Robust ANNs-Based WSN Localization in the Presence of Anisotropic Signal Attenuation

We propose a novel range-free localization algorithm for wireless sensor networks that is robust against the anisotropic signal attenuation induced by fading, shadowing, interference, etc., present in any wireless channel, and hereby, develop a new distance estimation (DE) approach able to efficiently derive distances' estimates in closed form. Exploiting artificial neural networks, we also develop a power-efficient DE correction mechanism that properly accounts for anisotropic signal attenuation. Simulation results show that the proposed algorithm significantly outperforms most representative range-free localization algorithms, not only in accuracy but also in robustness against anisotropic attenuation.