1

CHAPTER 2: THE CELLULAR SYSTEM

 Overview
2

 Cellular system
 Cell shape
 Frequency reuse
 Cell capacity and reuse
 Channel assignment strategies
 Handoff
 Interference and system capacity
 Trunking and grade of service
 Improving capacity
 Cellular System - Architecture
3

Radio tower

PSTN
Telephone
Network
Mobile Switching
Center
 Cellular System ….
4

 High capacity is achieved by limiting the coverage of each base

stations to a small geographic region called a cell
 Single, high power transmitter (large cell) are replaced with many low
power transmitters (small cells)
 A portion of the total number of channels is allocated to each cell or
base station
 Available group of channels are assigned to a small number of
neighboring base stations called cluster
 Near by base stations are assigned d/t groups of channels to minimize
interference
 Same channels (frequencies/timeslots/codes) are reused by spatially
separated base stations
▪ Reuse distance and frequency reuse planning?
 Cellular System ….
5

▪ A switching technique called handoff enables a call to proceed from

one cell to another
 As demand (or # of users) increases, the number of base stations
may be increased to provide additional capacity
▪ Smaller cells, e.g., Microcells, Picocell, Femtocell
▪ Also cell sites in trucks to replace downed cell towers after natural
disasters, or to create additional capacity for large gatherings (football
games, rock concerts)
▪ Transmission power reduction => interference decreases
 Typical power transmitted by the radios in a cell system
 Base station: Maximum Effective Radiated Power (ERP) is 100W, or
up to 500 W in rural areas
 Mobile station: Typically 0.5 W. For CDMA, transmit power is
lowered when close to a BS
 Cell Shape – Why hexagon?
6

• The hexagonal shape is a simplistic assumption

 (a) is theoretical coverage area and (b) measured coverage area where
red, blue, green, and yellow indicate signal strength, in decreasing
order
 Footprint: Actual radio coverage and obtained experimentally
 Actual shape is random that depends on the environment
 Circular (theoretical): If path loss was a decreasing function of
distance, say, 1/dn, where d is the distance b/n BS & MS
 Cell Shape – Required
7

• Geometric shape that

approximates the
theoretical shape?
• Shape which is non-
overlapping and
repetitive placement
covers an entire region?
• Possible shapes
• Triangles, squares,
hexagons
• Which one to choose? Has dead zones
 Cell Shape . . .
8

R
R
R

aR = 2R2 aT = 33/2 R2/4 aH= 33/2 R2/2

 Hexagonal cell is conceptual, however, it is universally
adopted for most theoretical treatment because:
 Hexagons are a geometric shape that approximates a
circle (for Omni-directional radiation)
 Using a hexagon geometry, fewest number of cells can
cover the entire geographic region
 Cell Shape . . .
9

 When using hexagon to model coverage areas

 Center-excited cell: Base station (BS) depicted as being
in the center of the cell
◼ Omni-directional antenna is used

 Edge-excited cell: on three of the six cell vertices

◼ Sectored direction antenna is used
 Frequency Reuse – Example
10

 Example: Consider a
cluster of 7 cells

 Same color labeled

cells use the same
frequency
 Frequency reuse factor is
1/7 since each cell
contains one-seventh of
the total available
channels
 V

Geometry of Hexagons
y
U
11

D
 Geometry of Hexagons …
12

 Using these equations, to locate the co-channel cells, start from a

reference cell and move
 i-hexagons along the U-axis and
 j-hexagons along the V-axis
 The distance, D, between co-channel cells in adjacent clusters is given
by
D = i 2 + ij + j 2
 The number of cells in a cluster, N, is given by
N = i 2 + ij + j 2
where i and j are non-negative integers
 There are only certain cluster sizes and layouts possible
 Typical values of N are 1, 3, 4, 7, 12, ……
 Example
13

Re-use Coordinates Number of Cells in Normalized repeat

the cluster distance
i j N D
1 0 1 1

1 1 3 1.732

2 1 7 2.646

2 2 12 3.464

1 3 13 3.606

3 2 19 4.359

1 4 21 4.583
Locating Co-Channel Cells: Example N=19, i=3, j=2
14
 Re-use Factor
15

D  For Hexagonal cells, the

re-use distance is given
by:
D=R 3N
R
 Where R = cell size and N
= cluster size

BS1  Re-use factor is defined

BS1 as: D
q= = 3N
R
 Cell Capacity and Reuse
16
 Consider a cellular system
 Which has S duplex channels available for re-use
 Each cell allocated a group of k channels
 Let the S channels be divided among N cells (unique and
disjoint) then,
S = kN
 Cluster: N cells, which collectively use the complete set
of available frequencies
 If a cluster is replicated M times in the system, the
total number of duplex channels, C, as a measure of
capacity is C=M k N =M S
 Cell Capacity and Reuse . . .
17

 If cluster size N is reduced while cell size is kept constant

M C 
 More clusters are required to cover area of interest, i.e.,
 So capacity is directly prop. to replication factor for fixed area
 However, small cluster size means co-channel cells are located much
closer together
 Results in larger co-channel interference
 May result in lower Quality of Service (QoS)
 Conversely, large cluster size indicates that co-channel cells are far
from each other
 Less co-channel interference and frequency utilization
 The value of N is a function of how much interference a mobile or BS
can tolerate
 Cell Capacity and Reuse: Example 1
18

 Assume that:
 50 MHz is available for
forward channels
 GSM is deployed
 Each channel is 200 kHz
 In GSM, TDMA is used so that
8 simultaneous calls can be
made on each channel

 How large is k?
 How many forward calls can
be made simultaneously for
the deployment containing
28 cells as in the figure?
 Cell Capacity and Reuse: Solution
19

 Solution:
 There are 50 MHz / 0.2 MHz = 250 channels per
cluster
 With N = 4, then k = 250/4 = 62.5

 With 62.5 channels, 8(62.5) = 500 simultaneous calls

can be made in each cell

 There are 28 cells on the cell map in Figure, so the total

forward calls is 28(500) = 14×103 calls can be made
simultaneously
 Cell Capacity and Reuse: Example 2
20

 Suppose 33 MHz BW allocated to particular FDD cellular system,

where two 25 KHz simplex channel to provide full-duplex for
voice/data
 Compute the number of channels per cell if a system uses
 Four-cell reuse
 Seven-cell reuse
 Twelve-cell reuse

 Solution: Given that

 Total BW = 33 MHz, channel BW = 25 KHz x 2 = 50 KHz/duplex
channel
 S = 33,000/50 = 660 channels
◼ For N = 4, k = 660 / 4 ≈ 165 channels
◼ For N = 7, k = 660 / 7 ≈ 95 channels
◼ For N = 12, k = 660 / 12 ≈ 55 channels
 Handoff
21

 The process of transferring a call, which is in progress from one

channel or BS to another is called handoff or handover

 Handoff is required when a MS moves into a different cell

 MSC facilitates the transfer

 In general, handoff involves

 Identifying the new BS
 Allocation of voice and control channels in the new BS

 Prioritize handoff requests over call initiation requests

when allocating unused channels in a cell site
 Handoff Region – Power Strength
22

BS-1 P1(x) P2(x) BS-2

 By looking at the variation of signal strength from either base station, it is

possible to decide on the optimum area where handoff can take place
 Handoff
23

 Handoffs must be performed

 Successfully

 Asinfrequently as possible, and

 Must be imperceptible to the user

 To meet these requirements, a minimum usable

signal level must be specified for acceptable voice
quality at the base station
◼ If the received power drops too low prior to handoff, the call will be
dropped so that users complain about dropped calls
 Handoff Region . . .
24
 Handoff Margin
25

 Consider the following two power levels

 Pr,min. usable
be the minimum received power in dB, below
which a call cannot be received
◼A handoff has to be initiated much prior to this point

 Pr,handoffbe a higher threshold in dB at which the MSC

initiates the handoff procedure
◼ Handoff is made when the received signal at the BS falls
below the threshold

 Define handoff margin in dB as

∆ = Pr,handoff − Pr,min. usable
 Handoff Margin …
26

 How much margin is needed to handle a mobile at driving speeds?

 The margin ∆ should not be too large or too small
 The handoff threshold power is selected such that it is slightly greater
than the minimum usable signal power for an acceptable voice quality

 If ∆ is too large, it may lead to unnecessary handoffs which may

burden the MSC
 The call may be headed over to the neighboring BS when the MS is well
inside the home cell

 If ∆ is too small, there may be insufficient time to complete a

handoff before a call is lost due to weak signal conditions
 Handoff Margin …
27

 A call drop can also happen when there is an excessive delay

by the MSC in assigning a channel
 E.g., during high traffic conditions

 To effect handoff, it is important to ensure that the mobile is

actually moving away from the serving base station
 The measured signal level drop may be due to momentary fading
 In order to ensure this, BS monitors signal level for a certain period
of time before a handoff is initiated
 The length of monitoring depends on the speed of mobile units
 Where to get information about the mobile speed?

 At high mobile speeds, handoff needs to happen quickly

◼ In GSM, handoff is typically within 1-2 seconds
◼ In AMPS, this was 10 seconds (higher potential for dropped calls!)
 Handoff Margin – Example
28

 Assume that
A mobile moving at a speed of v = 35 mps (~125
Kph)
 Path-loss exponent n = 4

 Cell radius of 500 meters (the distance at which the call

is dropped)
 2 second handoff

 What is the required handoff margin?

 Handoff Margin - Solution
29

 Assume the mobile is driving directly away from the BS

 So distance d changes by 70 meters in two seconds
 Consider the received power at the two times
Pr,min. usable = 0 − 10nlog10d
Pr,handoff = 0 − 10nlog10(d−70)

 Taking the difference of the 2nd and the 1st equations,

∆ = 10nlog10d − 10nlog10(d − 70) = 10n log10(d/(d − 70))
 Taking that the call is dropped at d = 500 meters, we have
∆ = 40 log10(500/430) = 2.6 dB
 Note: In this example, the propagation equation used is for “large
scale path loss” only, which changes slowly
 Handoff Strategies
30

1. MSC controlled
 Used in the 1st generation analog cellular systems
 Signal strength measurements are made by the BS and
supervised by the MSC
 A spare receiver in each BS, called the location receiver,
is used to determine signal strengths of mobile users
which are in neighboring cells (and appear to be in
need of handoff)
 Handoff Strategies ….
31

2. Mobile-assisted hand-off (MAHO)

 Used in the 2nd generation systems
 MSs measures the received power from surrounding BSs
and report the results to home BS
 Handoff is initiated when the received power at the MS
from the neighboring BS begins to exceed the home BS
by a certain level for a certain period of time
 MAHO performs at a much faster rate, and is
particularly suited for micro cellular environments
 Handoff Strategies ….
32

 Intersystem handoff
 When a mobile user moves from one cellular system to
a different cellular system controlled by a different
MSC
 It may become a long-distance call and a roamer

 Compatibility between the two MSCs need to be

determined
 Handoff Strategies - Prioritizing Handoffs
33

 Having a call abruptly terminated while in the middle of a

conversation is more annoying than being blocked
occasionally on a new call attempt
 Concept of guard channels
 A fraction of the total available channel is reserved for handoff
requests, which then are not offered to mobiles making new calls
 It may reduce the total carried traffic
 However, it offers efficient spectrum utilization when dynamic
channel assignment strategies are used

 Queuing of handoff requests

 Does not guarantee a zero probability of forced termination
 Handoff Strategies - Umbrella Cell
34

 By using different antenna heights (often at the same

building or tower) and different power levels, “large”
and “small” cells are co-located at a single location
 Minimizes the number of handoffs for high speed users
and provides additional microcell channels for
pedestrian users
 Handoff Strategies – Hard Handoff
35

 Hard handoff: The channel in the source cell is released

only when the channel in the target cell is engaged
 I.e., assign different radio channels during a handoff

 For 1st generation analog systems, if takes about 10 seconds

and the value for ∆ is on the order of 6dB to 12dB
 For 2nd generation digital systems, typically requires only 1
or 2 seconds, and ∆ usually is between 0 dB and 6 dB
 In 2nd generation systems, handoff decision is also based on
a co-channel and adjacent channel interference levels
Handoff Strategies – Soft Handoff in CDMA
36

 The channel in the source cell is retained and used for a

while in parallel with the channel in the target cell
 Used in CDMA system
 In CDMA, users share the same channel in every cell
 Consequently, handoff does not mean a physical change in
the assigned channel, rather that a different base station
handles the radio communication task
 By simultaneously evaluating the receiver signals from a
single subscriber at several neighboring base stations, the
MSC may actually decide which version of the user’s signal
is best at any moment in time
 Interference
37

 Interference is a major performance limiting factor in

cellular radio
 It limits capacity thereby increasing the number of dropped
calls
 Interference are difficult to control in practice largely
due to random propagation effects
 Sources of interference include
 Another mobile in the same cell or in a neighboring cell
 Other BSs operating in the same frequency band
◼ Any cellular (e.g., from competing cellular carriers) or non-cellular system
which inadvertently leaks energy into the cellular frequency band
◼ …
 Interference – Effects
38

 Interference in the voice channels causes crosstalk

 A subscriber hears interference in the background due to an
undesired transmission
 Interference in the control channels causes error in
digital signaling which causes
 Missed calls
 Blocked calls
 Dropped calls

 Interference is more severe in urban areas, due to the

greater RF noise floor and the large number of base
stations and mobiles
 Interference – Types
39

 There are two major types of Interferences:

 Co-channel Interference (CCI)
 Adjacent channel Interference (ACI)

▪ CCI is caused due to the cells that reuse the same

frequency set
▪ These cells using the same frequency set are referred to
as co-channel cells

▪ ACI is caused due to signals that are adjacent in

frequency
 Co-Channel Interference – First-tier
40
Interference
 Co-Channel Interference – First-tier
41
Interference
First-tier co-channel BSs

D1
D6

D5 D2

D4
D3

Serving Base Station

 Co-Channel Interference …
42

 Unlike thermal noise, CCI cannot be overcome by increasing

the carrier power of a transmitter
 This is because, any increase in the transmitter power also
increases the interference to other co-channel cells

 Instead, co-channel cells must be physically separated by a

minimum distance to provide sufficient isolation due to
propagation
 To reduce CCI the co-channel cells must be sufficiently
separated

 Co-channel interference is a function of

 The radius of the cell, R, and
 The distance to the center of the nearest co-channel cell, Di
 Co-Channel Interference …
43

 For a hexagonal geometry, the co-channel reuse ratio, Q is related

to the cluster size
D
Q = = 3N
R
 It determines the spatial separation relative to the coverage
distance of the cell
 N small gives Q small
 Provides a larger capacity (i.e., can re-use more), but higher CCI
 N large means Q large
 Better transmission quality due to a small level of co-channel
interference but small capacity

 Hence there is capacity vs. interference tradeoff

 Co-Channel Interference …
44
 Signal-to-Interference Ratio
45

 Signal-to-interference ratio (S/I) for a mobile which monitors a

forward channel is S S
= m

I
I
j
j =1

 Where S: desired signal power, Ij: interference caused by the jth co-
channel cell, and m: first-tier co-channels cells
 The average received power at a distance d from the transmitting
antenna is approx. by
−n
d d
Pr = Po   or Pr ( dB ) = P0 ( dB ) − 10n log( )
 do  d0
 Where Po is the received power at a close-in reference distance in the
far-field and n is the path-loss exponent
 The path loss exponent, n, ranges between 2 and 6
 Signal-to-Interference Ratio …
46

 If Di is the distance of the ith interferer, the received power is

proportional to ( Di ) − n
 If transmit power of each BS is equal & n is the same
throughout the coverage area, S/I for a mobile is approx. as
S R −n
= m

 (D )
I −n
i
i =1

 To simplify, assume all first-tier interferers are equidistance

S ( R) ( )
D n n
3N
= =
I m m
 This relates S/I to the cluster size, and in turn determines the overall capacity of
the system
 Puts limits on how low we may set N
 Signal-to-Interference Ratio …
47

 For a hexagonal cluster of R

cells with the MS situated
D
at the edge of the cell D
D
( )
n
S 1D 1
=   = 3N n

I 6 R  6 D
D
D

 As long as all cells are of the

same size, S/I is independent of
the cell radius, R
 Signal-to-Interference Ratio - Example 1
48

 Design parameters:
 Desired S/I = 15dB
 Path-loss exponent n = 4

 Assume that there are six co-channel cells in the first tier and
all of them are at the same distance from the mobile

 What is the required re-use factor and cluster size that

should be used for maximum capacity?
 Signal-to-Interference Ratio – Example 1 …
49

• Six co-channel cells

in the first tier
 Signal-to-Interference Ratio - Example 1 …
50
• Let’s try for N= 4. The co-channel • Let’s try: N= 7
re-use ratio is
D D
= 3.46 = 4.58
R R
• And the signal-to-interference  = (4.58)
S 1 4

I 6
ratio is = 73.5 = 18.66 dB

= (3.46 ) = 24 =13.8 dB
S 1 4
• Which is greater than
I 6
the desired
• Smaller than the desired 15 • Hence, N=7 can be
dB used
• We must move to the next reuse • The frequency reuse
distance factor = 1/7
 Example 2 - Repeat Example 1 for n = 3
51

• Solution
• Let’s try for a seven-cell reuse pattern, i.e. N= 7. Like the previous
example D
= 4.58 and
S 1
=
3
(4.58) = 16.04 = 12.05 dB
R I 6

• Which is smaller than the desired 15 dB, hence we need to use

larger N
• Let us try N=12
= (6.00 ) = 36 = 15.56 dB
D S 1
= 6.00 
3
and
R I 6

• Since this is greater than 15 dB, N=12 can be used

• Note: n=3 is typical value for sub-urban area

• Exercise: Try for n=2, which represents rural area!
 Adjacent Channel Interference (ACI)
52

 Results from signals that are adjacent in frequency to the

desired signal
 Due to imperfect receiver filters, that allow nearby frequencies to
leak

 Near-far effect: The adjacent channel interference is

particularly serious. This occurs when:
 When an interferer close to the BS radiates in the adjacent
channel, while the subscriber is far away from the BS
◼ The BS may not discriminate the desired mobile user from the “bleed
over” caused by the close adjacent channel mobile
 Or, an interferer which is in close range to the subscriber’s
receiver, is transmitting while the receiver receives from the BS
 Adjacent Channel Interference …
53

 If a subscriber is at a distance d1 and the interferer is

d2 from the base station, then SIR (prior to filtering) is:
−n
S  d1 
=  
I  d2 

 Example
 Suppose a subscriber is at d1 = 1000m from the BS and an
adjacent channel interferer is at d2 = 100m from the BS
 Assume: Path-loss exponent is n = 3
 The signal-to-Interference ratio prior to filtering is then

−n −3
S  d1   1000 
=   =  = 10 −3 = − 30dB
I  d2   100 
 Trunking
54

 Trunking: How to accommodate a large number of users in

a limited radio spectrum?
 Trunking refers to sharing a fixed and small number of channels among
a large and random user community
 Each user demands access from a pool of channel
infrequently & at random times
 A channel is allocated on a per call basis and a channel is
returned to the pool up on termination of a call
 So a dedicated channel for each user is not required
 If U be number of users and C be number of channels, for any
C < U, possibility of more requests than channels
 Trunking exploits statistical behavior of users so that a fixed number of
channels accommodate a large, random user
 Trunking …
55

 Trunking accommodates large & random users:

 By providing access to each user on demand from a pool of
available channels
 When a user requests service and if all channels are in use
1. The user is blocked, or denied access to the system
2. In some systems, a queue may be used to hold the requesting
users until a channel becomes available
 Upon termination of the call, the previously occupied channel is
immediately returned to the pool
 Designing a trunked system, that can handle a given
capacity at a specific “grade of service”, requires trunking
and queuing theories
 Trunking – Definition of Terms . . .
56

 Setup time: The time required to allocate a radio channel to

a requesting user
 Users request may be blocked or have to wait
 Blocked Call: A call that cannot be completed at the time of
request due to congestion
 Also called lost call => lost revenue, e.g., pick hours, holidays, …
 Holding time: Average call duration in seconds, denoted H
 Depends on users and operator's tariff
 Request (or call) rate: Average number of calls per unit time,
denoted λ seconds-1
 Typically taken to be at the busiest time of day
 Depends on type of users community: Office, residential, call
center, …
 Trunking – Definition of Terms . . .
57

 Traffic Intensity: A measure of channel time utilization

 Is the average channel occupancy measured in Erlang, denoted by A
 Load: Traffic intensity across the entire trunked radio system
 Measured in Erlang

 Erlang: A “unit” of measure of usage or traffic intensity

 A channel kept busy for one hour is defined as having a load of one
Erlang
 Grade of Service (GoS): Measure of congestion (or ability of a user to access a
trunked system) during the busiest hour
 Typically given as likelihood that a call is blocked, called Erlang B or
 The likelihood of a call experiencing a delay greater than a certain
amount of time, called Erlang C
 Trunking …
58

 Average arrival rate, λ:

 Average number of MSs requesting service (call request/time)
 Average hold-time, H
 Average duration of a call (or time for which MS requires
service) A = H (Erlangs)
u

 An average traffic intensity offered (generated) by each

user
2
Au = 3 min = 0.1Erlang
60 min
 Example 1: If a user makes on average two calls per hour,
and that a call lasts an average of 3 minutes
 Trunking …
59

 Example 2: In a cell with 100 MSs average of 30

requests are generated in an hour with average
holding time of 6 minutes
 The arrival rate:  = 30 requests / sec
3600

 Offered load is: Au =

30 Calls
*
360 Seconds
= 3 Erlangs
3600 Seconds Call
 Trunking …
60

 The total offered traffic intensity for U users A = UAu

 Note: A is not necessarily the traffic carried by the trunked system

 In a C channel trunked system, if traffic is distributed equally

among channels, then traffic intensity per channel

UAu A
AC = =
C C

 In Example 1, assume that there are 100 users and 20 channels

 Then A = 100(0.1)= 10 and Ac = 10/20 = 0.5

 Note: Ac is a measure of the efficiency of channels utilization

 Trunking and GoS
61

 Offered traffic is not necessarily the traffic carried by the

trunked system, only that is offered to the system
 Maximum possible carried traffic is the total number of channels,
C, in Erlangs
 AMPS system is designed for a GOS of 2% blocking
 Channel allocations for cells are designed so that 2 out of 100
calls will be blocked due to channel occupancy during the busiest
hour

 What do we do when a call is offered (requested) but all

channels are full?
 Blocked calls cleared? Offers no queuing for call requests,
Erlang B
 Blocked calls delayed? Erlang C
 Trunking – Erlang B Formula
62

 The probability of an arriving call being blocked is:

Ac
Pr [blocking ] = C C! k = GOS
A
k = 0 k!
 Where C: number of trunked channels and A: total offered traffic
 Erlang B is a measure of the GOS for a trunked system which
provides no queuing for blocked calls
 Setting the desired GOS, one can derive
 Number of channels needed
 The maximum number of users we can support as A = UAU or
 The maximum AU we can support (and set the number of minutes on our
calling plans accordingly)
 Since C is very high, it is easier to use table or graph
 Erlang B Formula - Table Form
63
 Erlang B Formula - Graphical Form
64
 Trunking – Blocked Calls Delayed
65

 Instead of clearing a call, put it in a queue and have it wait until a

channel is available
 First-in, first-out line: Calls will be processed in the order received

 There are two things to determine here

1. The probability a call will be delayed (enter the queue), and
2. The probability that the delay will be longer than t seconds
 The first is no longer the same as Erlang B
 It goes up, because blocked calls aren’t cleared, they “stick around” and
wait for the first open channel
 Meaning of GOS
 The probability that a call will be forced into the queue AND it will wait
longer than t seconds before being served (for some given t)
 Trunking - Blocked Calls Delayed …
66

 Additional assumptions:
1. The queue is infinitely long: Translates to infinite memory
2. No one who is queued gives up/hangs up (rather than wait)

 The probability of an arriving call not having an immediate

access to a channel (or being delayed) is given by Erlang C
Formula
Ac
Pr [delay  0] =
A C −1 Ak
A + C!(1 − )
c

C k = 0 k!

 It is typically easiest to find a result from a chart

 Trunking - Calls Delayed …
67

 Once it enters the queue, the probability that the delay is

greater than t (for t > 0) is given as
 C−A 
Pr [delay  t delay  0] = exp − t
 H 

 GOS: The marginal (overall) probability that a call will be

delayed AND experience a delay greater than t is then
Pr [delay  t ] = Pr [delay  0]Pr [delay  t | delay  0]
 C−A 
= Pr [delay  0] exp − t
 H 
 The average delay for all calls in a queued system
H
D = Pr [delay  0]
C−A
 Erlang C Formula - Graphical Form
68
 Trunking - Example 1
69

 Consider a system with

 100 cells
 Each cell has C = 20 channels
 Generates on average λ = 2 calls/hour
 The average duration of each call (H) = 3 Minutes
 How many number of users can be supported if the allowed
probability of blocking is 2%?
 Solution:
 From Erlang B Chart, total carried traffic = 13 Erlangs
 Traffic intensity per user AU = λH = 0.1 Erlangs
 The total number of users that can be supported by a cell = 13/0.1 =
130 Users/cell
 Therefore, the total number of users in the system is 13,000
 Trunking - Example 2
70

 Consider a system with

 100 cells, each cell has C = 20 channels
 Generates on average λ = 2 calls/hour
 The average duration of each call (H) = 3 Minutes
 How many number of users can be supported if the allowed
probability of blocking is 0.2%?
 Solution
 Again from Erlang B Chart, total carried traffic = 10 Erlangs
 Traffic intensity per user AU = λH = 0.1 Erlangs
 The total number of users that can be supported by a cell = 10/0.1 =
100 Users/cell
 Therefore, the total number of users in the system is 10,000
 We support less number of users
 Improving Capacity
71

 A network may need to expand because of

 Increase in traffic or demand for service
 Or because of a change in the environment (e.g., a new
building)
 As traffic increases, the channels originally assigned to
each cell will be congested
 System designers have to provide more channels per
unit coverage area
 Common techniques
 Cell splitting, sectoring, microcell zoning, and use of
repeaters