论文部分内容阅读
研究利用人工神经网络求解多级交换网路由重排问题，提出了一种求解路由重排问题的模糊控制神经网络算法。大量的计算机模拟结果表明，该算法与其它同类算法相比较具有收敛速度快，结果有效性高，自适应性强等特点。利用ＴＭＳ３２０Ｃ３０数字信号处理器构成了虚拟硬件系统，实时实现了多级交换网路由重排控制器，理论计算与测试结果一致。该工作为神经网络与模糊系统相结合探索了一条新的途径。
In this paper, the artificial neural network (ANN) is used to solve routing rearrangement in multilevel switching networks. A fuzzy control neural network algorithm for routing rearrangement problem is proposed. A large number of computer simulation results show that compared with other similar algorithms, the proposed algorithm has the characteristics of fast convergence, high result validity and strong adaptability. The use of TMS320C30 digital signal processor constitutes a virtual hardware system, real-time multi-level switching network routing rearrangement controller, the theoretical calculation and test results. This work explores a new way for the combination of neural network and fuzzy system.