A Multiobjective Evolutionary Tracking Indoor Positioning Algorithm for Smart Space

This paper is mainly devoted to identify an evolutionary approach based on search strategy, namely multiobjective evolutionary algorithm for indoor positioning (MEIP). Each subproblem is optimized by information from its several neighboring subproblems, which makes MEIP lower computational complexity at each generation and be capable of determining the user position with high accuracy. Experimental results have demonstrated that MEIP is able to achieve accuracy significantly better than the current WLAN location determination systems. It has been shown that MEIP using objective normalization can deal with disparately scaled objectives, and MEIP with an advanced decomposition method can generate a set of very evenly distributed solutions for n-objective test instances. The ability of MEIP to track a large number of users and to be used with large areas, the scalability and sensitivity of MEIP have also been experimentally investigated in this paper.