Companding = Compression + Expansion
Companding is the process of compression and then expansion. With companded system, the higher amplitude analog signals are compressed (amplified less than lower amplitude signals) prior to transmission and then expanded (amplified more than the lower amplitude signals) in the receiver.
Or we can say For audio analog signals, the amplitude of weak signals is raised and the amplitude of strong signals is decreased, thereby altering (compressing and expanding) the dynamic range of the signals. The technique is helpful in improving the quality of amplified voice and musical instrument sounds. Dolby and dbx noise reduction also employ companding.
Companding is employed in telephony and other audio applications such as professional wireless microphones and analog recording
Figure shows the basic process of Compression & Expansion
This diagram shows that the amount of amplifier gain is reduced as the level of input signal is increased. This keeps the input level to the modulator to a relatively small dynamic range. At the receiving end of the system, an expanding system is used to provide additional amplification to the upper end of the output signal. This recreates the shape of the original input audio signal.
For digital audio signals, companding is used in pulse code modulation (PCM). The process involves decreasing the number of bits used to record the strongest (loudest) signals. In the digital file format, companding improves the signal-to-noise ratio at reduced bit rates. For example, a 16-bit PCM signal may be converted to an eight-bit ".wav" or ".au" file.
Compression and Expansion of Dynamic Range
Why we need to compress data?
The data rate is important in telecommunication because it is directly proportional to the cost of transmitting the signal. Saving bits is the same as saving money. Companding is a common technique for reducing the data rate of audio signals by making the quantization levels unequal. If the quantization levels are equally spaced, 12 bits must be used to obtain telephone quality speech. However, only 8 bits are required if the quantization levels are made unequal, matching the characteristics of human hearing.
The human ear is more sensitive to quantization noise in small signals than large signals. A-law and m-law coding apply a logarithmic quantization function to adjust the data resolution in proportion to the level of the input signal. Smaller signals are represented with greater precision – more data bits – than larger signals. The result is fewer bits per sample to maintain an audible signal-to-noise ratio (SNR).
Companding can be carried out in three ways:
(1) run the analog signal through a nonlinear circuit before reaching a linear 8 bit ADC.
(2) use an 8 bit ADC that internally has unequally spaced steps.
(3) use a linear 12 bit ADC followed by a digital look-up table (12 bits in, 8 bits out).