Analysis of autowave characteristics for competitive pulse coupled neural network and its application
Analysis of autowave characteristics for competitive pulse coupled neural network and its application
复制标题
DOI：
10.1016/j.neucom.2008.12.008
复制
发表时间：
2009-06
期刊：
影响因子：
6
通讯作者：
Dongming Zhou;Rencan Nie;Dongfeng Zhao
中科院分区：
文献类型：
--
作者：
Dongming Zhou;Rencan Nie;Dongfeng Zhao
针对网络最短路径问题，分析了竞争脉冲耦合神经网络(CPCNN)的脉搏波传播特性。推导出脉搏波沿网络布线图最短路径树(SPT)路径传播的神经元参数设定值。仿真结果表明，当网络路由图中含有多个节点时，该方法是一种有效的算法。随着网络节点数的增加，算法的迭代次数基本保持不变。该方法具有较好的计算性能和优势。
The characteristics of pulse wave propagation of competitive pulse coupled neural network (CPCNN) are analyzed for the solution of the network shortest routing. The setting terms of the neuron parameters for the pulse wave propagation along the routing of the shortest path tree (SPT) of the network routing graph are educed. The simulation results show that the proposed method is an effective algorithm when the network routing graph holds many nodes. With the nodes number of the network increasing, the number of iterations of our algorithm will basically hold the line. The method shows better computational performance and dominance.