An ACO-GA Optimization Scheme for Route Discovery in Cellular Networks

Abstract

The recent advancements in mobile communication has triggered mankind in every part of the world to use wireless devices viz., cell phones, laptops, PDAs, etc., To provide excellent services to the mobile users, every service provider expands his network to cover a wide coverage area. To reduce the establishment cost of service providers, infrastructure sharing among service providers is becoming popular. This requires existing connectivity/ topology in a Cellular Network (CN) to be explored and discovered. In this paper, a novel method for topology discovery of CN using Ant Colony Optimization (ACO) is proposed. In this approach, the Base Station(BS)s simulate the way ants forage for food to find out the routes to other BSs. The route discovered by each ant is associated with trail (pheromone) strength which in turn decides whether the route is the best or not. ACO applied to CN gives all the existing routes between the various BSs in a CN. Genetic Algorithm (GA) based optimization is further applied to get the optimal path satisfying multiple constraints viz., number of hops, Poisson traffic distribution, buffer capacity, link delay, queuing delay, and residual bandwidth from the set of paths given by ACO. Our simulation results show that different network models viz., Random model, unidirectional ring, bidirectional ring, star, and tree are explored faster using ACO and our scheme using ACO-GA outperforms the scheme without GA with respect to Call Service Rate and Call Dropping Rate.