3.320 UL RR Measurement Report 3.750 DL RR System Info 5 3.800 UL RR Measurement Report 4.210 DL RR System Info 6 4.260 UL RR Measurement Report 4.700 DL RR System Info 5 4.750 UL RR Measurement Report 5.170 DL RR System Info 5 5.220 UL RR Measurement Report 5.660 DL RR System Info 6 5.700 UL RR Measurement Report 6.120 DL RR System Info 5 6.190 UL RR Measurement Report
Channel Organization
Most important combinations TS0 TS0
TS0 TS0
01234567
In 51-TDMA Frame we have 9 paging Blocks (DL) In a complete paging cycle the maximum number of 51-TDMA frames is 9 In Uplink only RACH is used
GSM radio optimisation . Figure 12
Alternative Mapping on TS 0
Used in Low Signaling Traffic Cells
1st 51-TDMA-frames
2nd 51-TDMA-frames
TDMA Frame
Used on TS1-7
TDMA Frame
Um interface
Abis interface
A interface
OSI Model
Section B
UM interface UM interface
Topic Outline
Um Protocol Stack
MS in Dedicated Mode
MS in Idle Mode
Exercises
Q&A
Um Protocol Stack
Upper Layers
User
MM
Layer 3
CC
RR
Layer 1 & 2
Transmission
Layer 2
LAPDm Protocol
Transmission
Layer 1
Radio
RR
CM
SS
BTS
CC Call Proceeding CC Alerting CC Connect CC Connect Ack
Modify, Modify Reject, Modify Complete
Establishment
Active State
Hold, Hold Ack, Hold Reject Retrieve, Retrieve Ack, Retrieve Reject
Release
CM support
Support MS mobility
Location Update (normal, periodic, IMSI attach) Registration messages (IMSI detach, LA Update Accept, LA Update Reject, LA Update Request)
MS
MM
User
Support User Confidentiality Authentication Security messages (Authentication Reject /Request/Response, Identity Request/Response)
Layer 3: RR Management
MS
MM
MOC
BTS
Dedicated Mode
RR
Idle Mode
Layer 2 service
MOC: Establishment of RR connection initiates via MM MTC: by the peer entity on the NS via the paging channel
Release
CM MM RR RR RR
CM MM RR
Layer 3
CM MM RR
Connection Management Sub-layer Mobility Management Sub-layer Radio Resource Management Sub-layer
Layer 2
BTS
MS scans for BCCH frequencies stop scanning the selected PLMN MS tries cells by Rx level
Suitable cells are not barred and have C1>0 (with some other criteria).
Cell Reselection
C1>0?
Highest C1?
Cell Selection
No
No
GSM radio optimisation . Figure 32
<>11111 (Encourage)
C1 + CELL_RESELECT_OFFSET
=11111 (Discourage)
C1 CELL_RESELECT_OFFSET
Yes
Yes
Cell Reselection
LA updates generate a significant amount of load on MSCs. LA borders should be put in areas with clear cell borders
GSM radio optimisation . Figure 34
Neighbors Interval
10 20 8s 15s
Note MS may select wrong cell: Fast moving users, long neighbor lists, high BS_PA_MFRMS
Compare
No DSC - 4
DSC +1
DSC<=0?
Yes
Up to 90/BS_PA_MRFMS
DL signaling Failure
Cell Reselection
GSM radio optimisation . Figure 36
BA-IND: Differentiate between different BA lists EXT-IND (for 5 & 5bis only): Indicates complete message or extension MS reports measurements to the network every 480 ms
23
4 5 6 7 0 1
23
4 5 6 7 0 1 2 3 4 5 6 7 0 1
23
4 5 6 7
DL
Rx
MS MS measures at measures at least 25 least 25 Nbor SS Nbor SS measureme measureme nt samples nt samples in 26 TDMA in 26 TDMA frame frame Multiframe Multiframe
Rx Tx
0 1
Rx Tx
Measures BSIC
Tx 23
4 5 6 7
23
4 5 6 7 0 1
23
25
4 5 6 7 0 1 2 3 4 5 6 7 0 1 Idle
UL
24
Nbor BSIC is Nbor BSIC is measured during the measured during the Idle frame using the Idle frame using the sliding window sliding window technique technique
Implementation
MS-implementation is specified in GSM 05.08. Based on the successful decoding of messages on the downlink SACCH. BTS-implementation is manufacturer-specific. Based on successful SACCH decoding and/or RxLev/RxQual measurements.
GSM radio optimisation . Figure 42
S is initialised on every channel change (SDCCHTCH, intra-cell HO, HO failure with return, etc.) RR System Info Type 6 Downlink SACCH Cell Id: MCC-MNC-LAC-CI: 123-20-401-15243 MaxRadioLinkTimeout: 16 multi-frames
SACCH Multiframes
GSM radio optimisation . Figure 43
BTS
Radio resource (RR) establishment. Request and assign a radio resource for signaling (SDCCH) Service request
SDCCH
CC Setup CC Call Proceeding RR Assignment Command Assign a traffic channel (TCH) RR Assignment Complete Call initiation
TCH
CC Setup (MS to NW) bearer capability, BCD called party number CC Call Proceeding RR Assignment Command RR Assignment Complete CC Alerting CC Connect CC Connect Ack Connected number, Connected subaddress Connect acknowledge
GSM radio optimisation . Figure 45
RR Channel Request (RACH) Establishment Cause: Mobile originating call Random Reference: 3 RR Immediate Assignment (AGCH) Establishment Cause: Mobile originating call Random Reference: 3 Time: T1': 29, T2: 19, T3: 46
Use of time stamping, e.g. MS1 and Use of random reference MS3 & MS4
. RxLev Resolution invalid RACH due to interference. both mobiles must re-attempt different random reference: Strongest MS will get channel (with correct random reference in Imm. Assign) MS3>>MS4 same random reference: both MS seize the channel. The weakest MS will fail to establish a layer-2 link with the network due to interference from strongest MS Similar
...
MS3
MS4
MS
Paging Request - PCH
BTS
restarted
S+Tx TDMA frames
Channel Request (resent, IA not received)
restarted
Up to MaxRetransmissions Times
Stopped
BTS
RR establishment
SDCCH
CC Setup CC Call Confirmed RR Assignment Command Assign a traffic channel (TCH) RR Assignment Complete CC Alerting CC Connect CC Connect Ack Call accepted Call initiation
TCH
User alerting
BTS
New TCH
Physical Information
Handover Complete
New TCH
Handover Access . MS starts T3124 MS timer T3124 expires, or lower layer failure
Handover Access . MS starts T3124 MS timer T3124 expires, or lower layer failure BSC timer T3103 expires, or lower layer failure MS and BSC release all radio resources
Old TCH
Handover Failure
Um intra-cell handover
Intra-cell TCH-TCH handovers are initiated by the RR Assignment Command
Success
Note In general, the Assignment Command is used to instruct the mobile to change dedicated channel within a cell, whereas the Handover Command is used to change channel across cells.
Failure, return Assignment Failure is sent on old channel to old channel MS lost during none of these messages are sent. The MS channel doesnt manage to seize the new channel, change and fails to return to the old channel. The MS will return to idle mode
GSM radio optimisation . Figure 53
Um call clearing
Initiated by MS MS
CC Disconnect CC Release CC Release Complete RR Channel Release Messages CC Disconnect Release of RR Call clearing
BTS
Causes (Note abnormal releases in CAUSE analysis) Normal call clearing, User busy, No user responding, User alerting - no answer, Call rejected, Pre-emption, Invalid number format (incomplete number), Normal - unspecified, No circuit/channel available, Network out of order, Temporary failure, switching equipment, congestion, etc.
CC Release & CC Release Complete: optional information element cause RR Channel Release normal release, abnormal release - unspecified, abnormal release - timer expired, etc.
GSM radio optimisation . Figure 54
Um dropped call
Normally the cause for dropped calls cannot be inferred from drive test data. Typical dropped calls:
Radio link timeout at MS
DL_Lev MS goes to idle mode and starts analysing Sys Info (type 3)
DL_RxQual
Cause in RR channel release message Cause in CC Disconnect, CC Release, CC Release Complete message.
Section C
Topic Outline
CS call benchmarking analysis
CS call analysis
CS Call Analysis
Identify Problems
Negative Events
Dropped Call Setup Failure Handover Failure
Poor Level
Rx Lev Analysis Missing Handover Missing Neighbour
Poor Quality
Poor Rx Qual in poor Level Interference Poor RX Qual in good level Excessive or ping-pong handover
Is pointed to the following file Please load the logfile: Nemo Drive test 1. dt1
GSM radio optimisation . Figure 59
RX Lev Analysis
We analyse the DL Rx level to indicate the coverage Right click the level from GSM-Downlink measurements & place on the Map
We can see from the map that generally the level is good but in the area circled the level is bad. Lets look into this in more detail
Rx Lev Analysis
RX Qual Analysis
Right click the DL Serving Rxqual & put on the map
Rx Qual Analysis
Correlation between areas of poor level and poor quality
This is expected, and is more pronounced as traffic volumes increase
We can see other areas of poor quality where coverage was good
Interference? Excessive or ping-pong handover?
Last CI
Interference
Timing Advance
Handover Analysis
Import the HO OK Event query & display on the statistics explorer
Handover Analysis
Using a map we can look into handovers in more detail
Handover Analysis
The following Analysis pack gives some useful over view figures regarding handovers & call set up.
Handover Analysis
CS Call Benchmarking
Why Benchmark?
Overview of Performance on certain routes KPI performance can give picture of user perception High Level Analysis can help distinguish between area-specific and network-level problems
It is useful to run benchmark reports prior to detailed analysis to get a feel for the problems on the route
GSM radio optimisation . Figure 87
CS Call Benchmarking
Actix design validation & quick analysis application pack provides a series of reports & statistics that can be used to bench mark your network Right click the stream & open the design validation & quick analysis application pack The statistics window displays a count of various KPIs per Cell this can then be used to start to identify problem cells
CS Call Benchmarking
Call details Report
This gives details of the of the main call events such as call set up & call drops negative events are highlighted in red.
CS Call Benchmarking
The Design Validation report
This report looks at the level quality & interference with your network & compares to a series of Matrix to decide if the area driven is of good or bad design
CS Call Benchmarking
Measurement Charts report
Shows distribution of Level, Quality and Timing Advance
CS Call Benchmarking
Neighbor level
This report looks into the dominance of your server