Quantitative Comparison of Survivability Models for Wireless Ad Hoc Networks

Network survivability is defined as the ability of a network keeping connected under failures and/or attacks. In this paper, we propose two stochastic models, binomial model and negative binomial model, to quantify the network survivability and compare them with the existing Poisson model. We give the mathematical formulae of approximate network survivability and use them to carry out the sensitivity analysis on model parameters. Through a numerical example it is shown that the network survivability can change drastically when the number of nodes is relatively small under a severe attack mode which is called the Black hole attack.