Traffic Scheduling Optimization in Cognitive Radio based Smart Grid Network Using Mini-Batch Gradient Descent Method

Smart Grid (SG)network comprises of wide variety of applications with diverse Quality of Service (QoS)requirements. Each SG application has its own requirements of latency, throughput and reliability. Cognitive Radio (CR), which offers better spectrum utilization through opportunistic spectrum access and spectrum sharing, is considered as a promising wireless technology for SG networks. Traffic scheduling and optimization is one of major challenges in CR-based SG communication network. In this paper, we propose a novel usage of mini-batch gradient descent method for the optimization of QoS-based differential scheduling in CR-based SG network. A 2-class based priority scheduling model with emergency and interrupt handling capability is used. It is shown through simulation results that mini-batch gradient descent method achieves better results as compared to gradient descent method in terms of the fast convergence and minimization of the overall cost function.