Automatic planning of UMTS release 4.0 networks using realistic traffic

Since Universal Mobile Telecommunications System (UMTS) networks are very expensive to deploy, network optimization is an important issue for service providers to remain competitive. As a result, this paper proposes a new mathematical programming model in order to plan UMTS Release 4.0 networks. Given realistic traffic profile as input, the model simultaneously considers voice and data services as well as network signaling. More precisely, the model finds the number, the type and the location of the network elements (NE), selects the number and the type of links and interfaces and finally designs the network topology. The objective of the model is to minimize network cost. An illustrative example of the model is described and a set of randomly generated problems are solved using a commercial implementation of the branch-and-bound algorithm.