A Priority-Driven Adaptive Genetic Algorithm for Time-Sensitive Network Traffic Scheduling

Abstract

As a network technology specifically designed to meet the demands of real-time data transmission, Time-Sensitive Networking (TSN) plays a crucial role in applications such as vehicular networks, industrial IoT, and 5G ultra-reliable low-latency communication. TSN can efficiently schedule real-time data streams. TSN traffic scheduling requires fast and precise algorithms; however, existing exact solution methods, due to their high computational complexity, struggle to meet the real-time requirements in large-scale joint scheduling scenarios. Genetic algorithms are suitable for addressing scheduling problems in TSN. The challenge lies in designing the genetic algorithm to find the optimal or near-optimal solution for various complex problems, while considering both performance and quality in the scheduling table. This paper proposes a priority-driven adaptive genetic algorithm for TSN traffic scheduling, which jointly considers routing and TSN constraints. The method employs a dual-vector encoding scheme, and prioritizes traffic characteristics by assigning priority scores to initialize the population. By combining a joint sequential crossover operator and a dynamic adaptive mutation strategy, the algorithm enhances both search capability and scheduling performance. Experimental results demonstrate that, compared to other genetic algorithm-based TSN traffic scheduling methods, the proposed method shows significant advantages in terms of convergence speed, task completion time for TSN traffic scheduling, and scheduling stability.