Performance improvements of wire fault diagnosis approach based on time-domain reflectometry

The aim of this study is to improve the performance of a wire fault diagnosis approach based on time-domain reflectometry (TDR). In this approach, the TDR response collected from measurements for a given (faulty) coaxial cable network is used as a reference for the cable network model response computed by an accurate analytical transmission line (ATL) method. The ATL model is iteratively tuned according to the outcome of the optimisation algorithm proposed. The advantage of the proposed ATL method over the more classic finite difference time-domain method is the extremely faster computational time which represents a significant factor for the diagnosis of faults in wiring networks. The proposed ATL method accurately provides the features of the resistance-inductance-conductance-capacitance-based model of a RG-58 coaxial cable employed as a device under test. The ATL model has been validated by comparison with measurement based and circuit simulations. Furthermore, seven experiments have been investigated in order to evaluate the performance of the ATL method for the diagnosis of wiring networks. The obtained results reveal that the ATL method together with the efficient optimisation algorithm is a reliable, efficient and fast method for the diagnosis of wiring networks.