Genetic algorithms and the location area partitioning problem in cellular networks

Abstract

As the subscriber population grows and the network capabilities are enhanced, mobility management and resource management become increasingly critical in (micro-) cellular networks. Moreover, coverage areas are increasingly enlarged, possibly requiring the adoption of partitions to facilitate management activities. Location areas constitute an important strategy of location management, used to reduce signaling traffic caused by location updating and paging messages in cellular networks. Due to the very large state spaces to be searched, the determination of optimal LA's represents a NP-hard combinatorial optimization problem. In this paper, genetic algorithms are used in order to group cells in an efficient way, while preserving bandwidth. Elitism, linear normalization of chromosoma and edge-based crossover are used to speed up the convergence time, allowing near-optimal solutions to be obtained in an acceptable computation time.