Sensor localization in NLOS environments with anchor uncertainty and unknown clock parameters

Abstract

In this paper, joint sensor localization and synchronization in non-cooperative wireless sensor networks (WSNs) using time-of-arrival (TOA) measurements is studied. In addition to zero-mean errors in TOA measurements we consider other sources of error such as non-line-of-sight (NLOS) propagation and anchor uncertainty to make our technique more useful in practice, where the presence of these errors is inevitable. The proposed technique is based on semi-definite programming (SDP) relaxation which can be solved in polynomial time and guarantees convergence to the global minimum. It is shown that the optimal accuracy is obtained by discarding the NLOS measurements and applying the maximum likelihood (ML) technique to jointly estimate the positions of anchor nodes, and the position and clock parameters of the sensor node. The results show that the proposed SDP technique, which does not require prior identification of NLOS links, is robust against NLOS errors and its performance is close to that of optimal accuracy.