Adaptive iteration localization algorithm based on RSSI in wireless sensor networks

Based on the maximum likelihood method, one novel adaptive iterative localization algorithm is proposed based on the steepest gradient descent. The algorithm regards the cost function as the target function, within the range of the gradient error, the target position can be localized. To improve the convergence speed and the localization accuracy of the algorithm, a searching algorithm of variable step based on sigmoid function is present. To demonstrate the efficiency of the proposed scheme, we carry out large numbers of experiments to learn the performance trend with various network settings. With all the simulations on localization criteria (localization accuracy and localization coverage), we make the comparison on the localization energy. Based on the simulations, the proposed algorithm has certain practical significance to meet the requirement of localization accuracy.