Product category: Design and Development Software
News Release from: Adept Scientific | Subject: Maple
Edited by the Electronicstalk Editorial Team on 27 September 2002

Software models telecomms capacity

Nortel has used Maple to develop a generic capacity model for its radio products, specifically to model the effect of the introduction of higher bandwidth services to traditional PSTN systems

Nortel has used Maple to develop a generic capacity model for its radio products, specifically designed to model the effect of the introduction of higher bandwidth services to traditional PSTN radio systems. Nortel is one of the worlds leading telecomms products manufacturers and has its principal European research and development facility at Harlow in England where over 1200 research engineers work on the design and development of telecomms systems.

Boris Sedacca talks to Ben Freeman at Nortel, about the development of the model and the choice of Maple as a platform.

Currently there are many different systems, in the world, which offer PSTN services via radio systems.

Some are fixed wireless access, which offer PSTN services to homes via a radio link; other are mobile, such as GSM, the European digital mobile radio standard.

The vast majority of the established systems were originally designed for voice only (PSTN) and at the time of their design the demand for higher bandwidth services, such as Internet access, was not envisaged.

One such standard for higher-datarate services is ISDN, and demand for such a service has driven designers of fixed wireless access and mobile personal communications systems to look for means of offering higher bandwidth services in addition to PSTN on existing radio systems.

ISDN was designed by the United Nations with the ultimate goal of allowing any communications equipment to plug into any phone jack anywhere in the world.

While this goal is still many years from realisation, ISDN service is proving to be an extremely reliable and fast means of transmitting digital voice and data over existing copper wires, fibre optics, radio and satellite channels.

Many radio systems, such as GSM, operate on a frequency and time slot air interface structure.

Any one user making a call is assigned a timeslot and a frequency, which may be fixed for the duration of the call.

The radio interface protocols were designed to have a flexible timeslot structure to be able to cope with the provision of different types of service (eg voice, data etc) although, as the majority of calls were expected to be PSTN, the single timeslot capacity was optimised for voice traffic.

Data traffic therefore needs to occupy multiple timeslots and/or frequencies.

For example, if each individual time slot has enough capacity for a PSTN voice call, say 32Kbit/s, then a higher bandwidth services such as ISDN would require multiple time slots to meet its capacity demand, eg 64Kbit/s for ISDN basic rate interface.

Nortel needed an improved capacity model which could be used to analyse the effects that mixtures of traffic types would have on their systems, especially how higher bandwidth traffic would affect the service offered to existing PSTN users.

The study of traffic modelling is a well-established discipline in telephone engineering.

The basic factors involved are the nature of the demand from users on the system - ie the call attempt rate, call holding time or duration, and various system parameters such as the number of channels available in the system and minimum acceptable grade of service (GOS).

GOS is often defined as the probability that a user attempting a call will be blocked or delayed due to the system running at full capacity.

It is commonly expressed as the fraction of calls failing to receive immediate service (blocked calls), or the fraction of calls waiting longer than a given service time (delayed calls).

As Nortel designs and manufactures a wide range of radio systems, involving many different standards and protocols, a generic model was required; one which could be easily adapted to different systems offering different services.

This generic structure would allow the blocking, or GOS, to be broken down by type of service request - eg PSTN voice, PSTN fax, ISDN data etc - as well as being adapted to take account of specific channel selection algorithms implemented in the radio hardware as well as the different call demand statistics for each of the service types.

Ben Freeman of Nortel has developed just such a universal analytic blocking model that takes advantage of Maple's symbolic computational features.

The model builds symbolic multidimensional Markov finite state models, which are used to model the traffic on the radio system.

As an example of some of the more subtle complications of offering higher bandwidth services to a primarily PSTN only system, take the case of a six frequency by 10 time slot system, which can maintain up to 60 simultaneous single time slot calls.

In this example each time slot has a data capacity of 32Kbit/s, which is adequate for digital PSTN voice.

An ISDN call, requiring 64Kbit/s, needs two time slots to obtain this data capacity.

However, the choice of timeslots has restrictions because of the design of the radio system.

Cost constraints on mobile-type equipment prevent either receiving or transmitting on multiple frequencies at the same time, so for example, if the first timeslot were (T2,F4) then the second timeslot could not be (T2,F3).

"If you need two time slots for ISDN you cannot just pick any two free timeslots because of the system constraints.

If somebody wants to make an ISDN connection, it is not sufficient just to have two time-slots free.

They have to be at an appropriate separation", says Freeman.

The type of blocking which occurs when there is free capacity, but not in the right place, is called 'soft blocking' to distinguish it from 'hard blocking' when there is no free capacity and the system is working at full capacity.

Soft blocking can only affect multiple-time-slot calls and the probability of its occurrence gets progressively higher as the system nears capacity, ie as the airside capacity 'fills up'.

Designers of a radio system need to accurately quantify this affect and ensure that the radio system intelligently allocates capacity to minimise 'soft blocking', which will be experienced by the user as a refused call attempt by their ISDN radio modem.

Another complication in traffic modelling is that different types of users have different call profiles so any model must be adaptable to allow the population to be defined in terms of user profiles, each with their own call statistics.

For example, a home Internet user will typically make longer calls than a home phone user.

"We wanted a model that would allow us to look at the performance of these systems by specifying the total capacity in a cell and running a model which would indicate what the GOS is likely to be", says Freeman, "The model we have developed allows the user to plan capacity before installing any equipment and hence to optimise their revenue for their system, whilst ensuring that all users experience a good GOS".

Traditional methods of calculating GOS relied on Monte-Carlo simulation techniques, but the problem with this technique on traffic modelling problems is that over 100,000 iterations would be needed to approach realistic results, because of the burst nature of the traffic blocking.

Therefore, Nortel's model was designed to be analytic, providing accurate results within a few seconds running on modern PC platforms.

Another advantage of analytic models is that the computation time is the same, regardless of the probabilities of the events being modelled.

"With an analytical model, you need to put in a little more effort into understanding the problem, but once you have worked out your method, it is easier to compute", adds Freeman.

One of the key advantages of using symbolic mathematics is that the programmer does not need to specify the solution exactly in terms of all parameters but by a set of equations, each relating one part of the system to another.

There is no rigid structure for specifying the problem.

There are also advantages of reduced coding time over lower-level languages like C or Modula-2, such as built in support for complex variable types like sets, tables and matrices.

Debugging and testing also takes less time, with the effort applied to checking algorithms rather than the code syntax.

Ben Freeman argues that with Maple, any mathematical concept can be tackled directly without having to write tools to deal with them.

The price of an interpretive system's slower run times, compared with a compiled C program, is more than outweighed by the savings in development time.

Maple software is supplied and supported by Adept Scientific.

Request a free brochure from Adept Scientific....

• Contact details and Company Profile for Adept Scientific
• Other news from Adept Scientific
• Email this article to a colleague
• Register for the free Electronicstalk email newsletter
• Electronicstalk Home Page

Search the Pro-Talk network of sites

Put your news on Electronicstalk  |   Advertise  |   Get the free Electronicstalk newsletter
Electronicstalk Home  |   About Electronicstalk  |   Copyright © 2000-2007 Pro-Talk Ltd, UK