Expectation maximization-based passive localization relying on asynchronous receivers:: Centralized versus distributed implementations

This paper considers a passive localization scenario relying on a single transmitter, several receivers and multiple moving targets to be located. The so-called “passive” targets equipped with RFID reflectors are capable of reflecting the signals from the transmitter to the receivers. Existing approachesassume that the transmitter and receivers are synchronous or quasi-synchronous, which is not always realistic in practical scenarios. Hence, an asynchronous wireless network is considered, where different clock offsets are assumed at different receivers. We propose a centralized expectation maximization-based passive localization method for asynchronous receivers(EMpLaR) by treating the clock offsets as hidden variables. Thereby, the proposed algorithm makes use of Taylor expansions to arrive at a closed-form maximization. Furthermore, to improve the robustness to link failures and to reduce the energy consumption, we propose a distributed localization approachbased on average consensus formulation to locate the target at each receiver. By applying a quadratic polynomial approximation of the function on which consensus has to be reached, both the computational complexity and the communications overhead are significantly reduced. The Cram ́ r-Rao bound of the target location is derived as a benchmark of our proposed algorithms. Our simulation results show that the proposed centralized and distributed EMpLaR algorithms match the Cram ́ r-Rao bound and significantly improve the localization performance compared to the conventional methods