基于改进进化规划的RBF网络二相码旁瓣优化

RBF Neural Network Based on Improved Evolutionary Planning Used in Binary-coded Side-lobe Optimization

在研究雷达脉冲压缩信号中的二相编码距离旁瓣压缩问题基础上,采用改进的进化规划算法优化径向基函数网络,提出了一种基于改进进化规划算法的计算方法。该算法将进化规划算法和神经网络结合起来,用径向基RBF(Radial Basis Function)神经网络结构取代多层前馈网络MFNN(Multilayer Feedforward Neural Nerworks)结构,用改进进化规划取代反向传播算法BP(Back Propagation),可以以较大概率快速的收敛到要求的旁瓣抑制指标。通过对13位巴克码和31位M编码的仿真实验,表明新的方法在旁瓣抑制能力和运算速度等性能方面,都有较大的提高。

Based on researching binary-coded range side-lobe suppression problem for radar pulse compressed signal, the radial basis function (RBF) neural network was optimized with the improved evolutionary planning algorithm (IEPA), a new algorithm based on the improved evolutionary planning algorithm was presented. IEPA was combined with neural network by the new algorithm, multi-layer feed-forward neural network structure was replaced with RBF neural network structure, and back propagation (BP) algorithm was replaced with IEPA, then, expected side-lobe suppression quota was rapidly waisted. The emulation experiments for the 13-elements baker code and 31-elements M code show that side-lobe suppression ability and operating speed was improved with the new method.