Task Scheduling in Intelligent Edge of ۵G -Enabled AI -Industrial IoT Using DL-LSTM -AQL

Artificial Intelligence (AI) is a great technology for Industrial Internet of Things (IoT), which has led to the emergence of AI -Industrial IoT to support Industrial IoT intelligent applications. The complex AI tasks of AI -based applications require complex servers to process. This causes high Energy Consumption (EC). Therefore, in AI -Industrial IoT equipped to ۵G (۵G -Enabled AI -Industrial IoT) and edge computing Computing (IECo), increase the processing speed of intelligent services at lower cost than the cloud -based IoT, which leads to reduced delay and reduced EC. This paper deals with the EC of edge devices and heterogeneous architecture and cloud services in processing of ۵G -Enabled AI -Industrial IoT tasks, processing delay and Task Scheduling (TSch) problem formulation in edge and cloud using Deep Learning Long -Short -Term Memory (LSTM) Attention Q -Learning (DL -LSTM -AQL). The performance evaluation based on extensive simulation indicated that this framework outperforms other methods with less EC, and it can be less delay using ۵G.

کلیدواژه ها:

Industrial Internet of Things ، Intelligent Edge Computing ، Task Scheduling ، Artificial Intelligence ، Long -Short -Term - Memory and Q -Learning