We present a method for a fast prediction of the coverage area on the uplink of a Universal Mobile Telecommunication System (UMTS) network with non-homogeneous traffic. The coverage radius of a base station depends on the interference distribution at this base station since it determines the transmit powers required from the mobile stations (MSs). The interference comprises both the own cell interference and the other cell interference which again depend on each other. Our proposed model for the computation of the interference distribution is based on an iterative solution of fixed-point equations which describe the interdependence of the interferences at neighboring base stations. Furthermore, we develop an efficient algorithm based on lognormal approximations to compute the mean and standard deviation of the other cell interference. We show that our model is accurate and fast enough to predict efficiently the coverage area which is one of the key tasks in the planning process of UMTS networks.