Multi-damage causes much more complex scattering phenomena in captured signals thanmono-damage does. Examination of an individual signal may fail to provide sufficientinformation to identify all instances of damage. Upon comparative evaluation of theperformance of forward and inverse inferences for damage identification, a data fusionscheme was developed for predicting multi-damage in a structure with the aid of a sensornetwork. The approach, conducted hierarchically by activating different sensors in a sensornetwork, fused an extracted signal feature, time-of-flight (ToF), at different levels, toprovide an overall consensus as to all possible instances of damage. This consensus waspresented in an intuitional contour map indicating the probability of damage occurrence.Benefiting from the sensor network, the dependence of identification processes on a specificsensor was minimized, and the need for interpreting complex signal scattering bymulti-damage was avoided as much as possible. To facilitate the extraction of ToFfrom raw signals, a signal processing approach, scale-averaged wavelet power(SAP) analysis, was introduced. As validation, the proposed identification schemewas employed to gauge dual delamination in a CF/EP woven laminate with abuilt-in active piezoelectric sensor network. The results have demonstrated theexcellent capability of the approach in evaluating multiple structural damagesites.