Abstract: A method enabling a multiple sensor equalization in a radioelectrical receiver including a specified number of reception antennas, the method including the steps of estimating the transmission channel on each of the antennas; estimating the background noise component plus interference on each of the antennas on the basis of the estimation of the transmission channel; estimating the spatial correlation matrix referenced R.sub.b of the background noise component plus interferences from the received signal; computing a spatial-temporal filter formed for each discrete temporal element of the estimated multiple-sensor channel of a spatial filter; achieving a temporal filtering of the data elements on the different sensors by the spatial-temporal filter; and equalizing the signal at output of the spatial-temporal filter by one-dimensional equalization at a symbol rate deciding the symbols transmitted.

Abstract: The process according to the invention consists, in order to reduce the number of coefficients of the filters of the spatial part and the temporal part connected to the output of the spatial part, in jointly adapting the coefficients of the filters of each part using an adaptive algorithm according to the paths selected on the basis of a determined criterion. The coefficients are periodically recalculated at the rate of the known symbols in the learning sequences in order to minimize the estimation error apparent between a response signal (d(t)) and the receiver output signal (z(t)).