Surveillance by means of a random sensor network: a heterogeneous sensor approach

Abstract

A distributed approach to the surveillance is presented. A clustering architecture is modelled and the behavior of the corresponding heterogeneous random network with self-organizing capability is investigated. Two types of sensors, simple and complex, spread out over the surveillance area. Simple sensors can only compute binary information (yes or no detection); complex sensors, instead, are able to form a target track. A two-way efficient local communication among sensors is hypothesized. From it, a global coherent behaviour of the network emerges, and the resulting network is able to track the moving objects. Benefits and drawbacks of our solution are analysed by means of Monte Carlo simulations.