C-V2X (Cellular Vehicle-to-Everything) standard introduced in 3GPP Release-14 is emerging as a potential technology for Cooperative Awareness Message (CAMs) dissemination among connected vehicles. But to attain its full potential, optimal resource scheduling and interference management must be enforced. To address this issue, we propose a weighted exponential averaging based Context-Aware Resource Reselection scheme (CARRs), enabling the periodic exchange of CAMs in a vehicular network. The proposed strategy allows autonomous resource selection by performing continuous power sensing and cooperative learning with the neighbors within the safety zone. CARRs is a two-stage learning process. In the first stage, it learns the exponential weighing factor for each resource available in the Vehicle-to-Vehicle resource pool and performs resource reselection. In the second stage, it learns to select transmit power level based on the interference experienced over the reselected resource. It is established through numerical results that CARRs outperforms the existing strategies. Packet reception ratio, average error rate, average blocking rate and update delay are considered as the performance metrics. CARRs improves packet reception ratio by 3.1% while reducing the average error rate by 28.2% and lowering update delay by 4.5% in comparison to existing schemes.

3GPP Release-14中引入的C-V2X（蜂窝车辆到一切）标准正在成为一种潜在的技术，可用于在连接的车辆之间传播协作意识消息（CAM）。但是要充分发挥其潜力，必须实施最佳的资源调度和干扰管理。为了解决这个问题，我们提出了一种基于加权指数平均的上下文感知资源重选方案（CARR），可以在车辆网络中定期交换CAM。所提出的策略允许通过与安全区内的邻居执行连续的功率感测和协作学习来自主选择资源。CARRs是一个分为两个阶段的学习过程。在第一阶段 它为车对车资源池中可用的每个资源学习指数加权因子，并执行资源重选。在第二阶段，它学习根据在重新选择的资源上遇到的干扰来选择发射功率电平。通过数值结果可以确定，CARR优于现有策略。分组接收率，平均错误率，平均阻塞率和更新延迟被认为是性能指标。与现有方案相比，CARR将数据包接收率提高了3.1％，同时将平均错误率降低了28.2％，并将更新延迟降低了4.5％。通过数值结果可以确定，CARR优于现有策略。分组接收率，平均错误率，平均阻塞率和更新延迟被认为是性能指标。与现有方案相比，CARR将数据包接收率提高了3.1％，同时将平均错误率降低了28.2％，并将更新延迟降低了4.5％。通过数值结果可以确定，CARR优于现有策略。数据包接收率，平均错误率，平均阻塞率和更新延迟被视为性能指标。与现有方案相比，CARR将数据包接收率提高了3.1％，同时将平均错误率降低了28.2％，并将更新延迟降低了4.5％。