Hypergraph-based Performance Optimization in IRS-UAV-assisted Ultra-dense Network
Zhiqing Yang, Hui Ping Liang, Guobin Zhang, Yanfeng Zhang, Wei Zhang · 2024
Ultra-dense networks (UDN) are pivotal for the practical implementation of 6 G technologies. However, they often suffer from a lack of control over the wireless propagation channel and worse communication performance for edge users. Intelligent reflecting surfaces (IRS) is regarded as an emerging technology that can overcome these challenges. In this work, we introduce multiple unmanned aerial vehicles (UAVs) carried IRSs in a UDN to serve multiple users by reflecting signals from macro base station (MBS), which improves the communication rate of edge users by extending the coverage of the communication network. To maximize the minimum user rate in the network, we consider the optimization variables including IRS-UAV association matrix, UAV’s deployment location, and the phase shift of the IRS elements. We propose a weighted hypergraph model to transform the problem into a clustering problem and the machine learning is applied to solve this problem. Simulation results show that the IRS-UAV employment improves the performance of edge users in the network compared with the deployment of small base stations (SBSs) and fixed IRS.