Noise reduction
Noise suppression method ( engl. Noise Reduction, Noise Cancellation or squelch ) are technical procedures for analog transmission and storage of speech and music, reduce unwanted noise. There is a distinction between dynamic and static methods. In digital technology, so for example for audio CDs or music stored in MP3 format, such methods are necessary because of the Inherent almost complete freedom from noise only in special cases, such as high compression. Main applications of analog noise reduction procedures were the soundtrack on vinyl records, tape recorders and cassette recorders. In the fields of broadcasting and voice, which still partly based on analog technology, such methods are still widely used.
Pre-Emphasis
The hearing threshold is between 1 kHz and 5 kHz particularly low, therefore acts noise in this area is especially bothersome. The distance between the signal level and the noise may be increased, if one emphasizes the upper portion of the Tonfrequenzspektrums during recording. When playing the original frequency response of the sound signal is restored and there with lowered noise. In FM broadcasting is working using this method.
Another example of this is the RIAA curve for analogue records, by this method, high frequency disturbances can be significantly reduced, and therefore reminds the remaining, more low-frequency noise to a rumble.
Compander
In the compander certain frequencies of its input levels to different extents depending raised before they are stored or transmitted and equalized accordingly during playback. The achieved noise reduction effect is significant. This process is described by the well-known noise reduction Dolby NR, B, C and S are used ( in the studio even Dolby A and SR ) as well as the lesser known HighCom, HighCom II, Super D and dbx and achieved worldwide distribution, especially in cassette recorders. A high importance in the analogue satellite TV transmission was also awarded Wegener Panda 1 method.
Reduction of the quantization noise
The phone speech is first converted in an analog- to-digital converter into numerical values which are then transmitted. Here, at low amplitude arises as disturbing quantization noise can be reduced if small signal levels with small steps and large signal levels are only quantized coarsely. The non-linearity is defined by the A-law process.
Radio
This is ready to receive analog walkie-talkies not rushing in pauses or at very low coverage, they have a noise reduction ( " squelch " ) that mutes the audio portion when the signal received by the antenna signal falls below an adjustable threshold.
The license-free walkie-talkies ( PMR - Private Mobile Radio ) by default operate with noise reduction enabled, but can be switched off.
Fourier analysis, filtering and Fourier synthesis
A very effective way to reduce noise is the Fourier analysis followed by selective filtering and back-transformation by means of the Fourier synthesis. The method is relatively computationally intensive and time consuming, but can with modern microprocessors or PLDs, at not too high bandwidth, still be carried out in real time ( see also SDR). For implementation on digital computer systems may be used in the discrete Fourier transform (see FFT).
In the Fourier analysis of the amplitudes and phases are determined using a Fourier transform for all frequencies studied. Noise signals differ from useful signals usually by continuous frequency response or lie in certain frequency ranges. It may be helpful to make an analysis of a noise signal with no useful signal in order to gain certainty about the spectral composition of the noise. The noise components can be removed by suitable computational methods in the Fourier spectrum. In the subsequent inverse transform only the useful components of the signal are composed ( synthesis).
In the following example, this is the basis of a one-dimensional signal, as occurs for example in audio engineering, explains: