A Probabilistic Codebook Technique for Fast Initial Access in 6G Vehicle-to-Vehicle Communications

The Third Generation Partnership Project (3GPP) has recently introduced millimeter Waves (mmWaves) in Release 16, with Vehicle-to-Everything (V2X) communications as the primary use-case. To overcome the strong attenuation at mmWaves, beam-based communications are used, although they require a beam selection procedure at each initial access or after detecting a link failure. The 5G New Radio (NR) standard selects the best beam by implementing an omnidirectional spatial scanning, which increases the communication delay and overhead. Therefore, this may not be suitable for the more challenging Vehicle-to-Vehicle (V2V) scenario, characterized by frequent re-alignment due to vehicle mobility and blockage. In this paper, we propose a solution for beam selection that emerges as a valid candidate for beyond 5G V2V communications. The constraints coming from road topology enforce a set of predefined mobility patterns, which result in a non-uniform distribution of communication directions. We leverage this information to build a Probabilistic Codebook (PCB) of prioritized beams. The PCB is trained to perform realistic simulations. We observed that our approach fasten the beam selection time by 67% (on average) compared to conventional methods that do not take advantage of the environmental characteristics and over 90% in some scenarios.