Improved models and algorithms for UMTS radio planning

Classical coverage models based on signal predictions, adopted for second generation cellular systems, are not suitable for planning the universal mobile telecommunication system (UMTS) base station location since the area actually covered by each base station depends on the traffic distribution, the power control mechanism as well as the signal quality constraints. In a previous paper Amaldi, Capone and Malucelli (see. Proceedings of IEEE VTC Spring 2001, 2001) presented a discrete optimization model for the UMTS base station location problem. In this paper we propose enhanced models which consider the signal-to-interference ratio (SIR) as quality measure and capture at different levels of detail the specific power control mechanism of the CDMA air interface. Moreover, we propose a tabu search algorithm for the uplink (mobile to base station) direction. The different models and algorithms are compared on realistic instances generated using classical propagation models.