Chapter 6

Telephone Traffic
6.1 Network Traffic load and parameters

6.2 Loss System: Grade of service (GOS) and Blocking

probability

6.3 Delay System: Queuing theory

6.4 Routing

6.5 Numbering Plans, Charging Plans

 Basic Concept

• Traffic in a communications network refer to the

aggregate of all user requests arrive randomly and
usually require unpredictable service times.

• Traffic analysis is the characterization of traffic

arrivals and service times in a probabilistic framework.

• The effectiveness of a network can be evaluated in

terms of how much traffic it carries under normal or
average loads and how often the traffic volume exceeds
the capacity of the network.
Traffic Engineering Techniques are used most often to determines:
Telecommunication Traffic Engineering standard Terminologies:
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Telecommunication Traffic Engineering standard Terminologies: …
Traffic Theory:
Blocking Formulas – Infinite sources
Blocking Formulas – Finite sources
 Performance analysis methods applied to
telephony are usually referred to as
traffic engineering

 Motivated by two factors

» Unpredictable behavior of users
–You never know when they call!
» Users have to share resources
– Users have to be happy!
 Statistical Analysis
 We cannot exactly predict what users do!

 We can statistically tell what users may do!

 Looking at the traffic pattern we can see

» Number of callers arrived in one hour = 20/hr
» Average length of calls = 3 min/call

 Statistics on voice and data communications are

very different
» Data tends to be burst
» Voice is continuous with known average length
 Call Arrivals
 Calls arrive randomly! Call Arrivals
I=Call Interval
» But how?

Time
 Looking at call arrivals
» Average call interval Call Arrivals/Interval
(I=2sec/call)
» Number of calls per unit 1 min Interval

interval (A=10call/min) Time

» Note that I is inversely
proportional to A
More Calls  Shorter time between successive calls
 Objective of Traffic Engineering
 Given expected traffic (+ growth assumptions
 Provision resources (trunks, switches)
 To minimize cost
 Subject to minimum acceptable quality of service
requirements.

 Available tools:
» Mathematical tools
– Equations and formulas / Statistical tools
» Simulation tools
 What is Blocking?

Call
Calls arriving Blocked
randomly ! No
Blocking
Call !
Center Trunk Call Duration

Trunk
We need to figure out statistically Call
what the probability of blocking is! Center
 What is the grade of service! Trunk
The challenge
 Traffic Model Tool
 Calls arrive randomly
 All calls are independent
 Traffic models
» Idea: telephone usage and sizing the network
» Difference: What to do when blocking happens
» Objective: calculate the Grade of Service (GOS)
– If GOS is 0.01 (P.01) 1 call will be blocked out of 100 attempted calls!
– Defined formally as the number of lost calls over offered calls
 Two common models
» Erlang C
– Call arrival behavior follows a Poisson distribution
– blocked calls may be retried at anytime
» Erlang B Model
– When a call is blocked it is cleared and will only be tried later
– The load does not consider blocked calls
 Traffic Characteristics
 It is all about sharing effectively:
» Telephone traffic is the aggregate of
telephone calls over a group of circuits or
trunks with regard of the number and
duration of calls.
 Units:
» Call arrival rate / hour (A) Switch 1 Switch 2
500 Users 500 Users
» Average service time or hold time (tm) /
hour Trunks

– Duration of the call

» An Erlang (E) is a unit of
telecommunications traffic measurement.
– Strictly speaking, an Erlang represents the
continuous use of one voice path.
– In practice, it is used to describe the total
traffic volume of one hour
– Expresses the traffic intensity
(dimensionless)
– E = A x tm (in Erlang)
 Traffic Characteristics
 Common units used in North America:
» CS : calls-second per second
» CCS : hundred (centrum) calls-second per hour
» 1 Erlang = (60)(60)/100 = 36 CCS = 3600 CS

 Capacity of a single channel is one Erlang

» Interpretation: a telephone that is busy 10% of the time
represents a load of 0.1 Erlang on that particular line

 Example: 1000 calls/hour each has an average

length of 5 min: 1000x(5/60)=83.33 Erlang;
 Using Erlang B Formula
 Assumptions:
» Poisson arrivals (infinite # of sources)
» equal traffic density per source
» lost calls cleared
 Probability of blockage at the switch
due to congestion or “all trunks Es

busy”:
PBlocking  s s! k
» E is the mean of the offered traffic E
[Erlangs], s is the number of trunks 
k  0 k!
 Using Erlang B Formula
Example: Suppose our expected Trunks P=0.001 P=0.005 P=0.01 P=0.05
traffic in the busy hour is 10 erlangs 1 .001 .005 .01 .05

and we require blocking probability 5

10
.76
3.09
1.13
3.96
1.36
4.46
2.22
6.22
of 1 call in 200 (or better). 15 6.08 7.38 8.11 10.63

• What is the grade of service? 20

25
9.41
12.97
11.09
15
12.03
16.12
15.25
19.99
•How many trunks should we use?
• What is the efficiency of the
system?

1/200 = P0.005 = GOS

20 Trunks
Efficiency =
Erlang/Number of Tunks (x100)
= 10/20 (x100) = 50%
 Erlang B Behavior
 The Y axis is typically
expressed in log scale
 S is the number of trunks Probability of Blocking
(log Scale)
 Blocking is between 0-1 S = 10
 Note that eventually blocking
will approach 1 S = 20

 Larges S (more trunks result

in lower blocking)
Erlang (linear scale)
 As the load increases (higher
Erlang) more blocking is
expected
 Higher load can be due to
longer hold time or more call
arrivals
Delay Formulas – Infinite sources
Loss System
 The Kendall-Lee Notation for
Queuing Systems
 Standard notation used to describe many queuing
systems.
 The notation is used to describe a queuing system
in which all arrivals wait in a single line until one of s
identical parallel servers is free. Then the first
customer in line enters service, and so on.
 To describe such a queuing system, Kendall
devised the following notation.
 Each queuing system is described by six characters:
1/2/3/4/5/6
 The first characteristic specifies the nature of
the arrival process. The following standard
abbreviations are used:
M = Interarrival times are independent, identically
distributed (IID) and exponentially distributed
D = Interarrival times are IID and deterministic
Ek = Interarrival times are IID Erlangs with shape
parameter k.
GI = Interarrival times are IID and governed by some
general distribution
 Thesecond characteristic specifies the
nature of the service times:
M = Service times are IID and exponentially
distributed
D = Service times are IID and deterministic
Ek = Service times are IID Erlangs with
shape parameter k.
G = Service times are IID and governed by
some general distribution
 The third characteristic is the number of parallel
servers.

 The fourth characteristic describes the queue

discipline:
» FCFS = First come, first served
» LCFS = Last come, first served
» SIRO = Service in random order
» GD = General queue discipline
 The fifth characteristic specifies the maximum
allowable number of customers in the system.
 The sixth characteristic gives the size of the
population from which customers are drawn.
 In many important models 4/5/6 is GD/∞/∞.
If this is the case, then 4/5/6 is often
omitted.

 M/E2/8/FCFS/10/∞ might represent a health

clinic with 8 doctors, exponential interarrival
times, two-phase Erlang service times, a
FCFS queue discipline, and a total capacity
of 10 patients.
Birth-Death Processes

A birth-death process is a
continuous-time stochastic
process for which the
system’s state at any time
is a nonnegative integer.
 subsequently use birth-death processes to answer
questions about several different types of queuing
systems.
 define the number of people present in any queuing
system at time t to be the state of the queuing
systems at time t.
 πi the steady state, or equilibrium probability, of
state i.
 The behavior of Pij(t) before the steady state is
reached is called the transient behavior of the
queuing system.
The User Datagram Protocol (UDP) is one of the
core members of the Internet protocol suite (the set
of network protocols used for the Internet). With
UDP, computer applications can send messages, in
this case referred to as datagrams, to other hosts
on an Internet Protocol (IP) network without prior
communications to set up special transmission
channels or data paths. The protocol was designed
by David P. Reed in 1980 and formally defined
in RFC 768.
For Poisson processes the property is often
referred to as the PASTA property (Poisson
Arrivals See Time Averages) and states that the
probability of the state as seen by an outside
random observer is the same as the probability
of the state seen by an arriving customer. The
property also holds for the case of a doubly
stochastic Poisson process where the rate
parameter is allowed to vary depending on the
state.
In probability theory, a Cox process, also
known as a doubly stochastic Poisson
process or mixed Poisson process, is a
stochastic process which is a generalization
of a Poisson process where the time-
dependent intensity λ(t) is itself a stochastic
process. The process is named after
the statistician David Cox, who first published
the model in 1955
 Queuing system

Customer Departure of
arrivals served customers
Departure of impatient
customers

• A queuing system can be described as:

"customers arrive for a given service, wait if the service
cannot start immediately and leave after being served"

• The term "customer" can be men, products,

machines, ...
• The theory of queuing systems was developed to
provide models for forecasting behaviors of systems
subject to random demand.
• The first problems addressed concerned congestion of
telephone traffic (Erlang, "the theory of probabilities
and telephone conversations ", 1909)
• Erlang observed that a telephone system can be
modeled by Poisson customer arrivals and
exponentially distributed service times
• Molina, Pollaczek, Kolmogorov, Khintchine, Palm,
Crommelin followed the track
Queuing theory found numerous
applications in:

– Trafic control (communication networks,

air traffic, …)

– Planing (manufacturing systems,

computer programmes, …)

– Facility dimensioning (factories, ...)

Characteristics of simple queuing systems

• Customer arrival processes

• Service time

• Service discipline

• Service capacity

• Number of service stages

 Notation of Kendall
T/X/C/K/P/Z with
– T: probability distribution of inter-
arrival times
– X: probability distribution of service
times
– C: Number of servers
– K: Queue capacity
– P: Size of the population
– Z: service discipline
 Customer arrival process
T/X/C/K/P/Z

• T can take the following values:

– M : markovian (i.e. exponential)
– G : general distribution
– D : deterministic
– Ek : Erlang distribution
–…

• If the arrivals are grouped in lots, use the notation T[X] where X
is the random variable indicating the number of customers at each
arrival epoch
– P{X=k} = P{k customers arrive at the same time}

• Some arriving customers can leave if the queue is too long

 Service times
T/X/C/K/P/Z
• X can take the following values:
– M : markovian (i.e. exponential)
– G : general distribution
– D : deterministic
– Ek : Erlang distribution
–…
k exponential servers with
parameter m

Erlang distribution Ek with parameter m

 Number of servers
T/X/C/K/P/Z

In simple queuing systems, servers are identical

 Queue capacity
T/X/C/K/P/Z

Loss of customers if the queue is full

Capacity K
 Size of the population
T/X/C/K/P/Z

The size of the population can be

either finite or infinite

For a finite population, the customer

arrival rate is a function of the
number of customers in the system:
l(n)
 Service discipline
T/X/C/K/P/Z
Z can take the following values:
• FCFS or FIFO : First Come First Served
• LCFS or LIFO : Last Come First Served
• RANDOM : service in random order
• HL (Hold On Line) : when an important customer arrives, it takes the
head of the queue
• PR ( Preemption) : when an important customer arrives, it is served
immediately and the customer under service returns to the queue
• PS (Processor Sharing) : All customers are served simultaneously
with service rate inversely proportional to the number of customers
• GD (General Discipline)
 The concept of customer classes

A queuing system can serve several

classes of customers characterized by:
• different arrival processes
• different service times
• different costs
• service priority according to their class
 Simplified notation

the simplified notation T/X/C when it

consider a queue where:

• The capacity is infinite

• The size of the population is infinite
• The service discipline is FIFO

• Hence T/X/C = T/X/C///FIFO

Little's law
Some transient performances
THe(T) THs(T)
L(T)

W(T)
• A(T) : number of customers arrived from 0 to T
• D(T) : number of departures between 0 to T
• THe(T) = A(T)/T : average arrival rate between 0 to T
• THs(T) = D(T)/T : average departure rate between 0 to T
• L(T) : average number of customers between 0 to T
• Wk: sojourn time of k-th customer in the system
A T 
1
• W T   
A T  k 1
Wk average sojourn time between 0 to T
 Stability of the queuing system

THe(T) Queueing THs(T)

system

A queuing system is said stable if the

number of customers in the system remains
finite. lim TH T   lim TH T 
e s
T  T 

Implication of D T 
lim 1
the stability: T  A T 
 Little's law
For a stable and ergodic queuing system,
L = TH×W Queuing system
TH TH
L

where W
• L : average number of customers in the
system
• W : average response time
• TH : average throughput rate
Single stage queuing systems
Single stage queuing systems

M/M/1 queue
N(t) : number of customers
in the system

l
Poisson
Exponentially distributed
arrivals
service time
 Stability condition of M/M/1 queue

The M/M/1 queue is stable if l<m

or equivalently r<1

where
r  l/m is called the traffic ratio or traffic
intensity.
The number of customers in the system is
unlimited and hence there is no steady state when
the system is not stable.
Markov chain of the M/M/1 queue

When the system is stable, stationary

probability distribution exists as is
irreducible.

Let  n  lim P N  t   n , n  0
t 
 Steady state distribution of M/M/1 queue

state 0 : 1m   0l

state 0-1:  2 m   1l

Balance equations 
state 0-1-...-n:  n 1 m   n l



Normalization equations:   n  1
n=0

With r  l/m,
0  1 -  n = rn 0
 Performance measures of M/M/1 queue
(online proof and figures)

Ls = Number of customers in the queue = r/1-r) = l/(m-l)

Ws= Sojourn time in the system = 1/1-r)m = 1/(m-l)
Lq= queue length = l2/(m-l)m  Ls - r
Wq= average waiting time in the queue = l/(m-l)m  Ws -
1/m
TH= departure rate = l
Server utilization ratio = r
Server idle ratio = P0 = 1 - r
P{n > k} = Probability of more than k customers = rk+1
 M/M/C queue
Exponentially distributed
service time

l
Poisson
N(t) : number of customers
arrivals
in the system

N(t) is a birth and death process with

• The birth rate l.
• The deadth rate is not constant and is equal to N(t)m if
N(t)  C and Cm if N(t) > C.
Stability condition : l< cm.
 Steady state distribution of M/M/C queue
Distribution :
r l/m n = rn/n! 0,  0 < n  C
-1
n  C -1 r n rC 
r 0     
 n C     C , n  0  n 0 n! C !1 - r C  
C  

m 2m 2m 2m

0 1 2 3

l l l l

Markov chain of M/M/2 queue

Performance mesures of M/M/C queue
Ls = Number of customers in the system
= Lq + r

Ws = Sojourn time in the system

= Wq + 1/m
r C
Lq = Average queue length = C
1 - r C 
¨2

Wq = Average waiting time

= Lq / l

 = Average number of busy server,  = r

U = Waiting probability
= C + C+1 + ...
= C/(1-r/C)
 M/M/C with impatient customers

• Similar to M/M/C queue except the loss of

customers which arrive when all servers are
busy.
m 2m

0 1 2

l l

Markov chain of M/M/2 queue with impatient customers

 M/M/C with impatient customers

Steady state distribution :

-1
r  l/m  C rn 
Pn = rn/n! P0,  0 < n  C P0    
 n 0 n! 
 
Percentage of lost customers = PC

Server utilization ratio = (1 – PC) l/Cm

Insensitivity of Erlang Loss system M/GI/C without queue:

Pn depends on the distribution of service time T only
through its mean, i.e. with m = E[T]
 M/G/1 queue

Service time Ts
Poisson arrival
 M/G/1 queue: Pollaczek-Khinchin formula

• Pollaczek-Khinchin formula or PK formula

r r 2
Ls  
1 - r 2 1 - r 
 - 1
cv 2

• From the PK formula, other performance

measures such as Ws, Lq, Wq can be easily
derived.

• From PK formula, can observe that randomness

always hurt the performances of a system. The
larger the randomness (i.e. larger cv2), the
longer the queue length is.
 G/G/1 queue

• Inter-arrival times An between customer n and n+1 :

E[An] = 1/l

 A2  Var  An 
• Service time Tn of customer n :
E[Tn] = 1/m

  Var Tn 
2
T

• Waiting time Wn in the queue of customer n (Lindley

equation)
Wn+1 = max{0, Wn + Tn – An}
 G/G/1 queue

• Bounds of Waiting time

1
l T2 -  2 - r  l  A2   T2 
m
 E W  
2 1 - r  2 1 - r 

• If E[A - t | A > t] < 1/l, then

l  A2   T2  1  r l  A2   T2 
-  E W   tightless check with Lq
2 1 - r  2l 2 1 - r 

• Waiting time approximation (Kingman's equation or VUT

equation)

E W  
 2
A   T   r  1
2

 
2  1- r  m
Variability Utilization Time
Queuing networks
Definition of queuing networks

A queuing network is a system composed of several

interconnected stations, each with a queue.

Customers, upon the completion of their service at

a station, moves to antoher station for additional
service or leave the system according some routing
rules (deterministic or probabilitic).
Example of deterministic routing

Shortest queue rule

Open network or closed network

Open network

N customers

Closed network
Multi-class network
 A production line

Raw Finished
parts parts
 Open Jackson Network

An open Jackson network (1957) is

characterized by:
• One single class of customers
• A Poisson arrival process at rate l equivalent to independent
external Poisson arrival at each station)
• One server at each station
• Exponentially distributed service time with rate mi at station i
• Unlimited capacity at each queue
• FIFO service discipline at all queues
• Probabilistic routing
 Open Jackson Network
routing

• pij (i ≠0 and j≠ 0) : probability of moving to

station j after service at station i
• p0i : probability of an arriving customer
joining station i
• pi0 : probability of a customer leaving the
system after service at station i
 Open Jackson Network
stability condition

• Let li be the customer arrival rate at station i, for

i = 1, ..., M where M is the number of stations.
• The system is stable if all stations are stable, i.e.
li < mi, i = 1, ..., M

• Consider also ei the average number of visits to

station i for each arriving customer:
ei = li/l
 Open Jackson Network
arrival rate at each station

• These arrival rates can be determine by the

following system of flow balance equations
which has a unique solution.
 Open Jackson Network
Are arrivals to stations Poisson?

as the
departure
process of
M/M/1 queue
is Poisson.

Feedback
keeps
memory.
 Open Jackson Network
State of the queueing network

• Let n(t) = (n1(t), n2(t), …, nM(t)), where ni(t) is

the number of customers at station i at time t
• The vector n(i) describes entirely the state of
the Jackson network
• {n(t)}t≥0 is a CTMC
• Let (n) be the stationary probability of being
in state n
• Notation: ei = (0, …, 0, 1, 0, …, 0)
i-th position
 Open Jackson Network
Underlying Markov Chain

Attention: Some
transitions are not
possible when ni = 0, for
some i
 Open Jackson Network
Stationary distribution - Product form solution

Theorem: The stationary distribution of a

Jackson queuing network has the following
product form : M
  n     i  ni 
i 1
li
 i  ni   r 1 - ri  , ri 
ni

mi
i

where i(ni) is the stationary distribution of a

M/M/1 queue with arrival rate li and service
rate mi, i.e.
 Open Jackson Network
Performance measures

TH i  li
Performance ri
Lsi 
measures of each 1 - ri
M/M/1 queue ri
Wsi 
mi 1 - ri 
TH  l
M
Performance Ls   Lsi
measures of the i 1

queuing network Ws  M e  Ws  Ls
i i i 1 TH
 Open Jackson Network
Extension to multi-server stations

• Assume that each station i has Ci servers

• The stability condition is
li < Cimi, i = 1, …, M
• The stationary probability distribution still has the
product form: M
  n     i  ni 
i 1

where i(ni) is the stationary distribution of a M/M/Ci

queue with arrival rate li and service rate mi.
 Closed Queuing Network
Definition

• Similar to Jackson network but

• with a finite population of N customers
• without extern arrivals.
As a result,
• l=0
M

• p
j 1
ij  1, i  1,..., M

•  n t   N , t  0
i 1
i
 Closed Queuing Network
Arrival rates

• The arrival rates li satisfy the

following flow balance equations
M
li   l j pij , i  1,..., M
j 1

• Unfortunately, the above system of

flow balance equations has one
free variable.
 Closed Queuing Network
Product form solution

Product form solution of Gordon and Newell (1967)

1
  n1 , n2 ,..., nM   r1n1 r2n2 ...r MnM 1
CN 
C  N  n1 ,..., nM
r1 r
n1 n2
2 ...r M 1
nM

where
• ri = li/mi with li obtained from the solution of the flow
balance equations with a free constant chosen arbitrarily
• C(N) is a normalizing constant such that the sum of
probability equals 1, i.e.

Direct computation of C(N) is very tedious when the

state space is large.
 Closed Queuing Network
Computation of the normalization constant C(N)

Buzen's algorithm (1973) uses relations

Ci(k) = Ci-1(k) + riCi(k-1), i=2, ..., M, k = 2, ..., N
where
Ci  k   
n1 ... ni  k
r r ...r
n1
1
n2
2 i
ni

with initial conditions

C1(k) = (r1)k, Ci(1) = 1
from which C(N) is obtained as
C(N) = CM(N)
 Closed Queuing Network
Computation of the normalization constant C(N)

It can be shown that the utilization of station i is given by

C  N - 1
1 -  i  0   ri
CN

The marginal distribution can be determined as follows :

rik
P ni  k  C  N - k  - ri C  N - k - 1 
CN
N CN - k
E  ni    r k

CN
i
k 1
 Closed Queuing Network
Mean Value Analysis (MVA)

• Suppose we are only interested in throughput THi and mean number of customers at
station i Li (i.e. Lsi)and mean system time Wi (i.e. Wsi)
• The MVA method of Reiser and Lavenberg (1980) bypasses te computation of C(N).
• It relies on the following simple relations :
1 1
Wi   Li
mi mi
where TH i  aN li
Wi is the average system time experienced by a customer arriving at i

Li is the average queue length seen by a customer arriving at i

li is any solution of the flow balance equation
aN is the missing factor.
 Closed Queuing Network
Mean Value Analysis (MVA)
• It can be shown that Li is the same as the average queue length at i in a
network with (N-1) customers.
• Let Li(N), THi(N), Wi(N) be the queue length, the throughput and the system
time of a network with N customers.
• The following system can be iteratively solved to obtain the results:
1  L  N - 1  , i  1,..., M , N
1
1 Wi  N  
mi
i

(2) Li  0   0, i  1,..., M
N
(3) N   Li  N , N
i 1

(4) Li  N    li aN Wi  N  , i  1,..., M , N

where equation (4) is from Little's law for station i.

At each iteration N = 1, 2, ..., (1) is used to determine Wi(N), combination of (3)

& (4) determines aN and THi(N), (4) gives Li(N).
Closed Queuing Network

p m2
m1
1- m3
p

p = 0.5, m1 = 4, m2 = 1, m3 = 2
N= 2, 3, 4
 Routing
&
flow control
 Routing Algorithms

 Main function of Network Layer:

» Routing of packets form the source

machine to the destination machine.

 Routing Algorithms
 Routing Algorithm:
» Network Layer Software responsible for
deciding which output line an incoming
packed should be transmitted on.
–Datagrams:
 require computation of decision making tables for each
packed

–Virtual Circuit:
 routing decision are made only when a new virtual circuit is
being set up.

–Session Routing:
 data packets follow the same routing for the entire session.
» Routing vs. Forwarding:
–Routing:
 Filling and Updating routing tables

–Forwarding:
 making the decision which routes to use based on routing tables.

» Adaptive vs. Non-Adaptive Algorithms.

–Non-Adaptive Algorithms:
 Routing decision is based on pre-computed measurements or
estimates and do not update the table based on current traffic and
topology

–Adaptive Algorithms:
 Change their routing decisions to reflect changes in the topology
and traffic.
 Shortest Path Routing
 Optimization criterion:
» Distance,

» Bandwidth,

» Average Traffic

» Communication cost,

» Mean Queue Length,

» Measured Delay, …
 Distance Vector Routing
 Static Routing Algorithms
» Do not take into account actual network load.

 Dynamic Routing Algorithms

» Taking into account actual network load

» Distance Vector Routing: Each router maintain a table

with the best known distance to each destination and
which line to use to get there. Tables updated by
exchanging information with the neighbors.

» Link State Routing

 Link State Routing
 Distance Vector Routing was used in the ARPANET
until 1979 – when it was replaced by link state routing.
» Delay Metric was Queue Length thus did not take into
account line bandwidth when choosing routes.

1. Problem when line bandwidth changed for some

bands from 56 kbps to 230 kbps or 1.544 Mbps.

2. Algorithm took to long to converge (the count-to-

infinity problem).

– Solution: Link State Routing

 Link State Routing …

 Each router must do the following:

1. Discover its neighbors and learn their network addresses.

2. Measure the delay or cost to each of its neighbors.

3. Construct a packet telling all it has just learned.

4. Send this packet to all other routers.

5. Compute the shortest path to every other router.

 Complete topology and all delays are experimentally measured and

distributed to every router. Dijkstra’s algorithm can be run to find the
shortest path to every other router.
 Building Link State Packets
 Packet Format: A B C
» Identity of Sender Seq. Seq. Seq.
» Sequence Number
Age Age Age
» Age
» List of Neighbors B 4 A 4 B 2
» Corresponding Delay E 5 C 2 D 3

F 6 E 1

B 2 C

4 3
D D E F

A 6 Seq. Seq. Seq.

1 Age Age Age

5 4 C 3 A 5 B 6

F 4 C 1 D 4
E 8 F F 8 E 8
 Packets easily built – problem with knowing when to built
them.
 Hierarchical Routing
 Large Networks:
» Proportionally large routing tables are required for each
router

» More CPU time is needed to scan them

» More bandwidth is needed to send status reports.

» At certain point network may grow so large where it is

no longer feasible for every router to have an entry for
every other router.

» Solution: Routing has to be done hierarchically.

 Hierarchical Routing …
 Routers divided in Regions (as in telephone network):
» Each router knows how to route packets to destinations within its own region.

» However, router does not have any information regarding the topology of the network
of other regions.

 When different networks are interconnected they are regarded as a separate

region in order to free the routers in one network form having to know the
topological structure of the other ones.

 Huge networks will require more than two-level hierarchy.

 How many hierarchical levels are optimal.

» Kamoun and Kleinrock (1979): optimal number for an N router subnet is ln(N),
requiring total of e*ln(N) entries per router.
 Two Level Hierarchical Routing
Example
Full table for 1A Hierarchical table for 1A

Dest. Line Hops Dest. Line Hops

2A 2B 1A - - 1A - -
1B
1B 1B 1 1B 1B 1
Region 2 1C 1C 1 1C 1C 1
Region 1 2A 1B 2 2 1B 2
1A 1C
2C 2D 2B 1B 3 3 1C 2

2C 1B 3 4 1C 3

2D 1B 4 5 1C 4

3A 1C 3

3B 1C 2

5B
4A 1C 3

4B 1C 4
3A 3B 5A 5C
4A
4C 1C 4

5A 1C 4
Region 3 Region 44C Region 5 5B 1C 5

5E
5C 1B 5
4B
5D
5D 1C 6

5E 1C 5
 Broadcast Routing
 Sending a packed to all destinations simultaneously is called
Broadcasting.
» Direct Method: Source sends a distinct packet to each destination
routers in the subnet:
1. Wasteful of the bandwidth.

2. It requires source to have a list of all destinations.

– In practice this may be the only feasible solution.

» Flooding:
– Ordinarily ill suited for point-to-point communication:
 Generates to many packets, and

 Consumes to much bandwidth.

 Broadcast Routing …
» Multi-destination Routing
– Each packets contains:
 A list of designations, or

 A bit map indicating the desired destinations.

– When packet arrives at a router:

 The router checks all the destinations to determine the set of output lines that will be
needed.

 Generates a new copy of the packed for each output line to be used and includes in each
packet only those destinations that are to use the line.

 After a sufficient number of hops, each packed will carry only one destination and can be
treated as normal packet.

– Multi-destination routing is like separately addressed packets, except that when

several packets must follow the same route, one of them pays full fare and the rest
ride free.
 Broadcast Routing …
 Spanning Tree:
» It is a subset of the subnet that includes all routers but contains no loops.

» Each router knows which of its lines belong to the spanning tree, it can
copy an incoming broadcast packet onto all the spanning tree lines
except the one it arrived on.
– Makes excellent use of bandwidth (generates absolute minimum number of
packets necessary to do the job)

– Must have knowledge of some spanning tree for the method to be applicable.
 Information available in some instances (e.g., link state routing)

 Information not available (e.g., distance vector routing)

 Broadcast Routing …
 Reverse Path Forwarding:
» Router checks if the broadcast packet arrived on the line that is
normally used for sending packets to the source of the broadcast.

» If so, there is excellent chance that the broadcast packet itself

followed the best route from the router and is therefore the first
copy to arrive at the router. The router forwards copies of it onto
all lines except the one it arrived on.

» If the broadcast packet arrived on a line other than the preferred

one for reaching the source, the packet is discarded as a likely
duplicate.
 Broadcast Routing …

 Example of Reverse path Forwarding

B C
B C
A
D
A
D F
F
E
E I G
I G

J
J H
H L
L
N
O
N
O
K
K

M
M

A subnet A sink tree for

router I
 Broadcast Routing …

 Reverse Path Forwarding in spite of not being

optimal procedure:
» Efficient and easy to implement Algorithm

» It does not require routers to know about spanning trees.

» Does not have the overhead of destination list or bit map in

each broadcast packet (as multi-destination addressing).

» It does not require any special mechanism to stop the

process as flooding does
 Multicast Routing
 Application that require separate processes (i.e., each from separate
location) access and ability to work on the same data.
» Small group can use point-to-point messaging to accomplish this task.

» Broadcasting can be used but communicating with 1000 “interested”

machines out of million-node network is inefficient.

 Need a mechanism that would send messages to well-defined groups

that are numerically large in size but small compared to the network as a
whole.
» Sending a message to a such a group is called multicasting.

» Corresponding routing algorithm is called multicast routing.

 Multicast Routing …
 Requirements:
» Create and Destroy Groups

» Nodes should be able to Join and Leave Groups, etc.

– Group Management.

» When a process joins a group it informs its host.

» Routers must know which of their hosts belong to which group.

– Host must inform their routers about changes in group membership, or

– Routers must query their hosts periodically.

» Information shared with Neighboring Routers (propagation of

information through the subnet).
 Multicast Routing …
 Each Router Computes Spanning Tree Covering all other routers.

 Example of a network with nodes belonging to two groups (1 & 2).

2 1
2 1
1,2 1,2
1,2 1,2 2
2
2 2
2 1
1 1
1 Spanning Tree for the leftmost Router
Network

1 2

1 1 2 2
2

2
1
1
A Multicast Tree for group 1 A Multicast Tree for group 2
 Routing for Mobile Hosts
 Increasing number of users of Portable
Computers and Personal Computer Devices.
They require access to:
» E-mail
» File System, etc.
 In order to route a packet to a mobile host, the
network first has to find it.
 World Model of communication network:
» WAN consisting of routers and hosts,
» LAN’s connected to WAN, and
» MAN’s connected to WAN.
 Routing for Mobile Hosts …
 A WAN to which LAN’s, MAN’s and wireless cells are attached:
Wireless cell
Home Agent

Mobile
Host
Home LAN

Foreign
Agent

Foreign LAN

WAN MAN
 Routing for Mobile Hosts …
 Stationary Hosts:
» Hosts that never move.

 Migratory Hosts:
» Stationary hosts who move from one fixed site to another from
time to time but use the network only when they are physically
connected to it.

 Roaming Hosts:
» Need to maintain their connections as they move around.

 Mobile Hosts:
» Migratory and Roaming Hosts – that is all host that are away from
home and still want to be connected.
 Routing for Mobile Hosts …

 All hosts are assumed to have:

» A permanent home location, and
» A permanent home address:
– Used to determine their home location (analogous to
telephone number; e.g., 1-212-555-1212).
 Routing goal in systems with mobile hosts:
» To make possible to send packets to mobile hosts
using their home address, and
» Have the packets efficiently reach them wherever they
may be.
 Trick is off course to find them first.
 Routing for Mobile Hosts …
 world is divided up (geographically) into small units – areas.
» Areas are typically LANs or wireless cells.
» Each area has one or more
– Foreign agents:
 Processes that keep track of all mobile hosts visiting the area.
– Home agent:
 Keeps track of hosts whose home is in the area, but who are
currently visiting another area.

 REGISTRATION with Foreign Agent: When a new host enters an

area, either by connecting to it (e.g., plugging into LAN), or wandering
into the cell it must register itself with the foreign agent of that area.
 Registration Procedure of Mobile Hosts
1. Each foreign agent broadcasts periodically a packet announcing its existence and
address.
 Newly-arrived mobile host may:
1. Wait for one of these message, or if none arrives quickly enough
2. Can broadcast a packet “saying”: Are there any foreign agents around?

2. The mobile host registers with the foreign agent:

 Gives its home address,
 Current data link layer address, and
 Some security information.

3. The foreign agent contact the mobile host’s home agent and informs it about a mobile
host in his area. This message contains:
 The foreign agent’s network address,
 Security information (“to convince the home agent that the mobile host is really there”).

4. The home agent authenticates security information containing:

 Timestamp (to prove that it was generated within the past few seconds), etc.
 Acknowledges foreign agent by indicating to proceed if everything checks out.

5. Foreign agent registers and informs the mobile host.

 Checking out when done (typically mobile hosts just turn-off their computers).
 Packet Routing for Mobile Hosts

 Example: Sender wants to send a packet to a host in New York.

» Packets sent to the mobile host on its home LAN in Kathmandu are intercepted by the
home agent (step 1).

» Home agent looks up mobile host’s new (temporary) location and finds the address of
the foreign agent handling the mobile host (i.e., Pokhara.)
» Home agent does:

1. It encapsulates the packet in the payload field of an outer packet and sends the
latter to the foreign agent (step 2). This mechanism is called tunneling.
After getting the encapsulated packet, the foreign agent removes the original
packet from the payload field and sends it to the mobile host as a data link frame.

2. The home agent tells the sender to henceforth send packet to the mobile host by
encapsulating them in the payload of packets explicitly addressed to the foreign
agent instead of just sending them to the mobile host's home address (step 3.)
Subsequent packets can now be routed directly to the host via foreign agent (step
4.), bypassing the home agent entirely.
 Routing in Ad Hoc Networks
 Ad Hoc networks refers to the extreme case of mobility
where not only the hosts are mobile but routers as well.
Examples:

1. Military vehicles on a battlefield with no existing

infrastructure.

2. A fleet of ships at sea.

3. Emergency workers at en earthquake that destroyed

the infrastructure.

4. A gathering of people with notebook computers in an

area lacking 802.11 (IEEE Wireless LAN protocol).
 Routing of Ad Hoc Networks …

 Each node consists of a router and a host – usually on the

same computer.

 Networks of neighboring nodes are called ad hoc

networks or MANETs (Mobile Ad hoc NETworks).

 Features of such networks:

» Routers come and go or appear in new places all the time.

» Consequently, Network Topology may be changing all the time.

» Validity of paths can thus change spontaneously.

 Ad hoc On-demand Distance Vector
routing (AODV) Algorithm

 AODV takes into account limited bandwidth

and low battery life of devices.

 It is a on-demand algorithm; that is it

determines a route to some destination only
when there is a demand to send a packed to
a destination.
Ad hoc On-demand Distance Vector routing …

 Route Discovery
» Ad-hoc network can be described by a graph of the nodes
(routers + hosts).

» Two nodes are connected (depicted via an connecting arc

in the graph) if they can communicate directly using their
radios.

» For simplicity connection is assumed symmetric (e.g., case

where node A has a more powerful transmitter then node
B thus node A can communicate to B but not vice versa is
not considered).
 Ad hoc On-demand Distance Vector
routing Example
Range of A’s
Broadcast

A A A A
B C B C B C B C
D D D D
E E E E
F G F G F G F G

H I H I H I H I
Range of A’s Broadcast After B and D have received of A’s After C,F and G have received of After E,H and I have received of
Broadcast A’s Broadcast A’s Broadcast
 Congestion Control
Algorithms
 The situation when to many packets are present in subnet it is
called congestion.

Congestion Traffic Performance Chart as functino of Number of Packets

Sent

1.2

Maximum Carrying capacity of

subnet
1
Packets Delivered [%]

0.8

Desirable Performance
0.6
Congested Performance

Ideal
0.4

0.2

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Number of Packets in Mil.

 Congestion Factors
 Streams of inputs packets arriving from multiple lines (3-4 or more) needing the same output line
=> queue buildup.
» Adding more memory may help to a point but – a study by Nagle suggest that increasing memory to ∞ to
accommodate larger queues congestions gets worse not better due to time required to get to the front of the
queue (timed out).

 Slow Processors:
» Queue build up due to routers slow CPU’s at performing bookkeeping task:
– Queueing buffers,

– Updating tables, etc.

 Low bandwidth lines.

 Typically the problem is that upgrading one part of the system shifts the bottleneck to the other
someplace else. The real problem frequently is a mismatch parts of the system. The problem will
persist until all components of the system are in balance.
 Congestion Control vs. Flow Control

 Congestion Control has to due with making sure the subnets are able to
carry the offered traffic. Thus it is a global issue involving the behavior of
» all the hosts,

» all the routers,

» the store-and forwarding processing within the routers, and

» all the other factors that tend to diminish the carrying capacity of the subnet.

 Flow Control relates to point-to-point traffic between a given sender a

given receiver. Its job is to make sure that a fast sender cannot continually
transmit data faster than the receiver is able to absorb it.
» It frequently involves some direct feedback from the receiver to the sender to
tell the sender how things are doing at the other end.
 Congestion Control vs. Flow Control
Example

 Flow Control Problem: Consider a fiber optic network with a capacity of

1000 Gbps on which a supercomputer is trying to transfer a file to a
personal computer at 1 Gbps. Although there is not congestion the
supercomputer has to frequently stop in order to allow PC to catch up.

 Congestion Control Problem: Consider a store-and-forward network with

1 Mbps lines and 1000 large computers, half of which are trying to
transfer files at 100 kbps to the other half creating a traffic of 500x100
kbps = 50000 kbps = 50 Mbps. The problem here is that this traffic
exceeds the capacity of what the network can handle.
Numbering Plans,

Charging Plans
 Numbering

Ensures equal and simple access to

the various networks and services
 Numbering

• A number (string of decimal digits) uniquely indicates

the public network termination point and contains the
information to route the call to this termination point
at national or international level (with country code).

• The National Numbering Plan is the set of numbers

identifying any termination points of fixed or mobiles
networks and telephony services, used for routing calls,
or access to the external resources.
 Numbering …

• The National Numbering Plan is a part of the

worldwide numbering plan (E 164). Gov draws
up and manages the National Numbering Plan
(ITU - the worldwide numbering plan).

• There are others plans dealing with addressing

(IP addresses, DNS, etc.).
 Effective Numbering activities:

• Gov. pursues its numbering activities in connection at

the international level.

• A consultative numbering committee, chaired by Gov,

serves as a forum for all sector participants affected by
numbering issues: operators, service providers,
manufacturers, users, etc.

• Gov adopted rules defining the actual procedures for

reservation, allocation, control, abrogation, publication,
Effective Numbering activities: …

• Gov allocates, on a case-by-case basis, the resources

(prefixes, numbers and numbering ranges) required
to ensure fair competition between network
operators;
• The conditions under which these numbering
resources may be used are fixed, either in the
operator's schedule of conditions or in the decision
transmitted to the operator by Gov.
• The operators may pay a fee which covers the cost
of managing the numbering plan and monitoring its
use.
 Effective Numbering activities: …

• Gov draw up and manages the National Numbering plan:

– principle of long-distance carrier selection using a single digit or a four-
digit prefix.
– definition of non-geographic numbers to identify service categories by
the tariff charged to the caller : freephone numbers, shared cost
numbers and shared revenue numbers.
– Freephone numbers are free of charge for the caller.
– Shared cost numbers, the cost is divided between the called and
calling parties.
– Shared cost numbers, the called party is reimbursed by the
telecommunications service provider.
– opening up the 0XXX numbering range for Internet dial-up access.
 Shared Revenue Services
Code = (0)900

Called Party Calling Party

Revenue = X Charge = 100+X

 Network 
Network Conveyance
Charge = 100
 Shared cost Services
Code (0)800

Called Party Calling Party

Charge = 50 Charge = 50

 Network 
Network Conveyance
Charge = 100
 Freephone services
Code = (0)800

Called Party Calling Party

Charge = 100 Charge = Free

 Network 
Network Conveyance
Charge = 100
 Before liberalization

• Numbering plan for the existing network

• Operator responsible for numbering

• New operators need new numbers

• Conflict of interest

• Need for independent administration

 What is E.164?

 ITU-T Recommendation E.164: “The

international public telecommunication
numbering plan”
» Tied to treaty obligations (specific roles and
obligations defined for ITU Member States and
TSB Director)
» Defines number structure and functionality for four
principal categories of numbers:
– Geographic Areas
– Global Services
– Networks
– Groups of Countries (“GoCs”)
Examples of E.164 Resources
 Geographic areas
» ITU Member States, including integrated
numbering plan involving more than one (e.g., +1)
 Global Services
» e.g., Universal International Freephone Numbers
(+800)
 Networks
» Global Mobile Systems (+881 + 1 digit IC)
» Shared code for Networks (+ 882 + 2 digit IC)
 Groups of Countries
» e.g., ETNS
 Related to E.164
 ITU-T Recommendation E.164.1: Criteria and procedures for the
reservation, assignment and reclamation of E.164 country codes
and associated Identification Codes (ICs);
 ITU-T Recommendation E.164.2: E.164 numbering resources for
trials (to be published);
 Determined Recommendation E.164.3: Principles, criteria and
procedures for the assignment and reclamation of E.164 country
codes and associated identification codes for Groups of
Countries (determined at January 2001 meeting of SG2);
 ITU-T Recommendation E.190: Principles and responsibilities for
the management, assignment and reclamation of E-series
international numbering resources;
 E.195: ITU-T International numbering resource administration
 Key Elements/ ITU-T Structure E.164

Format:-

00 +CC+ NSN (CC +NSN must not exceed 15 characters)

NSN definition: NDC + SN

NDC=Area Code

SN= Subscriber Number (for local dialling)

 Key Elements/ ITU-T Structure E.164

Format:-

00 +CC+ NSN (CC +NSN must not exceed 15 characters)

00 965 1 8187912
NSN definition: NDC + SN
1 8187912
NDC=Area Code
1
SN= Subscriber Number (for local dialling)
8187912
 Key Elements Numbering Plan
 Existing structure of Numbering Plan maintained

Prefix Designation
00 International Code
01
02
03
04
05
06
07 Mobile Services
08 Special Services (Free Phone, Fixed Cost and Shared Cost)
09 Premium Rate Services (PRS)
 Key Elements/ Guidelines

 Implementation principles

 Description of the Gov Numbering

Database

 Who can apply for numbers?

 Reserving numbers or codes

 Assessment of applications
 Key Elements/ Guidelines ...

 Response times and publicity

 Conditions placed on allocations

 Withdrawal of reservations and allocations

 Appeals procedure

 Annual report, audit and long range

forecasts
 Numbering, Licenses & Laws

• Primary legislation on telecommunications

• A license for the existing network operator

• License condition on numbering

• Review of existing Numbering Plan

• Update of Numbering Plan

• Move to independent administration.

 Function of Numerating Unit

• Set-up and maintain the data base

• Filter and process applications

• Make reservations and allocations

• Publication current status details-

– Designated codes and number ranges

– Reservation (service details confidential)

– Allocation with planned in service dates.

 Ongoing Joint Responsibilities

• Maintain accurate statues records

• Measure utilization
• Prepare 5-year forecasts of demand
• Conserve capacity where necessary
• Identify and solve potential shortages
• Publication annual report on numbering
• Periodic review and update of plan
 Unusable numbering capacity

• 10% - Digit 0 – National dialing prefix

• 10% - Digit 1xx(x) – Short code access
• 10% - Digit Y – Reserved for expansion
• 10% - Protected after previous use
• Balance = 60%
• Also – local network structure may use some
number-blocks better than others
 Numbering –Principles

• Conformance with International

Recommendations of ITU-T
• Numbering Plan - longevity
• Clear and understandable
• Capable of expansion for new operators
• Reflect services and location
• Be innovative
• Neutral as regards competition
• Objective and non-discriminatory
Numbering Plan of Nepal
Charging Plans
• The charges are defined keeping other
competitors and regulatory in mind.

• T here could be different types of charges to

be applied for a product and associated
services.

• For a given product, an operator can define

one or more of the following charges, but
they are not limited to only these charges,
there could be some other type of charges
depending on country, location and business
situation.
Product Initiation Charges:
These are one-time charges, which can be taken from
the customer as a part of installation, activation,
service or initiating a connection.
Product Periodic Charges:
These are the charges, which can be applied on
monthly or bi-monthly or yearly basis as a rental of
the product and service provided.
Product Termination Charges:
These are the charges, which can be applied on
termination of the product and service.
Product Suspension Charges:
These are the charges, which can be applied if a
product is suspended because of some reason; for
example, non-payment.
Product Suspension Periodic Charges:
There could be a requirement to charge a
customer periodically even if a customer is
suspended because of some reason.
Product Re-activation Charges:
Assuming a product was suspended due to some
reason and now it needs its activation, an operator
can apply re-activation charges for this service.
Product Usage Charges:
This is most important type of charge, which
would be applied based on the usage of the service.
For example, call per minute or per second, data
download per MB, etc.
 In-Advance & In-Arrear Charges:

When an operator would like to charge their customers

in advance for some services and in the end of every
month for some services.

Charges taken in advance before providing the services

are called in-advance charging and charges taken after
providing the services are called in-arrear charges.
For in-arrears charging , the product charges are
applied for a period up to at least the day before the
current nominal bill date (or bill request date for non-
periodic bills).
So while configuring different charges, billing
system should give a provision to configure
charges in advance and it is always optional for
the operators if they want to configure a
particular price in-advance or in-arrears.

Usage charges can not be taken in advance until

they are lump-sum because you never know
how much usage a customer is going to generate
in coming month. If they are lump-sum amount,
then you can take that amount in advance and
let the customer use unlimited based on their
requirement.
 Proratable & Non-Proratable Charges:

when a customer takes phone connection in the

middle of the month and his invoice needs to be
generated on 1st of every month.

If prices are non-proratable, billing system would

charge the customer for the whole month which
would not be fair with the customer. Same apply at
the termination, if customer terminates a service in
the middle of the month, then operator may not be
willing to charge the customer for rest of the month.
Pro-ratable pricing means that they would apply only
for the number of days customer is going to use the
service. For example, if monthly product rental is $30
and customer used this product for 10 days only, then
billing system should charge the customer only $10 for
those 10 days.

So billing system should provide an option to configure

particular prices to be pro-ratable as well as non-
proratable and let the operator choose what suites them
best.
Refundable & Non-Refundable Charges:

let us consider a situation where an operator is

charging a customer in advance for the whole month,
but customer leaves in the middle of the month after
using a service for 10 days.

If prices were configured as non-refundable, then

they would not be refunded to the customer, but if
they were configured as refundable, then they
would be refunded to the customer.
 Charge Overriding Option:
A good billing system provides an option to
override base prices at the time they are given to
the customer.
For example, for a particular product base prices in the catalogue
are defined $30 per month but customer is not ready to pay $30 per
month, and based on some bargaining, he is ready to pay $25 per
month. In such situation, customer service representative (CSR)
should be able to override defined base price $30 and add them as
$25 at the time of customer creation in the system.
Billing system should give an optional provision to the operators if
a particular price can be overridden or not and let the operators
decide if they want to override some charges at the time of sale or
they are fixed in all the situation.
 Revenue Segregation by
Revenue Codes:
All the operators would like to know how much
they have earned using a particular product, its
rental, suspension or usage, etc.
While defining different prices in the catalogue,
billing system should give a provision to associate
some kind of revenue codes or keywords with
different types of charges. This helps in generating
different reports based on the codes associated with
the revenue.
 Tariffs Classification:
An operator may define different tariffs, which can
be offered to different people having different credit
classes.

For example, a 5mbps data line at a cost of $100 per

month can be offered to a customer having monthly
income more than $1000/month and a 1mbps data
line can be offered to a customer having minimum
monthly income $500/month.
All the billing systems give options to define
different credit classes, which can be assigned to
customers based on their credit history and income
and may be based on some other parameters defined
by the operator.

All the products and services can have different

tariff plans, which can be offered to different classes
of people ranging from general class to VIP class.
 Parameters for Usage Charges:

There are number of parameters, which can be used

while defining usage charges. For example:

• Calls in daytime, usually called peak time, will be

charged on higher rate and in night time, i.e., off
peak time rate will be relatively low.

• If calls are terminating within the same network,

usually called on-net calls, would be charged at
relatively low prices.
• Calls during weekend, i.e., Saturday would be
charged at low prices.
• Calls to a particular destination would be charged
at high prices.
• Calls during some festival would be charged at
special prices.
• Data download from a particular site would be free
of cost.
• Sending SMS to a particular code would be
charged at high rate.
• Calls with-in a particular group of numbers,
usually called closed user group (CUG), would
be charged at zero prices.
• Sending international or national MMS would be
charged at the same prices.

Billing systems provide lots of flexibility to define

various such rules to charge voice, data, SMS or
MMS usage generated by the customer.