Elsevier

Applied Soft Computing

Volume 11, Issue 1, January 2011, Pages 410-427
Applied Soft Computing

Differential evolution for solving the mobile location management

https://doi.org/10.1016/j.asoc.2009.11.031Get rights and content

Abstract

In this work we present two new approaches to solve the location management problem, respectively, based on the location areas and the reporting cells strategies. The location management problem corresponds to the management of the network configuration with the objective of minimizing the costs involved. We use the differential evolution algorithm to find the best configuration for the location areas and the reporting cells strategies, which principally considers the location update and paging costs. With this work we want to define the best values to the differential evolution configuration, using test networks and also realistic networks, as well as compare our results with the ones obtained by other authors. These two new approaches applied to this problem have given us very good results, when compared with those obtained by other authors.

Introduction

The use of mobile networks is growing every day and being applied to the most of newly and renovated applications for data transfer, voice and fax services among many other mobile services. Because of this, communication networks [1] must support a big number of users and their respective applications maintaining a good response without losing quality and availability. With the goal that mobile networks keep this quality it is necessary to consider the mobility management when making design of the network infrastructure.

Mobility management is a very important point because it includes the process of hand off management and location management. The process of hand off management enables the mobile network to locate roaming mobile terminals. The process of location management enables the mobile network to find the current location of a mobile terminal in order to make or receive calls, from any location and at any time of the day.

We are principally concerned about the location management because their requests normally occur when a mobile terminal changes its location or when the quality of the received signal becomes deteriorated, so this process becomes even more important for the current and future generations of mobile networks. One of the major objectives of location management is to minimize the involved costs associated to the user movements and their tracing, and this will be also our major goal.

There exist several strategies of location management and we will apply the location area and reporting cell schemes, which are two of the more common ones.

This article proposes two new approaches that use a differential evolution (DE) based algorithm to solve, respectively, the location areas and reporting cells problems. The main goal of each of these problems is to optimize the configuration, for mobile networks, that minimizes the involved costs.

With the objective of testing our approaches we have used test networks and also realistic networks as SUMATRA [2]. The application of the algorithm is described in detail for both approaches and the parameters were studied intensively to define the most adequate values.

In conclusion, our contributions have the objective of introducing the DE algorithm for solving these important location management problems, outperforming the results of the existing works.

The article is organized as follows. In the next section it is explained the location management and in more detail the location areas and reporting cells strategies. In Section 3, the DE based algorithm is described, as well as its parameters and different possible schemes. Section 4 includes the implementation details for both approaches. In Section 5, the experimental results are presented, analyzed and compared with the results of other authors. Finally, Section 6 includes the conclusions and future lines of work.

Section snippets

Location management problem

In cellular network systems it is very important to keep track the location of the users, even when they move around without making or receiving calls, so as to consequently, be able to route calls to the users regardless of their location.

Location management involves two elementary operations: location update and location inquiry (or terminal paging). The location update corresponds to the notification of current location, performed by mobile terminals when they change their location in the

Differential evolution algorithm

The differential evolution (DE) is a population-based algorithm, created by Price and Storn [14], whose main objective is functions optimization. It is one strategy based on evolutionary algorithms with some specific characteristics.

The DE algorithm’s main strategy is to generate new individuals by calculating vector differences between other randomly-selected individuals of the population. This algorithm uses four important parameters: population size, mutation, crossover and selection

Implementation details

In this section we intent to explain the considerations that must be taken before the implementation of experiments.

We will divide this section in two major subsections that will include, respectively, the implementation details for the location areas and the reporting cells schemes.

Experimental results and analysis

In this section we expose the different experiments performed, the results obtained and the respective analysis and conclusions taken.

We will divide this section in two major subsections that will include, respectively, the results and analysis for the location areas and the reporting cells schemes.

Conclusions and future work

In this paper we present two approaches based on the differential evolution (DE) algorithm applied to the location management problem with the objective of minimizing the involved costs. Each approach is specified, respectively, for the location areas (LAs) and the reporting cells (RCs) strategies of location management problem.

Considering the LAs based approach, we have shown that it improves the results obtained with other classical location management strategies as always-update and

Acknowledgments

This work was partially funded by the Spanish Ministry of Science and Innovation and FEDER under the contract TIN2008-06491-C04-04 (the MSTAR project). Thanks also to the Polytechnic Institute of Leiria, for the economic support offered to Sónia M. Almeida-Luz to make this research.

Sónia M. Almeida-Luz is assistant professor of Programming and Information Systems in the Dept. of Computer Engineering, School of Technology and Management, Polytechnic Institute of Leiria, Leiria, Portugal. She is currently making investigation to her PhD in evolutionary computing and optimization. Her main research interests are information systems, applications of artificial intelligence, nature inspired optimization methods and evolutionary computing.

References (24)

  • P. Demestichas et al.

    Computationally efficient algorithms for location area planning in future cellular systems

    Computer Communications

    (2000)
  • K. Pahlavan, A. Levesque, Wireless Information Networks, John Wiley & Sons Inc., Wiley-Interscience, New York,...
  • SUMATRA, 1998, Stanford University Mobile Activity TRAces (online), InfoLab – Stanford University, Stanford. Available...
  • V. Wong et al.

    Location management for next generation personal communication networks

    IEEE Network

    (2000)
  • P. Gondim

    Genetic algorithms and the location area partitioning problem in cellular networks

  • J. Taheri et al.

    A genetic algorithm for finding optimal location area configurations for mobility management

  • J. Taheri et al.

    A simulated annealing approach for mobile location management

  • J. Taheri et al.

    Clustering techniques for dynamic mobility management

  • R. Subrata et al.

    A comparison of three artificial life techniques for reporting cell planning in mobile computing

  • J. Taheri et al.

    A modified Hopfield network for mobility management, in: Wireless Communications and Mobile Computing, vol. 8, no. 3

    (2008)
  • R. Subrata et al.
  • A. Bar-Noy et al.

    Tracking mobile users in wireless communications networks

    IEEE Transactions on Information Theory

    (1993)
  • Cited by (35)

    • Analysis and comparison of mobility management strategies in public land mobile networks from a multiobjective perspective

      2021, Journal of Network and Computer Applications
      Citation Excerpt :

      In Berrocal-Plaza et al. (2014b), we proposed two multiobjective evolutionary algorithms (NSGAII and SPEA2) to solve the registration areas planning problem with a simultaneous paging in the set of mobile networks used in Taheri and Zomaya (2008, 2007b, 2005, 2007a) and Almeida-Luz et al. (2011). In that work, we showed that both NSGAII and SPEA2 were able to provide very good Pareto sets and, at the same time, outperform the single-objective metaheuristics proposed in Taheri and Zomaya (2008, 2007b, 2005, 2007a) and Almeida-Luz et al. (2011). Furthermore, we realized that NSGAII provided better results than SPEA2 in terms of hypervolume.

    • Bio-inspired algorithms: principles, implementation, and applications to wireless communication

      2020, Nature-Inspired Computation and Swarm Intelligence: Algorithms, Theory and Applications
    • Reporting cell planning-based cellular mobility management using a Binary Artificial Bat algorithm

      2019, Heliyon
      Citation Excerpt :

      The implementation flowchart of the BABA is given in Fig. 3 and the pseudo-code is presented in Table 1. Simulations are carried out for both reference as well as realistic networks of sizes 4 × 4, 6 × 6 and 8 × 8 to gauge the ability of the proposed BABA with respect to the existing ones, such as BDE-PEO (Almeida-Luz et al., 2011; Bettencourt, 2004; Swayamsiddha et al., 2017a,b) and BPSO (Kim et al., 2012). The control parameters used in the BABA, BPSO, and BDE-PEO are outlined in Table 2.

    • A stop-and-start adaptive cellular genetic algorithm for mobility management of GSM-LTE cellular network users

      2018, Expert Systems with Applications
      Citation Excerpt :

      On the other hand, the proposed 2SA-cGA is outperformed in 2 out of 5 instances (networks of sizes 6 × 6 and 8 × 8 cells). On the basis of the results of Tables 2–4, we compare the proposed 2SA-cGA against three of the most efficient state-of-the-art algorithms designed to solve the mobility management problem: the GPSO proposed in Alba et al. (2008), the OI-GA devised in Dahi et al. (2016) and the DE proposed in Almeida-Luz et al. (2011). Table 9 shows the numerical results of this comparison.

    • Optimizing the mobility management task in networks of four world capital cities

      2015, Journal of Network and Computer Applications
      Citation Excerpt :

      In this work, we provide to the research community four novel mobile activity traces generated from real mobile networks. These mobile activity traces avoid the weaknesses of the test networks presented and/or studied in previous works (Subrata and Zomaya, 2003; SUMATRA; Taheri and Zomaya, 2004, 2005a,b, 2007; Almeida-Luz et al., 2008, 2011, 2009a,b; Berrocal-Plaza et al., 2013, 2012a,b), and combine their goodnesses: high number of heterogeneous network cells (Subrata and Zomaya, 2003; SUMATRA; Almeida-Luz et al., 2008, 2011, 2009a; Berrocal-Plaza et al., 2013), and user׳s call and mobility patterns close to those that we can find in current mobile networks (Taheri and Zomaya, 2004, 2005a,b, 2007; Almeida-Luz et al., 2011, 2009b; Berrocal-Plaza et al., 2012a,b). For it, we have blended the works published in Cowling (2004) and Taheri and Zomaya (2009).

    View all citing articles on Scopus

    Sónia M. Almeida-Luz is assistant professor of Programming and Information Systems in the Dept. of Computer Engineering, School of Technology and Management, Polytechnic Institute of Leiria, Leiria, Portugal. She is currently making investigation to her PhD in evolutionary computing and optimization. Her main research interests are information systems, applications of artificial intelligence, nature inspired optimization methods and evolutionary computing.

    Miguel A. Vega-Rodríguez is Professor of Computer Architecture and, at present, Head of the Dept. Technologies of Computers and Communications, University of Extremadura, Spain. He received a PhD degree in Computer Science from the University of Extremadura. Dr. Vega-Rodríguez has authored or co-authored more than 280 publications including journal papers, book chapters and peer-reviewed conference proceedings. Furthermore, he is editor and reviewer of several international ISI journals. Dr. Vega-Rodríguez’s main research interests are parallel and distributed computing, reconfigurable computing, and also evolutionary computing.

    Juan A. Gómez-Pulido is Professor of Computer Architecture in the Dept. Technologies of Computers and Communications, University of Extremadura, Spain. He received a PhD degree in Computer Science from the Complutense University of Madrid in 1993. He has authored or co-authored 25 ISI journals, and many book chapters and peer-reviewed conference proceedings. His current research interest is the application of the high-performance reconfigurable computing to speedup evolutionary algorithms in large optimization problems.

    Juan M. Sánchez-Pérez is Professor of Computer Architecture in the Dept. Technologies of Computers and Communications, University of Extremadura, Spain. He received a PhD degree in Physics from the Complutense University of Madrid in 1976. His research interests are artificial intelligence, logic design and modern computer architectures.

    View full text