Delay and Total Network Usage Optimisation Using GGCN in Fog Computing

Network performance and throughput is affected by network congestion, which is caused by unnecessary bandwidth over-utilisation, expanding transmission delays, and increase in cost. Fog computing has emerged as a promising solution to overcome these shortcomings by provisioning computational resources to the network's edge. However, selecting suitable fog nodes can pose challenges due to increased latency and high energy consumption, leading to unnecessary bandwidth utilisation. This study proposes a deep learning mechanism called gated graph convolution neural networks (GGCNs) for resource scheduling management in fog computing to improve the average loop delay and the total network usage of the system. Our deep learning mechanism promotes energy-efficient collaborative intelligence among IoT devices while optimising resource utilisation. Reducing energy consumption not only promotes but also enhances sustainability and scalability in IoT networks. Our proposed mechanism shows improved results compared with several benchmark algorithms, such as first come first serve, shortest job first, and particle swarm optimisation. Our results demonstrate that the proposed model will resolve the problem of application placement and present a noticeable reduction in delay and bandwidth. The results can prove to be a standard benchmark in the IoT-Fog computing discipline and used to enhance the quality of service in wide-ranging heterogeneous applications located at distributed locations.