期刊文献+

硬件实现基于BP神经网络设计的带阻FIR滤波器 被引量:1

Implementation of band-block FIR filters designed based on BP neural network
下载PDF
导出
摘要 针对带阻FIR滤波器传统设计方法的不足和缺陷,在改进BP神经网络设计带阻FIR滤波器方法的基础上,提出了BP神经网络设计FIR滤波器的硬件实现方法.通过神经网络训练使得实际滤波器的幅度响应逼近理想滤波器的幅度响应,训练得到带阻滤波器系数,再利用数字信号处理芯片(DSP),设计适当的FIR滤波器结构以及硬件结构,从硬件上实现带阻FIR滤波器.试验结果显示,用该方法设计的带阻FIR滤波器克服了传统方法的主要缺陷,具有良好的幅频特性及衰耗特性,并且边界频率控制精确. A new method of implementing the band-block FIR filters was presented to solve the main problems of the conventional methods, based on the improved method of band-block FIR filter design using BP neural network. In the first step, the coefficient of the filters was obtained through the training of the BP neural network. Then based on digital signal process CMOS chip, the architectures of both the block FIR filters and the hardware were designed to implement the band-block finite impulse response (FIR) hatters. The experimental results showed that it overcame the main defects of the conventional methods, and that the filter had good amplitude, frequency and attenuation performance, and that the edge frequency could be set exactly.
出处 《浙江大学学报(工学版)》 EI CAS CSCD 北大核心 2006年第7期1146-1149,1163,共5页 Journal of Zhejiang University:Engineering Science
关键词 BP神经网络 衰耗特性 数字信号处理器 带阻滤波器 BP neural network attenuation performance digital signal process band-block filters
  • 相关文献

参考文献12

  • 1PREUSS K P.Novel approach for complex chebyshevapproximation with FIR filters using the Remez exchange algorithm[C]// Proceedings-ICASSP,IEEE International Conference on Acoustics,Speech and Signal Processing.United States:IEEE,1987:872-875.
  • 2KOBAYASHI T,IMAI S.Design of Arma digital filters with Arbitrary log magnitude function by WLS techniques[C] // Proceedings-ICASSP,IEEE International Conference on Acoustics,Speech and Signal Processing.United States:IEEE,1998,3:1447-1450.
  • 3TRISTAN C.Combining FIR filters and artificial neural networks to model stochastic processes[C] // Conference Record of the Asilomar Conference on Signals,Systems and Computers.United States:IEEE,2002,1:333-336.
  • 4ZHANG S,SALANI E.Reducing artifacts in coded images using a neural network aided adaptive FIR filter[J].Neurocomputing,2003,50:249-269.
  • 5BHATTACHARYA D.Design of equiripple FIR filters using a feedback neural network[J].IEEE Transactions on Circuits and Systems Ⅱ:Analog and Digital Signal Processing,1998,45(4):527-531.
  • 6KYPRAIOS I I.Performance assessment of unconstrained hybrid optical neural network (U-HONN) filter for object recognition tasks in clutter[J].Proceedings of SPIE-The International Society for Optical Engineering,2004,5437:51-62.
  • 7KYPRAIOS I I.Object recognition within cluttered scenes employing a hybrid optical neural network filter[J].Optical Engineering,2004,43 (8):1839-1850.
  • 8孙宗瀛,谢鸿琳.TMS320C5X DSP原理设计与应用[M].北京:清华大学出版社,2003.
  • 9HAGANMT DEMUTHHB BEALEMH.Neural networks design[M].北京:机械工业出版社,2003..
  • 10RAMANATHAN S,NANDY S K,VISYANATHAN V.Reconfigurable filter coprocessor architecture for DSP applications[J].Journal of VLSI Signal Processing,2000,26(3):333-359.

二级参考文献2

  • 1彭启棕.TMS320C54X使用教程[M].成都:电子科技大学出版社,1998.5-34.
  • 2哈里Y-F拉姆 陆延丰(译).模拟和数字滤波器的设计与实现[M].北京:人民邮电出版社,1985.586-695.

共引文献3

同被引文献12

引证文献1

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部