Studying the reporting cells planning with the non-dominated Sorting Genetic Algorithm II

Abstract

This manuscript addresses a vital task in any Public Land Mobile Network, the mobile location management. This management task is tackled following the Reporting Cells strategy. Basically, the Reporting Cells planning consists in selecting a subset of network cells as Reporting Cells with the aim of controlling the subscribers’ movement and minimizing the signaling traffic. In previous works, the Reporting Cells Planning Problem was optimized by using single-objective metaheuristics, in which the two objective functions were linearly combined. This technique simplifies the optimization problem but has got several drawbacks. In this work, with the aim of avoiding such drawbacks, we have adapted a well-known multiobjective metaheuristic: the Nondominated Sorting Genetic Algorithm II (NSGAII). Furthermore, a multiobjective approach obtains a wide range of solutions (each one related to a specific trade-off between objectives), and hence, it gives the possibility of selecting the solution that best adjusts to the real state of the signaling network. The quality of our proposal is checked by means of an experimental study, where we demonstrate that our version of NSGAII outperforms other algorithms published in the literature.