Assigning cells to switches in cellular mobile networks is an NP-hard problem which, for real size mobile networks, could not be solved using exact methods. In this context, heuristic approaches like memetic algorithms can be used. This paper proposes a memetic algorithm (MA) to solve this problem. The implementation of this algorithms has been subject to extensive tests. The results obtained confirm the efficiency and the effectiveness of MA to provide good solutions for moderate- and large-sized cellular mobile networks, in comparison with tabu search and Merchant and Sengupta's heuristics. This heuristic can be used to solve NP-hard problems, like designing and planning, in the next-generation mobile networks.

A memetic algorithm for assigning cells to switches in cellular mobile networks

Evolutionary design of intelligent systems is gaining much popularity due to its capabilities in handling several real world problems involving optimization, complexity, noisy and non-stationary environment, imprecision, uncertainty and vagueness. This edited volume 'Engineering Evolutionary Intelligent Systems' deals with the theoretical and methodological aspects, as well as various evolutionary algorithm applications to many real world problems originating from science, technology, business or commerce. This volume comprises of 15 chapters including an introductory chapter which covers the fundamental definitions and outlines some important research challenges. Chapters were selected on the basis of fundamental ideas/concepts rather than the thoroughness of techniques deployed.

Engineering Evolutionary Intelligent Systems

The infrastructure efficiency of cellular wireless networks

Cellular genetic algorithms (cGAs) are a kind of genetic algorithm (GA) — population based heuristic — with a structured population so that each individual can only interact with its neighbors. The existence of small overlapped neighborhoods in this decentralized population provides both diversity and opportunities for exploration, while the exploitation of the search space is strengthened inside each neighborhood. This balance between intensification and diversification makes cGAs naturally suitable for solving complex problems. In this chapter, we solve a large benchmark (composed of 160 instances) of the Capacitated Vehicle Routing Problem (CVRP) with a cGA hybridized with a problem customized recombination operation, an advanced mutation operator integrating three mutation methods, and two well-known local search algorithms for routing problems. The studied test-suite contains almost every existing instance for CVRP in the literature. In this work, the best-so-far solution is found (or even improved) in 80% of the tested instances (126 out of 160), and in the other cases (20%, i.e. 34 out of 160) the deviation between our best solution and the best-known one is always very low (under 2.90%). Moreover, 9 new best-known solutions have been found.

A Hybrid Cellular Genetic Algorithm for the Capacitated Vehicle Routing Problem

Multi-service traffic engineering has a strong impact in Mobile Broadband Systems (MBS) revenues, allowing one to obtain merit functions for optimisation purposes, a key aspect in cellular planning A net cost model is presented for the design trade-offs between re-use pattern, K , coverage distance, R , and spectral efficiency, Sef (R) It allows for the determination of the revenue per basic channel, R384, that achieves a given value for the annual profit per kilometre Fast terminal mobility has an important impact in handover failure probability, hence, in system capacity While in the business city centre and other urban scenarios mobility has no significant effect, it strongly affects the supported traffic in main roads Comparing the urban with the roads scenarios, a reduction up to 54% may come as a consequence, from (Sef)TOT = 322 to 152%, for R = 100 m and K = 2; R384 has to be 0005 and 0045 /min, respectively, ie, the prices in the roads scenario have to be around one order of magnitude higher than in the urban one As time goes by, and the use of MBS evolves, the operator will be able to choose different cell coverage distances, in order to support a different number of users, whilst maximising profit In a medium term scenario, while the number of foreseen users is less than 70% of the number of users in the mature phase, cells with R = 200 m will be used

Optimisation of mobile broadband multi-service systems based in economics aspects

This paper proposes a model for configuring cellular networks to study the dynamics of mobility between a single cell and its adjacent cells. It differs from most models considered in the literature by explicitly incorporating the dependency between the handoff rate and the system state. Besides, the handoff rate is also a function of cell size and subscriber mobility. Extensive computational experiments were done to study the impact of various input parameters on specific performance measures. Several observations are made regarding the system performance and as to how they are affected by the complex interaction between subscriber mobility, cell size, number of channels and the mean call initiation rate. The results of these experiments show that the proposed model, where handoff rates are state-dependent, captures additional traffic due to mobility when compared to the traditional method of modeling handoffs using information about the average behavior. Finally, the economic impact of mobility on system configuration decisions is analyzed. Though an approximation, the above work provides interesting insights about the impact of mobility in configuring cellular networks.

/pdf/the-impact-of-mobility-on-cellular-network-configuration-3fotyxjdi4.pdf

Suresh Sridhar

Papers

The impact of mobility on cellular network configuration

O(n) algorithms for load balancing in distributed computing systems