Millimeter-wave (mmWave) communication is considered to be a promising candidate to enable multi-Gbps data rates in future vehicle-to-everything (V2X) communication. Beam alignment is quite crucial for beam-based mmWave communication and the beam sweeping method is widely adopted for the alignment at present. However, this kind of method will cause large overhead in beam alignment and is inefficient in high mobility environment due to Doppler spread. In this paper, we design an overhead-free vehicular-position-based beam alignment scheme for mmWave V2V communication between neighbor vehicles in highway scenarios. In the proposed beam alignment scheme, the beam is directly steered to the estimated vehicular position without any searching steps in beam training. To avoid beam misalignment caused by localization errors and to maximize transmission throughput, the problem is formulated as a tailored beamwidth optimization problem. A Monte Carlo based Beamwidth Optimization (MCBO) method is developed to divide this optimization into two phases and solve this problem statistically. Simulation results demonstrate that, comparing to the widely-adopted beam-sweeping based beam alignment schemes, the proposed vehicular-position-based overhead-free beam alignment scheme with MCBO method can provide significant throughput improvements in general car-following scenarios on the highway.
