Piggyback-Based Collaborative Resource Selection in C-V2X

Cellular Vehicle-to-Everything (C-V2X) was introduced as the enabling vehicular communication technology for the first time in 2017 by the Third Generation Partnership Project (3GPP) in its Release 14. 3GPP defines two modes, mode 3 and mode 4, for resource scheduling in V2X. Mode 4 is used for decentralized resource scheduling, where the vehicles select the resource by themselves based on local observations. Resource scheduling in mode 4 is challenging due to resource conflicts arising from the limited spectrum, as competing vehicles may select the same resource. Additionally, the aperiodic nature of Cooperative Awareness Message (CAM) generation increases resource contention. Each year, 3GPP introduces new features to enhance the enabling V2X technology. In 3GPP Release 17, a physical sidelink feedback control mechanism is also introduced to assist resource scheduling through feedback control. In this work, we utilized the Physical Sidelink Feedback Control Channel (PSFCH) and reinforcement learning to enhance resource scheduling in C-V2X. The proposed scheme has two objectives: to address the aperiodic CAM generation and to improve resource scheduling.