基于混沌神经网络的最短路径路由算法

飞速发展的计算机网络对路由算法的反应速度提出了更高的要求．神经网络作为一种新的组合优化计算工具。在网络路由方面的应用得到较大关注．与传统的采用串行执行方式的算法相比，神经网络路由算法以其固有的并行执行方式，以及潜在的硬件实施能力，将成为这一领域的有力竞争者．由此提出了一种基于混沌神经网络的最短路径路由算法．仿真结果表明，该算法能有效克服Hopfield神经网络易陷入局部最优解的缺点，并且在收敛速度方面有了很大改进．

A Shortest Path Routing Algorithm Based on Chaotic Neural Networks

With the rapid development of computer networks, routing algorithms are required to achieve higher response speeds. As a new computational tool for solving combinational optimization problems, neural networks attract many attentions in the application for network routing. Compared with traditional algorithms adopting serial way, neural networks may become a powerful competitor in this field, relying on the parallel architecture and hardware approach. A shortest path routing algorithm based on chaotic neural networks is presented. Simulation results show that the algorithm can efficiently avoid being trapped in a local minimum compared with the Hopfield model, and has obvious improvement in convergence speed.