Advertisement

Circuits, Systems, and Signal Processing

, Volume 33, Issue 1, pp 257–274 | Cite as

Multichannel Time-Delay and Signal Model Estimation with Missing Observations

  • Jui-Chung HungEmail author
Article
  • 211 Downloads

Abstract

In this paper, we propose a maximum likelihood (ML) estimator in the frequency domain for estimating multichannel time delay and parameters with missing observations. The missing observations are described by a random Bernoulli pattern. In this context, the ML estimator for missing observations is highly sensitive to the initial conditions and complexity of a given problem. In conventional calculations, the complexity of problems will often make it difficult to obtain the optimal results. Thus, we adopted an iterative method using a genetic algorithm (GA) to develop an ML estimator for a model signal, the time delay, and the missing probability in the frequency domain.

Several simulation examples were analyzed to evaluate the performance of the proposed method. The simulation results show that the performance is significantly improved if the effect of missing observations on the ML estimator is considered.

Keywords

Maximum likelihood Missing observations Time delay Genetic algorithm 

Notes

Acknowledgements

The author is grateful to the Editor-in-Chief, Prof. M.N.S. Swamy and anonymous referees, whose constructive and helpful comments led to significant improvements in the manuscript. This research was supported by the National Science Council under grant number NSC 101-2221-E-133-005.

References

  1. 1.
    H. Akaike, A new look at statistical model identification. IEEE Trans. Autom. Control 19(6), 716–723 (1974) CrossRefzbMATHMathSciNetGoogle Scholar
  2. 2.
    K.J. Astrom, B. Wittenmark, Computer Controlled Systems: Theory and Design (Prentice Hall, Englewoods Cliffs, 1984) Google Scholar
  3. 3.
    P.M.T. Broersen, S. de Walele, R. Bos, Application of autoregressive spectral analysis to missing data problems. IEEE Trans. Instrum. Meas. 53(4), 981–986 (2004) CrossRefGoogle Scholar
  4. 4.
    O. Capp’e, S.J. Godsill, E. Moulines, An overview of existing methods and recent advances in sequential Monte Carlo. Proc. IEEE 95(5), 899–924 (2007) CrossRefGoogle Scholar
  5. 5.
    J.M. Chen, B.S. Chen, System parameter estimation with input/output noisy data and missing measurement. IEEE Trans. Signal Process. 48(6), 1548–1558 (2000) CrossRefGoogle Scholar
  6. 6.
    B.S. Chen, J.C. Hung, A global estimation for multichannel time delay and signal parameters via genetic algorithm. Signal Process. 81(5), 1061–1067 (2001) CrossRefzbMATHGoogle Scholar
  7. 7.
    J. Chiasson, J.J. Loiseau, Application of Time-Delay Systems (Springer, Berlin, 2007) CrossRefGoogle Scholar
  8. 8.
    R.C. David, E.M. Petriu, S. Preitl, M.B. Radac, Fuzzy control systems with reduced parametric sensitivity based on simulated annealing. IEEE Trans. Ind. Electron. 59(8), 3049–3061 (2012) CrossRefGoogle Scholar
  9. 9.
    J. Feng, Z. Wang, M. Zeng, Recursive robust filtering with finite-step correlated process noises and missing measurements. Circuits Syst. Signal Process. 30(6), 1355–1368 (2011) CrossRefzbMATHMathSciNetGoogle Scholar
  10. 10.
    D.E. Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning (Addison-Wesley, Boston, 1989) zbMATHGoogle Scholar
  11. 11.
    J.H. Holland, Adaptation in Natural and Artificial Systems (University of Michigan Press, Ann Arbor, 1975) Google Scholar
  12. 12.
    J.C. Hung, Modified particle swarm optimization structure approach to direction of arrival estimation. Appl. Soft Comput. 13(1), 315–320 (2013) CrossRefGoogle Scholar
  13. 13.
    A.J. Isaksson, A recursive EM algorithm for identification subject to missing data, in Proc. IFAC Symposium on System Identification, Copenhagen, Denmark (1994), pp. 679–684 Google Scholar
  14. 14.
    R.H. Jones, Maximum likelihood fitting of ARMA model to time series with missing observations. Technometrics 20(3), 389–395 (1980) CrossRefGoogle Scholar
  15. 15.
    S.M. Kay, Modern Spectral Estimation: Theory and Application (Prentice Hall, Englewoods Cliffs, 1988) zbMATHGoogle Scholar
  16. 16.
    Yu.S. Kharin, A.S. Huryn, Plug-in statistical forecasting of vector autoregressive time series with missing values. Aust. J. Stat. 34(2), 163–174 (2005) Google Scholar
  17. 17.
    C. Knapp, G. Carter, The generalized correlation method for estimation of time delay. IEEE Trans. Acoust. Speech Signal Process. 24(4), 320–327 (1976) CrossRefGoogle Scholar
  18. 18.
    Y. Liu, F. Ding, Y. Shi, Least squares estimation for a class of non-uniformly sampled systems based on the hierarchical identification principle. Circuits Syst. Signal Process. 31(6), 1985–2000 (2012) CrossRefzbMATHMathSciNetGoogle Scholar
  19. 19.
    R. Lu, H. Li, Y. Zhu, Quantized H filtering for singular time-varying delay systems with unreliable communication channel. Circuits Syst. Signal Process. 31(2), 521–538 (2012) CrossRefzbMATHMathSciNetGoogle Scholar
  20. 20.
    E.A. Parzen, On spectral analysis with missing observations and amplitude modulation. Sankhya 25(4), 323–392 (1963) MathSciNetGoogle Scholar
  21. 21.
    H. Raghavan, A.K. Tangirala, R.B. Gopaluni, S.L. Shah, Identification of chemical processes with irregular output sampling. Control Eng. Pract. 14(5), 467–480 (2006) CrossRefGoogle Scholar
  22. 22.
    Y. Rosen, B. Porat, Optimal ARMA parameter estimation based on the sample covariances for data with missing observations. IEEE Trans. Inf. Theory 35(2), 342–349 (1989) CrossRefzbMATHMathSciNetGoogle Scholar
  23. 23.
    M. Segal, E. Weinstein, B.R. Musicus, Estimate maximize algorithm for multichannel time delay and signal estimation. IEEE Trans. Signal Process. 39(1), 1–16 (1991) CrossRefzbMATHGoogle Scholar
  24. 24.
    R.M. Udrea, N.D. Vizireanu, S. Ciochi, An improved spectral subtraction method for speech enhancement. Digit. Signal Process. 18(4), 581–587 (2008) CrossRefGoogle Scholar
  25. 25.
    R.M. Udrea, N. Vizireanu, S. Ciochina, S. Halung, Nonlinear spectral subtraction method for colored noise reduction using multi-band Bark scale. Signal Process. 88(5), 1299–1303 (2008) CrossRefzbMATHGoogle Scholar
  26. 26.
    D.N. Vizireanu, R.O. Preda, Is “five-point” estimation better than “three-point” estimation. Measurement 46(1), 840–842 (2013) CrossRefGoogle Scholar
  27. 27.
    Z. Xiang, C. Liang, M.S. Mahmoud, Robust l 2l filtering for switched time delay systems with missing measurements. Circuits Syst. Signal Process. 31(5), 1677–1697 (2012) CrossRefzbMATHMathSciNetGoogle Scholar
  28. 28.
    A. Zeira, Realizable lower bounds for time-delay estimation. 2. Threshold phenomena. IEEE Trans. Signal Process. 42(5), 1001–1007 (1994) CrossRefGoogle Scholar
  29. 29.
    R.F. Zgheib, G.A. Fleury, E. Lahalle, Lattice algorithm for adaptive stable identification and robust reconstruction of nonstationary AR processes with missing observations. IEEE Trans. Signal Process. 56(7), 2746–2754 (2008) CrossRefMathSciNetGoogle Scholar
  30. 30.
    L. Zhang, X. Wu, On the application of cross correlation function to subsample discrete time-delay estimation. Digit. Signal Process. 16(6), 682–694 (2006) CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  1. 1.Department of Computer ScienceTaipei Municipal University of EducationTaipeiTaiwan

Personalised recommendations