a- law compression
Hi,
Why compresion???
compression is used whenever there is non-linearity. especially they use natural logarithm.
for eg EAR response is a logarithmic response. I uses compression technique.
If one drops a pin in a silent atmosphere it takes its sound level as such into the ear. Mathematically take log 10 =1
If one drops atom bomb in a noisy envirnment it compress the very large
amplitude level to a smaller value and it takes its sound level in to the ear.
Mathematically take log 1000 =3
Usualy compressor will be kept infront of Non-Uniform quantizer only then expander is employed. (exapnder does inverse operation of compressor)
Note: compressor + Exapander ===> compander.
Non uniform quantizer will usually exploits the human hearing characterisitcs in speech signals
i.e., assigning more representation levels to weak passages(noise) at the expense of loud passages.
Non uniform quantizer will usually exploits the human visualizing characterisitcs in video signals
assigning more representation levels to low & medium freq content ( as human eye can visualize tht signals), at the expense of the higher freq content.(difficult to see)
Mu-law companding uses compressor characteristics which assigns linear dependence to low i/p signals and Non linear dependence(logarithmic) one to high i/p levels. especially for speech signals.
this is applicable for video signals as well.
dis adv of mu-law companding?????
If U know pl do reply
Happy learning
similarly compresion in video image one can visuallize small and medium freq content so it will deploy linear dependence. for very freq content signal our eyes cant see that so we assign nonlinear dependence.