Perceptual data importance and freshness aware transmission in millimeter wave vehicular networks

Abstract

The extensive sharing of perceptual data between vehicles and between vehicles and roads has significantly enhanced the performance of intelligent transportation system (ITS). The current vehicular networks using sub-6 GHz struggle to meet the demands for high-rate, low-latency, and highly reliable communication. To address this issue, this paper proposes a perceptual data sharing strategy based on millimeter-wave (mmWave) communication technology. This strategy takes into account the characteristics of vehicular perceptual data, i.e., the importance and freshness of the data, and constructs a mixed-integer nonlinear sum-of-ratios optimization problem. To meet the stringent real-time decision-making requirements of vehicular networks, we leverage the transmission slot characteristics of the Time Division Multiple Access (TDMA) Medium Access Control (MAC) architecture to transform the nonlinear original problem into a series of approximate integer linear programming (ILP) problems. Then we employ maximum weight matching in graph theory to further reduce computational complexity, enabling the problem to be solved in polynomial time. Additionally, we have designed a brute-force algorithm to ensure the global optimum is achieved, thereby validating the performance of our proposed algorithm. Comparative simulation studies with the brute-force algorithm, the ILP solver, the edge coloring algorithm, our previously developed parameterization-based iterative algorithm (PIA), and the First-Come-First-Serve (FCFS) scheduling scheme verify the effectiveness of our proposed algorithm.