Research on Optimization of Underground Pipe Gallery Remote Antenna Deployment and Resource Allocation Based on Machine Learning

With the continuous advancement of the construction of the Internet of Things (IoT) in China's power sector and the increasing number of cable channels, the demand for cable channel monitoring is becoming increasingly urgent. There is a pressing need for low-cost, high-efficiency digital operation and maintenance methods to support this demand. Currently, there is no mature wireless sensor network solution suitable for cable channels. Therefore, it is necessary to develop a customized low-power wireless sensor network system that supports coverage enhancement and flexible topology networking, forming an application deployment plan for wireless sensor network systems in cable channels and laying the foundation for the construction of digital cable channels. To address this, our paper proposes a wireless sensor network model for cable channels. This model comprises convergence nodes, remote radio units, and equipment within the cable channels, aiming to maximize the total system throughput. To solve this problem, we use deep reinforcement learning methods in machine learning. Through theoretical analysis and numerical simulations, we verify that this method performs better in terms of total system throughput compared to other benchmark methods.