In addition, Figure 6 illustrates NMSE
performance of the proposed scheme.
In order to get a better insight of the relative complexity of the two models, the NMSE
is plotted as a function of the polynomial function's number of coefficients for the FTNTB model and the AFTNTB model.
Correlation FB FAC2 Temperature ([degrees]C) 0.04 0.86 0.002 96.12 Relative humidity (%) 0.13 0.60 -0.18 90.72 Wind speed (m/s) 0.39 0.60 0.05 61.43 TABLE 3: Comparison of fog occurrence and observed time.
The residual errors [e.sub.out] are compared with the NMSE
[e.sub.in] of the error-corrupted data received at the sink node.
To show the effects of this parameter on final reconstructions, the curves of NMSE
with respect to parameter [lambda] for the 10th frame of the first dynamic cardiac dataset at R = 3 and 4 were plotted in Figure 5.
= 1/[absolute value of [bar.R]] [square root of [1/n] [n.summation over (i=1)] [([R.sub.COSMO,i] - [R.sub.CONTAM,i]).sup.2]] (15)
= [1/[([absolutely value of [[bar.F].sub.a]])([absolutely value of [[bar.F].sub.b]])]][[1/n][n.summation over (i=1)][([F.sub.[a,i]] - [F.sub.[b,i]]).sup.2]] (1)
9 gives the relationship between the NMSE
and training times in the training process.
(ii) Sparse Spectrum Recovery Performance of NMSE
versus SNR and Compression Ratio.
It is clear that, after every step, the quality of the deconvolved image becomes better, and finally, we can get a fine quality of the deconvolved images based on the minimum value of the NMSE
. Image B is degraded by blur kernel of sizes 7 x 7, 19 x 19, and 23 x 23 with a values of 4.66, 11.33, and 15.33, respectively.
To check whether the model performance is acceptable or not, some studies  gave the following limit values for the metrics mentioned above: -0.3 < FB < 0.3, NMSE
< 1.5, and R > 0.8.