A hilbert transform approach in source identification via multiple-input single-output modeling for correlated inputs

Several approaches can be employed to tackle source/path identification problems via multiple-input single-output modeling. When the measured inputs are correlated to each other, partial coherence function approach can be,used in which conditioning of the inputs is performed before calculating the contribution of each input. To ensure that the conditioning analysis is successful, correct priorities of the correlated inputs must be provided first, which can be often resolved based on the assumption of causality between a pair of correlated inputs. The contribution of inputs is frequently assessed in the frequency domain and, furthermore, the priorities of the inputs may be frequency dependent. Here, Hilbert transform approach is employed to determine effectively such priorities in the frequency domain and is then applied to an acoustical system. The results are compared with those by other approaches. (C) 1998 Academic Press.