基于长短期记忆网络的数字孪生移动通信网络环境生成技术

Digital Twin Environment Generation for Mobile Communication Networks:A Long Short-Term Memory Approach

未来6G时代将进入虚拟化的数字孪生世界,实现“万物智联,数字孪生”的愿景。移动通信网络的数字孪生需要建立物理通信网络与孪生通信网络之间的高精度映射,保证孪生虚拟数据与实际物理数据保持高度匹配。针对上述数字孪生环境生成的问题,采用了数据驱动的长短期记忆网络的方法,对移动通信网络中的无线信道、基站业务流和用户移动轨迹进行时序预测与生成,从而构建出高精度数字孪生环境。仿真结果表明,所提算法能精准生成未来时序数据,与物理通信网络真实数据保持高度一致。

The future 6G era will enter a virtual digital twin world and realize the vision of wisdom connection and digital twin.The digital twin of mobile communication network needs to establish a high-precision mapping between the physical communication network and the twin communication network to ensure that the virtual data and the physical data are highly matched.To solve the problem of digital twin environ-ment generation,the method of data-driven long short-term memory network is adopted to predict the time sequence of wireless channel,base station traffic flow and user mobility trajectory in mobile communication network,so as to construct a high-precision digital twin envi-ronment.Simulation results show that the proposed algorithm can accurately predict future time series data.