ALPS: An Adaptive Link-State Perception Scheme for Software-Defined Vehicular Networks
IEEE Transactions on Vehicular Technology
The software-defined vehicular networking (SDVN) paradigm alleviates the deficiencies brought on by distributed vehicular. The separation of the control plane and the data plane allows the controller to manage the network based on global information. Most existing routing schemes in SDVN obtain the link-state through which vehicles periodically send beacon messages to the controller. However, due to the high mobility of the vehicles and the dynamic communication environment, the link-state changes within the beacon interval. In this case, the controller may select an expired link to transmit data during routing calculation, which will undoubtedly result in packet loss. Therefore, it is important for the controller to timely obtain the link-state during the beacon interval. If the controller can timely obtain the information after a link becomes unavailable, the risk of selecting unavailable links can be significantly reduced. In this paper, we propose an adaptive link-state perception scheme (ALPS) for SDVN, which enables the controller timely obtain the link-state within the beacon interval. We obtain the link-state by detecting the loss of packets on a link. A link quality evaluation method based on fuzzy logic is present to evaluate the possibility of link failure. After the link evaluation, we present an adaptive threshold adjustment method to dynamically adjust the detection range to decrease the detection cost. Simulation results demonstrate that ALPS can effectively reduce the packet loss ratio at a low cost.
Electronic mail, fuzzy logic, Fuzzy logic, link-state perception, Packet loss, Reinforcement learning, reinforcement learning, Routing, Software-defined vehicular network, VANET, Vehicle dynamics, Vehicular ad hoc networks
N. Lin, D. Zhao, L. Zhao, A. Hawbani, M. Guizani and N. Kumar, "ALPS: An Adaptive Link-State Perception Scheme for Software-Defined Vehicular Networks," in IEEE Transactions on Vehicular Technology,, pp. 1-12, 2022, doi: 10.1109/TVT.2022.3214660.