4-1

4. RF Coverage and Capacity

Overview

Purpose
The design for RF coverage and capacity for stand-alone deployment begins with
the calculation of the Reverse link (uplink) and forward link (downlink) budgets
to determine the base station coverage area for a desirable data rate at the cell
forward edge. If the 1xEV-DO base station is being overlaid in a mixed 3G-
1X/IS-95 environment, Alcatel-Lucent strongly recommends that the 1xEV-DO
base station coverage be aligned to the 3G-1X/IS-95 coverage area, thereby
greatly simplifying the link budget task associated with base station deployment.
The objective of reverse link budget analysis is to calculate the maximum path
loss value permitted that will result in a quality signal at the receiver. The result of
this calculation is a dB value that represents the maximum amount of attenuation
the AT signal is permitted to encounter as the AT user travels away from the base
station. If the AT user continues to travel away from the base station, assuming
that no candidate sectors are able to accept a handoff, the maximum path loss
value is exceeded, and the signal quality will be diminished to a point at which the
call is dropped.
Although link budget calculation and analysis for 1xEV-DO is similar to those
performed in 3G-1X and IS-95, a number of differences must be considered.

As in all RF technologies, coverage estimates for planning purposes can be
obtained through the use of link budget tools. This section will examine the link
budgets for the reverse link. Reverse link budget analysis is used to establish the
cell footprint for a given data rate.





4-2


4.1 Reverse Link Budget Analysis

Overview

Purpose
A link budget analysis is a series of mathematical calculations, signal gains, and
losses as it travels from transmitter to receiver. In a typical duplex wireless
system, two link budget calculations exist: a forward link or downlink from the
base station to the AT or mobile unit, and a reverse link or uplink from the AT or
mobile unit to the base station. Link budgets are used to derive maximum path
losses for forward and reverse wireless communication links that meet design
criteria for reliability and performance.

4.1.1 Reverse link description

Introduction

A link budget calculation is a full accounting of the RF signal level gains and
losses as the signal travels from transmitter to receiver. This accounting is a
budget of signal gains and losses with respect to interference and noise levels to
obtain the maximum path loss permitted that will result in an acceptable signal
quality at the receiver. A balance must be achieved between gains and losses so
that the transmit signal received by the base station from an AT at the edge of the
coverage area is minimally above the total noise and interference experienced at
its receiver to ensure acceptable quality.
CDMA systems trade-off between signal quality, coverage (cell radius), and
capacity (data throughput). Traditionally, in IS-95 and 3G-1X systems, voice
quality is the ultimate criterion to consider when determining the link budget.
Once a voice quality objective is defined, this trade-off is narrowed between
coverage and capacity. However, because 1xEV-DO eliminates voice
transmission and the real-time restriction associated with high voice signal, high
quality associated with voice is also eliminated, essentially reducing the trade-off
between coverage and capacity. As capacity increases, coverage decreases. For
data, capacity is measured as the data throughput on a carrier.

Reverse Link Similarity with 3G-1X
4-3



The reverse link of a 1xEV-DO carrier is similar to the reverse link of a 3G-1X
carrier. Unlike the forward link, which is time-shared with each active user, the
1xEV-DO reverse link is CDMA code-shared with embedded pilot pulses for
coherent detection, and has similar power control and data rate (9.6 to 153.6 kbps)
schemes with 3G-1X. In addition, the 1xEV-DO reverse link enables soft handoff
similar to 3G-1X.
However, the 1xEV-DO reverse link differs from 3G-1X in that 1xEV-DO does
not have fundamental and supplemental channels, and that the reverse link data
rate is dynamically controlled by the base station based on sector loading. The AT
initiates its transmission data rate at 9.6kbps and may incrementally increase or
decrease its data rate after every 26.67-ms frame following a transition probability
based on RAB (Reverse Activity Bit) set by the base station. The data rate
selected by the AT is reported to the base station via a data rate control (DRC)
channel. The 1xEV-DO reverse link data rate is indicated by an RRI (Reverse
Rate Indicator) channel on the reverse link that is used to inform the base station
of the rate that the AT is transmitting.

4.1.2 Maximum Path Loss

Maximum Path Loss Components

The reverse link budget analysis is performed to compute the maximum allowable
path loss between the Access Terminal (AT) transmit antenna and the cell site
receive antenna. If forward link analysis indicates that the forward link can
support performance at the same loss, the maximum path loss can be used on a
market-by-market basis in the RF design. This design process employs algorithms
that map loss into cell radii via consideration of local variables such as tower
height, terrain, and clutter.
The allowed point-to-point path loss is determined by considering the terms that
dictate net loss from the AT to the cell. Components of the net loss are indicated
in Figure 4-1, Components of Net Path Loss from AT to Base Station

4-4


Figure 4.1 Components of Net Path Loss from AT to Base Station


Maximum Path Loss Calculation

The terms characterizing the net loss are captured in the following relation:

Figure 4-2 Equation 1

Where:
Xmax = Maximum AT transmit power (EIRP) out of the antenna (in dBm)
HL = Head and body loss (in dB)
BL+VL = Building and vegetation (and other) penetration loss (in dB)
PL = Average path loss between AT antenna and cell site antenna (in dB)
fade = Fade at AT location (in dB)
AG = Cell site antenna gain (in dBi)
4-5


CL = Cell site cable loss (in dB)
Smin = Base station receiver sensitivity (in mW, converts to dBm).

The maximum AT transmit power (Xmax) must be sufficient to overcome the
maximum path loss so that the signal power received at the base station transmit
I/O port J4 port (antenna connector) is equal to or exceeds the base station
receiver sensitivity, Smin.
The above expression is rewritten for the allowed maximum dB path loss. This
value dictates the edge (boundary) of the cell coverage.
Figure 4-3 Equation 2


The above expression can be viewed as constructing the allowed maximum path
loss as a dB sum of credits (e.g., AT transmit power) and deficits (e.g., cable
loss). This dB process is captured in the reverse link budget.

Maximum AT Transmit Power

The Effective Isotropic Radiated Power (EIRP) or Effective Radiated Power
(ERP) is the power out of the antenna and equals the sum (in dB) of the AT
transmit power and the AT antenna gain. The difference between EIRP and ERP
is 2.15dB. Table 4-1 shows the maximum transmit power for 850 MHz and 1900
MHz as defined by the standard.

Table 4-1 Maximum AT Transmit Power
Frequency Band [MHz] Maximum AT EIRP [dBm]
850 25
1900 23

4.1.3 Shadow Fading
4-6



The uplink maximum path loss is the maximum loss in signal strength permitted
as an AT signal is propagated outward in space. As illustrated in Figure 4-1,
Components of Net Path Loss from AT to Base Station, in an actual
application, the AT signal does not always travel in free space, and the
propagation path between transmitter and receiver will be obstructed. Losses
attributed to obstructions in the signal propagation path are referred to as shadow
fading or slow fading losses, which result in the dispersion of the received signal
strength at a fixed distance from the cell site.
The obstructions, primarily from tall buildings and heavy vegetation, cast RF
shadows on the paths leading away from the AT. Other losses are body losses; the
user may be positioned between the AT and base station antenna. Normally, the
shadow paths are not completely darkened due to RF signal reflection from other
surrounding buildings. Signal reflection from a large number of buildings, which
is typical in an urban environment, causes random in-phase reinforcement and
interference with the RF signal. Reflected signals may reinforce each, producing a
gain. As a result, the actual path loss or gain at any point in such an environment
will vary as a function of the predictable path loss and unpredictable shadow
fading loss.

4-7


4.1.4 Reverse Link Budget

Introduction

The objective of reverse link budget analysis is to calculate the maximum uplink
path loss value permitted that will result in a quality signal at the receiver. The
result of this calculation is a dBi value that represents the maximum path loss
attenuation (with respect to an isotropic antenna) an AT transmitted signal is
permitted to encounter as the AT user travels away from the base station. If the
AT user continues to travel away from the base station, assuming that no
candidate sectors are able to accept a handoff, the maximum path loss value is
exceeded, and the received signal quality will be diminished to a point that the
call is dropped.

Typical link budget analysis

A typical link budget analysis form for 90% area coverage at different data rates
is shown in Table 5-2, PCS Reverse Link Budget Spreadsheet (p. 5-9) , in Rev
0 and Tables 5-3 and 5-4, in Rev A. By accounting for sources of path loss, noise
interference, and margins for specified signal quality and loading, which is the
amount of traffic on a carrier, as well as sources for signal gains, the maximum
allowable path loss for the reverse link can be determined.
After the maximum allowable path loss is determined, its value is inserted into a
propagation model or propagation tool to determine the cell radius for a given
quality. Multiple propagation models and tools are available, which address a
variety of environmental and geographic situations and base station antenna
heights.
In an ordinary urban area under the 800 MHz band, the coverage probability in
the area is 95% (the edge coverage probability is 87%), the antenna height is 25 m
(the feeder length is 35 m), the antenna gain is 15 dBi, the reverse load rate is
75%, and other parameters take on default values. The reverse link budge is
described in Table 4-2 Reverse link budget


Table 4-2 Reverse link budget
4-8


Reverse link Budget Detail
Information
Cell Edge
service
rate
(kbps)
for DoA
Cell Edge
service rate
(kbps) for
DoB 2X/
per carrier
Cell Edge
service rate
(kbps) for
DoB 3X/
per carrier
Remark
Reverse Service Data
Rate (kbps)
76.8 38.4 25.60
cell edge service
rate
AT Max Transmitting
power (dBm)
23.00 20 18.23
A= 10x
log(200mw/N),
where N is the
Number of bound
Carriers
AT Feeder Cable &
Connector Loss (dB)
0.00 0.00 0.00 b
AT Antenna Gain (dBi) 0.00 0.00 0.00 c
AT Body Loss (dB) 0.00 0.00 0.00 d
AT EIRP (dBm) 23.00 20 18.23
e=a-b+c-d
Background Thermal
Noise Density (dBm/Hz)
-174.00 -174.00 -174.00 f
BS Noise Figure (dB) 4.00 4.00 4.00 g
Required Eb/Nt For
Reverse Investigated
service (dB)
1.71 2.36 3.00 h
Reverse Processing Gain
(dB)
12.04 15.05 16.81 i=10*log(W/R)
BS Receiver Sensitivity
(dBm)
-119.43 -121.79 -122.92
j=10*LOG(10^(f/
10)*W)+g+h-i
BS Antenna Gain (dB) 15.00 15.00 15.00 k
BS System Feeder Cable
Loss (dB)
1.27 1.27 1.27 l
BS System Jumper Loss
(dB)
0.13 0.13 0.13 m
BS Total Connector Loss
(dB)
0.50 0.50 0.50 n
Required Minimum Received
Signal Strength(dBm)
-132.53 -134.89 -136.02 o=j-(k-l-m-n)
Soft HandOver Gain
Again Slow Fading (dB)
4.66 4.66 4.66 p
Shadow Fading Margin
(dB)
10.72 10.72 10.72 q
Interference Margin (dB) 6.02 6.02 6.02 r
4-9


Building Penetration Loss
(dB)
18.00 18.00 18.00 s
Max Allowed Propagation
Loss For Cell Radius (dB)
125.45 124.82 124.17 t=e-o+(p-q-r-s)
Morphology Urban Urban Urban u
Propagation Model
Okumura
Hata
Okumura
Hata
Okumura
Hata
v
System Carrier Center
Frequency (MHz)
825.00 825.00 825.00 w
BS Effective Height (m) 25.00 25.00 25.00 x
AT Effective Height (m) 1.50 1.50 1.50 y
Reverse Link Cell Radius
(km)
0.93 0.90 0.86
z=function(t,u,v,
w,x,y)

As shown in Table 4-2 Reverse link budget, when the cell edge service rate is
the same (76.8 kbps), the reverse cell radius of DORB 2X or DORB 3X is smaller
than that of DORA. The DORB terminal reduces the number of carriers from
three to two or one automatically when its power is insufficient. Therefore, in
fact, the reverse coverage of DORB may be considered to be the same as that of
DORA.
Note: The reduction of carriers is decided by the terminal. The transmit power of
the terminal is unknown to the system. All DORB terminals have the function,
which is implemented by Qualcomm chips.

4.1.5 Receiver Sensitivity


Description

As stated in the previous section, the base station receiver sensitivity is the
receivers ability to discriminate the signal from noise and interference.
Specifically, in reference to the CDMA link budget analysis form, receiver
sensitivity, entered as link budget Item m, can be regarded as a parameter that
determines the power (in dBm) required at the input of the CDMA receiver to
maintain a desired frame error rate (FER). This power level is equal to the power
4-10


level at the base station antenna plus the antenna gain less the antenna cable and
connector loss:

Figure 4-4 Equation 3

Where:
S
min
= Receiver input power level
P
ant
= Base station antenna gain input power level from a single AT
G
ant
= Base station antenna gain
L
cab
= Antenna cable and connector loss.

Smin Signal Quality

The quality of the Smin signal is determined by its signal-to-noise ratio. In
CDMA, this ratio is expressed as energy per bit divided by the total ambient noise
and interference level (Eb/Nt, commonly pronounced as eb-no). The energy per
bit is calculated by dividing the receive power, Smin, which expressed in
Joules/second rather than watts, by the data bit rate, R:

Figure 4-5 Equation 4

The noise refers to the receiver noise floor calculated for Item j. The receiver
sensitivity must also account for signal bit levels above the RF ambient noise
level, which is computed as energy per bit divided by the total ambient noise and
4-11


interference level (Eb/Nt) and the receive data rate, and is computed by summing
Items j through l.

4.1.6 Required Eb/Nt, Item l

Description

The quality of the signal received at the base station is determined by the strength
of the carrier signal level (that is, its bit energy) above the noise and, more
importantly, interference levels. As stated earlier in this chapter, in CDMA the
signal to noise relationship is measured as the bit energy bit-to-total noise ratio, or
Eb/Nt. The larger the Eb/Nt value, the higher the signal quality. In 1xEV-DO, the
signal quality can be measured by the packet error rate (PER), which is the
percentage of packet that must be transmitted because its data could not be
recovered. The disadvantage of transmitting at a high Eb/Nt value is that it
consumes more AT battery power; even worse, it creates a higher-level of RF
interference to other users in the environment. Therefore, the design objective is
to create a system that requires the lowest Eb/Nt value for a target PER.
The required Eb/Nt for a given AT is a function of its mobility, the multipath
environment, and target packet error rate (PER). The required Eb/Nt values listed
in Table 4-3, Reverse Link Required Eb/Nt Values are estimates at each data
rate considering all the power the AT radiates. This includes the power from the
non-traffic channels such as DRC, pilot/RRI, and ACK channels. The Eb/Nt
values are based on link layer simulations.
Reverse Link Required Eb/Nt Values

Table 4-3 Reverse Link Required Eb/Nt Values
Data Rate (kbps) Required Eb/Nt (dB)
Mobility Stationary
4.8 6.8 5.8
9.6 4.9 3.9
19.2 3.9 2.9
28.8 3.4 2.4
38.4 3.3 2.3
4-12


57.6 2.7 1.7
76.8 2.6 1.6
153.6 2.1 1.1
230.4 1.7 0.7
307.2 1.8 0.8
460.8 3.4 2.4

Vehicle Speed Effect on Eb/Nt Value

Higher Eb/Nt values are required when the AT is operating from a moving
vehicle because shadow or slow fading will occur more frequently when the AT is
motion. In 1xEV-DO, the influence of shadow fading is minimized in two ways:
fast power control and bit interleaving. The relationship between vehicle speed
and Eb/Nt value is shown in Figure 4-6, Relationship Between Vehicle Speed
and Eb/Nt Value.
Figure4-6 Relationship Between Vehicle Speed and Eb/Nt Value

Consequences of Power Control at Low Vehicle Speeds

4-13


At low vehicle speeds, the AT reaction to the base station power control is fast
enough to respond to most shadow fading conditions. In addition, the AT is more
likely to remain in a multipath environment for a longer period to help maintain a
low bit error rate (BER). As the vehicle speed increases moderately, a worst-case
condition is approached at speeds between a narrow range.

Consequences of Bit Interleaving at High Vehicle Speeds

At higher vehicle speeds, the fast fading durations are smaller, enabling better
data recovery from bit interleaving. Bit interleaving is operated in conjunction
with the turbo coder. The turbo coder uses convolution coding, which is very
effective for recovering from corrupted bits scattered over the received bit stream.
However, during a fade, a cluster of consecutive transmitted bits are corrupted,
rendering turbo coding ineffective. To prevent fading from rendering turbo coding
ineffective, bit interleaving is performed by the AT after turbo coding. As a result,
the stream of data bits to be transmitted is pseudo-randomly scattered out of
sequence. If a fade is encountered, resulting in the corruption of a burst of
consecutive transmitted bits, when the bit stream is de-interleaved at the receiver,
rather than being clustered together, the corrupted bits will be scattered
throughout the de-interleaved bit stream, enabling turbo coding recovery of the
transmitted message.
The duration of the fast fade primarily depends on the vehicle speed. If the fade is
too long, a large number of corrupted bits are created so that even after de-
interleaving, consecutive corrupted bits remain in the bit stream to lower the turbo
coding efficiency. The duration of the fade will shorten as the vehicle speed
increases, thus increasing the turbo coder data recovery efficiency. Therefore, at
higher vehicle speeds, the required Eb/Nt value will decrease.

Data Rate Effect on Eb/Nt Value

The required Eb/Nt includes power from channels other than traffic channels such
as pilot, DRC, and ACK channels, and represents the total amount of power
required to transmit traffic information. The higher data rates have lower Eb/Nt
requirements because the Pilot, DRC, and ACK channels occupy a lower
percentage of overhead relative to the traffic channel as the rate increases.
However, at the 153.6 kbps rate, the Eb/Nt requirement is higher due to the
4-14


weaker turbo coding rate used at this data rate. The code rate identifies the ratio of
the number of information bits to the total number of information bits plus
overhead correction bits transmitted. To achieve the higher data rate, a code
rate is used for 153.6 kbps transmission versus the code rate used for 9.6, 19.3,
38.4, and 76.4 kbps data rates. This means that at the 153.6 kbps data rate, each
information bit is sent twice instead of four times as with the lower data rates. The
lower repetition rate at the 153.6 data rate offers fewer opportunities for turbo bit
correction, therefore requiring transmission at a higher Eb/Nt value.

4.1.7 Soft Handoff Gain

Description

Soft handoff occurs in the soft handoff zone, which is the outer edge of two or
more cells as the AT moves from the domain of one base station to the domain of
another. In AMPS and TDMA, the signal received from the cell outer edge is the
weakest, and their link budgets reflect this weakness through the increase in the
maximum path loss value, thus reducing the cell coverage radius. The opposite is
true in CDMA. When an AT enters the soft handoff zone, a signal gain is
experienced because the AT is communicating with the two or more base stations
that share the soft handoff zone. The overall result of an AT operation at the cell
outer edge is to reduce the maximum path loss value, allowing a lower Eb/Nt per
link value for the same coverage area.
Therefore, an advantage due to soft handoff exists that results in effectively
lowering the fade margin required to obtain a specific probability of edge
coverage, as compared to other technologies. For a CDMA system that admits
soft handoff, for any given reverse frame, the better or alternatively stronger of
two or more base stations reception will be utilized at the frame selector, typically
at the switching center.

4-15


Example

Assuming an 8 dB standard deviation and 50% partially correlated two-way
handoff, the soft handoff gain numerically works out to about 4 dB when edge
coverage probability is 90 percent. Due to the soft handoff feature, the excess link
margin requirement has dropped by 4 dB. This is the advantage due to soft
handoff that results in increased coverage.

4.1.8 Path loss

Log-Normal Fade Margin, Item p

The influence of shadow fading is shown in Figure 4-7, Propagation Loss. The
X axis is expressed as 10 times the log of the distance in miles; therefore, the 0
point on the graph represents a 1 mile (1.6 km) distance from the antenna and the
-3 point is one-half mile (.8 km) from the antenna. The mean value line shows
that as the distance from the antenna increases, the path loss increases. At any
given distance, there exists a significant variation in the path loss about its mean
value. A change in receiver position by a meter can result in a change in path loss
by as much as 20 dB. Such large changes in path loss typically occur when an
obstacle (e.g., a building or a hill) obstructs the RF path. For example, in the data
shown in Figure 4-7, Propagation Loss , the path loss at a distance of 2 miles
(3.2 km); that is, at the horizontal axis value of 3 which is the log of 2, varies
from as much as 147 dB to as low as 130 dB. The distribution of the path loss
follows the log-normal distribution, with the standard deviation of typically 8 dB.
As networks are designed using the mean path loss at a certain distance, you must
factor in a margin for this variation in path loss to assure adequate signal strength.
In the link budget, this term is called fade margin.

Propagation Loss diagram

Figure 4-7 Propagation Loss
4-16


Figure 4-7 Propagation Loss


Fade Probability

The probability of 90 and 75 percent cell edge coverage vs. fade margin is given
in Table 4-4, Probability Of Edge Coverage vs. Fade Margin . These values are
derived from the standard tables of the normal distribution for the extra margin
needed to obtain the desired probability of edge coverage.





4-17



Table 4-4 Probability of Edge Coverage vs. Fade Margin
Probability of edge
coverage
Probability of area
coverage
Margin in terms of
standard
deviation of fading
()
Fade Margin for
= 8dB [dB]
90% 95% 1.28 10.3
75% 90% 0.67 5.4

The above margin is interpreted in network design in the following way. Let us
assume that 148.1 dB is the maximum median allowable path loss. If the median
path loss of 148.1 dB is obtained after using a fading margin of 10.3 dB, and then
from a design point of view, such a network would be expected to have a 90%
probability of edge coverage, assuming the standard deviation is 8 dB.
The above calculation of fading margin is independent of technology; it would
apply to CDMA, AMPS, TDMA, and GSM systems.

Building/Vehicle Penetration Loss, Item q

The building/vehicle penetration loss is the amount of attenuation introduced due
to obstacles in the RF line-of-sight path. The AT transmit signal will be attenuated
by the over-the-air attenuation, and also by the building/vehicle penetration loss
and fading. These terms must be added to the reverse link over the air loss to
provide the total attenuation the signal will experience.

Maximum Path Loss with Respect Isotropic Antennas, Item r

The over-the-air maximum allowable path loss for any particular reverse link data
rate is calculated by subtracting and adding the following link budget items:
r = c - d - m + e - f + o + n - p q
Considering that the maximum allowable path loss for 3G-1X voice service is
150.5 dBi, about the same value as the lowest data rates for 1xEV-DO, which
indicates that a 1xEV-DO cell can be overlaid directly over a 3G-1X cell.
4-18


Remember that the reverse link maximum allowable path loss is the maximum
over the air path loss.

4-19



4.2 Forward Link Budget Analysis

Overview

Purpose

This section covers the process of forward link budget analysis.


4.2.1 Forward link description

Introduction

Classically, the objective of forward link budget analysis is to ensure that the
forward link has sufficient power to support the performance needed and the
desired throughput within the footprint established by the reverse link. In 1xEV-
DO, the forward link budget objective is to determine the percentage of the
coverage area established by the reverse link maximum allowable path loss that
can be achieved at each forward link data rate. As the data rate increases, the
percentage of area covered decreases.

Percentage of Area Covered Vs. Data Rate

The decreasing percentage of area covered at each data rate can be thought of as
ever-smaller concentric rings of coverage, as shown in Figure 4-8, Percentage of
Area Covered Vs. Data Rate. The outer-most (largest) ring represents over
95 percent of the cell footprint established by the reverse link, and is shaded to
show the maximum cell coverage achieved when operating at the lowest forward
data rate. The inner-most ring (closest to the base station) is shaded to represent
the highest data rates that can be achieved in the cell.

4-20


Figure 4-8 Percentage of Area Covered Vs. Data Rate


4.2.2 Forward Link factors

RF Conditions Evaluation by the AT

Unlike 3G-1X and IS-95, in 1xEV-DO the forward link is not code-shared to
distinguish each user within the sector. Instead, the forward link transmit signal is
dedicated to one user (at a time) on a time-shared basis. That is, the base station
communicates on the forward link traffic channel with each user during its
dedicated 1.667-millisecond time slot. To maximize the base station data
throughput, the forward link traffic data is then transmitted at full power to a
single AT selected by the scheduler algorithm. The transmit data rate is varied per
user based on feedback from the users on the RF condition that the AT is
experiencing. Essentially, the RF condition experienced by the AT is determined
by how well the AT can recover the turbo-coded packet information. If because
the AT moves away from the base station, the RF conditions deteriorate such that
the AT cannot recover the turbo-coded packet information at the current data rate
4-21


to maintain low BER, for the next time slot, the AT might request transmission at
lower data rate.

AT Minimum Performance Specification

The minimum performance specifications guide AT manufacturers on the
required Eb/Nt values for each forward traffic channel data rate. The Eb/Nt values
given in Table 4-5, Required Traffic Channel Forward Link Eb/No Value are
the values that are currently used for planning purposes. The minimum
performance specifications for the two highest data rates (1843.2 and 2457.6
kbps) are derived from requirement and objective values. The Eb/Nt values given
in this table for the two highest data rates are the linear average of their
requirement and objective values.

Table 4-5 Required Traffic Channel Forward Link Eb/No Value

Data Rate (kbps) Required Traffic Channel Eb/No (dB)
38.4 2.5
76.8 2.5
153.5 2.5
307.2 2.5
614.4 2.5
921.6 3.5
1228.8 5.0
1843.2 7.5 1
2457.6 10.5 1

Notes:
1. Linear average of requirement and objective specification values

4-22


Forward Link Signaling Channel

The Eb/Nt values given in Table 4-5, Required Traffic Channel Forward Link
Eb/No Value are for forward link traffic channels, as opposed to forward link
signaling channels, which are transmitted at the 76.8 kbps data rate. Rather that
waiting for optimum RF conditions to transmit on the traffic channels, data
transmission on the signaling channels occur at fixed schedule intervals. Because
the signal channels may not be transmitted during optimum RF conditions, the
required Eb/Nt level for signaling channels may be higher than the level required
for traffic channels.

Different Repetition Factors

The Eb/Nt values listed in Table 4-5, Required Traffic Channel Forward Link
Eb/No Value are kept to minimum levels by relying on the data recovery
techniques defined in the Physical Layer to correct bit errors resulting from the
low required Eb/Nt level. In other words, reliable data transmission is not only
dependent on meeting the required Eb/Nt value, but is also dependent on bit
recovery techniques such as turbo coding redundancy and the transmission
repetition factor that may vary at different data rates. Even though the required
Eb/Nt values for data rates 38.4 through 614.4 kbps are the same (2.5 dB), bear in
mind that the Physical Layer specifies different repetition factors for each data
rate in the form of the number time slot periods required to transfer each data
packet. For example, at the 38.4-kbps data rate, 1024-bit packets are turbo-coded
at a 1/5 code rate, producing 5120 bits (1024 X 5). The 5120 bits are QPSK-
modulated, resulting in 2560 2-bit symbols per packet. At the 38.4-kbps data rate,
the information in each packet is transmitted to the AT over 16 time slot periods.
Each slot contains 1600 chips for data, so 16 slots contain 25600 chips for data.
At this data rate, 1024 of those chips are used for preamble, leaving 24576 chips
for data. Therefore, the repetition factor is 9.6 (24576/2560), which mean that the
2560-bit data parcel can be transmitted 9.6 times within the allotted 16 slots.
When transmitting at the 76.8-kbps data rate, a 1024-bit-packet, which is also
turbo-coded at a 1/5 code rate, is also QPSK-modulated, resulting in 2560
symbols to be transmitted. However, when transmitting at 76.8 kbps, the
preamble size is reduced to 512 chips, and the packet is transmitted to the AT
over an 8-time slot period. Therefore, at 1600 chips per time slot, 12800 chips less
4-23


512 chips are used for data, resulting in a 4.8 [(12800-512)/2560] repetition
factor. Therefore, when reducing the transmission data rate from 76.8 kbps to 38.4
kbps, the repetition factor is doubled, decreasing the AT bit error rate (BER).
4.2.3 Link Budget Calculation

Percentage of the coverage area
The calculation for the 1xEV-DO forward link budget begins by determining the
percentage of the coverage area, derived from the reverse link that can be
achieved at each forward link data rate. At any particular forward traffic channel
data rate, the Eb/Nt value at the AT receiver antenna port must be equal to or
greater than the required traffic Eb/Nt value specified for that data rate in Table 4-
5, Required Traffic Channel Forward Link Eb/No Value . The value specified
in this table for any data rate is represented as d in the following expression:
Figure 4-9 Equation 5

Energy per bit

The energy per bit can be expressed in term of the AT receive power from its host
or serving sector (P
host
) divided by the bit rate. The total noise and interference is
the product of the receiver noise figure and the thermal noise density, plus the
power within the bandwidth of interest from the neighboring sectors. The
expression then becomes:
Figure 4-10 Equation 6

where:
R = Data rate
P
host
= Power from serving base station
F = Base station receiver noise figure
N
o
= Thermo noise density
P
other
= Power from neighboring sectors
W = Bandwidth which is reduced to account for traffic slots (~75% of total slots).

4-24


4.2.4 Forward Link Budge Spreadsheet

Overview
In an ordinary urban area under the 800 MHz band, the coverage probability in the area
is 95% (the edge coverage probability is 87%), the antenna height is 25 m (the feeder
length is 35 m), the antenna gain is 15 dBi, and other parameters take on default values.
The forward link budge is described in Table 4-6 Forward link budget.
Table 4-6 Forward link budget
Forward link Budget
Detail
Information
Cell Edge
service rate
(kbps) for DoA
Cell Edge
service rate
(kbps) for DoB
2X/ per carrier
Cell Edge
service rate
(kbps) for DoB
3X/ per carrier
remark
Forward Effective
Burst Data Rate (kbps)
300.00 150.00 100.00
Cell edge
data rate
BS Max Traffic Channel
Transmitting power
(dBm)
43.00 43.00 43.00 A
BS System Feeder Cable
Loss (dB)
1.27 1.27 1.27 b
BS System Jumper Loss
(dB)
0.13 0.13 0.13 c
BS System Connector
Loss +TMA Insertion
Loss (dB)
0.50 0.50 0.50 d
BS Antenna Gain (dBi) 15.00 15.00 15.00 e
BS System EIRP (dBm) 56.10 56.10 56.10
f=a-b-c-
d+e
Background Thermal
Noise Density (dBm/Hz)

-174.00

-174.00

-174.00

g
AT Noise Figure (dB) 8.00 8.00 8.00

h
Required C/I For
Forward
Investigated service (dB)
-3.37 -6.11 -7.61 i
Forward Processing
Gain (dB)
0.00 0.00 0.00
j=10*log(
W/R)
Terminal Receiver
Sensitivity
-108.48 -111.21 -112.72
k=10*LOG
(10^(g
4-25


(dBm) /10)*W)+h
+i-j
AT Atenna Gain (dB) 0.00 0.00 0.00 l
AT Feeder
Cable&Connector
Loss (dB)
0.00 0.00 0.00 m
AT Body Loss (dB) 0.00 0.00 0.00 n
Required Minimum
Received
Signal Strength (dBm)
-108.48 -111.21 -112.72
o=k-(l-m-
n)
Virtual SHO Gain (dB) 4.10 4.10 4.10 0
Shadow Fading Margin
(dB)
10.72 10.72 10.72 Q
Forward Interference
Margin
(dB)
7.43 2.49 1.60 R
Building Penetration
Loss (dB)
18.00 18.00 18.00 S
Max Allowed
Propagation
Loss For Cell Radius
(dB)
132.52 140.21 142.59
t=f-o+(p-q-
r-s)
Morphology Urban Urban Urban U
Propagation Model
Okumura
Hata
Okumura Hata Okumura Hata V
System Carrier
Center Frequency
(MHz)
875.00 875.00 875.00 w
BS Effective Height (m) 25.00 25.00 25.00 X

AT Effective Height (m)

1.50

1.50

1.50

y
Forward Link Cell
Radius (km)
1.41 2.32 2.70
z=function(
t,u,v,w,x,y)

Table 4-6 Forward link budget compares the forward link budge of DORB 2X
(two-carrier binding), DORB 3X (three-carrier binding) and DORA. When the
cell edge data rate is the same (300 kbps), the forward cell radius of DORB 2X or
DORB 3X is larger than that of DORA. Figure 4-11 Comparison of forward cell
radiuses gives a bar chart comparing the forward cell radiuses.
4-26


Figure 4-11 Comparison of forward cell radiuses

From the above forward link budget and cell radius comparison, it is known that
under a same edge rate (300 kbps), the forward cell radius of DORB 2X or DORB
3X is larger than that of DORA.

4.2.5 Summary

Considering the balance of forward and reverse links, under the same user
experience, DORB is better than DORA in terms of forward coverage. Owing to
the forward interference, the cell edge C/I of DORA is subject to an upper limit.
Accordingly, the forward edge rate also has an upper limit R edge. After the
multi-carrier DORB system is deployed, the forward edge service rate perceived
by the user is up to N x R edge. The inherent edge rate limit of the single-carrier
DORA is thus broken. The reverse rate is restricted by the power of the terminal.
The edge service rate perceived by a user is equivalent to that of DORA. Within a
cell where power restriction is not present, however, the service rate perceived by
a user is N times that of DORA.
4-27


4.3 Capacity Planning

4.3.1 Throughput per Sector

Because of the multi-carrier binding of DORB, the forward peak rate of
2X will rise to 6.2 Mbps and that of 3X will rise to 9.3 Mbps. The
distribution of forward rates of DORA, DORB 2X and DORB 3X is
shown in Figure 4-1, according to the emulation of Qualcomm.
Figure 4-12 Throughput per sector

As shown in Figure 4-1, the throughput per sector of DORB 2X and
DORB 3X is respectively increased to 2.5 Mbps and 3.8 Mbps relative to
DORA.
4.3.2 VoIP User Size

According to Qualcomm emulation, DORA allows 44 VoIP users and the
user size is restricted on the reverse link. According to the maximum
reverse capacity formula, the result of calculation is about 42. The formula
and calculation are as follows:
Figure 4-13 Equation 7

4-28


Where:
Nmax = the maximum number of users simultaneously accessed to a cell;
W/R = the spreading gain, where W = 1.2288 MHz and R = 9.6 kpbs;
= the voice activity factor which equals 0.45;
Eb/Nt = the required signal to noise ratio which equals 4.955 dB;
= the cell interference factor which equals 0.5.
The calculation result is Nmax=60.69. Allowing for a 5 dB margin
(68.37% load), the allowed number of users is 60.69 x 68.37% = 41.5

is obtained through calculation. According to the MSO model, there are
29% full rate channels, 4% 1/2 rate channels, 7% 1/4 rate channels, 6%
1/8 rate channels, and 54% idle channels. Considering a 22-byte voice
payload and an 8-byte overhead, the equivalent voice activity factor is
0.45. Therefore, in the case of 5 dB rise, the number of VoIP users is 42,
close to the emulation result of Qualcomm. For DORB, the supported
number of VoIP users is two times (2X binding) or three times (3X
binding) that of DORA, in particular, 84 and 126.

4.3.3 BE Throughput

Forward BE Throughput
Forward BE services are relevant to the terminal type, user distribution,
terminal movement, HARQ, radio channel environment and scheduling
algorithm. Figure 4-14 Forward BE throughput shows the emulation
result of forward BE throughput of DORA. Huawei's emulation result is
close to that of Qualcomm
4-29


Figure 4-14 Forward BE throughput

In capacity planning, to ensure the Internet experience of broadband users,
the recommended forward BE throughput of DORA is 1.2 Mbps. For
DORB 2X and DORB 3X, considering the scheduler gain of BE services,
the forward BE throughputs are respectively 2.5 Mbps and 3.8 Mbps.
Reverse BE Throughput
Reverse BE services are relevant to the terminal type, user distribution,
terminal movement, HARQ, and radio channel environment. Figure 4-15
Reverse BE throughput shows the emulation result of reverse BE
throughput of DORA.


4-30


Figure 4-15 Reverse BE throughput


In Figure 4-15 Reverse BE throughput, the reverse throughput per sector
is obtained by averaging the throughputs of 57 sectors. It is the largest
throughput among all sectors where the probability of ROT larger than 7
dB is below 1%. The emulation result of Huawei is consistent with that of
Qualcomm.

In capacity planning, to ensure the Internet experience of broadband users,
the recommended reverse BE throughput of DORA is 600700 kbps,
depending on the number of ongoing connections. For DORB 2X and
DORB 3X, because the service rate is two or three times that of DORA,
the reverse BE throughputs are respectively 1.21.4 Mbps and 1.82.1
Mbps.

4.3.4 Hybrid Service Planning
Figure 4-16 Forward hybrid throughput shows Qualcomm's emulation
of VoIP and BE hybrid throughput on the forward link.


4-31


Figure 4-16 Forward hybrid throughput


As shown in Figure 4-16 Forward hybrid throughput, when 30 VoIP
users are on the forward link, 50% of the forward load is occupied and the
forward throughput of BE will be 50% down. Therefore, for DORB 3X, in
the case of 10 VoIP users, the forward BE throughput is3.8 Mbps x 0.75 =
2.8 Mbps; in the case of 20 VoIP users, the forward BE throughput is 3.8
Mbps x 0.69 = 2.6 Mbps; in the case of 42 VoIP users, the forward BE
throughput is 3.8 Mbps x 0.4 = 1.5 Mbps.


4-32


4.3.5 Traffic Model

Table 4-7 Forward traffic model

Items Data & voice subscriber Voice subscriber
Proportion 100% 0%
Hierarchy Lower Medium Higher -
Sub-proportion 60% 25% 15% -
Traffic type
Data rate
(kbps)
- - - -
VOIP (erl) 9.6 0 0 0 0
VT (erl) 76.8 0 0 0 0
BCMCS (second) 204.8 0 0 0 -
PPP session Time (s) - 300 600 900 -
Packet call duty ratio - 10% 15% 20% -
115.2 kbps 115.2 20% 15% 10% -
230.4 kbps 230.4 30% 25% 20% -
460.8 kbps 460.8 25% 30% 35% -
921.6 kbps 921.6 25% 27% 30% -
Traffic specified (kbps) 3000 0% 3% 2% -
Traffic specified (kbps) 4500 0% 0% 3% -
Average PS data rake (kbps)
437.76 551.952 690.36 -

504.198



4-33


Table 4-8 Reverse traffic model describes the reverse traffic model of DORB.

Table 4-8 Reverse traffic model

Items Data & voice subscriber Voice subscriber
Proportion 100% 0%
Hierarchy Lower Medium Higher -
Sub-proportion 60% 25% 15% -
Traffic type
Data rate
(kbps)
- - - -
VOIP (erl) 9.6 0 0 0 0
VT (erl) 76.8 0 0 0 0
PPP session Time (s) - 300 600 900 -
Packet call duty ratio - 10% 15% 20% -
28.8 kbps 28.8 20% 15% 10% -
57.6 kbps 57.6 30% 25% 20% -
115.2 kbps 115.2 25% 30% 35% -
230.4 kbps 230.4 25% 27% 30% -
460.8 kbps 460.8 0% 3% 2% -
921.6 kbps 921.6 0% 0% 2% -
1382.4kbps 1382.4 0% 0% 1% -
Average PS data rake (kbps)
109.44 129.312 165.312 -
122.7888