Presentation 2002/3/13
FPGA Implementation of Pulsed Neural Network with GHA Learning Rule Based on ΔΣ Modulation
Yoshimitsu MURAHASHI, Shinji DOKI, Shigeru OKUMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Pulsed Neural Network model based on Delta-Sigma modulation (DSM-PNN) with GHA learning rule had been proposed and shown its availability by hardware simulation. In this article, we proposed hardware implementation methods of DSM-PNN which is suitable for digital hardware implementation. DSM-PNN was implemented in FPGA and evaluated its performance. Neural networks which consist of multi-bit commection (M-NN) also were constructed and compared to the proposed method from the viewpoint of circuit scale and signal processing. The experimental results of the proposed mnethod were comparable to one by floating-point calculation. And the proposed method could be implemented more compact than M-NN.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Delta-Sigma Modulation / Pulsed Neural Network / FPGA Implementation / PCA / GHA
Paper # NC2001-212
Date of Issue

Conference Information
Committee NC
Conference Date 2002/3/13(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) FPGA Implementation of Pulsed Neural Network with GHA Learning Rule Based on ΔΣ Modulation
Sub Title (in English)
Keyword(1) Delta-Sigma Modulation
Keyword(2) Pulsed Neural Network
Keyword(3) FPGA Implementation
Keyword(4) PCA
Keyword(5) GHA
1st Author's Name Yoshimitsu MURAHASHI
1st Author's Affiliation Dept.of Electrical Eng.,Graduate School of Eng.,Nagoya University()
2nd Author's Name Shinji DOKI
2nd Author's Affiliation Dept.of Electrical and Electronic Eng., Fac.of Eng.,Mie University
3rd Author's Name Shigeru OKUMA
3rd Author's Affiliation Dept.of Electrical Eng.,Graduate School of Ebg.,Nagoya University
Date 2002/3/13
Paper # NC2001-212
Volume (vol) vol.101
Number (no) 737
Page pp.pp.-
#Pages 8
Date of Issue