0% found this document useful (0 votes)
14 views

Communication System Engineering

The document discusses different concepts related to handling lost calls in telecommunication systems including lost calls held, lost calls cleared, and lost call delayed. It also discusses probability distributions, Erlang traffic formula, queuing systems, dimensioning telecommunication networks, and other related topics.

Uploaded by

precash12
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
14 views

Communication System Engineering

The document discusses different concepts related to handling lost calls in telecommunication systems including lost calls held, lost calls cleared, and lost call delayed. It also discusses probability distributions, Erlang traffic formula, queuing systems, dimensioning telecommunication networks, and other related topics.

Uploaded by

precash12
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 18

Telecommunication system

engineering
Introduction to Telephony (2)
Lecture # 3
24-02-2010
Handling of Lost calls
• Lost Calls Held (LCH)
This concept assumes that the telephone user will
immediately reattempt the call on receiving the congestion
message and will continue to redial. The user hopes to
seize the next available switching equipment is free.
• Lost Call Cleared (LCC)
On receiving the congestion signal, the user will hang up
and wait for a while for attempting the call. Such calls
usually disappear from the system.
• Lost Call Delayed
This concept assumes that the user request is automatically
put in a queue, where it waits for it’s turn to be routed.
Probability distributions curves for
traffic theory
• The origination calls at an exchange closely fit a family of probability
distribution curve following the a Poisson distribution.
• Most distribution curves are two parameters curves, mean and variance.
– mean is average
– variance is parameter for dispersion
Smooth, Rough and Random traffic
• VMR (variance to mean ratio)
It is the coefficient of over-dispersion, and characterizes the
traffic patterns.
• Smooth traffic: VMR < 1
• Random traffic: VMR = 1
• Rough traffic: VMR > 1
Erlang traffic formula
• Erlang traffic formula is commonly used for dimensioning the route (finding the
number of trunks required to accommodate traffic).
• The parameters that are basically dealt with are for computing the traffic are
– Call arrival and holding time distribution
– Number of traffic sources
– Availability of the traffic sources
– Handling of lost calls

• n is number of serving channels


• A is mean traffic offered
• EB is the grade of service
Erlang B table
Waiting systems (Queuing)
• Queuing system is required when the scenario
of lost call delay (LCD) is encountered.
• There are four ways (queuing discipline) to
select the waiting call from the queue.
– First come first serve
– Random selection
– Last come first serve
– Priority selection
Waiting systems (Queuing) cont’d
• Grade of service for the queuing system is
defined by the probability of delay.
• Other factors include
– Average delay on all calls
– Length of the queue
Dimensioning and efficiency
• Efficiency is rate of utilization of resources.
• 100 % efficiency will be achieved when all the
trunks are busy with calls for all the times.
• Practically achieving 100%, will mean that
there will be a lot of lost calls, hence
decreasing the grade of service.
• Optimize dimensioning will mean a trade off
between grade of service and efficiency.
Alternate routing
• One way to achieve higher efficiency is to use
alternate routing.
• Lost calls are routed through routes other then
the direct route to the destination.
Z

X Direct route Y

• The basic problem is to increase circuit group


efficiency.
Alternate routing through traffic graph
Efficiency versus circuit group size
• Increase in the circuit group (using the higher
number of trunks), increases the efficiency.
• For a grade of service p = 0.01
– 11 trunks will carry 5 Erlang of traffic, trunk to
Erlang ratio is more than 2:1.
– 30 trunks will carry 20 Erlang of traffic, trunk to
Erlang ratio is 3:2.
– 120 trunks will carry 100 Erlang of traffic, trunk to
Erlang ratio is 6:5.
Basis of network configuration

Tandem Exchange

Mesh Configuration Star Configuration


Usually applied for Usually applied for
higher traffic among lower traffic
exchanges among exchanges Double star configuration
Star connections utilizing an
Note: In practice most networks intervening(dominant) exchange
for the connection among them
are compromise between mesh selves
and star configurations
Hierarchical networks
• Hierarchal networks are systematic networks
developed to reduce trunk groups.
• This network gives orders of importance to
the exchanges making up the network.
1A 1B

2A1 2A2 2B1 2B2

3A1 3A2 3A3 3A4 3B1 3B2 3B3 3B4

The rank of the exchanges is based on the sizes of the boxes


Connection through hierarchical
networks
1A 1B

2A1 2A2 2B1 2B2 The case with no


congestion
between 2B1 & 2B2
3A1 3A2 3A3 3A4 3B1 3B2 3B3 3B4

1A 1B
The case with
congestion between
2B1 & 2B2. The
2A1 2A2 2B1 2B2 dotted line is
showing final route.

3A1 3A2 3A3 3A4 3B1 3B2 3B3 3B4

The route for the overflow traffic through the highest exchange level is the Final route.
The trend away from a hierarchical
structures
• There are trends that are pushing away the
hierarchal networks, but they will remain in
the system for foreseeable future.
• The reasons are
– Satellite communications allowed direct routes
between continents.
– Optic fibres offered higher performance and
higher bandwidths over long distances.
– Optimum routing through signalling system 7.
Routing method
• There are generally three routing methods between intermediate nodes
(switches).
• Right through routing
– The originating exchange determines the route of the route from source to
destination.
– Addition of a new exchange in the network will require the updated at all
exchanges.
• Own exchange routing
– This routing allows changes in routing as call proceeds to its destination.
– Minimal switch modification is required for the addition of new exchange.
– It is suited when alternate switching is mostly encountered.
• Computer controlled routing
– The switching is computer based.
– The route is mapped in memory based on network details.
Quality of service
• The most common measure of quality of service
is customer satisfaction.
• It is usually measured by how well the customer
can hear the calling party
• Other factors include
– Delay before receiving dialling tone
– Post dial delay
– Correctness of billing
– Reasonable cost of the customer services.
– Responsiveness to servicing requests.

You might also like