Power Communication Resource Allocation Algorithm Based on SPN

With the development of communication technology in the power system, there are higher demands for comprehensive support and assurance capabilities in communication specialties. Different application scenarios in power communication have varied requirements, such as ultra-large bandwidth, ultra-low latency, and massive device access. To allocate resources reasonably and adapt to the differentiated demands of various services, this paper proposes the use of SPN (Slicing Packet Network) technology to slice the resources of power communication networks. Four types of service slices are designed, and the resource allocation problem is modeled as a MDP (Markov Decision Process). Deep reinforcement learning is introduced to solve for the optimal strategy. Simulation results show that compared to other algorithms, the ADQNS (Adaptive DQN Slicing) algorithm, which is framed by DQN (Deep Q-Network) technology, meets the quality of experience requirements for multiple services, demonstrating better performance in terms of end-to-end throughput, end-to-end latency, resource utilization, and fairness metrics.