An Improved Genetic Algorithm of Efficient and Delay Sensitive Traffic Scheduling for Massive Data Transmission
Bo Yi, Yongrui Yang, Qiang He · 2023
The emergence of 5G and Time Sensitive Networking (TSN) have greatly promoted the development of Industry 4.0 with super-high bandwidth and service quality. However, they also produce massive delay sensitive traffic, such that the demand on ultra-low and deterministic end-to-end delay may not be satisfied. In this paper, an improved genetic algorithm is used to address the delay sensitive traffic scheduling problem, so as to satisfy the end-to-end delay demands of diverse new applications. Specifically, the proposed solution is designed from both the local and global perspectives. On one hand, the greedy strategy is used to calculate the earliest starting time for each delay sensitive traffic, so as to minimize the end-to-end delay in a local view. However, such greedy strategy may lead to the long waiting time for some traffic that are to be scheduled in a later time, so that the traffic may be dropped with a high probability. To avoid such situation, the genetic algorithm is applied to arrange a better scheduling order for delay sensitive traffic in a global view. Experiments indicate that the proposed solution achieves better performance in terms of the end-to-end delay.