A Distributed Deep Learning-Driven Edge Caching Strategy for Industrial IoT Networks

Industrial Internet of Things (IIoT) networks are a critical enabler of use cases in the manufacturing, logistics, and healthcare, sectors. Due to the need to support realtime and mission-critical use cases, IIoT networks have low tolerance for delay and require timely wireless content access. As such, we propose the use of an edge computing model, multi-access edge computing (MEC), and a distributed deep learning (DDL)-based edge caching strategy to jointly support ultra-reliable low-latency content access in IIoT. Specifically, a distributed framework was designed to harness on the computing power of edge servers to run deep learning models concurrently. Through simulation results, we show that the proposed strategy can increase cache hit rates by 5%-10% and reduce realtime latency by 10%-22 % as compared against the traditional baseline frameworks, such as least recently used and least frequently used.