Optimizing geographic routing protocols for urban VANETs using stigmergy, social behavior and adaptive C-n-F mechanisms: An optimized CLWPR

Abstract The most efficient routing protocols in Vehicular Ad Hoc Networks are geographic protocols. However, these kinds of protocols communicate a large amount of information which negatively affects the network overhead load. The second major drawback of geographic routing protocols is the Local Maximum Problem. To deal with such problem, some of protocols employ the Carry-and-Forward mechanism by caching data packets, and consequently increasing the packet delivery Fraction, however, this technique decreases the End-to-End Delay performance. These observations motivated our effort to develop three mechanisms to enhance routing in geographic protocols namely: information transport via Stigmergy, Social Behavior and Adaptive Caching Time of the Carry-and-Forward mechanism. Using the three proposed mechanisms, we present an optimized geographic routing protocol called O-CLWPR based on CLWPR which is a recent promising protocol. Simulation results using NS-3 simulator show not only that the overhead load is highly reduced (about 27%), but the number of Local Maximum Problem cases and the End-to-End Delay metrics are also significantly reduced without deteriorating the packet delivery ratio metric.