Title: Performance assessment of biogeography-based multi-objective algorithm for frequency assignment problem

Authors: Asma Daoudi; Karima Benatchba; Malika Bessedik; Leila Hamdad

Addresses: Laboratoire des Méthodes de Conception de Systèmes (LMCS), Ecole nationale Supérieure d'Informatique (ESI), BP 68M 16309, Oued-Smar, Algiers, Algeria ' Laboratoire des Méthodes de Conception de Systèmes (LMCS), Ecole nationale Supérieure d'Informatique (ESI), BP 68M 16309, Oued-Smar, Algiers, Algeria ' Laboratoire des Méthodes de Conception de Systèmes (LMCS), Ecole nationale Supérieure d'Informatique (ESI), BP 68M 16309, Oued-Smar, Algiers, Algeria ' Laboratoire de la Communication dans les Systèmes Informatiques (LCSI), Ecole nationale Supérieure d'Informatique (ESI), BP 68M 16309, Oued-Smar, Algiers, Algeria

Abstract: Solving frequency assignment problem (FAP) with multiple objectives is NP-hard. The network has many constraints to be respected when assigning frequencies. In this paper, we consider two objectives: reduce the number of frequencies and the interference caused by the violation of the network constraints. We use a new model based on a set-T-colouring graph which takes in consideration all constraints related to cellular network. We propose to use an evolutionary metaheuristic, biogeography-based multi-objective optimisation algorithm (BBMO), to solve the considered problem. As it is a bi-objective approach, our results are represented by a set of non-dominated solutions based on the definition of Pareto optimum front. A framework to study and measure the quality of the obtained Pareto fronts based on six performance indicators (ONVGR, HV, Spread, GS, GD, IGD) has been implemented. The obtained results show the efficiency of the BBMO on FAP based on a well known COST259 dataset. Moreover, a statistical analysis compares our approach BBMO-FAP to NSGA-II and identifies the best-performing method.

Keywords: frequency assignment problem; FAP; multi-objective optimisation; performance indicators; biogeography-based multi-objective optimisation; BBMO; Kruscal-Wallis statistical test.

DOI: 10.1504/IJBIC.2021.119981

International Journal of Bio-Inspired Computation, 2021 Vol.18 No.4, pp.199 - 209

Received: 22 Aug 2020
Accepted: 18 Feb 2021
Published online: 04 Jan 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article