Collaborative diffusive source localization in wireless sensor networks

We propose a collaborative, energy efficient method for diffusive source localization in wireless sensor networks. The algorithm is based on distributed and iterative maximum-likelihood (ML) estimation, which is very sensitive to initialization. As a part of the proposed method we present an approach for obtaining a "good enough" initial value for the ML recursion based on infinite time approximation and semidefinite programming. We also present an approach for determining the sensor node that initiates the estimation process. To improve the convergence rate of the algorithm, we consider the case where selected nodes collaborate with their neighbors. Simulation results are used to characterize the performance and energy efficiency of the algorithm. We also illustrate estimation accuracy/energy consumption trade-off by varying the communication radius of sensor nodes.