超密集无线通信网络最优接入点部署策略OA

This paper proposes an optimal access point(AP)deployment scheme based on stochastic gradient ascent to address the strong interference signals and high sensitivity of system performance to network topology in ultra-dense wireless communication networks.The deployment strategy for the APs is determined by sampling user positions at different time instances,which is particularly effective when the explicit analytical expression of the user spatial distribution is unavailable.Furthermore,the proposed method incorporates the root mean square propagation(RMSProp)algorithm,markedly enhancing the convergence performance of the deployment process.Simulation results demonstrate it achieves effective interference management and substantially improves the long-term average data rate of the system compared with both the uniform deployment scheme and the K-means clustering deployment scheme.The performance gains are particularly noticeable in scenarios with highly clustered users.