Optimal deployment of UAV'sfor communication using metaheuristic methods

Unmanned air vehicles (UAV) have brought a different approach in the engineering problems. Using them as a base stations in broadband wireless access is one of these approaches. Calculating the optimum positions of the UAVs is the important factor in the performance of the application. Because of the nonlinear structure of the problem and huge number of the optimized parameters, metaheuristic searching methods are used to solve the problem in a resonable of time. In this work, five methods for UAV positioning are compared. These are Particle Swarm Optimization(PSO), Simulated Annealing (SA), Tabu Search (TS), Biogeography Based Optimization (BBO) and Taguchi Method (TM). From these methods, PSO, SA, TS and BBO are typical metaheuristic algorithms that are based on random search. On the other hand, the last method, Taguchi method makes a search by systematically dividing the search space and shrinking them at each stage, using predefined experiment sets. A convenient property of these methods is that they can be terminated at any stage in order to obtain a (suboptimal) solution in a shorter amount of time. Simulation results reveal that PSO performs the best among the compared methods. Taguchi Method also proves to be convenient, as it reaches to a reasonable solution in a very short amount of time.