Question: Quantization by Truncation Set the parameter RMODE of fxquant ( ) to trunc and repeat Exercises and. What is the main difference between the results for rounding and those for truncation?
Exercises: Quantization Experiments Use fxquant ( ) to quantize the 8000 input speech samples from the file S5 . MAT. Using rounding and saturation, compute the quantization error sequences for I 0-, 8-, and 4-bit quantization. Use the program s triplet ( ) to plot these error sequences. What are the important differences among them? Do they look like they fit the white noise mode l? Make histograms of the quantization noise samples. Do they seem to fit the uniform amplitude distribution model?