Deep Learning-Based Scheduling Scheme for IEEE 802.15.4e TSCH Network

IEEE 802.15.4e time-slotted channel hopping (TSCH) is one of the most reliable resources Industrial Internet Things (IIoT). TSCH operates on slot-frame structure consisting multiple channel-offsets and slot-offsets. It gaining acceptance due to its simple architecture consume low power in industrial applications. The performance mainly dominated by media access control (MAC) mechanism, which covers refitment, enumeration, composition, data transmission. However, many cases, transmission schedules are not accurately prescribed. Therefore, researchers trying define pragmatic scenarios scheduling. Their fundamental approach schedule network a centralized way while framing scheduling based such as throughput delay. In this work, deep learning (DL)-based scheme has been proposed. for links cell assignment can be constructed maximum edge weighted bipartite matching approach. paper, we design weight composed delay, use Hungarian algorithm proper assignment. With algorithm, generate training train neural (DNN) accordingly. simulation, consider with 5 nodes where 12 formulated, 16 cells link simulation results show that proposed learning-based provide similar algorithm-based above 90% accuracy nearly 80% execution time reduction.