Companding

Under companding is understood in communications technology, a method by which one can attenuate audio signals for the adverse effects of a limited dynamic range in a transmission or storage process. Thus, the signal -to-noise ratio is improved and overdrive is avoided. The term is a portmanteau of " compression " and " expansion ". The method is used by the phone.

Demarcation

Companding is a nonlinear, static process in which the pre-distorted signal voltage by a fixed characteristic ( " compressed" ) and is then transmitted to the destination. In this form, the signal is very strongly distorted. When playing back the signal passes through an " expander " with complementary characteristic, where it is converted back to the original waveform. The loss of quality is usually low.

Companding is often confused with compression in which a control voltage is generated by an envelope detector changed, the gain of a ( linear response ) Voltage Controlled Amplifiers relatively slow ( in Hertz range ), that the output voltage remains at a desired frame. This makes it possible to avoid or reduce disturbing noise ( the Dolby system ) either override. The thereby partially clearly audible change in the original is here considered acceptable. The effects caused by a compressor amplitude changes can be made only partially reversed by an expander.

Principle of operation

The very large dynamic range of speech is pre-emphasized in telephone systems by means of A - law method. Since the characteristic is difficult to generate by analog means, it is performed after digitization. The analog- to-digital converter initially works with 12-bit accuracy, which corresponds to a dynamic range of ± 66 dB. The A-law method reduces the amplitude range of ± 42 dB, so the data can be transferred with 8-bit accuracy. At the destination, the data first, and then converted back to the original 12-bit accuracy converted from a digital to analog converter in the voice signal.

The described method is referred to as a pre-distortion of amplitude in the time domain. To reduce the high frequency noise pre-emphasis is used, which can be described as pre-distortion of amplitudes in the frequency domain.