We consider a wireless communication network with homogeneous cells where new calls and handover calls are generated according to independent Poisson processes. To model the behavior of new calls and handover calls in a cell we use phase-type distributions for the call dwelling times in a cell, which is a generalization of previous studies. As a performance measure, we consider the blocking probabilities of new calls and handover calls.
From our model, we construct a continuous time Markov chain, but the dimension of the M.C. could be very large and consequently its analysis could be very complex. To overcome this difficulty, we consider an identical finite source Jackson network to our model and obtain the closed form expression of the blocking probabilities. Our numerical studies show that the blocking probabilities depend on the distribution of the call dwelling time.