D2D-Based Vehicular Communication With Delayed CSI Feedback

Device-to-device (D2D) communications can combine the superiorities of both the cellular network and the Ad hoc network, and it is an efficient way to improve the capacity of short distance communication. D2D communications underlaying a cellular network can support vehicle-to-vehicle (V2V) communications. Channel state information (CSI) is crucial to the performance optimization of network. However, it is difficult to obtain the accurate CSI in the highly mobile vehicular environment. CSI of vehicular links is reported to the BS periodically. The delayed CSI feedback exists in the links of V2V and vehicle-to-CUE. The delayed message and the dynamic environment in the V2V network result in the uncertainty of channel gain. In order to improve the reliability of transmission, the vehicular delay time and the speed information are considered in the channel model. The sum-rate maximization problem under uncertainty constrains is formulated, and successive convex approximation is adopted to make the non-convex problem solvable. Two different forms of uncertain channel model are analyzed, and they are converted to the same form of optimization problem. The robust power allocation algorithm is given. The simulation results show the feasibility of the proposed scheme.