ASFALT: A Simple Fault-Tolerant Signature-based Localization Technique for Emergency Sensor Networks

Abstract

We consider the problem of robust node deployment and fault-tolerant localization in wireless sensor networks for emergency and first response applications. Signature-based localization algorithms are a popular choice for use in such applications due to the non-uniform nature of the sensor node deployment. But, random destruction/disablement of sensor nodes in such networks adversely affects the deployment strategy as well as the accuracy of the corresponding signature-based localization algorithm. In this paper, we first model the phenomenon of sensor node destruction as a non-homogeneous Poisson process and derive a robust and efficient strategy for sensor node deployment based on this model. Next, we outline a protocol, called Group Selection Protocol, that complements current signature-based algorithms by reducing localization errors even when some nodes in a group are destroyed. Finally, we propose a novel yet simple localization technique, ASFALT, that improves the efficiency of the localization process by combining the simplicity of range-based schemes with the robustness of signature-based ones. Simulation experiments are conducted to verify the performance of the proposed algorithms.