復(fù)數(shù)FIR DF設(shè)計的神經(jīng)網(wǎng)絡(luò)優(yōu)化方法
A NOVEL NEURAL NETWORK-BASED APPROACH FOR DESIGNING COMPLEX FIR FILTERS
-
摘要: 本文基于人工神經(jīng)網(wǎng)絡(luò)(ANN)能量函數(shù)優(yōu)化理論,提出了一種FIR數(shù)字濾波器(DF)神經(jīng)網(wǎng)絡(luò)優(yōu)化設(shè)計(NNO)方法的理論框架。該理論將實(shí)數(shù)與復(fù)數(shù)FIR DF設(shè)計工作統(tǒng)一起來。表征設(shè)計質(zhì)量的加權(quán)均方誤差被當(dāng)作ANN能量函數(shù),以此導(dǎo)出FIR-NNO的Lyapunov方程。文中說明了算法實(shí)現(xiàn)的基本原則,并給出了兩個實(shí)數(shù)線性相位和一個復(fù)數(shù)非線性相位FIR DF設(shè)計實(shí)例。通過與其它幾種方法的比較證明了該方法的有效性。Abstract: A novel complex FIR filter design approach based on Neural Network Optimiza-tion(NNO) technique is proposed in this paper. To demonstrate the feasibility of the NNO design approach, the weight least mean square criterion between the disired frequency response and the designed filter response is defined as the Lyapunov energy function of a continuous Hopfield network, and the network state equations are drived. The implementation of the NNO approach is described together with some design guidelines. A few design examples are given and the advantages of NNO approach over conventional methods are illustrated.
-
Xu D J, Daley M L. Design of optimal digital filter using a parallel genetic algorithm. IEEE Trans. on CAS- n，1995,42(10): 673-675.[2]Karam L J. Complex Chebyshev approximation for FIR filter design. IEEE Trans. on CAS-II, 1995, 42(3): 207-215.[3]Komodromes M Z, et al. Design of FIR filter with complex desired frequency response using a generalized Remez algorithm. IEEE Trans. on CAS- II，1995, 42(2): 274-278.[4]焦李成.神經(jīng)網(wǎng)絡(luò)計算.西安:西安電子科技大學(xué)出版社,1993,第二章.[5]Hopfield J J, Tank D W. Neural computation of decision on optimization problems. Bio. Cyb. 1985, 52(3): 141-152.[6]Lightstone M, et al. Efficient frequency-sampling design of one- and two-dimensional FIR filters using structural subband decomposition. IEEE Trans. on CAS- II，1994,41(3): 189-201.[7]Zhao H.[J].Yu J-B. A novel neural network-based approach for design digital filters, in Proc. ISCAS 97/IEEE, Hong Kong.2272,1997:- -
計量
- 文章訪問數(shù): 2612
- HTML全文瀏覽量: 171
- PDF下載量: 425
- 被引次數(shù): 0
下載:
