引入神经网络模型和算法来解决移动通信系统中信道分配问题，采用改进的Ｋｏｈｏｎｅｎ自组织模型，提出了在此模型下的信道分配神经网络算法，给出了仿真结果。结果表明神经网络算法解决分布式控制的动态信道分配门题是有效的，它对ＦＤＭＡ、ＴＤＭＡ、ＣＤＭＡ移动通信系统信道分配有重要参考价值，特别是对大容量、大地域的移动通信系统（如ＧＳＭ系统、移动卫星通信系统）的信道分配有特殊的意义。 The neural network model and algorithm are introduced to solve the channel assignment problem in mobile communication system. An improved Kohonen self-organizing model is proposed and the channel assignment neural network algorithm is proposed. The simulation results are given. The results show that the neural network algorithm is effective to solve the dynamic channel assignment of distributed control. It has important reference value for the channel assignment of FDMA, TDMA and CDMA mobile communication systems, especially for large capacity and large area mobile communication system GSM system, mobile satellite communication system) channel allocation is of special significance.