Abstract
Although various unmanned aerial vehicle (UAV)-assisted routing protocols have been proposed for vehicular ad hoc networks, few studies have investigated load balancing algorithms to accommodate future traffic growth and deal with complex dynamic network environments simultaneously. In particular, owing to the extended coverage and clear line-of-sight relay link on a UAV relay node (URN), the possibility of a bottleneck link is high. To prevent problems caused by traffic congestion, we propose Q-learning based load balancing routing (Q-LBR) through a combination of three key techniques, namely, a low-overhead technique for estimating the network load through the queue status obtained from each ground vehicular node by the URN, a load balancing scheme based on Q-learning and a reward control function for rapid convergence of Q-learning. Through diverse simulations, we demonstrate that Q-LBR improves the packet delivery ratio, network utilization and latency by more than 8, 28 and 30%, respectively, compared to the existing protocol.
Highlights
The vehicular ad hoc network (VANET), a special type of mobile ad hoc network (MANET), has been investigated to provide the infrastructure of a new service paradigm through self-organizing networks that exist between vehicles
Because the unmanned aerial vehicle (UAV) relay path is most likely to be the best approach in terms of link quality and the number of hops, it is highly likely that a bottleneck of the UAV relay node (URN) will occur from the existing routing protocol when the network is congested
We can state that most of the proposed routing protocol techniques designed for UAV-assisted VANET disregarded the traffic characteristics and dynamic load balancing in congested network environments
Summary
The vehicular ad hoc network (VANET), a special type of mobile ad hoc network (MANET), has been investigated to provide the infrastructure of a new service paradigm through self-organizing networks that exist between vehicles It still experiences difficulty in routing with disconnected features that are associated with dynamic wireless environments in mobile network topologies. Because the UAV relay path is most likely to be the best approach in terms of link quality and the number of hops, it is highly likely that a bottleneck of the URN will occur from the existing routing protocol when the network is congested. We propose a Q-LBR system for load balancing in a UAV-assisted VANET This provides a method for URN to handle the maximum traffic acceptable while maintaining a certain level of ground network load. Our evaluation results showed that Q-LBR achieved a significantly better packet delivery ratio (PDR) and network utilization and latency according to the traffic load conditions than existing algorithms and Q-LBR without Q-learning
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
