Comparison Of Meta-Heuristic Algorithms For Mobile Radio Path Loss Model Optimization
The main requirement in the design of the cellular networks is the determination of path losses. The accuracy of the path loss estimation is enhanced by optimizing the path loss model so that the tuned model effectively reflects the real time propagation medium. In this work, the path loss estimation of the semi empirical Walfisch Ikegami model is improved by optimizing the parameters using major Meta heuristic algorithms. The optimizations are performed with Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and the recently proposed Grey Wolf Optimization (GWO) algorithm. The model optimized path losses with the techniques are compared with mobile radio path losses obtained from real time measurements. The performances are evaluated in terms of, error statistics and accuracy of path loss estimation. It is inferred from the results, that Walfisch Ikegami model optimized by GWO gives a least MSE (0.0505), RMSE (0.2248) and a highest path loss estimation accuracy (98.17% ) followed by optimization with Particle Swarm and Genetic algorithms. The work presents an insight to the application of Meta heuristic algorithms to tune the semi empirical mobile radio path loss model and the highlight is the implementation of Grey Wolf Optimization to efficiently estimate the path loss.
Keywords - Genetic Algorithm, Grey Wolf Optimization, Path Loss, Particle Swarm Optimization, Walfisch Ikegami model