### Table 1. Quantitative Signal Recovery Comparison of the proposed algorithms

"... In PAGE 29: ... The qualitative simulation result summary is presented in Figure 7. See Table1 for a quantitative comparison. Pole-Zero Map of the Final Demixing Network Real Axis I m ag A x i s -1 -0.... In PAGE 31: ...5 3 x 104 (f) (g) Figure 8. State Space BSR for a non-minimum phase system: (a) Transmission poles-zero map for the mixing environment, (b) Environment Impulse Response, (c) Theoretical Environment Inverse, (d) Estimated Demixing Network, (e) Global Transfer Function Achieved, (f) Pole-zero map of Estimated Demixing Network (g) Convergence of MISI Performance Index A quantitative performance comparison of the (blindly) recovered communication signals using the presented BSR algorithms is provided in Table1 . For the table below, we have defined ... ..."

### Table 3. Error detection and recovery for discrete signals.

### Table 3. Example 3, performance: recovery of the postnonlinearly-mixed speech signals from figure 5 using nonlinearGeo and linear algorithms.

2003

"... In PAGE 6: ... Us- ing polar nonlinearGeo we separate these signals; the al- gorithm can be seen in figure 7. Table3 again shows the performance of the various algorithms; polar nonlinearGeo performs quite well again. We see that also this postnonlin- ear case is nicely modelled by nonlinearGeo.... ..."

Cited by 4

### Table 1. Measured signal intensities for an inversion recovery sequence in inner cylinder depending on inversion time and position in phantom with standard deviations. T1 shows the resulting T1 relaxation times and standard deviations obtained by curve fitting signal values from each position Position [cm]

### Table 2. Example 2, performance: recovery of the nonlinearly-mixed speech signals from figure 5 using non- linearGeo and linear algorithms.

2003

"... In PAGE 5: ... In figure 6 we can also see how the linear recoveries in the rings worked, and how polar nonlin- earGeo recovered the sources. Table2 again gives a com- parison of the different algorithms for this example; again polar nonlinearGeo performs best. In our last example we want to consider postnonlinear Algorithm covariance E polar nonlinearGeo 0:8255 0:0012 0:0275 0:7926 0.... ..."

Cited by 4

### Table 1. Example 1, performance: recovery of the nonlinearly-mixed gamma signals from figure 1 using non- linearGeo and also linear algorithms for comparison.

2003

Cited by 4

### Table 1: Averaged gain at time , and time derivative at as a function of the control signal power. The CW signal is turned-off at the time .

"... In PAGE 6: ... From the steadystate solution with a CW control signal Pc1 at the input, at time t = 0 we have turned off the signal. The gain recovery to the steady-state value is exponential and Table1 shows a faster recovery for larger signals. Let apos;s now fix the control signal apos;s power at 2 mW .... In PAGE 6: ... Let apos;s now fix the control signal apos;s power at 2 mW . The results from Table1 show that to optimize our device we have to focus the optical signal to achieve a high photon density. For this goal there is an optimum in the thickness of the active layer as displayed in the Table 2.... ..."

### Table 1. Peak Signal-to-Noise Ratio (PSNR) and Sum of Squared Errors (SSE) of the real MR image with corrupted regions (PSNR1, SSE1) and the same image after the recovery process (PSNR2, SSE2) reconstructed by the selected wavelet functions Wavelet function PSNR1 [dB] PSNR2 [dB] SSE1 SSE2

"... In PAGE 4: ... Results As the most efficient method of the MR image denoising the wavelet Symmlet of the 4th order has been used here for the decomposition into two levels. The best result of the MR image recovery has been obtained using the Daubechies wavelet function of the 8th order (see Table1 ) for the wavelet decomposition into one level and 350 steps. Coefficients in the ... ..."

### Table 2: The peak-signal-to-noise-ratio (PSNR) and sum of squared errors (SSE) of the simulated image containing cor- rupted regions (PSNR1, SSE1) and the same simulated im- age after the recovery algorithm (PSNR2, SSE2) recon- structed by the selected wavelet functions after 400 iterations

"... In PAGE 4: ... 6 together with the evolution of the peak-signal- to-noise-ratio during the iteration process based upon the use of Daubechies wavelet function of the 4th order for image de- composition and reconstruction. Table2 provides results of the study of different wavelet func- tions use for reconstruction of image regions. Daubechies wavelet function of the 8th order provided the lowest SSE and the highest PSNR in this case.... ..."