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

Table 2

Comparison of proposed results against true values of INCAR model for 20 dB SNR.

μ	Method	Design parameters								MSE
		𝑝1	𝑝2	𝑎1	𝑎2	𝑎3	𝑞2𝑎1	𝑞2𝑎2	𝑞2𝑎3
(10^−04, 10^−08)	VLMS	1.347467	−0.751995	1.185194	0.341222	0.235790	1.435438	1.067026	0.282885	2.19𝐸−02
	KLMS	1.352035	−0.745072	0.923361	0.244504	0.577959	1.563722	0.660224	0.673292	4.67𝐸−02
	FLMS1	1.351852	−0.750247	0.998773	0.499946	0.201665	1.679453	0.841320	0.336062	1.23𝐸−06
	FLMS2	1.351805	−0.750225	0.998823	0.499995	0.201691	1.679341	0.841338	0.336139	1.22𝐸−06
(10^−03, 10^−05)	VLMS	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
	KLMS	1.353428	−0.745015	1.141632	0.298110	0.300711	1.325873	1.163538	0.362182	3.77𝐸−02
	FLMS1	1.345754	−0.753166	0.996107	0.496821	0.201403	1.679157	0.838343	0.338860	8.36𝐸−06
	FLMS2	1.345854	−0.753357	0.996484	0.496661	0.201533	1.679222	0.838346	0.338957	8.30𝐸−06
True values		1.350000	−0.750000	1.000000	0.500000	0.200000	1.680000	0.840000	0.33600	0