作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (03): 253-255. doi: 10.3969/j.issn.1000-3428.2007.03.092

• 工程应用技术与实现 • 上一篇    下一篇

模糊CMAC的硬件结构分析及其FPGA实现

沈宪明,白瑞林,章智慧   

  1. (江南大学控制科学与工程研究中心,无锡 214122)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-02-05 发布日期:2007-02-05

Hardware Structure Analysis of Fuzzy CMAC
and Its FPGA Implementation

SHEN Xianming, BAI Ruilin, ZHANG Zhihui   

  1. (Research Center of Control Science and Engineering, Southern Yangtze University, Wuxi 214122)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-02-05 Published:2007-02-05

摘要: 提出了模糊CMAC的一种基于FPGA的硬件实现方法。与其它FPGA实现的神经网络相比,包含了可以用于在线学习的权学习算法。分析了模糊CMAC的模型结构及其相应的硬件模块;用VHDL实现基于上述模块的模糊CMAC;对该模糊CMAC进行硬件综合与测试。测试结果表明:该模糊CMAC的FPGA实现方法是可行的,硬件化后的网络具有速度快、精度高、占用器件资源少的特点,是在SOPC中实现模糊CMAC模块的一种有效方法。

关键词: 模糊CMAC, FPGA, VHDL, 函数学习

Abstract: This paper proposes a FPGA implementation structure of a fuzzy CMAC. Compared with other neural networks implemented by FPGA, it contains the learning algorithm which can be employed to realize the on-line learning. The model and the relevant hardware modules of fuzzy CMAC are analyzed. It implements the fuzzy CMAC based on the above hardware modules with VHDL. The design is synthesized and tested. The test result shows that the method of the hardware implementation of the fuzzy CMAC is feasible. The implemented network comprises the characteristics of high speed, good precision and little chip resource. It is an effective method in implementing the module of fuzzy CMAC in SOPC.

Key words: Fuzzy CMAC(cerebellar model articulation controller), FPGA, VHDL, Function identification