Identification of Wiener Model with Internal Noise Using a Cubic Spline Approximation-Bayesian Composite Quantile Regression Algorithm

Table 1

Comparison among CSA-BCQR, CSA-BQR, and CSA-RLS in the numerical simulation.

	Variable	CSA-BCQR	CSA-BQR	CSA-RLS
Linear block		−1.001	−1.01	−0.996
		2.99	2.95	2.93
		−2.0	−2.07	−2.11
	ARE	2.69E − 03	2.31E  − 02	3.49E − 02
Nonlinear block	MSE	3.121E − 04	5.675E − 04	9.554E − 04
Comprehensive error	CE	2.95	8.06	12.98