Cooperative localization using efficient Kalman filtering for mobile wireless sensor networks

We consider the problem of cooperative localization in mobile wireless sensor networks (WSNs). To be able to continuously localize the mobile network, we propose to exploit the knowledge of the location of the anchor nodes to linearize the nonlinear distance measurements with respect to the location of the unknown nodes. Based on this linearized measurement model, we estimate the location of the unknown nodes using a Kalman filter (KF) instead of a suboptimal extended KF (EKF) and try to estimate the corresponding unknown measurement noise covariance matrix using an iterative process. The simulation results illustrate that the proposed algorithm (only with a few iterations) attains the posterior Cramer-Rao bound (PCRB) of mobile location estimation and clearly outperforms related anchorless and anchored mobile localization algorithms.