Abstract
Urban vehicular ad hoc network (UVANET) is a modality of high-topology variation MANET where vehicles travel across constrained layout of the roads and tall buildings, which eventually results in shadowing effects. One of the critical challenges in UVANET is the handover process that happens when a vehicle changes its association with the roadside unit. Urban area suffers from high vehicular density at traffic signals, and vehicle slowdowns caused by speed limits, frequent link disconnection caused by signal attenuation due to obstacles. Thus, it is difficult for UVANET to disseminate the packets to their destinations located in a specified area. This paper presents an effective handover management that calculates the signal-to-interference-plus-noise ratio (SINR) to ultimately determine the optimal cluster head for seamless handover in UVANET. It also utilizes the software-defined network to resolve the problem of handover failure in UVANET. The finding in the experimental results shows that the developed association clustering mechanism based on SINR significantly outperforms the baseline association mechanism. More importantly, it can resolve different types of handover problems that improved the quality of service in terms of throughput, handover latency, packet loss ratio, and handover failure ratio, when compared to the baseline handover mechanisms.