Research Article
Subspace Identification of Hammerstein Model with Unified Discontinuous Nonlinearity
Table 2
Parameters estimation for the Hammerstein model with SNR = 10 and 8 dB.
| SNR | | Parameters for the nonlinear model | Denominator | Numerator | MSE | | p | a | b | | | | | | | | |
| Without noise | True | 2.0000 | 0.5000 | 0.0000 | 1.5000 | −1.0000 | 0.0000 | 0.0000 | −0.2000 | 0.3500 | 1.0000 | 0.5000 | 0.0004 | | 2.0000 | 0.5001 | 0.0000 | 1.5001 | −1.0000 | 0.0006 | 0.0007 | −0.2001 | 0.3500 | 1.0013 | 0.4998 | | 0.0001 | 0.0003 | 0.0001 | 0.0005 | 0.0000 | 0.0004 | 0.0007 | 0.0001 | 0.0000 | 0.0002 | 0.0002 |
| 10 dB | True | 2.0000 | 0.5000 | 0.0000 | 1.5000 | −1.0000 | 0.0000 | 0.0000 | −0.2000 | 0.3500 | 1.0000 | 0.5000 | 0.0437 | | 2.0027 | 0.4760 | 0.0326 | 1.4874 | −1.0546 | 0.0538 | −0.0368 | −0.2006 | 0.3496 | 1.0029 | 0.4898 | | 0.0709 | 0.0618 | 0.0208 | 0.0264 | 0.0188 | −0.0573 | 0.0189 | 0.0017 | 0.0010 | 0.0091 | 0.0073 |
| 8 dB | True | 2.0000 | 0.5000 | 0.0000 | 1.5000 | −1.0000 | 0.0000 | 0.0000 | −0.2000 | 0.3500 | 1.0000 | 0.5000 | 0.1596 | | 2.0741 | 0.5341 | 0.1018 | 1.4210 | −0.8515 | −0.0715 | −0.0290 | −0.1980 | 0.3488 | 1.0385 | 0.4887 | | 0.0388 | 0.0106 | 0.0287 | 0.0941 | 0.0963 | 0.1068 | 0.0327 | 0.0016 | 0.0022 | 0.0247 | 0.0783 |
|
|