Abstract: With the increasing demand, the available spectrum is becoming scarce in cellular mobile communication network. It is important to scheme frequency resources by applying a better optimized algorithm. To solve this problem, a master-
slave cellular model parallel genetic algorithm is presented, which is characterized by dynamic change of neighborhood units in the process of evolution. It can maintain the diversity of the population in the evolutionary process and prevent the algorithm from the local optimum premature convergence, and enhance the global search capability. Simulation results show that the proposed algorithm for channel allocation problem has a stronger search capability and faster convergence rate. It has certain value of practical use.

Key words: cellular network, channel allocation, cell model parallel genetic algorithm

摘要： 在蜂窝移动通信中，用户需求的不断增长使可用的频谱资源日益紧缺，采用较优的优化算法进行信道分配就显得非常重要。针对这一问题，首次提出了细粒度主从式并行遗传算法，其特点是在进化的过程中动态改变邻域单元，来维持进化过程中的种群多样性，防止算法进入局部最优而过早收敛，增强算法的全局搜索能力。实验仿真表明，该算法对信道分配问题有很强的最优解搜索能力，收敛速度也较快，具有一定的实用价值。

关键词: 蜂窝网络, 信道分配, 细粒度并行遗传算法