Efficient Rate-Splitting Multiple Access for the Internet of Vehicles: Federated Edge Learning and Latency Minimization

Rate-Splitting Multiple Access (RSMA) has recently found favour in the multi-antenna-aided wireless downlink, as a benefit of relaxing accuracy Channel State Information at Transmitter (CSIT), while achieving high spectral efficiency and providing security guarantees. These benefits are particularly important high-velocity vehicular platoons since their Doppler affects estimation CSIT. To tackle this challenge, we propose an RSMA-based Internet Vehicles (IoV) solution that jointly considers platoon control FEderated Edge Learning (FEEL) downlink. Specifically, proposed framework is designed for transmitting unicast messages within IoV platoon, well privacy-preserving FEEL-aided downlink Non-Orthogonal Unicasting Multicasting (NOUM). Given sophisticated framework, multi-objective optimization problem formulated to minimize both latency FEEL deviation vehicles platoon. efficiently solve problem, Block Coordinate Descent (BCD) developed decoupling main into two sub-problems. Then, solving these non-convex sub-problems, Successive Convex Approximation (SCA) Model Predictive Control (MPC) method FEEL-based respectively. Our simulation results show system outperforms popular Multi-User Linear Precoding (MU–LP) conventional (NOMA) system. Finally, BCD shown generate near-optimal solutions reduced complexity.