On the Design of Large-Scale Cellular Mobile Networks Using Tabu Search

Abstract

In this paper, we proposed a tabu search approach to design large-scale UMTS mobile networks and to specifically solve the problem of assigning Node Bs to RNCs in cellular mobile networks. Experiments were conducted to measure the quality of solutions provided by this algorithm. This approach was compared against genetic algorithm and simulated annealing. Computational results obtained confirm the efficiency and the effectiveness of the tabu search to provide better solutions than genetic algorithm and simulated annealing, especially for large-scale cellular mobile networks with a number of Node Bs varying between 100 and 400, and a number of RNCs oscillating between 5 and 8, meaning that the search space size ranges between 5100 and 8400 and that the average improvement rates are in the order of 2% and 7% respectively. This improvement represents a substantial reduction in maintenance and operations costs, which, for a 5 year period, amount to millions of dollars.

Related Papers

• → Evolving core networks from GSM to UMTS R4 version(2009)13 cited
• → A study on evolving the architecture of circuit switched domain in UMTS core networks(2010)2 cited
• → Optimizing the Traffic of Voice Core Domain in UMTS Network through RNC Re-homing(2010)1 cited
• → The Impact of Cell Site Re-Homing on the Performance of UMTS Core Networks(2010)1 cited
• → Network Architectures, Standardizations Process(2021)