Introduction of Orthonormal Transform into Neural Filter for Accelerating Convergence Speed(Special Section on Intelligent Signal and Image Processing)

スポンサーリンク

概要

• 論文の詳細を見る

• As the nonlinear adaptive filter, the neural filter is utilized to process the nonlinear signal and / or system.However, the neural filter requires large number of iterations for convergence.This letter presents a new structure of the multilayer neural filter where the orthonormal transform is introduced into all inter-layers to accelerate the convergence speed.The proposed structure is called the transform domain neural filter(TDNF)for convenience.The weights are basically updated by the Back-Propagation(BP)algorithm but it must be modified since the error back-propagates through the orthogonal transform.Moreover, the variable step size which is normalized by the transformed signal power is introduced into the BP algorithm to realize the orthonormal transform.Through the computer simulation, it is confirmed that the introduction of the orthonormal transform is effective for speedup of convergence in the neural filter.

• 社団法人電子情報通信学会の論文
• 2000-02-25

著者

• Nakanishi I

  Tottori Univ. Tottori‐shi Jpn

• ITOH Yoshio

  the Faculty of Engineering, Tottori University

• FUKUI Yutaka

  the Faculty of Engineering, Tottori University

• Itoh Y

  Faculty Of Engineering Tottori University

• Itoh Yoshio

  Faculty Of Engineering Tottori University

• Itoh Yoshio

  Fac. Of Engineering Tottori Univ.

• Fukui Y

  Faculty Of Engineering Tottori University

• Fukui Y

  Graduate School Of Engineering Tottori University

• NAKANISHI Isao

  the Faculty of Education & Regional Sciences, Tottori University

• Itoh Y

  Tottori Univ. Tottori‐shi Jpn

• Fukui Yutaka

  The Faculty Of Engineering Tottori University

関連論文

• Parallel Composition Based Adaptive Notch Filter : Performance and Analysis(Digital Signal Processing)
• An Adaptive Algorithm for Cascaded Notch Filter with Reduced Bias
• On the Necessity of Estimating the Transfer Level in an Allpass-FIR ADF by the Use of Lyapunov Criteria
• A Newton Based Adaptive Algorithm for IIR ADF Using Allpass and FIR Filter
• A Simple Algorithm for Adaptive Allpass-FIR Digital Filter Using Lattice Allpass Filter with Minimum Multipliers
• A New Linear Prediction Filter Based Adaptive Algorithm For IIR ADF Using Allpass and Minimum Phase System
• Robust DSP Type Squaring Loop withAdaptive Notch Filter Type Frequency Estimator and Adaptive BPF (Special Section on Noise Cancellation and Reduction Techniques)
• A Study on Semi-Blind Adaptive equalizer for PSK and QPSK Signals
• A Variable Step Size Algorithm for Speech Noise Reduction Method Based on Noise Reconstruction System
• Two-microphone system using linear prediction and noise reconstruction
• Method estimating reflection coefficients of adaptive lattice filter and its application to system identification
• A Noise Reduction System for Wideband and Sinusoidal Noise Based on Adaptive Line Enhancer and Inverse Filter (Digital Signal Processing)
• A Noise Reduction Method for Non-stationary Noise Based on Noise Reconstruction System with ALE(Digital Signal Processing)
• A Noise Reduction Method Based on Linear Prediction Analysis
• A New Noise Reduction Method Using Estimated Noise Spectrum(Special Section of Selected Papers from the 14th Workshop on Circuits and Systems in Karuizawa)
• Threshold Equalization for On-Line Signature Verification
• DWT Domain On-Line Signature Verification Using the Pen-Movement Vector(Information Security)
• Speech Noise Reduction System Based on Frequency Domain ALE Using Windowed Modified DFT Pair(Selected Papers from the 18th Workshop on Circuits and Systems in Karuizawa)
• Multi-Matcher On-Line Signature Verification System in DWT Domain(Information Hiding, Cryptography and Information Security)
• Introduction of Orthonormal Transform into Neural Filter for Accelerating Convergence Speed(Special Section on Intelligent Signal and Image Processing)
• Reduction of Computational Complexity in the IA Algorithm
• Performance Analysis and Improvement of the NACF Algorithm (Special Section on Digital Signal Processing)
• An Adaptive Algorithm with Variable Step-Size for Parallel Notch Filter (Digital Signal Processing)
• A Noise Reduction Method Based on Linear Prediction with Variable Step-Size(Selected Papers from the 17th Workshop on Circuits and Systems in Karuizawa)
• Isolation and characterization of a cDNA clone of UDP-galactose : flavonoid 3-O-galactosyltransferase expression in the seedling of Vigna mungo
• PLL Frequency Synthesizer with Binary Phase Comparison (Special Section of Selected Papers from the 12th Workshop on Circuit and Systems in Karuizawa)
• Prescaler PLL Frequency Synthesizer with Multi-Programmable Divider (Special Section of Selected Papers from the 12th Workshop on Circuit and Systems in Karuizawa)
• Prescaler PLL Frequency Synthesizer Introducing Multi-(N+1/2) Programmable Dividers
• A New Prescaler PLL Frequency Synthesizer with Multi-Programmable Divider
• Speedup of Frequency Switching Time in PLL Frequency Synthesizers Using a Target Frequency Detector (Special Section on Selected Papers from the 11th Workshop on Circuits and Systems in Karuizawa)
• PLL Frequency Synthesizer with Multi-Phase Detector (Special Section on Selected Papers from the 11th Workshop on Circuits and Systems in Karuizawa)
• A Multi-Loop PLL Frequency Synthesizer
• Speedup of Frequency Switching Time in PLL Frequency Synthesizers Using Target Frequency Detector
• A Novel (N+1/2) Pulse Swallow Divider for the PLL Frequency Synthesizer
• A New Lowpass Filter for the Fast Frequency Acquisition of the PLL Frequency Synthesizer
• PLL Frequency Synthesizer for Low Power Consumption (Special Section of Selected Papers from the 9th Karuizawa Workshop on Circuits and Systems)
• Cancellation Technique of Parasitics in Active Filter Design (Special Section on JTC-CSCC '92)
• A Realization of Multiple Circuit Transfer Functions Using OTA-C Integrator Loop Structure
• Electronically Tunable Current-Mode Biquad Using OTAs and Grounded Capacitors
• The Automatic Counting of Chlorella Using Image Processing and Neural Network(Special Section of Selected Papers from the 13th Workshop on Circuits and Systems in Karuizawa)
• Improvement of Active Net Model for Region Detection in an Image(Special Section of Selected Papers from the 13th Workshop on Circuits and Systems in Karuizawa)
• Region Extraction Using Color Feature and Active Net Model in Color Image (Special Section on Selected Papers from the 11th Workshop on Circuits and Systems in Karuizawa)
• 4d PHOTOIONIZATION OF Xe^+, Xe^, AND Xe^ IONS
• DEVELOPMENT OF THE MULTIPLE COINCIDENCE DATA ACQUISITION SYSTEM FOR USING TWO-DIMENSIONAL POSITION SENSITIVE DETECTORS

もっと見る 閉じる

スポンサーリンク