ISSN 0869-6632 (Print)
ISSN 2542-1905 (Online)

For citation:

Anosov O. L., Butkovskij O. J., Kravtsov Y. A. Reconstruction of dynamical systems from chaotic time series: short review. Izvestiya VUZ. Applied Nonlinear Dynamics, 2000, vol. 8, iss. 1, pp. 29-51.

This is an open access article distributed under the terms of Creative Commons Attribution 4.0 International License (CC-BY 4.0).
Full text PDF(Ru):
PDF icon 2000no1p029.pdf
(downloads: 0)
Language: 
Russian
Heading: 
Review of Actual Problems of Nonlinear Dynamics
Article type: 
Article
UDC: 
517.9

Reconstruction of dynamical systems from chaotic time series: short review

Autors: 
Anosov Oleg Lvovich, Institute for Nuclear Research
Butkovskij Oleg Jaroslavovich, Federal State Budget Educational Institution of Higher Professional Education "Vladimir Grigorievich Vladimir State University and Nikolai Grigorievich Stoletovykh"
Kravtsov Yury Aleksandrovich, Space Research Institute Russian Academy of Sciences
Abstract: 

The brief survey of problems accompanying the reconstruction of dynamical equations from chaotic time series is presented. The most frequently used procedures of reconstruction are described, including estimate for the dimension of a system, choice of the equation type, determination of parameters of nonlincar functions by means оf differential equation fitting to time series, deleting of unreliable coefficients, testing of global stability of the reconstructed system. Several examples of reconstruction are presented. 
New effective criterium for distinguishing dynamical and random processes is suggested, based оп the notion of degree of predictability. Principal limitations are pointed out imposed by the properties of instability in presence of noise оп time-interval of predictable behaviour («horizon of predictability»), on the length of a sample, on amount of coefficient to be determined. So named «discriminant» (two window) approach is outlined, which allows to reveal nonstationarity in dynamical system, and important role of low dimensional models for retrieval of nonstationarities in systems of higher dimension is discussed. In conclusion prospective areas of applicability of reconstruction procedures are pointed out.

Key words: 
-
Acknowledgments: 
The work was supported by the INTAS (grant 96-0305), RFBR (grant 99-02-16625) and Federal Target Program "Integration" (grant А-0030).
Reference: 
  1. Ljung L. System Identification: Theory for the User. New Jersey: Prentice Hall PTR;1999. 609p.
  2. Ivakhnenko AG. Long-term Forecasting and Management of Complex Systems. Moscow: Tehnika;1995. 312p.
  3. Eykhoff P, editor. Trends and Progress in System Identification. Oxford: Pergamon;1981. 418p.
  4. Balakrishnan AV. Kalman Filtering Theory. N.Y.: Optimization Software, Inc.;1984. 222p.
  5.  Anosov OL, Butkovsky OYa, Kravtsov YuA. Restoration of chaotic systems from time series: achievements and limitations. in: Abstracts of the V International School "Chaotic self-oscillations and the formation of structures (CHAOS'98)". 6-10 October 1998, Saratov, Russia. Saratov: Kolledzh;1998. P. 16-19 (in Russian).
  6. Cremers J, Hubler А. Construction of differential equations from experimental data. Z. Naturforschung A. 1987;42(8):797-802. DOI: 10.1515/zna-1987-0805.
  7. Grutchfield JP, McNamara BS. Equations of motion from а data series. Complex Systems. 1987;1(2):417-452.
  8. Breeden J, Hubler A. Reconstructing equations of motion from experimental data with unobserved variables. Phys.Rev.A. 1990;42(10):5817-5826. DOI:https:.doi.org/10.1103/PhysRevA.42.5817.
  9. Gouesber С. Reconstruction оf the vector fields of continuous dynamical systems from numerical scalar time series. Phys.Rev.A. 1991;43(10): 5321-5331. DOI: 10.1103/physreva.43.5321.
  10.  Brush JS, Kadrtke JB. Nonlinear signal processing using empirical global dynamical equations. In: Proceedings ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing. 23-26 March 1992, San Francisco, USA. IEEE;2002. P. 321. DOI: 10.1109/ICASSP.1992.226618
  11. Gribkov DA, Gribkova VV, Kravtsov YuA, Kuznetsov YuI, Rzhanov AG. Reconstruction of the dynamic system structure from time series. Radiotekhnika i Elektronika. 1994;39(2):269-277.
  12. Gribkov DA, Gribkova VV, Kravtsov YuA, Kuznetsov YuI, Rzhanov AG. Renewal of differential equations of autostochastic systems via time realization of one dynamic variable of the process. Tech. Phys. 1994;64(3):1-12.
  13.  Gribkov DA, Gribkova VV, Kravtsov YuA, Kuznetsov YuI, Rzhanov AG, Anosov OL, Butkovskii OYa. Dynamic equation reconstruction from the observed one dimensional time series. In: Proceedings Internat. Conf. Dynamical Systems and Chaos. 23-27 Мау 1994, Tokyo, Japan. Singapore: World Scientitfic;1995;2. P. 378.
  14.  Anosov OL, Butkovskii OYa, Kravtsov YuA, Surovyatkina ED. Predictable nonlinear dynamics: advances and limitations. In: Katz RA, editor. Chaotic, Fractal аnd Nonlinear Signal Processing. N. Y.: AIP Press;1995;375. P. 71.
  15.  Mees AI, Judd K. Parsimony in dynamical modelling. In: Kravtsov YuA, Kadtke JB, editors. Predictability оf   Complex Dynamical Systems. Berlin: Springer, 1996. P. 123-141. DOI:10.1007/978-3-642-80254-6_7
  16. Anosov OL, Butkovskii OYa, Kravtsov YuA. Strategy and algorithms for dynamical forecasting. In: Kravtsov YuA, Kadtke JB, editors. Predictability оf   Complex Dynamical Systems. Berlin: Springer, 1996. P. 105-121. DOI:10.1007/978-3-642-80254-6_6
  17.  Kravtsov YuA, Kadtke JB, editors. Predictability оf Complex Dynamical Systems. Berlin: Springer; 1996. 234p. DOI:10.1007/978-3-642-80254-6
  18.  Pavlov AH, Yanson NB. Application of the mathematical model reconstruction method to the analysis of electrocardiograms. Izvestiya VUZ. Applied Nonlinear Dynamics. 1997;5(1):93-108. (in Russian).
  19.  Anishchenko VS, Smirnova NB. Analysis аnd synthesis оf dynamical systems from experimental data. SPIE. 1994;2098:137-142.
  20.  Janson NB, Anishchenko VS. Modeling the Dynamical Systems on Experimental Data. In: Katz RA, editor. Proc. of the conference "Chaotic, fractal and nonlinear signal processing". 10-14 July 1995. N.Y.: AIP Press;1995;375. P.688-708.
  21.  Janson NB, Anishchenko VS. Modeling of dynamical systems from experimental data. Izvestiya VUZ. Applied Nonlinear Dynamics. 1995;3(3):112-121. (in Russian).
  22.  Packard NM, Crutchfield JP, Farmer JD, Shaw RS. Geometry from а time series. Phys.Rev.Lett. 1980;45:712-716. DOI:10.1103/PhysRevLett.45.712
  23. Takens F. Detecting strange attractor in turbulence. In: Rand D Young LS, editor. Dynamical Systems and Turbulence, Warwick 1980. Lecture Notes in Mathematics. Berlin: Springer, 1981;898. P. 366-381. DOI:10.1007/BFb0091924.
  24.  Noakes L. The Takens embedding theorem. Int.J.Bifurc.Chaos. 1991;1(1):867-872. DOI:10.1142/s0218127491000634
  25.  Casdagli M. Nonlinear prediction of chaotic time series. Physica D. 1989;35:335-356. DOI:10.1016/0167-2789(89)90074-2
  26.  Farmer JD, Sidorowich JJ. Predicting chaotic time series. Phys. Rev. Lett. 1987;59(8):845-848. DOI:10.1103/PhysRevLett.59.845.
  27.  Breeden JL, Packard NH. A learning algorithm for optimal representation оf experimental data. Int. J. Bifurc. Chaos. 1994;4(2):311-326. DOI: 10.1142/s0218127494000228.
  28. Gousbet G, Letellier С. Global vector-field reconstruction by using а multivariate polynomial L2 approximation оn nets. Phys. Rev. Е. 1994;49(6):4955- 4972. DOI: 10.1103/physreve.49.4955.
  29.  Yanson NB, Pavlov AN, Balanov AG, Anishchenko VS. Reconstructing a mathematical model as applied to an electrocardiogram. Tech. Phys. Lett. 1996;22 (8):669-671.
  30.  Breeden JL, Packard NH. Nonlinear analysis оf data sampled nonuniformly in time. Physica D: Nonlinear Phenomena. 1992;58(1-4):273-283.
  31.  Anishchenko VS, Janson NB, Pavlov AN. A method for reconstructing nonuniform attractors. Tech. Phys. Lett. 1996;22 (4):265-267.
  32.  Frazer AM, Swinney HL. Independent coordinates from mutual information. Phys. Rev.A. 1986;33(2):1134-1140. DOI: 10.1103/physreva.33.1134.
  33. Bransater А, Swinney HL. Strange attractor in weakly turbulent Couette-Taylor flow. Phys. Rev. А. 1987;35(5):2207-2220. DOI: 10.1103/physreva.35.2207.
  34.  Liebert W, Shuster HG. Proper choice of the time delay for the analysis of chaotic time series. Phys. Lett. А. 1989;142(2-3):107-111. DOI:10.1016/0375-9601(89)90169-2
  35.  Schuster HG. Deterministic Chaos: An Introduction. N.Y.:VCH;1988. 266p.
  36.  Anishchenko VS. Complicated Oscillations in Simple Systems: Appearance Routes, Structure and Properties of Dynamical Chaos in Radiophisical Systems. M.: Nauka;1990. 312p.
  37.  Farmer JD, Ott E, Yorke  JA. The dimension of chaotic attractors. Phisica D. 1983;7(1-3):153-180. DOI: 10.1016/0167-2789(83)90125-2
  38.  Frederickson P, Kaplan J, Yorke J. The Lyapunov dimension of strange attractors. J. Diff. Eqs. 1983;49(2):185-207. DOI:10.1016/0022-0396(83)90011-6
  39. Wolf А, Swift J. Progress in computing Lyapunov exponents from experimental data. In: Horton CW, Rcichl LE, editors. Statistical Physics and Chaos in Fusion Plasmas. N.Y.: Wiley; 1984. P. 111.
  40.  Pawelski K, Schuster HG. Generalized dimensions and entropies from а measured time series. Phys. Rev. А. 1987;35(1):481-484.DOI: 10.1103/physreva.35.481.
  41.  Havstad JW, Ehlers CL. Attractor dimension оf nonstationary dynamical systems from small data sets. Phys. Rev. A. 1989;39(2):845-853.DOI: 10.1103/physreva.39.845.
  42.  Theiler J. Estimating fractal dimension. J. Opt. Soc. Аm. А. 1990;7(6):1055-1073. DOI: 10.1364/JOSAA.7.001055
  43.  Sauer T, Yorke JA, Casdagli M. Embedology. J. Stat. Phys. 1991;65(3-4):579-616. DOI: 10.1007/BF01053745.
  44.  Маnе R. On the dimension оf the compact invariant set of certain nonlinear maps. In: Rang DA, Young LS, editors. Vol. 898 of Lecture Notes in Mathematics. Berlin: Springer; 1981. P. 230-242. DOI:10.1007/BFb0091916
  45.  Theiler J. Spurious dimension from correlation algorithms applied to limited time series data. Phys.Rev.A. 1986;34(3):2427-2432. DOI: 10.1103/physreva.34.2427.
  46.  Landa PS, Rosenblyum MG. One method of estimation of dimensionality of attractor enclosure based on results of the experiment. Tech. Phys. 1989;59(1):13-20. (in Russian).
  47.  Nerenberg МА, Essex С. Correlation dimension and systematic geometric effects. Phys. Rev. A. 1990;42(12):7065-7074.DOI: 10.1103/physreva.42.7065.
  48. Grassberger Р, Procaccia 1. Characterization оf strange attractors. Phys. Rev. Lett. 1983;50(5):346-349. DOI: 10.1103/PhysRevLett.50.346.
  49.  Broomhead DS, King GP. Extracting qualitative dynamics from experimental data. Physica D. 1986;20(2):217-236. DOI: 10.1016/0167-2789(86)90031-X
  50.  Schwarz С. Estimating the dimension оf а model. Ann. Statist. 1978;6(2):461-464. DOI: 10.1214/aos/1176344136
  51.  Baake E, Baake M, Bock HG, Briggs KM. Fitting ordinary differential equations to chaotic data. Phys.Rev. 1992;45(8):5524-5529.DOI: 10.1103/physreva.45.5524.
  52.  Mees АI. Modelling complex systems. In: Mees AI, Vincent T, Jennings LS, editors. Dynamics of Complex Interconnected Biological Systems. Boston: Birkhauser;1990. P. 104-124. DOI: 10.1007/978-1-4684-6784-0_6
  53.  Mees АI. Dynamical systems аnd tesselation: detecting determinism in data. Int. J.Bifurc. Chaos. 1991;1(4):777-794. DOI: 10.1142/S0218127491000579
  54.  Mees AI. Parsimonious dynamical reconstruction. Int. J. Bifurc. Chaos. 1993;3(3):669-675.
  55.  Mees АI. Nonlinear dynamical systems from data. In: Kelly FP, editor. Probability, Statistics and Optimization. Chichester: Wiley;1994. P. 225-237.
  56.  Mees АI. Reconstructing chaotic systems in the presence of noise. In: Yamaguti M, editor. Towards the Harnessing of Chaos. Amsterdam; N.Y.: Elsevier, 1994. P. 305.
  57.  Brown R, Rulkov NF, Tracy ER. Modelling and synchronizing chaotic systems from time-series data. Phys. Rev. E. 1994. Vol. 49, № 5. P. 3784-3800. DOI: 10.1103/physreve.49.3784.
  58.  Glover J, Mees АI. Reconstructing the dynamics оf Chua’s circuit. J. Curcuits, Systems and Computers. 1992;3(2):201-214.
  59.  Anishchenko VS, Pavlov AN, Janson NB. Global reconstruction in the presence of apriory information. Chaos Solitons&Fractals. 1998;9(8):1267-1278. DOI: 10.1016/S0960-0779(98)00061-7.
  60.  Anosov OL, Butkovsky OYa, Isakevich VB, Kravtsov YuA. Detection of nonstationarities from randomly similar signals of a dynamic nature. Radio Electronics. 1995;40(2):255. (in Russian).
  61.  Gribkov DA, Gribkova BB, Kuznetsov YuI. Reconstruction of external signal from the time distribution of а variable of the autostochastic system. Moscow University Physics Bulletin. 1995;50(1):71-73.
  62.  Bezruchko BP, Seleznev EP, Smirnov DA. Reconstruction of the equations of a non-autonomous nonlinear oscillator from a time series: Models, experiment. Izvestiya VUZ. Applied Nonlinear Dynamics. 1999;7(1):49-68. (in Russian).
  63. Srark J, Broomhead DS, Davies ME, Huke J. Takens embedding theorems for forced and stochastic systems. Nonlinear Analysis, Theory, Methods &Applications. 1997;30(8):5303-5314. DOI: 10.1016/S0362-546X(96)00149-6
  64.  Teodorescu D. Time series decomposition and forecasting. Int.J.Control.1989;50(5):1577.
  65.  Berge P, Pomeau Y, Vidal C. Order Within Chaos. N.Y.: Wiley;1986. P. 329.
  66.  Anosov OL, Butkovsky OYa, Kravtsov YuA. A minimal procedure for identifying chaotic systems from an observed time sequence. Radio Electronics. 1997;42(3):1.
  67.  Pavlov AN, Yanson NB, Anishchenko VS. Application of statistical methods to solve global reconstruction problems. Tech. Phys. Lett. 1997;23 (4):297-299. 
  68.  Arbanel HDI, Brown R, Kadtke JB. Prediction in chaotic nonlinear systems: methods for time series with broadband Fourier spectra. Phys. Rev. А. 1990;41(4):1782-1807. DOI: 10.1103/physreva.41.1782.
  69.  Smith LA. Identification and prediction оf low dimensional dynamics. Physica D. 1992;58(1-4):50-76.DOI: 10.1016/0167-2789(92)90101-R
  70. Kadtke J, Kremliovsky M. Signal classification using global dynamical models. In: Xatz BA, editor. Chaotic, Fractal and Nonlinear Signal Processing. 10−14 July 1995, Mystic, USA. N.Y.: AIP Press;1996;375. P. 189-202. DOI:10.1063/1.51029
  71.  Brush JS. Classifying transient signals with nonlinear dynamic filter banks. In: Xatz BA, editor. Chaotic, Fractal and Nonlinear Signal Processing. 10−14 July 1995, Mystic, USA. N.Y.: AIP Press;1996. P. 145-166.DOI: 10.1063/1.51024
  72.  Kravtsov YuA. Randomness, determinateness, and predictability. Sov. Phys. Usp. 1989;32:434–449. DOI: 10.1070/PU1989v032n05ABEH002718
  73.  Kravtsov YuA. Fundamental practical limits to predictability. In: Kravtsov YuA, editor. Limits of predictability M.: Centrocom;1997. P. 170-200.
  74.  Anosov OL, Butkovsky OYa, Kravtsov YuA. Predictability limits for linear autoregressive models. Radio Electronics. 1995;40(12):1866.
  75.  Anosov OL, Butkovskii O.Ya., Gribkov D.A., Gribkova V.V., Kravtsov Yu.A., Kuznetsov Yu.l., Rzhanov A.G. Discriminant analysis ав applied 10 revealing of nonstationarity in chaotic systems. In: Dynamical Systems and Chaos (Proc.Internat.Conf.). 23-27 May 1994, Tokyo, Japan. Singapore: World Scientific;1995;2. P. 370.
  76.  Anosov OL, Butkovskii OYa. A discriminant procedure for the solution of inverse problem for nonstationary systems. In: Kravtsov YuA, Kadtke JB, editors. Predictability оf Complex Dynamical Systems. Berlin: Springer Verlag; 1996. P. 67-77.
  77.  Anosov OL, Butkovskii OYa, Kravisov YuA. Nonlinear chaotic systems identification from observed time series. Math.Models аnd Methods in Appl.Sciences. 1997;7(1):49-59. DOI:10.1142/S0218202597000049
  78.  Anishchenko VS, Pavlov АМ. Global reconstruction in application to multichannel communication. Phys Rev E. 1998;57(2):2455-2457.
  79. Anosov OL, Butkovskii OYa, Kadtke JB, Kravtsov YuA, Protopopescu VV. Low dimenisonal model of heart rhythm dynamics as a tool for diagnosing the anaerobic threshold. In: Kadtke JB, Bulsara A, editors. Int. Conf. оn Applied Nonlinear Dynamics near the Millenium (ANDM’97). 7-11 July 1997, San Diego, USA. N. Y.: AIP Press; 1997. P. 359.DOI: 10.1063/1.54233.
  80. Anishchenko VS, Vadivasova TE, Astakhov VV. Nonlinear dynamics of chaotic and stochastic systems. Saratov: Saratov University Publishing;1999. P. 368 с. (in Russian).
  81. Oppenheim AV, editor. Digital signal processing. New Jersey: Englewood Cliffs;1975. 585 p.
  82. Otnes RK, Enochson L. Applied Time Series Analysis: Basic Techniques. N.Y.: Wiley; 1978. Мир, 1983. 444p.
  83. Tikhonov VI, Kharisov VN. Statistical analysis and synthesis of radio engineering devices and systems. М.:Radio I Svyaz;1991. 608 p.
Received: 
21.12.1998
Accepted: 
30.11.1999
Published: 
15.04.2000
  • 95 reads