Environment-aware localization for wireless sensor networks using magnetic induction

The Magnetic Induction (MI) communication techniques can enable or enhance many wireless applications in the complex environments where line-of-sight (LOS) links do not exist. The critical position information of each wireless device can also be derived by the same MI systems without additional hardware or infrastructure. However, while MI signals can penetrate most of the transmission media without significant attenuation or phase shifting, the obstacles with high conductivity can still influence the signal propagation, which incurs additional positioning errors in the MI-based localization. To address such challenge, this paper develops an environment-aware MI-based localization technique for wireless sensor networks in complex environments with significant amount of high-conductive obstructions. First, the system architecture of the MI-based environment-aware localization and the MI channel is introduced. The environment-aware capability is realized by analyzing the unique MI response information gathered by each MI-based sensor node. Then, a joint device localization and environment sensing algorithm is developed to estimated the position of each device in the network as well as the distribution of the high-conductive objects. Finally, the performance of the proposed solution is validated through both computer simulations and real-world experiments.