Identification of Input Nonlinear Control Autoregressive Systems Using Fractional Signal Processing Approach

Table 3

Comparison of proposed results against true values of INCAR model for 10 dB SNR.

μ	Method	Design parameters								MSE
		𝑝1	𝑝2	𝑎1	𝑎2	𝑎3	𝑞2𝑎1	𝑞2𝑎2	𝑞2𝑎3
(10^−04, 10^−08)	VLMS	1.350924	−0.748145	0.772281	0.747924	0.138729	1.481738	1.066460	0.281078	2.63𝐸−02
	KLMS	1.355082	−0.736640	0.657397	0.631919	0.529419	0.932575	1.184584	0.819516	1.44𝐸−01
	FLMS1	1.348795	−0.753541	0.999514	0.505212	0.194240	1.687344	0.839732	0.332946	1.72𝐸−05
	FLMS2	1.348876	−0.753596	0.999636	0.505522	0.193962	1.686987	0.839622	0.333021	1.74𝐸−05
(10^−03, 10^−05)	VLMS	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
	KLMS	1.339625	−0.751687	1.065474	0.302732	0.375939	1.634616	0.705183	0.515233	3.88𝐸−02
	FLMS1	1.346058	−0.732482	0.977546	0.494359	0.193604	1.691872	0.857421	0.345628	1.80𝐸−04
	FLMS2	1.346365	−0.732517	0.977584	0.494478	0.192037	1.688032	0.857678	0.347713	1.79𝐸−04
True values		1.350000	−0.750000	1.000000	0.500000	0.200000	1.680000	0.840000	0.33600