在源信号和传输信道未知情况下，只利用接收天线的观测数据抽取源信号，称为盲信号分离.盲信号分离不仅是信号处理界、而且也是神经网络界的研究热点课题，在无线数据通信、雷达、图像、语音、医学以及地震信号处理等领域都具有广阔的应用前景.采用自然梯度法和分阶段学习法。-The source signal and transmission channel is unknown circumstances, using only observational data extraction receiving a

基于温度场下两相分离的二维格子Boltzmann模型，用于研究两相分离过程。(Based on the two-dimensional lattice Boltzmann model of two-phase separation in temperature field, the two-phase separation process is studied.)