The Journal of the Korea institute of electronic communication sciences (한국전자통신학회논문지)

Korea Institute of Electronic Communication Science (한국전자통신학회)
권호 표지
DOI QR코드

DOI QR Code

Generation of Pattern Classifier using LFSRs

LFSR을 이용한 패턴분류기의 생성

  • 권숙희 (부경대학교 응용수학과) ;
  • 조성진 (부경대학교 응용수학과) ;
  • 최언숙 (동명대학교 정보통신공학과) ;
  • 김한두 (인제대학교 응용수학과) ;
  • 김나령 (부경대학교 응용수학과)
  • Received : 2014.04.28
  • Accepted : 2014.06.16
  • Published : 2014.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

The important requirements of designing a pattern classifier are high throughput and low memory requirements, and low cost hardware implementation. A pattern classifier by using Multiple Attractor Cellular Automata(MACA) proposed by Maji et al. reduced the complexity of the classification algorithm from $O(n^3)$ to O(n) by using Dependency Vector(DV) and Dependency String(DS). In this paper, we generate a pattern classifier using LFSR to improve efficiently the space and time complexity and we propose a method for finding DV by using the 0-basic path. Also we investigate DV and the attractor of the generated pattern classifier. We can divide an n-bit DS by m number of $DV_i$ s and generate various pattern classifiers.

패턴분류기 설계의 중요한 조건은 데이터 처리량이 크고 저장 공간은 작고 낮은 가격대로 구현하는 것이다. Maji 등에 의해 제안된 MACA 기반의 패턴분류기는 DV와 DS를 사용하여 복잡도를 $O(n^3)$에서 O(n)으로 줄였다. 본 논문에서는 효율적으로 시간과 공간의 복잡성을 개선하기 위해 LFSR 기반 패턴 분류기를 생성하고 0-기본경로를 이용하여 DV를 구할 수 있는 방법을 제안한다. 그리고 생성한 패턴분류기의 DV와 끌개에 대해 살펴본다. n-비트 DS=(11 ${\cdots}$ 11)를 m개의 $DV_i$로 분할할 수 있고 다양한 패턴분류기를 생성할 수 있다.

Keywords

References

  1. S. Chattopadhyay, S. Adhikari, S. Sengupta, and M. Pal, "Highly Regular, Modular, and Cascadable Design of Cellular Automata-Based Pattern Classifier," IEEE Trans. VLSI Systems, vol. 8, no. 6, 2000, pp. 724-735. https://doi.org/10.1109/92.902267
  2. P. Maji, C. Shaw, N. Ganguly, B. K. Sikdar, and P. P. Chaudhuri, "Theory and application of cellular automata for pattern classification," Fundamenta Informaticae, vol. 58, 2003, pp. 321-354.
  3. S.-J. Cho, U.-S. Choi, and H.-D. Kim, "Analysis of Complemented CA Derived from a linear TPMACA," Computers and Mathematics with Applications, vol. 45, 2003, pp. 689-698. https://doi.org/10.1016/S0898-1221(03)00028-2
  4. Y.-H. Hwang, U.-S. Choi, and S.-J. Cho, "D1-MACA based Two-Class Pattern Classifier," J. of The Korea Institute of Electronic Communication Sciences, vol. 3, no. 4, 2008, pp. 269-274.
  5. Y.-H. Hwang, S.-J. Cho, and U.-S. Choi, "Multiple Attractor CA Based Pattern Classifier," J. of The Korea Institute of Electronic Communication Sciences, vol. 5, no. 3, 2010, pp. 315-320.
  6. H.-D. Kim, S.-J. Cho, M.-J. Kwon, and H.-J. An, "A study on the cross-correlation sequences," J. of The Korea Institute of Electronic Communication Sciences, vol. 7, no. 1, 2012, pp. 61-67.
  7. S.-J. Cho, U.-S. Choi, H.-D. Kim, and H.-J. An, "Analysis of nonlinear sequences based on shrinking generator," J. of The Korea Institute of Electronic Communication Sciences, vol. 5, no. 4, 2010, pp. 412-417.
  8. U.-S. Choi and S.-J. Cho, "Design of binary sequences with optimal cross-correlation values," J. of The Korea Institute of Electronic Communication Sciences, vol. 6, no. 4, 2011, pp. 539-544.
  9. M. Yilmaz, K. Chakrabarty, and M. Tehranipoor, "Test-pattern grading and pattern selection for small-delay defects," In Proc., IEEE VTS, 2008, pp. 233-239.
  10. S. Golomb, Shift Register Sequences. Aegean Park Press, 1967.
  11. G. L. Mullen and D. Panario, Handbook of finite fields. Chapman and Hall/CRC, 2013.
  12. P. P. Chaudhuri, D. R. Chowdhury, S. Nandy, and S. Chattopadhyay, Additive Cellular Automata; Theory and Applications. IEEE Computer Society Press, vol. 1, 1997.
  13. S.-J. Cho, U.-S. Choi, H.-D. Kim, Y.-H. Hwang, and J.-G. Kim, "Analysis of 90/150 Two Predecessor Nongroup Cellular Automata," ACRI 2008, LNCS 5191, 2008, pp. 128-135.
  14. R. A. Horn and C. R. Johnson, Matrix analysis. Cambridge University Press, 1985.
  15. S.-J. Cho, H.-D. Kim, U.-S. Choi, S.-T. Kim, J.-K. Kim, S.-H. Kwon, and G.-T. Gong, "Generation of TPMACA for Pattern Classification," ACRI 2014(submitted), 2014.