URLLC-Aware Bi-Objective User Scheduling for D2D Enabled V2X Networks

Vehicle to Everything (V2X) communication enabled by Device-to-Device (D2D) technology serves as the key enabler of ultra-reliable and low latency (URLLC) service in 5G and 6G networks. In this context, resource allocation targeting simultaneous reliability, low latency, and efficiency remains as a complex multi-objective challenge. In this paper, we formulate a URLLC constrained unicast scheduling problem that jointly maximizes user satisfaction and overall profit of the network operator under channel quality, distance, and resource block constraints. We proved the NP-Hardness of the problem by a polynomial time reduction from the 0/1 Knapsack problem. A Non-Dominated Sorting Genetic Algorithm II (NSGA-II) is proposed that incorporates constraint aware dominance sorting, elitist preservation, and crowding distance measurements for a Pareto-optimal solution. We determine the computational complexity of this algorithm as O(GNpop(Edecode+ logNpop)). Further, the performance of NSGA-II is compared with two baselines Greedy and Genetic algorithm. In all simulation scenarios, the proposed algorithm performs better in terms of user satisfaction, network profit, and URLLC support.