Abstract: A neural network is provided for equalizing distorted data signals. The data signal to be equalized is coupled via time-delay elements to a group of networks for weighting. The output signals of the networks for weighting are coupled to the input terminals of a plurality of neurons whose outputs are coupled, via a respective amplifier, to input terminals of a further neuron having an output terminals where the equalized data signal can be tapped.

Abstract: Equalizers, used for equalizing distorted signals, are calibrated with the aid of the so-called "eye" pattern on the oscilloscope. Data bits are traced with their edges above one another by triggering to the amplitude so that an "eye" opening is produced. The "eye" is wide open and sharply delimited in the case of correct equalization. However, this method does not provide for automatic calibration of the equalizer. In order to automatically calibrate an equalizer in an optimal manner, a Lissajous figure is formed from in each case of two successive samples of the signal to be equalized. A fuzzy logic control loop compares the Lissajous figures generated with a circle in the case of a sine oscillation and with a square in the case of a square wave oscillation. From this comparison, the fuzzy logic control loop derives the parameters for the equalizer. The equalizer is optimally calibrated when a circle or, respectively, a square is formed as Lissajous figure.

Abstract: A method and apparatus for setting the parameters of an equalizer. Equalizers are used for equalizing distorted signals. However, optimum equalization is only achieved when the parameters of the equalizer are optimally set. To set the parameters of an equalizer optimally, the output signal of the equalizer is rectified in a full-wave rectifier. The output signal of the latter is integrated in an integrator. The parameters of the equalizer are set by a microprocessor according to an algorithm. In one embodiment, when the integration time is predetermined, the parameters of the equalizer is changed until the level of the integrated signal becomes a maximum. In another embodiment, when the maximum level of the integrated signal is predetermined, the parameters of the equalizer are changed until the integration time becomes a minimum.

Abstract: Apparatus requiring a plurality of control signals is provided with a fuzzy logic controller. The fuzzy logic controller is programmed with a plurality of control functions. Respective parameters determinative of the control signals are applied to the fuzzy controller in time division multiplex fashion, and concurrently respective control signals provided by the fuzzy controller are time division multiplex applied to respective control signal paths.