Location area management for mobile networks with evolutionary algorithms

Abstract

Location management is a very important problem in mobile networks. In general, registration and paging costs are associated with tracking the current location of a mobile user. Considering the whole network as a single location area (LA) maximizes the paging cost and minimizes the registration cost. On the other hand considering each cell as a separate LA maximizes the registration cost and minimizes the paging cost. Partitioning the whole network into location areas and assigning base stations to these location areas can minimize the total cost of registration and paging. In this work, three evolutionary methods for optimizing the tracking cost of a mobile user by finding an optimal network structure are given and their results are compared. Genetic Algorithms, Multi-Objective Genetic Algorithms and Memetic Algorithms are used to partition a given network into optimal location areas. Finding optimal network structure is known to be NP-Complete. Evolutionary algorithms are suitable for optimizations when normal search algorithms are inefficient. This work gives detailed explanation of implementation details for each algorithm and a comparative study about the performances of algorithms on this particular problem is given.