Abstract
Handoff failure probability in cellular systems is defined as the probability that a handoff request is denied for lack of resources, and premature call termination probability is defined as the probability that an accepted ongoing call is terminated due to lack of recourses, are two of the main parameters used to study and analyze several cellular system performance measures. Further, they are the main parameters that are used in quality of service studies and teletraffic analysis of such a network during the planning and development stages. To evaluate the handoff failure probability, premature call termination probability, and other system performance measures such as call dropping probability, several statistical distributions have been used in the literature to model the channel holding time distribution in 3rd and 4th generations cellular systems, such as exponential, Erlang, Gamma, and generalized Gamma, hence, the complexity of the analytical and simulation models introduced in literature varies based on the assumed channel holding time distribution in addition to other assumed parameters. In this paper the effect of the statistical distribution used to model the channel holding time on handoff failure probability and premature call termination probability are investigated using analytical and simulation models, where the main goal of the research is to determine the level of complexity of the statistical distribution used to model the channel holding time to accurately evaluate the handoff failure probability and premature call termination probability.