×

General approach to filtering with fractional Brownian noises – application to linear systems. (English) Zbl 0979.93117

The authors considered the process \[ Y_t(\omega)= \int^t_0 C(s,\omega) ds+ \int^t_0 B(s) dW_s(\omega) \] on the probability space \((\Omega, F,F_t,P)\), where \(W\) is a normalized fractional Brownian motion and \(C\) is \(F_t\)-adapted. Using an innovation approach, they obtained the integral representation theorem of the \({\mathcal Y}_t\)-martingale and the filtering equation of \(E(Z_t/{\mathcal Y}_t)\) for a process \(Z_t\), where \({\mathcal Y}_t\) is the \(\sigma\)-field generated by \(\{Y_s,s\leq t\}\). Applying these results to the filtering problem of Gaussian linear systems with fractional Brownian noises, they described the optimal filter in terms of Kalman type equations for its mean and for the variance of the filter error. Various specific cases are presented.
Reviewer: M.Nisio (Osaka)

MSC:

93E11 Filtering in stochastic control theory
60G20 Generalized stochastic processes
60G35 Signal detection and filtering (aspects of stochastic processes)
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Coutin L., Annals of Applied Probability 9 pp 1058– (1999)
[2] Decreusefond L., Potential Analysis 10 pp 177– (1999) · Zbl 0924.60034 · doi:10.1023/A:1008634027843
[3] Elliott R. J., Stochastic Calculus and Applications (1982) · Zbl 0503.60062
[4] Fujisaki M., Osaka J. Math 1 pp 19– (1972)
[5] Gripenberg G., Journal of Applied Probability 33 pp 400– (1997) · Zbl 0861.60049 · doi:10.2307/3215063
[6] Jacod J., Limit Theorems for Stochastic Processes (1987) · doi:10.1007/978-3-662-02514-7
[7] Kailath T., I.E.E.E. Trans, Automatic Control 13 pp 646– (1968) · doi:10.1109/TAC.1968.1099025
[8] Kalman R. E., J. Basic Eng. ASME 2 pp 33– (1960)
[9] Kalman R. E., J. Basic. Eng. ASME 83 pp 95– (1961) · doi:10.1115/1.3658902
[10] Kleptsyna M. L., Stochastic Analysis and Its Applications 16 (5) pp 907– (1998) · Zbl 0916.93076 · doi:10.1080/07362999808809569
[11] Kleptsyna M. L., Information Transmission Problems 34 (2) pp 65– (1998)
[12] Kleptsyna, M. L., Le Breton, A. and Roubaud, M. C. An elementary approach to filtering in systems with fractional Brownian observation noise. VSP/TEV Proceedings of the 7th Vilnius Conference-1998. Edited by: Grigelionis, B. pp.373–392. · Zbl 0999.60038
[13] Kleptsyna M. L., Statistics and Control of Stochastic Processes pp 179– (1985)
[14] Le Breton A., Statistics and Probability Letters 38 (3) pp 263– (1998) · Zbl 0906.62104 · doi:10.1016/S0167-7152(98)00029-7
[15] Le Breton A., C.R. Acad. Sci. Paris 326 (1) pp 997– (1998) · Zbl 0947.60044 · doi:10.1016/S0764-4442(98)80129-8
[16] Lin S. J., Stochastics and Stochastics Reports 55 (1) pp 121– (1995) · Zbl 0886.60076 · doi:10.1080/17442509508834021
[17] Liptser R. S., Statistics of Random Processes (1978) · Zbl 0369.60001 · doi:10.1007/978-1-4757-4293-0
[18] Liptser R. S., Theory of Martingales (1989) · Zbl 0728.60048 · doi:10.1007/978-94-009-2438-3
[19] Lyons T., Mathematical Research Letters 1 (4) pp 451– (1994) · Zbl 0835.34004 · doi:10.4310/MRL.1994.v1.n4.a5
[20] Mandelbrot B. B., SIAM Review 2 (10) pp 422– (1968) · Zbl 0179.47801 · doi:10.1137/1010093
[21] Norros I., Bernoulli 5 (4) pp 571– (1999) · Zbl 0955.60034 · doi:10.2307/3318691
[22] Pardoux E., Ecole d’ete de Probabilites de Saint-Flour XIX 1464 pp 67– (1991)
[23] Samko S. G., Fractional Integrals and Derivatives (1993)
[24] Samorodnitsky G, Stochastic Models with Infinite Variance (1994) · Zbl 0925.60027
[25] Williams, W. E. A class of integral equations. Proc. Cambridge Phil. Soc. Vol. 59, pp.589–597. · Zbl 0119.09501 · doi:10.1017/S0305004100037269
[26] Zakai M., Zeit. Wahr. Verv. Geb 11 pp 230– (1969) · Zbl 0164.19201 · doi:10.1007/BF00536382
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.