Recently, the number of mobile communication subscribers has explosively increased. Because the electromagnetic spectrum for the mobile cellular communication is limited, it is necessary that cellular systems use their frequency channels as efficiently as possible. The lack of capacity is one of the most important problem among a number of cellular system problems. There has been a great deal of effort to increase the capacity of the cellular radio systems. Some techniques are developed for this purpose. They are digital multiple access technology : for example sectorization, cell splitting, power control etc. If wireless communication system cannot serve sufficient users, it is necessary to build additional base stations (BSs).
In this paper we consider a problem of minimizing the construction cost of the additional BSs to cover increased traffic. We call this problem cell planning with capacity expansion (CPCE). We divide the area to be covered into a number of subareas which are called traffic demand areas (TDAs). It is assumed that each TDA has traffic demand. The TDA should be connected a BS which is able to serve it with sufficient quality. We define the potential service area (PSA) of each BS and the capacity of each BS. We formulate this problem and convert it into a graph partitioning problem.
We devised a tabu-search based heuristic algorithm for this problem and tested the efficiency of the proposed algorithm with some randomly generated examples. We obtained near-optimal solutions with the proposed algorithm.