ML estimation using low complexity metaheuristic algorithms

Metaheuristic algorithms are characterized by high computational complexity due to the large number of cost function evaluations. Joint maximum likelihood (ML) estimation of multiple parameters demands computationally-efficient methods. In this paper, we propose a simple and elegant modification for metaheuristic algorithms, which can drastically bring down their computational cost, while improving estimation performance. The method is developed for joint ML problems, with cost functions having asymptotic separability. The proposed method is successfully applied to some new heuristic algorithms. The advantages of the method are demonstrated through application to a contemporary signal processing problem. The modified metaheuristic algorithms are found to have lesser computational complexity and better mean square error (MSE) performance. The advantages of the method are demonstrated through extensive computer simulation studies.