Abstract
This thesis introduces a decentralized positioning and tracking method for wireless sensor networks by utilizing the received signal strength (RSS) measurements to estimate the unknown location of an emitted source. We formulate the problem of positioning as a nonlinear least-squares optimization problem and our new approach is based on the decentralized incremental subgradient (IG) optimization methods to estimate the location of a source iteratively. At first, we add an extra step to determine whether the IG algorithm is executed to estimate the location of the source or not. By this way, a more accurate estimation can be achieved by using this modified IG algorithm to estimate the location of a stationary source. Furthermore, in order to apply the modified IG algorithm to track a moving source in practice, we present a method to estimate the velocity of the moving source and then exploit this estimation of velocity to make the convergence behavior of the modified IG algorithm associate with this estimation. Computer simulation results demonstrate that our proposed decentralized positioning and tracking method has better convergence accuracy than other decentralized positioning and tracking methods.