The vehicular ad hoc networks (VANETs) can be exploited in many safety applications like emergency, firefighting, etc. to multicast safety information quickly to several destinations. Thus, the safety applications need to design multicasting schemes to select the optimal path with minimum delay. On another side, Software Defined Network (SDN) is a modern concept that presents availability, reliability, and flexibility and enhances vehicular network architecture. Moreover, the Fog Computing (FC) framework is appropriate to perform the multicast routing process in VANET because it supports the continuous mobility of vehicles and location awareness. The parked vehicles can be used as fixed FC nodes to transfer data with high link stability. This paper aims to produce a Delay efficient Multicasting approach depending on Parked vehicles, FC and SDN for VANET (DMPFS). In DMPFS, the bandwidth constraint is taken into consideration in discovering the optimal multicast route. Moreover, a scheduling approach based on the priority of application type is proposed to classify and schedule the multicast demands. The partitioning technique is exploited to reduce the overhead and time complexity of DMPFS. The simulation results illustrate that DMPFS is superior than adaptive Distributed Tree-based Multicast Routing protocol (DTMR) and Micro Artificial Bee Colony based Multicasting (MABC). In terms of the packet delivery ratio (PDR), DMPFS results in about 9% and 14% improvement compared with DTMR and MABC, respectively. Average end to end delay with DMPFS is approximately 33% and 41% less than DTMR and MABC, respectively. In terms of the communication overhead, DMPFS leads to about 22% and 28% reduction compared with DTMR and MABC, respectively.

Keywords