A new measure of survivability for a cellular network

Survivability is one of many words which were coined as a measure of degree to which a system is able to provide critical services. Many definitions have been proposed but the precise definition of survivability is still being debated. Especially the available measures of survivability for cellular...

Full description

Saved in:
Bibliographic Details
Published in2008 Fourth International Conference on Wireless Communication and Sensor Networks pp. 201 - 205
Main Authors Purohit, N., Tokekar, S.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2008
Subjects
Online AccessGet full text
ISBN1424433274
9781424433278
DOI10.1109/WCSN.2008.4772711

Cover

Loading…
More Information
Summary:Survivability is one of many words which were coined as a measure of degree to which a system is able to provide critical services. Many definitions have been proposed but the precise definition of survivability is still being debated. Especially the available measures of survivability for cellular networks do not cover many inherent features simultaneously. Many measures has been used to describe survivability of cellular systems such as the number of functioning units, the maximum traffic capacity, blocking probability and the service restoration time, etc. but a compact and effective measure of cellular network survivability is still required. In this work a new measure, survivability index (SI) has been introduced to fulfill the above requirement. It takes into account a few important features of cellular network viz. dropping probability, blocking probability, availability and voice quality, in terms of carrier to cochannel interference ratio (C/I). A weighted sum form is proposed and various desirable features of cellular network are mapped into constraint expressions. Then weights {w(k)} are evaluated using desired values of above features within the above constraints such that SI is maximized, which has been designated as SIref. The genetic algorithm (GA) has been used to get the optimum value of w(k) up to four place of decimal. QoS provided by the cellular network is proposed to be measured as % of SI provided by it above or below the SIref. An example cellular network is considered for discussions.
ISBN:1424433274
9781424433278
DOI:10.1109/WCSN.2008.4772711