Method of digital transmission of messages using dynamic-range-limited precoding

DETAILED DESCRIPTION An object of the invention is to reduce the dynamic range of the received signal in a digital communication system compared to the above-described prior art.
The invention achieves this object by limiting the number of different possible signal values for each transmission symbol to be transmitted.
Instead of an infinite periodic continuation of the signal constellation, this method only recognizes signal values whose amplitude does not exceed a preselected upper barrier.
This is achieved for example by using a nonlinear transfer element, in the predistortion system, that has a characteristic curve with only a few sawteeth with linear branches for input values of large magnitude.
Each saw-tooth corresponds to a signal value.
The characteristic curve has two to four saw-teeth, for example.
Instead of many signal values for each transmission symbol, this method therefore only uses a few in accordance with the upper barrier.
In each instance, the signal value whose instantaneous value is the smallest at the output of the linear predistortion filter is selected.
This selection is made by the non-linear transfer element.
Therefore, this method already considers on the transmitting side the properties of the signal on the receiving side.
The signal dynamic range is thereby held to the desired limits in the decoder.
Even further processing of the signal and the adaptive fine correction of the distortion, as well as the extraction of the clock pulse from the symbol, are considerably simplified.
Since very large amplitude values in the receiving side signal are avoided, the intersymbol interference is significantly reduced.
In this way, a much smaller data and coefficient word length is sufficient to represent the signal in the receiver for further digital processing by means of an analog-digital conversion, for example for adaptive fine correction.
This not only makes implementation significantly more cost-effective; it can also be used at much higher data rates.
The bandwidth of the phase control loop for obtaining the clock pulse of a symbol can be increased, since more clock pulse jitter can be tolerated