Coordinated Multicast and Unicast Robust Transmission in C-V2V Massive MIMO With Imperfect CSI

In this paper, we investigate the coordinated multicast and unicast transmission for cellular-based vehicle-to-vehicle (C-V2V) massive multiple-input multiple-output (MIMO) in the scenario where imperfect channel state information (CSI) is available at the base station side and the transmitter in each V2V communication pair. Assume that the base station side and the transmitter in each V2V transmission pair know the statistical CSI and the imperfect CSI of the user terminals. The statistical CSI is a joint correlation model with channel variance information only, and the a posteriori CSI after channel estimation is a joint correlation model with both channel mean and variance information. Based on this, the C-V2V massive MIMO downlink channel model is established. The precoding design problem is a weighted ergodic sum-rate maximization problem, and the objective function is the weighted sum of the achievable ergodic unicast rate for all the V2V pairs and the achievable ergodic multicast rate for all the cellular users. The minorize-maximize (MM) algorithm is utilized to equivalently convert the complex non-convex original optimization problem into a quadratic optimization problem that can be solved iteratively. Since the closed-form optimal solution of the quadratic optimization problem contains the expectation of the complex random matrix, the deterministic equivalents of the desired matrix expectation are introduced to reduce the computational complexity, and the large-dimensional matrix inversion problem is avoided by matrix inversion lemma to further reduce the complexity. In addition, we prove that the beam domain transmission is optimal for the scenario with zero mean in the a posteriori CSI. Linear precoding design algorithms are given in conjunction with the above procedure. The simulation results demonstrate the good convergence performance of the proposed algorithms and the significant improvements of the proposed robust coordinated communication method in the system spectral efficiency.