Multi-innovation extended stochastic gradient algorithm and its performance analysis. (English) Zbl 1196.94026

Liu, Yanjun; Yu, Li; Ding, Feng

Circuits Syst. Signal Process. 29, No. 4, 649-667 (2010).

Summary: This paper derives the multi-innovation extended stochastic gradient algorithm for controlled autoregressive moving average models by expanding the scalar innovation to an innovation vector and analyzes its performance in detail. Four convergence theorems are given for the multi-innovation extended stochastic gradient algorithm to show that the parameter estimates converge to their true values under the weak persistent excitation condition. The simulation results show that the proposed algorithm can produce more accurate parameter estimates than the traditional extended stochastic gradient algorithm.

MSC:

94A12	Signal theory (characterization, reconstruction, filtering, etc.)
93E10	Estimation and detection in stochastic control theory
94-04	Software, source code, etc. for problems pertaining to information and communication theory
93-04	Software, source code, etc. for problems pertaining to systems and control theory

Keywords:

recursive identification; parameter estimation; signal processing; multi-innovation identification; stochastic gradient; performance analysis

References:

[1]	F. Ding, T. Chen, Parameter estimation of dual-rate stochastic systems by using an output error method. IEEE Trans. Autom. Control 50(9), 1436–1441 (2005) · Zbl 1365.93480
[2]	F. Ding, T. Chen, Performance analysis of multi-innovation gradient type identification methods. Automatica 43(1), 1–14 (2007) · Zbl 1140.93488
[3]	F. Ding, D.Y. Xiao, T. Ding, Multi-innovation stochastic gradient identification method. Control Theory Appl. 20(6), 870–874 (2003), (in Chinese)
[4]	F. Ding, H.B. Chen, M. Li, Multi-innovation least squares identification methods based on the auxiliary model for MISO systems. Appl. Math. Comput. 187(2), 658–668 (2007) · Zbl 1114.93101
[5]	F. Ding, Y. Shi, T. Chen, Amendments to ”Performance analysis of estimation algorithms of non-stationary ARMA processes”. IEEE Trans. Signal Process. 56(10-1), 4983–4984 (2008) · Zbl 1390.94151
[6]	F. Ding, H.Z. Yang, F. Liu, Performance analysis of stochastic gradient algorithms under weak conditions. Sci. China Ser. F Inf. Sci. 51(9), 1269–1280 (2008) · Zbl 1145.93050
[7]	F. Ding, P.X. Liu, G. Liu, Auxiliary model based multi-innovation extended stochastic gradient parameter estimation with colored measurement noises. Signal Process. 89(10), 1883–1890 (2009) · Zbl 1178.94137
[8]	G.C. Goodwin, K.S. Sin, Adaptive Filtering Prediction and Control (Prentice-Hall, Englewood Cliffs, 1984) · Zbl 0653.93001
[9]	L.L. Han, F. Ding, Multi-innovation stochastic gradient algorithms for multi-input multi-output systems. Digit. Signal Process. 19(4), 545–554 (2009)
[10]	L.L. Han, F. Ding, Identification for multirate multi-input systems using the multi-innovation identification theory. Comput. Math. Appl. 57(9), 1438–1449 (2009) · Zbl 1186.93076
[11]	L. Ljung, System Identification: Theory for the User, 2nd edn. (Prentice-Hall, Englewood Cliffs, 1999) · Zbl 0949.93509
[12]	D.Q. Wang, F. Ding, Auxiliary models based multi-innovation generalized extended stochastic gradient algorithms. Control Decis. 23(9), 999–1003 (2008), 1010 (in Chinese) · Zbl 1174.93720
[13]	D.Q. Wang, F. Ding, Performance analysis of the auxiliary models based multi-innovation stochastic gradient estimation algorithm for output error systems. Digit. Signal Process. 20(3), 750–762 (2010)
[14]	C.Z. Wei, Adaptive prediction by least squares prediction in stochastic regression models. Ann. Stat. 15(4), 1667–1682 (1987) · Zbl 0643.62058
[15]	L. Yu, F. Ding, Multi-innovation stochastic gradient identification methods for CARMA systems. Sci. Technol. Eng. 8(2), 473–475 (2008), (in Chinese)
[16]	L. Yu, F. Ding, J.B. Zhang, Convergence of multi-innovation stochastic gradient identification methods. Sci. Technol. Eng. 7(21), 5474–5478 (2007), 5484 (in Chinese)
[17]	L. Yu, F. Ding, P.X. Liu, On consistency of multi-innovation extended stochastic gradient algorithms with colored noises, in Proceedings of IEEE International Instrumentation and Measurement Technology Conference, Victoria, Canada, 12–16 May 2008, pp. 1695–1700
[18]	L. Yu, Y.J. Liu, F. Ding, Convergence of multi-innovation extended stochastic gradient parameter estimation for CARMA models. Syst. Eng. Electron. 31(6), 1446–1449 (2009), (in Chinese) · Zbl 1212.93309
[19]	J.B. Zhang, F. Ding, Y. Shi, Self-tuning control based on multi-innovation stochastic gradient parameter estimation. Syst. Control. Lett. 58(1), 69–75 (2009) · Zbl 1154.93040