Vehicular Ad Hoc Network (VANET) is an important component of the Internet of Vehicles (IoV), providing a wide range of traffic safety, efficiency, and infotainment applications. These applications pose challenges to onboard computation and energy resources. To address these issues, Vehicle Edge Computing (VEC) has been introduced as a novel computing paradigm. Given the rapid and frequent changes in the topology of VANET networks due to the highspeed movement of vehicles, most VANET applications adopt broadcast communication, except for a few employing unicast communication. However, traditional task offloading models struggle to support broadcast applications. Therefore, this paper proposes a VANET broadcast task model to offload tasks from On-Board Units (OBUs) to Road-Side Units (RSUs) for the first time. To meet the requirement of low latency, an efficient polynomial-time heuristic algorithm is further proposed. Numerical results demonstrate the outstanding performance of the algorithm, with all instances being solved in less than 6ms, which significantly surpasses the performance of existing solvers. Moreover, the solutions obtained from the algorithm exhibit gaps below 1%, indicating their high acceptability and applicability in practical VANET applications. Furthermore, extensive numerical experiments provide valuable insights into the practical implementation of task offloading, suggesting an appropriate ratio of 6 between the number of OBUs and RSUs.
Offloading Model and Algorithm for VANET Broadcast Applications
IEEE Transactions on Intelligent Transportation Systems ; 26 , 5 ; 6070-6083
01.05.2025
7279336 byte
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
V2V2I VANET Data Offloading Path Using Reinforcement Learning
Springer Verlag | 2025
|