Time-to-collision-based awareness and congestion control for vehicular communications

Vehicular wireless communications require both congestion control to guarantee the availability of a fraction of the bandwidth for safety-related event-driven messages in emergency cases, and awareness control to adapt the beaconing activity to the application needs and surrounding traf_c situation. Most current approaches either ignore the traf_c situation and only adapt the beaconing rate to the channel congestion state or override the congestion control limits, leading to questionable results in both cases. In this paper, we conceive and validate a novel approach, combining both aspects. Based on distributed Network Utility Maximization (NUM), our algorithm satis_es the constraints on channel availability, whereas the safety of the surrounding traf_c situation is captured with a time-to-collision metric, used to assign priorities in the optimal allocation problem. The performance of the proposed approach is validated and compared to other popular algorithms. Results show that our proposal automatically anticipates a potential increase in rate due to a critical safety situation, but does not interfere with the reserved bandwidth for safety applications.This work was supported in part by Spanish MECD/AEI/FEDER under Grant TEC2016-76465-C2-1-R, in part by Spanish MINECO/AEI/FEDER under Grant TEC2017-84423-C3-2-P (ONOFRE-2), in part by e-DIVITA under Grant 20509/PDC/18 (Proof of Concept, 2018), and in part by ATENTO Projects under Grant 20889/PI/18 (Fundación Seneca, Región de Murcia). The work of J. Aznar-Poveda was supported by the MECD under Grant BES-2017-081061