Abstract: A method to discriminate a real echo peak from an aliased echo peak comprises: computing ‘a correlation between N samples of a digital data and a copy of the same N samples delayed by a time delay ?1,0 to obtain a first correlation result, time delay ?1,0 being equal to a time interval T between a first and a second distinct power peaks of an estimated channel impulse response, the first peak being the highest power peak within the temporal window, and/or computing—a correlation between the N samples and a copy of the same N samples delayed by a time delay TIFFT-?1,0 to obtain a second correlation result, and deciding whether the second peak is a real echo peak or an aliased echo peak based on the first and/or second correlation results.