Mobile network deployment under electromagnetic pollution control criterion: An evolutionary algorithm approach P. García-Díaz, S. Salcedo-Sanz ⇑ , J.A. Portilla-Figueras, S. Jiménez-Fernández Department of Signal Theory and Communications, Universidad de Alcalá, Spain article info Keywords: Mobile network deployment Network optimization algorithms Electric field minimization Evolutionary algorithms abstract Electromagnetic pollution due to mobile telephony is one of the most concerning problems arising since the spreading of this technology. Different studies have shown the relationship between continuous exposition to electromagnetic fields and different kinds of pathologies. Despite this, the electromagnetic danger for exposition is not taken into account in recent mobile network deployments. In this paper we propose a novel evolutionary algorithm for mobile networks deployment, which takes into account the control of the electromagnetic emission from the base stations as one of the key design parameters. The proposed evolutionary approach is a variable-length algorithm, able to produce solutions with differ- ent number of base stations. We detail the encoding, operators and a repairing procedure applied to obtain good solutions in terms of coverage, cost and electromagnetic pollution. The algorithm has been tested in a real problem of mobile network deployment in Alcalá de Henares, Madrid, Spain, and compare with a greedy (constructive) approach and a meta-heuristic algorithm (Harmony Search), obtaining very good results. Ó 2012 Elsevier Ltd. All rights reserved. 1. Introduction Mobile communications are one of the technologies that have affected the most our way of living in the last few years. Nowadays, it is complicated to find a person in the 14–65 age range that does not have at least one mobile terminal in Europe. Mobile technology penetration in other parts of the world such as norther America or Asia is quite similar (Mao, Tsai, & Chen, 2008). In parallel with this boom of mobile communications, a social concern on the adverse effects of electromagnetic emissions from base stations, antennas and terminals, has grown all around the world (Kundi & Hutter, 2009). There are quite different opinions on this topic, from those that link the relationship between contin- uous exposition to electromagnetic fields to different kinds of pathologies (Augner et al., 2010), to those who deny these effects or claim for more definitive studies before making a complete deci- sion (Challies, 2007). In fact, there is no a definitive conclusion about this topic because a short time has passed since the massive mobile telecommunications introduction. However, and for pre- vention issues, several governments have warned about possible health risks due to mobile phones and electromagnetic fields expo- sition (Barnett, Timotijevic, Shepherd, & Senior, 2007). In spite of this, electromagnetic emission is still not considered as a key factor in mobile networks deployment problems. Instead, companies and researchers have focussed on the development of efficient algorithms for mobile network deployment mainly based on coverage and cost (Akella et al., 2005, Akella, Batta, Sudit, Rog- erson, & Blatt, 2008; Chu, Lin, & Wang, 2011; Glasser, Reith, & Voll- mer, 2005; Gódor & Magyar, 2005; Leung, Ng, Chan, Chu, & Li, 2003, 2010; Mathar & Schmeink, 2001; Muhammad, Neskovic, & Magill, 2005; Qi, Wu, Li, & Shu, 2007). On the other hand, evolu- tionary computation techniques have been successfully applied to different problems of network deployment: problems of WiFi network design (Agustín-Blas, Salcedo-Sanz, Vidales, Urueta, & Portilla-Figueras, 2011), optical networks (Wang, Chen, & Yuan, 2004; Xu, Salcedo-Sanz, & Yao, 2005), UMTS networks (Zola & Bar- celó, 2004) and of course mobile cellular networks (Weicker, Szab- o, Weicker, & Widmayer, 2003; Talbi, Cahon, & Melab, 2007). In this paper, we propose a novel evolutionary algorithm (EA) for a problem of network deployment in mobile cellular networks, that includes the control of electromagnetic field as an important part of the optimization process. The coverage provided by the net- work and the deployment cost are the other factors which are ta- ken into account. The proposed EA is defined as a variable length meta-heuristic, in order to manage solutions with different num- ber of base stations. Different repairing procedures are defined in order to obtain feasible or more effective layouts. We have tested the performance of the proposed approach in a real problem of net- work deployment in Alcalá de Henares, Madrid, Spain, where an 0957-4174/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.eswa.2012.07.050 ⇑ Corresponding author. Address: Department of Signal Theory and Communi- cations, Universidad de Alcalá, 28871 Alcalá de Henares, Madrid, Spain. Tel.: +34 91 885 6731; fax: +34 91 885 6699. E-mail address: sancho.salcedo@uah.es (S. Salcedo-Sanz). Expert Systems with Applications 40 (2013) 365–376 Contents lists available at SciVerse ScienceDirect Expert Systems with Applications journal homepage: www.elsevier.com/locate/eswa