Subspace tracking with adaptive threshold rank estimation

  • Aleksandar Kavčić
  • Bin Yang
Article

Abstract

In frequency and direction of arrival (DOA) tracking problems, singular value decomposition (SVD) can be used to track the signal subspace. Typically, for a problem sizen, only a few, sayr dominant eigencomponents need to be tracked, wherern. In this paper we show how to modify the Jacobi-type SVD to track only ther-dimensional signal subspace by forcing the (n-r)-dimensional noise subspace to be spherical. Therby, the computational complexity is brought down fromO(n2) toO(nr) per update. In addition to tracking the subspace itself, we demonstrate how to exploit the structure of the Jacobi-type SVD to estimate the signal subspace dimension via a simple adptive threshold comparison technique. Most available computationally efficient subspace tracking algorithms rely on off-line estimation of the signal subspace dimension, which acts as a bottleneck in real-time parallel implementations. The noise averaged Jacobi-type SVD updating algorithm presented in this paper is capable of simultaneously tracking the signal subspace and its dimension, while preserving both the low computational cost ofO(nr) and the parallel structure of the method, as demonstrated in a systolic implementation. Furthermore, the algorithm tracks all signal singular values. Their squares are estimates of the powers in the orthogonal modes of the signal. Thus, applications of the algorithm are not limited to only DOA and frequency tracking where information about the powers of signal components is not exploited.

Keywords

Singular Value Decom Singular Vector Systolic Array Minimum Description Length Signal Subspace 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    V.F. Pisarenko, “The retrieval of harmonics from a covariance function,”Geophys. J. Roy. Astron. Soc., Vol. 33, pp. 347–366, 1973.CrossRefMATHGoogle Scholar
  2. 2.
    R.O. Schmidt, “Multiple emiter location and signal parameter estimation,” inProc. RADC Spectrum Estimation Workshop, pp. 243–258, Oct. 1979.Google Scholar
  3. 3.
    R.O. Schmidt, “Multiple emiter location and signal parameter estimation,”IEEE Trans. Antenn. Propagat., Vol. AP-34, pp. 276–280, March 1986.CrossRefGoogle Scholar
  4. 4.
    R. Kumaresan and D.W. Tufts, “Estimating the angles of arrival of multiple plane waves,”IEEE Trans. Aerosp. Electron. Syst., Vol. 19, pp. 134–139, 1983.CrossRefGoogle Scholar
  5. 5.
    R. Roy and T. Kailath, “ESPRIT—Estimation of signal parameters via rotational imvariance techniques,”IEEE Trans. ASSP, Vol. 37, 1989.Google Scholar
  6. 6.
    M. Viberg, B. Ottersten, and T. Kailath, “Detection and estimation in sensor arrays using weighted subspace fitting,“IEEE Trans. Signal Processing, Vol. 39, pp. 2436–2449, 1991.CrossRefGoogle Scholar
  7. 7.
    P. Comon and G.H. Golub, “Tracking a few extreme singular values and vectors,”Proc IEEE, pp. 1327–1343, Aug. 1990.Google Scholar
  8. 8.
    R.J. Bunch, C.P. Nielsen, and D.C. Sorensen, “Rank-one modification of the symmetric eigenproblem,”Numer. Math, Vol. 31, pp. 31–48, 1979.MathSciNetCrossRefGoogle Scholar
  9. 9.
    N.L. Owsley, “Adaptive data orthogonalization,” inProc. IEEE ICASSP'78, pp. 109–112, 1978.Google Scholar
  10. 10.
    I. Karasalo, “Estimating the covariance matrix by signal subspace averaging,”IEEE Trans. ASSP, Vol. 34, pp. 8–12, Feb. 1986.MathSciNetCrossRefGoogle Scholar
  11. 11.
    J. Yang and M. Kaveh, “Adaptive eigensubspace algorithms for direction of frequency estimation tracking,”IEEE Trans. ASSP, Vol. 36, pp. 241–251, 1988.CrossRefGoogle Scholar
  12. 12.
    M. Moonen, P. Van Dooren, and J. Vandewalle, “Updating singular value decomposition. A parallel implementation,” inProc. SPIE Advanced Algorithms and Architectures for Signal Processing, San Diego, Aug. 1989, Vol. 1152, pp. 80–91.Google Scholar
  13. 13.
    R.D. DeGroat, “Noniterative subspace tracking,”IEEE Trans. Signal Processing, Vol. 4, pp. 571–577, March 1992.CrossRefGoogle Scholar
  14. 14.
    G.W. Stewart, “An updating algorithm for subspace tracking,”IEEE Trans. Signal Processing, Vol. 40, pp. 1535–1541, June 1992.CrossRefGoogle Scholar
  15. 15.
    S.Y. Kung,Digital Neural Processing, Englewood Cliffs; Prentice-Hall, 1993.Google Scholar
  16. 16.
    B. Yang and F. Gersemsky, “An adaptive algorithm of linear computational complexity for both rank and subspace tracking,” inProc. IEEE ICASSP'94, Adelaide, April 1994, pp. IV33–IV36.Google Scholar
  17. 17.
    B. Yang, “Gradient based subspace tracking algorithms and systolic implementation,”J. of High Speed Electronics and Systems, Vol. 4, pp. 203–218, 1993.CrossRefGoogle Scholar
  18. 18.
    E.M. Dowling, L.P. Amman, and R.D. DeGroat, “A TQR-iteration based adaptive SVD for real time angle and frequency tracking,”IEEE Trans. Signal Processing, Vol. 42, pp. 914–926, April 1994.CrossRefGoogle Scholar
  19. 19.
    E.M. Dowling, R.D. DeGroat, D. A. Linebarger, and H. Ye, “Sphericalized SVD updating for subspae tracking,” inSVD and Signal Processing: Algorithms, Applications and Architectures III, North Holland Publishing Co., 1995.Google Scholar
  20. 20.
    A. Kavčié and B. Yang, “A new efficient subspace tracking algorithm based on singular value decomposition,” inProceedings IEEE ICASSP'94, Adelaide, April 1994, pp. IV485–IV488.Google Scholar
  21. 21.
    A. Kavčić and B. Yang, “Simultaneous subspace tracking and rank estimation,” inProc. SPIE Advanced Signal Processing Algorithms, San Diego, July 1995, Vol. 2563, pp. 20–217.Google Scholar
  22. 22.
    S. Haykin, “Adaptive Filter Theory, Englewood Cliffs: Prentice Hall, 2nd ed., 1991.MATHGoogle Scholar
  23. 23.
    K.V. Fernando and S.J. Hammarling, “Systolic array computation of the SVD of complex matrices,” inProc. SPIE Advanced Algorithms and Architectures for Signal Processing, Vol 696, pp. 54–61, 1986.Google Scholar
  24. 24.
    G.W. Stewart, “A Jacobi-like algorithm for computing the Schur decomposition of a non-Hermitian matrix,” Tech. Report 1321, Computer Science Department, University of Maryland, 1983.Google Scholar
  25. 25.
    F.T. Luk, “A triangular processor array for computing the singular value decomposition,” Tech. Report 84-625, Department of Computer Science, Cornell University, 1984.Google Scholar
  26. 26.
    J.-P. Charlier and P. Van Dooren, “A Jacobi-like algorithm for computing the generalized Schur form of a regular pencil,”J. Comp. Appl. Math., Vol. 27, pp. 17–36, 1989.CrossRefMATHGoogle Scholar
  27. 27.
    A. Akaike, “A new look at the statistical model identification,”IEEE Trans. Automatic Control, Vol. 19, pp. 716–723, 1974.MathSciNetCrossRefMATHGoogle Scholar
  28. 28.
    J. Rissanen, “Modelling by shortest data description,”Automatica, Vol. 14, pp. 465–471, 1978.CrossRefMATHGoogle Scholar
  29. 29.
    G. Schwartz, “Estimating the dimension of a model,”Ann. Stat., Vol. 6, pp. 461–464, 1978.CrossRefGoogle Scholar
  30. 30.
    M. Wax and T. Kailath, “Detection of signals by information theoretic criteria,”IEEE Trans. ASSP, Vol. 33, pp. 387–392, 1985.MathSciNetCrossRefGoogle Scholar
  31. 31.
    T.W. Anderson,An Introduction to Multivariate Statistical Analysis, New York: John Wiley and Sons, 2nd ed., 1984.MATHGoogle Scholar
  32. 32.
    H.L. Van Trees,Detection, Estimation, and Modulation Theory, New York: John Wiley and Sons, Vol. I, 1968.MATHGoogle Scholar
  33. 33.
    M. Moonen,Jacobi-Type Updating Algorithms for Signal Processing, Systems Identification and Control, Ph.D.thesis, Department of Electrical Engineering, Katholike Universiteit Leuven, Heverlee, Belgium, 1990.Google Scholar
  34. 34.
    E.M. Dowling, R.D. DeGroat, D.A. Linebarger, and Z. Fu, “Real time architecture for sphericalized SVD updating,” inSVD and Signal Processing: Algorithms, Applications and Architectures III, North Holland Publishing Co., 1995.Google Scholar
  35. 35.
    L.L. Scharf,Statistical Signal Processing, New York: Addison-Wesley Publishing Company, 1991.MATHGoogle Scholar

Copyright information

© Kluwer Academic Publishers 1996

Authors and Affiliations

  • Aleksandar Kavčić
    • 1
  • Bin Yang
    • 2
  1. 1.Department of Electrical and Computer EngineeringCarnegie Mellon UniversityPittsburgh
  2. 2.Department of Electrical EngineeringRuhr University BochumBochumGermany

Personalised recommendations