Comparative analysis of optimization techniques for optimizing the radio network parameters of next generation wireless mobile communication

One of the primary aims of Optimization is to hunt the best entity from the available set of choice without explicitly assessing them. A new design, efficient outcome and cost effective solution is the outcome of optimization. Radio network planning is the basic seed required for configuring the base station parameters to attain the desired Quality of Service. The tuning of the radio network parameters like power control parameter, radius, tilt, azimuth orientation, cost etc. is a composite task, as majority of parameters are interdependent on each other. Optimization of these radio network parameters is the solution for the effective deployment of the network. Conventional optimizing algorithms like non-linear programming, Steepest Ascent, Golden Search, Newton Raphson, Quadratic programming etc are local search techniques which may stuck at local optimum. These techniques do not work for multi-modal functions and leads to degraded outcome if input parameters are misconfigured. To overcome this condition, modern optimization techniques like Genetic Algorithm, Particle Swarm Optimization, Simulated Annealing and Taguchi's Method of Optimization have been discussed in this paper. These meta-heuristic search techniques have been compared based on the exploration of search space, performance characteristics and computational intricacy. It is shown that Taguchi's method of Optimization has a comparable performance in terms of less iterations and control outcome.