A Mixed-Norm Approach Using Simulated Annealing with Changeable Neighborhood for Mobile Location Estimation

In this paper, the mobile location estimation problem in urban areas is considered. The time of arrival (TOA) technique is employed to obtain distance measurements. The position of the mobile station (MS) under consideration is estimated according to the position coordinates of the base stations (BSs) and the corresponding distance measurements. The Manhattan and euclidean norms are used to deal with the line-of-sight (LOS) and non-line-of-sight (NLOS) radio propagation. A mixed-norm approach using simulated annealing with changeable neighborhood (SA-CN) is proposed. The position of the MS is located by using the SA-CN algorithm to search for good solutions of a mixed-norm optimization problem. During the search process, the SA algorithm simulates the annealing of solids, whereas the CN method changes the search step size. Simulations show that the concept of SA-CN outperforms the original SA method with fixed neighborhood. The proposed approach can be applied to the urban area with various shapes of obstructions and can work without the knowledge of noise statistics.