中文摘要：

为寻求自适应阵列天线技术在波束形成准确性和收敛速度上的突破，尝试引入智能控制理论中的优化算法．提出一种染色体长度可调的编码方式，用这种编码方式下的遗传算法同时训练RBF网络的网络参数和拓扑结构，将训练后的RBF网络用于上行链路的自适应波束形成．理论分析和仿真结果表明，染色体长度可调编码方式下的遗传算法是有效的，该算法能够比较准确地逼近理论上的最优权值，提高了算法的收敛速度．

英文摘要：

In seek of breakthrough the veracity of beamforming and rapid rapidity of convergence for adaptive array antennas, we try to introduce the intelligent control theory to optimize the algorithms. A novel coding scheme with adjustable length of chromosome is presented. The Genetic algorithms can train the network parameters and topology structure of RBF networks under the proposed scheme. Then the trained RBF networks are used to build the uplink adaptive beamforming. Theoretical analysis and simulation results show that the genetic algorithms based on adjustable chromosome length encoding method is effective, the proposed algorithms can approach the theoretical optimal weights exactly, moreover, the rapidity of convergence is improved significantly.