Slice Management Mechanism based on Dynamic Weights for Service Guarantees in Smart Grid

With the rapid development of the power grid, the variety of power services has become more and more diverse, resulting in different service needs. As one of the important technologies of 5G, network slicing aims to accommodate different services over the same physical network. In this paper, after briefly analyzing the background of smart grid, we investigate the related research of network slicing. Each kind of services in smart grid has its own bandwidth, reliability and delay tolerance requirements. Therefore, for the purpose of service guarantees in smart grid, we propose a slice management mechanism for resource allocation based on 5G network slicing technology. Due to the good self-learning and adaptive capabilities, we use BP(back-propagation) neural network to train the appropriate weights. Correspondingly, a heuristic algorithm for dynamic weights based on BP neural network to solve the optimization problem is proposed. Finally, simulation results demonstrate the effectiveness of our algorithm within the mechanism.