COMSATS University Islamabad

Lahore Campus, Defence Road, Off Raiwind Road Lahore

Ph: 042-99203109 Fax: 042-99205253
Department of Electrical and Computer Engineering
Spring 2020
Course Title: Wireless Communication Systems Course Code: EEE464 Credit Hours: 3(3,0)
Course Dr. Saleem Akhtar Programme Name: BSETE, BSEEE
Semester:
Instructor/s: 8th Batch: Fall 2016 Section: A Due Date: July 26, 2020
Assignment 5 Maximum Marks: 30
Important Instructions / Guidelines:

• Show work and explain reasoning. The description must be supported by relevant figures and
mathematical expressions, where required. All the figures should be legible and neatly drawn, where
applicable.
• Attempt all problems. However, grading will be done for Problems # 4-13 only.
_

1. If 20 MHz of total spectrum is allocated for a duplex wireless cellular system and each simplex channel
has 25 KHz RF bandwidth.
a. Compute the number of duplex channels.
b. For a frequency reuse cellular system, compute the total number of channels per cell site, if N =
4 cell reuse is used.

Answers:

a. Number of Duplex channels = 400

b. total number of channels per cell 𝐶 = 100

2. a) Compute the noise floor (in dBm) for mobile receivers which implement the following standards: (i)
AMPS, (ii) GSM, (iii) DECT, (iv) IS-95. Assume all receivers have a noise figure of 10 dB.

b) If a base station provides a signal level of –90 dBm at the cell fringe, compute the SNR in dBs for each
of the mobile receivers described in Part (a).

Answers:

Part (a)

i) AMPS, BW= 30 𝐾𝐻𝑧, 𝑁𝑜𝑖𝑠𝑒 𝐹𝑙𝑜𝑜𝑟 = −118.9 𝑑𝐵𝑚

Page 1 of 17
 ii) GSM, BW= 200 𝐾𝐻𝑧, 𝑁𝑜𝑖𝑠𝑒 𝐹𝑙𝑜𝑜𝑟 = −110.6 𝑑𝐵𝑚

iii) DECT, BW= 1.728 𝑀𝐻𝑧, 𝑁𝑜𝑖𝑠𝑒 𝐹𝑙𝑜𝑜𝑟 = −101.3 𝑑𝐵𝑚

iii) IS-95, BW= 1.25 𝑀𝐻𝑧, 𝑁𝑜𝑖𝑠𝑒 𝐹𝑙𝑜𝑜𝑟 = −102.7 𝑑𝐵m

Part (b)

i) AMPS 𝑆𝑁𝑅(𝑑𝐵) = 28.9

ii) GSM 𝑆𝑁𝑅(𝑑𝐵) = 20.6

iii) DECT 𝑆𝑁𝑅(𝑑𝐵) = 11.3

iv) IS-95 𝑆𝑁𝑅(𝑑𝐵) = 12.7

3. A cellular service provider decides to use a digital TDMA scheme which can tolerate a signal-to-
interference ratio of 15 dB in the worst case. Assuming omnidirectional antennas

a. Plan the optimal value of cluster size, 𝑁, for a path loss exponent of 𝑛 = 4.
b. If an intensive propagation measurement campaign showed that the mobile radio channel provided
a propagation path loss exponent of 𝑛 = 3 instead of four, appraise how would your design decision
in Part (a) change?

Answers:

Part (a) N=7

Part (b) N = 12

4. A cellular communication system with a hexagonal cell layout and base stations placed in the middle of
the cell (with omni directional antennas) has a total of 100 channels at its disposal. The distance between
the two closest base stations is 2 km. The minimum required SIR for good signal quality is 11 dB. The
system is designed such that this requirement is not only met, but an additional 8 dB fade margin is
added. The propagation path loss is assumed to be proportional to the fourth power of the distance and
thermal noise is negligible.

a. Plan the optimum cluster size, N, of the system meeting the SIR requirements.

Page 2 of 17
 b. What will be the area capacity of this system in terms of channels/Km2, if one uses the optimum
cluster size of planned in Part (a)?

Answers

Part (a) cluster size 𝑁 = 9

Part (b) The area capacity in of channels/Km2 = 3.2 𝑐ℎ𝑎𝑛𝑛𝑒𝑙𝑠 𝑝𝑒𝑟 𝐾𝑚2

5. In an analog cellular system, the base station transmitter provides 15 W using 0 dB antenna gain. The
receiver antenna has a gain of 0 dB and the receiver bandwidth is 30 kHz. The system has to operate
with the minimum 𝑆𝑁𝑅 requirement of 20 dB. If the receiver system noise figure is 8 dB and the carrier
frequency is 900 MHz. Assuming 𝑇 = 17℃

a. Plan the optimum cluster size, 𝑁, and the maximum cell radius that will ensure the required
𝑆𝑁𝑅 for 95% of the time.
b. The analog system above is being modernized for digital transmission using 900 MHz GSM. As a
consequence, the receiver bandwidth is 200 KHz but the requirement of 𝑆𝑁𝑅 can be reduced
to 12 dB. Assuming the same transmitter power, antenna gains and 6 dB noise figure, plan the
optimum cluster size, 𝑁, and the maximum cell radius that will ensure the required 𝑆𝑁𝑅 in this
case for 95% of the time.
c. Comparing the results obtained in Part (a) and Part (b), would you support the modernization
of the cellular system by adopting GSM?

Answers

Part (a)

𝑅𝑚𝑎𝑥 = 10.96 𝐾𝑚

cluster size N = 9

Part (b)

𝑅𝑚𝑎𝑥 = 12 𝐾𝑚

cluster size N = 4

Page 3 of 17
6. A receiver in an urban cellular radio system detects a 1 mW signal at d = do = 1 meter from the
transmitter. In order to mitigate co-channel interference effects, it is required that the signal received
at any base station receiver from another base station transmitter which operates with the same
channel must be below -100 dBm. A measurement team has determined that the average path loss
exponent in the system is n = 3.

a. Design the major radius of each cell if a seven-cell reuse pattern is used.
b. What will be the major radius of each cell if a four-cell reuse pattern is used instead of seven?

Answers with hints:

Part (a)

Co-channel Interference I ≤ -100 dBm

𝐷
or 𝐼𝑑𝐵𝑚 = 10 𝑙𝑜𝑔10 𝑃𝑟 (𝑑0 ) − 10𝑛 𝑙𝑜𝑔10 (𝑑 ) ≤ −100 dBm
0

𝐷 ≥ 2151.4 m

𝐷
Using the relationship between cluster size 𝑁, re-use distance D and cell radius 𝑅 : √3𝑁 = 𝑅

Cell Radius 𝑅 = 469.5 𝑚

Part (b)

Cell Radius 𝑅 = 621 𝑚

7. Consider seven-cell frequency reuse. Cell B1 is the desired cell and B2 is a co-channel cell as shown in
the Figure below. For a mobile located in cell B1 at X, plan the minimum cell radius R to give a forward
C/I ratio of at least 15 dB at least 90% of the time.

Page 4 of 17
 Figure: Co-channel interference geometry between B1 & B2

Assume the following:

Co-channel interference is only due to base B2 only.

Carrier frequency, 𝑓𝑐 = 1800 𝑀𝐻𝑧.

Reference distance 𝑑0 = 1 𝑘𝑚 (assume free space propagation from the transmitter to 𝑑0 ).

Assume omnidirectional antennas for both transmitter and receiver, where Gbase = 6 dBi and Gmobile = 3
dBi.

Transmitter power, 𝑃𝑡 = 20 𝑊 (assume equal transmitted power for all base stations)

PL(dB) between the mobile and base B1 is given as

𝑑1
𝑃𝐿(dB) = 𝑃𝐿(𝑑0 ) + 10(3.5) 𝑙𝑜𝑔( ) + 𝑋𝜎𝑠ℎ 𝜎𝑠ℎ = 0 dB.
𝑑0

PL(dB) between the mobile and base B2 is given as

𝑑2
𝑃𝐿(dB) = 𝑃𝐿(𝑑0 ) + 10(4.5) 𝑙𝑜𝑔( ) + 𝑋𝜎𝑠ℎ 𝜎𝑠ℎ = 7 dB.
𝑑0

Answer with hints:

𝑃𝑟 (𝑑𝑜 ) = −45.5 𝑑𝐵𝑚

At the boundary of B1, signal power

𝑅
𝐶̅ = 𝑃𝑟 (𝑑𝑜 ) − 10 × 3.5 𝑙𝑜𝑔10 ( )
𝑑𝑜

Page 5 of 17
 At the boundary of B1, Interference from B2
𝑑2
𝐼 ̅ = 𝑃𝑟 (𝑑𝑜 ) − 10 × 4.5 𝑙𝑜𝑔10 ( )
𝑑𝑜

̅̅̅̅̅
𝐶 ⁄𝐼 = 25 + 10 𝑙𝑜𝑔10(𝑅)

C/I threshold = 15 𝑑𝐵

𝐶
Prob. [𝐶 ⁄𝐼 ≥ 𝐼 threshold] = 90%

Due to lognormal shadow fading

15 − ̅̅̅̅̅
𝐶 ⁄𝐼
𝑃𝑟𝑜𝑏. [𝑃𝑟 (𝐶 ⁄𝐼 ) ≥ 15] = 𝑄 ( )
𝜎𝑠ℎ

𝜎𝑠ℎ = √0 + 72

𝑅 ≥ 0.787 𝐾𝑚

8. Suppose that a mobile station is moving along a straight line between base stations BS1 and BS2, as
shown in Figure below.

Figure Q 1: Cellular system with two base stations

The distance between the base stations is 𝐷 = 2000 𝑚. For simplicity, assume small scale fading is
neglected and the received power (in dBm) at base station 𝑖, from the mobile station, is modeled as a
function of distance on the reverse link

Page 6 of 17
 𝑑
𝑃𝑟 (𝑑𝑖 ) = 𝑃𝑟 (𝑑0 ) − 10𝑛𝑙𝑜𝑔10 ( ) 𝑑𝐵𝑚, 𝑖 = 1,2
𝑑0

Where 𝑑𝑖 is the distance between the mobile and the base station 𝑖, in meters. 𝑃𝑟 (𝑑0 ) is the received
power at distance 𝑑0 from the mobile antenna. Assume that. 𝑃𝑟 (𝑑0 ) = 0 dBm and 𝑑0 = 1 𝑚. Let 𝑛 denote
the path loss which is assumed to be equal to 2.9.

Assume the minimum usable signal level for acceptable voice quality at the base station receiver is 𝛾 =
−88 𝑑𝐵𝑚, and the threshold level used by the switch for handoff initiation is 𝑃𝑟,𝐻𝑂 . Consider that the
mobile is currently connected to BS1 and is moving toward a handoff with speed 𝑣 = 80 km/hr. Time
required to complete a handoff, once that received signal level reaches the handoff threshold 𝑃𝑟,𝐻𝑂 . is
∆𝑡 = 4.5 seconds.

a. Evaluate the minimum required margin ∆𝑃 = 𝑃𝑟,𝐻𝑂 − 𝛾 to assure that calls are not lost due to
weak signal condition during handoff. Assume that the base station antenna heights are negligible
compared to the distance between the mobile and the base stations.
b. Explain the effects of the margin ∆𝑃 = 𝑃𝑟,𝐻𝑂 − 𝛾 on the performance of cellular systems.

Answers and hints:

Part (a)

Calls will not lost due to weak signal condition during handoff if:

𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑡𝑟𝑎𝑣𝑒𝑙𝑙𝑒𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 ℎ𝑎𝑛𝑑𝑜𝑓𝑓 𝑑𝑚𝑖𝑛 − 𝑑𝐻𝑂

= ≥ 4.5 𝑠𝑒𝑐
𝑚𝑜𝑏𝑖𝑙𝑒 𝑠𝑝𝑒𝑒𝑑 𝑣

where

𝑑𝑚𝑖𝑛 = distance at which the received power at BS1 reaches γ = -88 dBm

𝑑𝐻𝑂 =distance at which the received power at BS1 reaches handoff threshold level 𝑃𝑟,𝐻𝑂

𝑑𝑚𝑖𝑛
𝛾 = 𝑃𝑟 (𝑑0 ) − 10𝑛 log10 ( )
𝑑0

𝑑𝑚𝑖𝑛 = 1083 𝑚𝑒𝑡𝑒𝑟𝑠

𝑑𝐻𝑂
𝑃𝑟,𝐻𝑂 = 𝑃𝑟 (𝑑0 ) − 10𝑛 log10 ( )
𝑑0

Page 7 of 17
 𝑃𝑟,𝐻𝑂
𝑑𝐻𝑂 = 10− 29 𝑚𝑒𝑡𝑒𝑟𝑠

𝑑𝑚𝑖𝑛 − 𝑑𝐻𝑂
≥ 4.5 𝑠𝑒𝑐
𝑣

𝑃𝑟,𝐻𝑂 ≥ −86.8 𝑑𝐵𝑚

Therefore

∆= 𝑃𝑟,𝐻𝑂 − 𝛾 = 1.2 𝑑𝐵

Part (b)

If we set ∆ too large, several unnecsaary handoffs will be requested and performed,
increasing the signaling traffic.

On the other hand, if ∆ is too small, that is 𝑃𝑟,𝐻𝑂 is only slightly greater than γ, there will
not be enough time to complete the handoff especially for high speed mobiles and calls may
be lost due to weak signal condition.

9. Assume that a GSM cell has 7 RF Duplex Carriers, with an effective radiated power of 32 W and a cell
radius of 10 km. The path loss is 40 dB per decade. The grade of service is established to be a probability
of blocking of 5% (assuming blocked calls are cleared). Assume the average call length is two minutes,
and each user averages two calls per hour. Further, assume the cell has just reached its maximum
capacity and must be split into four new microcells to provide four times the capacity in the same area.

a. Compute the current capacity in terms of Erlangs per cell, Erlangs per Km2 and subscribers per cell
before splitting.
b. Evaluate the radius and transmit power of the new cells after splitting.
c. Appraise, how many channels are needed in each of the new cells to maintain frequency reuse
stability in the system.
d. If traffic is uniformly distributed, what is the new traffic load and the probability of blocking in each
new cell if we split the larger cell into four new microcells?
e. If the new cells are to be operated with call blocking probability of 2% (blocked calls are cleared),
what will be the capacity in terms of Erlangs per cell, Erlangs per 𝐾𝑚2 and subscribers per cell?

Answers

Page 8 of 17
 Part (a):

Available capacity in terms of Erlangs per cell = 𝐴 = 50.543 Erlangs/cell

Number of subscribers per cell ≅ 758

Available capacity in terms of Erlangs per 𝐾𝑚2 = 0.164

Part (b)

𝑅𝑛𝑒𝑤 = 5 𝐾𝑚

Part (c):

Each new smaller cell will get the number of RF carriers equal to the original large cell (i.e., 7
Duplex TDMA Carriers) once the cell splitting process is completed.

Part (d):

Traffic Offered/Intensity in the Cell = 𝐴𝑐𝑒𝑙𝑙 ≅ 12.7

Loss probability = 𝑃𝑏 << 0.001%

Part (e):

Available capacity in terms of Erlangs per cell = 𝐴 = 45.875 Erlangs/cell

Available capacity in terms of subscribers per cell ≅ 688

Available capacity in terms of Erlangs per 𝐾𝑚2 = 0.584

10. Suppose that you work with a cellular service provider and the cellular radio system your company
deployed in a given service area just reached its maximum system capacity. Your boss then asks you to
carry out a study to analyze the application of cluster size reduction technique combined with sectoring,
aiming to increase the carried traffic of the system. The current deployed system employs the AMPS
system, with 300 voice channels, cluster size 𝑁 = 7, and omnidirectional antennas at the base stations.
Base stations are located at the center of the cells and transmit the same power on the forward link.
The designed blocking probability is 2% and all voice channels have been used. The maximum carried

Page 9 of 17
traffic per cell is, therefore, 32.8 Erlangs. The minimum SIR (worst case) on the forward link can be
computed using the expression

𝑛
(√3𝑁)
𝑆𝐼𝑅 = 10 log10 [ ] (in dB),
𝑖0

where 𝑛 is the path loss exponent, 𝑁 is the cluster size, and 𝑖0 is the number of interfering base stations
in the first tier. For cluster size 𝑁 = 7, omnidirectional base station antennas (𝑖0 = 6), and 𝑛 = 4, we
find that 𝑆𝐼𝑅 = 18.7 𝑑𝐵.

When the cluster size is reduced, the maximum carried traffic per cell increases, at the expense of SIR
degradation. Sectorized base station antennas can then be used in order to increase SIR and guarantee
that the link quality is maintained. In other words, the minimum SIR achieved when cluster size is
reduced and sectorized antennas are used must be equal to or exceed the minimum SIR achieved in the
current deployed system (𝑆𝐼𝑅 = 18.7 𝑑𝐵).

In your analysis of cluster size reduction and sectoring, consider forward link only. Two sectorized
antennas are available: BW = 60° for six sectors per cell, and BW = 120° for three sectors per cell, as
shown in Figure a below. Assume that all cells have hexagonal shape, with radius R.

Figure a: Radiation patterns for BW = 60° and BW = 120°

We want to determine the maximum carried traffic per cell when cluster size is reduced to N = 3 and N
= 4, using three sectors (BW = 60°) and six sectors (BW = 120°) (four possible configurations). Assume

Page 10 of 17
that all sectorized antennas are installed at the same time. Consider the first tier of co-channel cells
only.

Figure b: Cluster size N = 3, three and six sectors.

Figure c: Cluster size N = 4, three and six sectors.

a. Compute the minimum SIR at the mobile (i.e., when the mobile is located at the cell boundary, as
indicated in Figure b and c below, for cluster sizes 𝑁 = 3 and 4, with three and six sectors. Determine
which configurations (cluster size N, number of sectors) are feasible regarding co-channel
interference (i.e., configurations where the minimum SIR is equal to or exceeds 18.7 dB). Note that
the number of interferers in the first tier depends on the cluster size used and the number of sectors
per cell. Use the expression above to compute the minimum SIR.
b. For each configuration (N = 3, 4 and three and six sectors per cell), compute the maximum carried
traffic per cell at blocking probability of 2% and 300 voice channels available in the system. Assume

Page 11 of 17
 that users are uniformly distributed over the service area and, therefore, all sectors are assigned an
equal number of channels.
c. Having both results of SIR and Erlang capacity for four different configuration in Part(a) and Part (b),
conclude your final recommendation for a suitable configuration to be adapted for upgradation.

Answers

Part (a)

Let 𝑖0 be the number of co-channel interfering cells, then the relationship between cluster size 𝑁 and
𝑆/𝐼 is given by

𝑛
(√3𝑁)
𝑆𝐼𝑅 = 10 log10 [ ] (in dB),
𝑖0

1
𝑆 ⁄𝐼 = (3𝑁)𝑛/2
𝑖𝑜

𝑃𝑎𝑡ℎ𝑙𝑜𝑠𝑠 𝑒𝑥𝑝𝑜𝑛𝑒𝑛𝑡 𝑛 = 4

Cluster Size 𝑁 𝑆𝐼𝑅 (120° 𝑆𝑒𝑐𝑡𝑜𝑟𝑖𝑛𝑔, 𝐵𝑊 = 120°) 𝑆𝐼𝑅 (60° 𝑆𝑒𝑐𝑡𝑜𝑟𝑖𝑛𝑔, 𝐵𝑊 = 60°)
3
4

Part (b)

Case 1: Cluster Size 3, 120° Sectorization

𝐶𝑙𝑢𝑠𝑡𝑒𝑟 𝑆𝑖𝑧𝑒 = 𝑁 = 3

𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑆𝑒𝑐𝑡𝑜𝑟𝑠 𝑝𝑒𝑟 𝑐𝑒𝑙𝑙 = 3

No of Duplex FDMA Carriers per cluster = 300

With uniform distribution of the RF Carriers

Number of Duplex FDMA carriers per cell = 100

Number of Duplex FDMA carriers per sector = 33

Page 12 of 17
 Number of traffic channels per sector (ignoring control channels) = 𝐶 = 33 × 1 = 33

Loss probability = 𝑃𝑏 = 2 %

Using Erlang B Table

Available capacity in terms of Erlangs per sector = 𝐴 = 24.63 Erlangs/sector

Available capacity in terms of Erlangs per cell = 𝐴 = 3 × 24.626 = 73.89 Erlangs/cell

Cluster 𝐸𝑟𝑙𝑎𝑛𝑔𝑠 (120° 𝑆𝑒𝑐𝑡𝑜𝑟𝑖𝑛𝑔 𝐸𝑟𝑙𝑎𝑛𝑔𝑠 (60° 𝑆𝑒𝑐𝑡𝑜𝑟𝑖𝑛𝑔)

Size 𝑁
3
4

Part (c):

11. A GSM-1800 licensee plans to build out a 6 MHz license in the DCS band from 1718.9 MHz to 1724.9
MHz (reverse link) and 1813.9 MHz to 1819.9 MHZ (forward link). The city has a circular shaped area of
2500 sq km. A preliminary study of propagation conditions in this area has found that the log-distance
path loss exponent is equal to four, standard deviation of shadow fading equal to 8 dB and the GSM-
1800 system can be deployed using four-cell reuse.

a. If each base station uses 10 Watts transmitter power and 10 dB gain omni directional antenna, plan
the maximum cell radius of each cell with 90% area coverage in each cell. Assume that 𝑑0 = 1 km
and each mobile is equipped with 0 dBi gain antenna and requires a minimum received signal level
of – 90 dBm.
b. If each base station costs $100,000 and each radio channel within the base station costs $10,000,
what will be the initial build out cost of the system?
c. If the company plans to design the system for 5% blocking probability at start-up, what is the
maximum number of users that can be subscribed at the start-up if each user can generate a traffic
of 0.05 Erlangs and each TDMA carrier (radio channel) supports eight users.
d. What is the average cost per user needed to recoup 10% of the initial system build out cost after
one year if the number of subscribers is static during the year 1?

Answers and hints:

Page 13 of 17
Part (a)

From Figure relating Area Coverage Probability and Boundary Converge Probability

Boundary Coverage Probability = 0.75

𝛾− 𝑃(𝑅) ̅̅̅̅̅̅̅
Boundary Coverage Probability = 𝑃𝑟𝑜𝑏. [ ̅̅̅̅̅̅̅
𝑃(𝑅) ≥ 𝛾] = 𝑄 ( 𝜎 ) = 0.75

̅̅̅̅̅̅̅ = −84.64 𝑑𝐵𝑚

𝑃(𝑅)

̅̅̅̅̅̅̅ = 𝑃𝑟 (𝑑𝑜 ) − 10 × 𝑛 𝑙𝑜𝑔10 ( 𝑅 )

𝑃(𝑅)
𝑑𝑜

𝑃𝑟 (𝑑𝑜 ) = 1.72 × 10−8 𝑊 = −47.63 𝑑𝐵𝑚

𝑅 = 8.4 𝐾𝑚

Area of a cell = 𝐴𝑐𝑒𝑙𝑙 = 2.6𝑅2

𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐶𝑒𝑙𝑙𝑠 = 14
Part (b)
𝑇𝑜𝑡𝑎𝑙 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑅𝐹 𝐶ℎ𝑎𝑛𝑛𝑒𝑙𝑠 = 30
Number of RF Carriers per cell = 7

Cost of a base station (US Dollars) = 500000 + (7 × 50000)

Total Cost of the system = 11900000 USD

Part (c)
Number of Duplex TDMA carriers per cell = 7

Number of traffic channels per cell (ignoring control channels) = 𝐶 = 7 × 8 = 56

Loss probability = 𝑃𝑏 = 2 %

Using Erlang B Table

Available capacity in terms of Erlangs per cell = 𝐴

Erlangs per user = 𝐴𝑢

𝐴
Number of users per cell = ≈ 917
𝐴𝑢

Page 14 of 17
 Therefore, number of users that can be subscribed at the start-up = 12838

Part (d)

Avg cost per subscriber = 92.7 𝑈𝑆𝐷

12. A cellular service provider decides to use a digital TDMA scheme which can tolerate a signal to-
interference ratio of 15 dB and GoS of 2 % in the worst case. Assume that the path loss exponent n is 4,
the average per user call rate λ = 1 per hour, average calls lasts for two minutes and that the blocked
calls are cleared from the system. Assume also that the service provider has acquired license of
bandwidth equivalent to 36 TDMA carriers in uplink and downlink each with availability of 8 time slots
of traffic channels on each TDMA carrier.

a. Plan the optimal value of cluster size N for omnidirectional antennas (360° sectoring) and compute
cell capacity in terms of subscribers per cell for 2% GoS.
b. What will be the optimum value of the cluster size N, if instead of 360° sectoring, the company
decides to use 60° sectoring? Appraise, this modification from 360° sectoring to 60° sectoring.
Defend or criticize the use of 60°sectoring.
c. Describe other alternatives that may also be considered to increase the capacity instead of using
60° sectoring.

Answers and hints:

Part (a):

Cluster Size = 𝑁 = 7

With uniform distribution of the RF Carriers

Total TDMA Carriers

Number of Duplex TDMA carriers per cell = ⌈ ⌉
𝐶𝑙𝑢𝑠𝑡𝑒𝑟 𝑆𝑖𝑧𝑒

Number of traffic channels per cell (ignoring control channels) = 𝐶

= Number of Duplex TDMA carriers per cell × 8

Loss probability = 𝑃𝑏 = 2 %

Using Erlang B Table

Available capacity in terms of Erlangs per cell = 𝐴

Page 15 of 17
 Erlangs per user = 𝐴𝑢

𝐴
Number of users per cell = ≅ 930
𝐴𝑢

Part (b)

Cluster Size = 𝑁 = 3
No of Duplex TDMA Carriers per cluster = 36

With uniform distribution of the RF Carriers

No of Duplex TDMA Carriers per cluster

Number of Duplex TDMA carriers per cell = ⌈ ⌉
𝐶𝑙𝑢𝑠𝑡𝑒𝑟 𝑆𝑖𝑧𝑒
Number of Duplex TDMA carriers per cell
Number of Duplex TDMA carriers per sector = ⌈ ⌉
Sectors per cell

Number of traffic channels per sector (ignoring control channels) = 𝐶

= Number of Duplex TDMA carriers per sector × 8

Loss probability = 𝑃𝑏 = 2 %

Using Erlang B Table

Available capacity in terms of Erlangs per sector = 𝐴𝑠 Erlangs/sector

Erlangs per user= 𝐴𝑢 Erlangs

𝐴𝑠
Number of users per sector =
𝐴𝑢

Number of users per cell = Number of users per sector × 𝑖𝑜 = 1770

capacity improvement = 90.3 %

13. Consider a mobile system supporting 100 channels per cell. A call blocking probability of 1% is desired.
Mobile users typically use their cell phones once per 10 minutes, on the average, their calls lasting an
average of 10 minutes. Say the system is concentrated in an urban area with a density of 500 cell phones
per km2.

a. Plan the required cell radius to meet this subscribers density.

Page 16 of 17
b. Modify the cell radius if the mobile users “stay on the line” for 4 minutes, on the average.
c. What happens to the cell size if, in addition, they start using their phones more often?

Answers
Part (a)
User density = 500 users per km2
Area of a hexagon = 2.6R2
The cell radius should be chosen to maintain the maximum user capacity per cell.
𝑐𝑒𝑙𝑙 𝑎𝑟𝑒𝑎 × 𝑠𝑢𝑏𝑠𝑐𝑟𝑖𝑏𝑒𝑟𝑠 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 = 𝑠𝑢𝑏𝑠𝑐𝑟𝑖𝑏𝑒𝑟𝑠 𝑐𝑎𝑝𝑎𝑐𝑖𝑡𝑦 𝑝𝑒𝑟 𝑐𝑒𝑙𝑙
𝑅 = 0.254 𝑘𝑚 𝑜𝑟 254 𝑚𝑒𝑡𝑒𝑟𝑠
Part (b):
𝑅 = 0.402 𝑘𝑚 𝑜𝑟 402 𝑚𝑒𝑡𝑒𝑟𝑠
Part (c):
If the users start using their phones more often, the traffic load per person will increase.
Hence, fewer users can be accommodated per cell and the cell size should be reduced
accordingly so as to maintain the same total traffic and call blocking probability.

Page 17 of 17