Solving the Location Areas Scheme in Realistic Networks by Using a Multi-objective Algorithm

The optimization of the management tasks in current mobile networks is an interesting research field due to the exponential increase in the number of mobile subscribers. In this paper, we study two of the most important management tasks of the Public Land Mobile Networks: the location update and the paging, since these two procedures are used by the mobile network to locate and track the Mobile Stations. There are several strategies to manage the location update and the paging, but we focus on the Location Areas scheme with a two-cycle sequential paging, a strategy widely applied in current mobile networks. This scheme can be formulated as a multi-objective optimization problem with two objective functions: minimize the number of location updates and minimize the number of paging messages. In previous works, this multi-objective problem was solved with single-objective optimization algorithms by means of the linear aggregation of the objective functions. In order to avoid the drawbacks related to the linear aggregation, we propose an adaptation of the Non-dominated Sorting Genetic Algorithm II to solve the Location Areas Planning Problem. Furthermore, with the aim of studying a realistic mobile network, we apply our algorithm to a scenario located in the San Francisco Bay (USA). Results show that our algorithm outperforms the algorithms proposed by other authors, as well as the advantages of a multi-objective approach.