A Novel MAC Scheduling Based on Cross-layer Algorithm and Deep Learning

Abstract

MAC scheduling plays an important role in affecting the performance of wireless communication system and the fairness among users. A significant number of algorithms have been put forward for increasingly complex scenarios in recent years. In this paper, we designed a novel MAC scheduling algorithm for intelligent 6G wireless network based on cross-layer mathematical algorithm and deep learning with Multi-Layer Perceptron (MLP). We define the cross-layer mathematical algorithm part as HI (Human Intelligence) engine and the deep learning part as AI (Artificial Intelligence) engine. Also, we respectively designed the coordination mechanism, required data, respective workflows and specific algorithms for the two engines. To demonstrate the superior performance gains of our proposed methodology, the use case was analyzed with simulation in various traffic scenarios and made performance comparison with traditional solution.