In order to achieve constant Signal to Quantization Noise Ratio (
SQNR) in the wide range of input variances, quasilogarithmic (QL) [mu]-law of compression can be applied
1 and 2 show
SQNR and bit rate R of the adaptive and non-adaptive three-level Lloyd-Max's scalar quantizer, estimated for the supposed Gaussian source in the wide dynamic range of variances |B| = 40 dB (B=[-20 dB, 20 dB]) and L = 32 levels [Q.sub.LU].
From the simulation results, the radix-2 DIT FFT algorithm has better accuracy in term of Signal-To-Quantization-Noise Ratio (
SQNR).
In order to keep the
SQNR of these final results above 80 dB, the valueof Paccmln had to be such that the result after this division would occupy at least 14 bits, giving 86 dB
SQNR.
Figure 2 shows the impact of the normalized saturation level [A.sub.M]/[sigma] and the number of quantization m on
SQNR. As can be observed, there is an optimum normalized saturation level for each value of m, since the quantizer's saturation becomes too frequent if [A.sub.M]/[sigma] is small and the quantization interval becomes too high when [A.sub.M]/[sigma] is high.
However, increase of parameter A leads to decrease of maximum
SQNR. Optimal values for these parameters are numerically calculated using MATLAB.
2 shows
SQNR obtained for the Lloyd-Max quantization with N = 2 and N = 4 levels having bit rates R = 1 bit/sample and R = 2 bit/sample, respectively [2], [3].
In order to provide the appropriate theoretical analysis in a wide dynamic range of the input variances, we define the distortion and
SQNR for the particular variance:
Logarithmic quantizers are robust and especially suitable for application in the case of high variance range of input signals, providing output of approximately constant
SQNR [18].
All support regions [x.sub.max.sup.[gamma]] will influence quantizer's performances, among which for us the most important are:
SQNR and the average bit rate [bar.R].
SQNR for proposed coding scheme is calculated by averaging
SQNR value for all frames [7]-[13]
Following the distortion it can be calculated Signal-to-Quantization-Noise Ratio (
SQNR) source