Interference Control Mechanism Based on Deep Reinforcement Learning in Narrow Bandwidth Wireless Network Environment

With the expansion of application scenarios of communication network and the complexity of communication network structure, in specific practical application scenarios, communication network often needs to consider all available communication network resources while providing services for business needs. The abstract modeling of network communication resources, the standardized description of network communication resources and the basis and mechanism of virtual network communication resource pool construction play an important role in the unified management and scheduling of network resources. At the same time, according to the business requirements in specific special scenarios, how to quickly build the virtual network, ensure the demand interconnection among network members and meet the differentiated transmission needs of the business in specific application scenarios. The network communication resources are scheduled in a unified and coordinated manner on demand. The communication network demands of different businesses are taken as the optimization objective, and the optimal scheduling and configuration of the network communication resources are realized under the condition that the business needs are met. This paper aims at the communication network under special application scenarios, based on the virtualization modeling of communication network resources, and realizes the optimal scheduling and configuration of network communication resources through deep reinforcement learning optimization algorithm.