User Mobility Modeling in Cellular Communications Networks

3 To all these who believe in, hope for, and work towards a more human technology, I dedicate my work. Abstract Mobility management is the cornerstone of cellular philosophy. Mobility analysis gives a deep insight on the impact of the terminal mobility on the cellular system performance. In third-generation mobile communication systems, the influence of mobility on the network performance will be strengthened, mainly due to the huge number of mobile users in conjunction with the small cell size. In particular, the accuracy of mobility modeling becomes essential for the evaluation of system design alternatives and network implementation cost issues. Currently available mobility models tend to be either too simplifying or too sophisticated. For mobility modeling under realistic traffic and environmental conditions, this thesis introduces a novel representation technique which uses the distribution functions of street length, direction changes at crossroads, and terminal velocity. The parameters required, e.g. mean and variance of street length, user velocity, and direction changes distributions, can be easily derived by observation and measurement. Other important factors influenced by user mobility concern the mobile user calling behavior expressed by the incoming/outgoing call arrival rate and average call duration. This work thus brings together teletraffic theory and vehicular traffic theory. This is capable its to describe the user behavior in detail, and is applied for the characterization of the traffic in individual single cells of the mobile network. The effect of mobility has been analyzed in terms of the local performance measures like probability of handover and call blocking probability (for new and handover calls). Additionally, this model has been used to calculate the distribution of channel holding times. The performance of new call handling algorithms are evaluated. The global performance criteria of interest are call dropping probability for all calls, call processing time dependent forced termination of handovers, and channel utilization. Thus the average number of functions per call for information handling systems with different hierarchical structures can be computed, too. All these parameters are expressed as a function of the user calling and mobility behavior. To assess the accuracy of the proposed mobility model a simulation tool has been constructed. The tool takes into account the user traffic and mobility behavior over different environments (high density city center, outskirts, etc.). Theoretical results, simulation trials, and measurement data coincide, indicating the excellent accuracy the analytically described mobility model provides. Additionally, an approach for Space Division Multiple …