Numerical Estimation of the Impact of Interferences on the Localization Problem in Sensor Networks

Abstract

Abstract.In this paper we numerically analyze the impact of interferences in wireless networks. More particularly, the impact on the probability of success of a localization algorithm. This problem is particularly relevant in the context of sensor networks. Actually, our numerical experiments provide information about this algorithm, even when we do not consider interferences. Indeed, in this case, sharp threshold is observed on the probability of success of the algorithm. Moreover, our numerical computations show that the main harmful interferences are the ones occurring between sensors which get localized at the same time and send in turn their own location. This is demonstrated by varying the time span of the random waiting time before the emissions. We then observe that the longer the waiting time the closer the curves are to the ones obtained without considering interferences. Hence, this proves to be an efficient way of reducing the impact of interferences on the localization algorithm. Moreover, our numerical experiments demonstrate that among the sectors of disk with same area, the one with the smaller radius of emission and larger angle of emission is the more appropriate to the localization algorithm.