Circuit configuration and method for setting the switching points of a decision maker

A circuit configuration for correcting an analog input signal having a DC element, an AC element, and upper and lower signal peaks whose ratio to one another is fixed but variable, the analog input signal being corrected by compensation for the DC element, the circuit configuration comprising:

• a decision maker receiving the input signal;
• a calibration device, including:
• an analog/digital converter actuated by the input signal and connected to said decision maker;
• peak-value detectors for determining the upper and lower signal peaks of the input signal, said peak-value detectors disposed downstream from said analog/digital converter;
• a controllable reference unit providing a reference value;
• a computation unit for determining a mean value from at least one signal minimum and at least one signal maximum and disposed downstream from said peak-value detectors, the mean value being one of an arithmetic mean of the upper and lower signal peaks and mean weighted using the fixed but variable ratio;
• a comparison unit for determining a signal position of the input signal by comparing the reference value with the mean value determined by said computation unit, said comparison unit connected to said computation unit and said controllable reference unit;
• a first regulating unit having an input connected to said comparison unit and an output linked to said decision maker, said first regulating unit compensating for the DC element in the input signal if an unbalanced signal position is detected by said comparison unit, with the compensation being carried out by subtraction of a value which is determined from an output of said comparison unit from the input signal;
• a second regulating unit having an input connected to said comparison unit and an output connected to said controllable reference unit, said second regulating unit readjusting the reference value in an opposite sense to an output signal from said comparison unit, in which case a difference between the mean value and the reference value is used to form a new reference value;
• and
• means for scaling the mean value disposed between said computation unit and one of said comparison unit and an output side of said comparison unit.