Vehicular communications have grown in interest over the years and are nowadays recognized as a pillar for the Intelligent Transportation Systems (ITSs) in order to ensure an efficient management of the road traffic and to achieve a reduction in the number of traffic accidents. To support the safety applications, both the ETSI ITS-G5 and IEEE 1609 standard families require each vehicle to deliver periodic awareness messages throughout the neighborhood. As the vehicles density grows, the scenario dynamics may require a high message exchange that can easily lead to a radio channel congestion issue and then to a degradation on safety critical services. ETSI has defined a Decentralized Congestion Control (DCC) mechanism to mitigate the channel congestion acting on the transmission parameters (i.e., message rate, transmit power and data-rate) with performances that vary according to the specific algorithm. In this paper, a review of the DCC standardization activities is proposed as well as an analysis of the existing methods and algorithms for the congestion mitigation. Also, some applied machine learning techniques for DCC are addressed.

多年来，车载通信越来越受到关注，如今已被公认为智能交通系统 (ITS) 的支柱，以确保有效管理道路交通并减少交通事故数量。为了支持安全应用，ETSI ITS-G5 和 IEEE 1609 标准系列都要求每辆车在整个社区内定期发送感知信息。随着车辆密度的增加，场景动态可能需要大量的消息交换，这很容易导致无线电信道拥塞问题，然后导致安全关键服务的降级。ETSI 定义了分散拥塞控制 (DCC) 机制，以减轻作用于传输参数（即消息速率、传输功率和数据速率），其性能因特定算法而异。在本文中，提出了对 DCC 标准化活动的回顾以及对现有拥塞缓解方法和算法的分析。此外，还讨论了一些用于 DCC 的应用机器学习技术。