한국정보통신학회논문지 (Journal of the Korea Institute of Information and Communication Engineering)

  • 제6권1호
  • /
  • Pages.14-24
  • /
  • 2002
  • /
  • 2234-4772(pISSN)
  • /
  • 2288-4165(eISSN)
한국정보통신학회 (The Korea Institute of Information and Commucation Engineering)
권호 표지

펄스폭변조 기법을 이용한 신경망회로 설계

A Neural Network Design using Pulsewidth-Modulation (PWM) Technique

  • 발행 : 2002.02.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 학습과 정정 기능을 갖는 PWM 뉴럴네트워크를 설계하였다. 설계된 PWM 뉴럴시스템에서, 네트워크의 입력과 출력 신호들은 PWM 신호에 의해서 표현되어진다. 뉴럴네트워크에서 곱셈은 가장 많이 사용하는 동작이다. 승산과 합산의 기능은 PWM 기술과 간단한 혼합모드 회로기술에 의해서 실현된다. 그러므로 설계된 뉴럴네트워크는 단지 소규모의 칩상에서 구현될 수가 있다. 하나의 뉴런과 세개의 시냅스, 연관된 학습회로로 설계된 네트워크회로는 양호한 선형성과 넓은 범위의 동작범위를 가지고 있다. PWM을 이용한 신경망회로의 학습능력을 검증하기 위해, 델타 학습 규칙을 적용하였다. AND 기능과 OR 기능 학습 예측 HSPICE 시뮬레이션을 통해서 설계한 신경망회로의 기능이 성공적임을 증명하였다.

In this paper, a design of the pulsewidth-modulation(PWM) neural network with both retrieving and learning function is proposed. In the designed PWM neural system, the input and output signals of the neural network are represented by PWM signals. In neural network, the multiplication is one of the most commonly used operations. The multiplication and summation functions are realized by using the PWM technique and simple mixed-mode circuits. Thus, the designed neural network only occupies the small chip area. By applying some circuit design techniques to reduce the nonideal effects, the designed circuits have good linearity and large dynamic range. Moreover, the delta learning rule can easily be realized. To demonstrate the learning capability of the realized PWM neural network, the delta learning nile is realized. The circuit with one neuron, three synapses, and the associated learning circuits has been designed. The HSPICE simulation results on the two learning examples on AND function and OR function have successfully verified the function correctness and performance of the designed neural network.

키워드

참고문헌

  1. R. Woodburn, H. M. Reekie, and A. F. Murray, 'Pulse-stream circuits for on-chip learning in analogue VLSI neural networks,' in Proc. 1994 IEEE Int. Symp. Circuits Systems, vol. 4, pp.103-106
  2. R. P. Lippmann, 'An introduction to computing with neural nets.' IEEE Acoust.,Speech, Signal Processing Mag., pp. 4-22, Apr. 1987
  3. A. F. Murray and A. V. W. Smith, 'Asynchronous VLSI neural networks using pulse stream arithmetic.', IEEE J. Solid-State Circuits, vol. 23, pp.688-697, June 1988 https://doi.org/10.1109/4.307
  4. J. E. Tomberg and K. Kaski, 'Pulse-density modulation technique in VLSI implementations of neural network algorithms, ' IEEE J. Solid- State Circuits, vol. 25, pp.1277-1286, Oct. 1990 https://doi.org/10.1109/4.62152
  5. A. F. Murray, D. Del Corso, and L. Tarassenko, 'Pulse-stream VLSI neural networks mixing analog and digital techniques,' IEEE Trans. Neural Networks, vol 2, pp. 193-204, Mar. 1991 https://doi.org/10.1109/72.80329
  6. J. Meador, A. Wu, C. Cole, N. Nintunze, and P. Chintrakulchai, 'Programmable impulse neural circuits,' IEEE Trans. Neural Networks, vol 2, pp. 101-109, Jan. 1991 https://doi.org/10.1109/72.80295
  7. A. Hamilton, A. F. Murray, D. J. Baxter, S. Churcher, H. M. Reekie, and L. Tarassenko, 'Integrated pulse stream neural networks: Results, issues, and pointers,' IEEE Trans. Neural Networks, vol 3, pp. 404-413, May. 1992 https://doi.org/10.1109/72.129413
  8. B. A. De Cock, D. Maurissens, and J. Cornelis, 'A CMOS pulse-width modulator/pulse-amplitude modulator for four-quadrant analog multipliers,' IEEE J. Solid-State Circuits, vol. 27, pp.151-154, Sept. 1992
  9. J. Ramirez-Angulo and I. Grau, 'Wide gm adjustment range, highly linear OTA with linear programmable current mirrors', Proc. IEEE Int. Symp. Circuit & System, vol. 3, pp. 1372-1375, 1992
  10. B. Gilbert, 'A precise four quadrant multiplier with subnanosecond response', IEEE J. Solid- State Circuits, vol. SC3, pp. 365-373, Dec. 1968 https://doi.org/10.1109/JSSC.1968.1049925