摘要
5G低流量区域特点鲜明、精准规划选址难度大。提出采用多层感知器神经网络和遗传算法组合的机器学习方法,结合5G现网运行数据,对低流量覆盖区域无线网络精准规划选址。采用MATLAB/Simulink搭建规划平台,仿真结果表明,在保障现网运行指标的前提下,设计平台能够有效指导5G低流量覆盖区域工程实践和节约站址资源,为无线网络精准规划提供一种新思路。
The 5G low-flow area has distinct characteristics,and it is difficult to select the location for precise planning.In this paper,a machine learning method based on multi-layer perceptron artificial neural network and genetic algorithm is proposed.MATLAB/Simulink is used to build the planning platform,and the verification results show that it can effectively guide the 5G low-flow area engineering practice and save the site resources,under the premise of guaranteeing the current network operation index,it provides a new idea for accurate planning of wireless network.
作者
丁瑜
邓敏茜
卢洁珍
DING Yu;DENG Minqian;LU Jiezhen(Nanning Vocational and Technical College,Nanning 530008,China;Guangxi Construction Vocational and Technical College,Nanning 530007,China)
出处
《通信电源技术》
2023年第11期44-47,共4页
Telecom Power Technology
关键词
5G
移动通信
无线网络规划
人工神经网络
遗传算法
5G
mobile communication
wireless network planning
artificial neural network
genetic algorithm