Evaluation of handover related statistics and the applicability of mobility modelling in their prediction

This paper examines the way handover-related parameters such as the handover margin, averaging window and measurement period affect the handover procedure and the statistics of its associated parameters like cell residence time, number of handovers and channel holding time. Furthermore, it compares these statistical results with the predictions of the mobility modelling and other assumptions commonly made. In a nutshell, mobility modelling results regarding cell residence time could be used for relatively large handover margins of 10 dB to 15 dB, whilst the averaging window equals 15 s, and generally exponential distribution approximations should be avoided. In contrast, under the same circumstances, the channel holding time could be assumed to be exponentially distributed.