On the resiliency of wireless sensor networks with circulation graphs

Abstract

One of the challenges in military wireless sensor networks is the determination of an information collection infrastructure that minimizes battery power consumption while being highly resilient against sensor and link failures. In our previous work we have proposed a heuristic for constructing an information flow graph in wireless sensor networks based on the mammalian circulatory system, with the goal of minimizing the energy consumption. In this paper we focus mainly on the resilience benefits that can be achieved when constructing such information flow graphs. We analyze the resilience of circulatory graphs constructed on top of regular as well as random topologies. We assume two modes of failure, random and targeted attacks, and we compare the resilience of the circulatory graphs against tree graphs.