Sub-6 GHz V2X-assisted MmWave optimal scheduling for vehicular networks

Abstract

Sub-6 GHz assisted millimeter wave (mmWave) technology is a promising solution to address the ever-increasing data exchange demand in Vehicle-to-Everything (V2X) communication. The omni-directional sub-6 GHz channels exchange vehicle and control information in the vehicular network, while the directional mmWave technology is used for high-rate data transmission. Consider the huge amounts of vehicular data, the transmission constraints imposed by mmWave, and the unique vehicular ad hoc network (VANET) data features, a sub-6 GHz assisted-mmWave scheduling algorithm is needed to facilitate V2X data transmission. Existing scheduling algorithms in the literature fall short of addressing realistic V2X data features. Therefore, in this paper, we propose a novel objective to maximize the network utility that succinctly captures realistic V2X data features, including varying data sizes, data importance and their freshness decay with transmission delay. We formulate the scheduling problem into a novel mixed-integer sum-of-ratios optimization problem. To solve the challenging NP-complete problem, we first derive an equivalent parametric mixed-integer linear programming problem with existence and uniqueness proofs. Next, we provide the necessary condition for the optimality of the equivalent problem. A novel Parameterization-based Iterative Algorithm (PIA) is then developed to learn the global optimal solution with convergence analysis. Comparative simulation studies with the brute-force algorithm, the widely used branch and bound (BB) method, the greedy algorithm, and our previously developed maximum weight matching based iterative algorithm (MWMIA) are conducted to verify the correctness and effectiveness of the proposed PIA scheduling algorithm.