Technique for achieving the theoretical coding gain of digital signals

Abstract

The theoretical coding gain associated with an error correction technique can be realized in a digital communications system employing this error correction by providing the equalization in the transmitter instead of the receiver. In a first embodiment, an automatic equalizer in the receiver of a digital communications system is adjusted using a training sequence. The equalizer coefficients for optimum adjustment are then transmitted back to the transmitter and supplied to an equalizer therein. After the training period is over, this transmitter equalizer, now having optimum coefficients, is activated and the receiver equalizer is disabled. In a second embodiment, the use of a training sequence can be eliminated and the receiver equalizer can be adjusted in response to actual data using 'blind' equalization. The generated coefficients, as in the first embodiment, are then transmitted back to the transmitter. In either embodiment, using the transmitter equalizer to predistort the transmitted signal in a manner opposite to that of the communications medium connecting the transmitter and receiver assures that the noise at the input to the receiver decoder is both Gaussian and white. As a result, the theoretical coding gain associated with the employed error correction is substantially realized.