Distributed sensor diagnosis for wire fault of complex topology wired networks based on chaos-TDR

The growing need for wiring in avionics, automotive, telecommunications, nuclear plants, buildings, etc., has caused the increase of the complexity of wired networks. This also leads to the increase of the difficulty of their maintenance that becomes not only problematic but also expensive. The need of embedded diagnosis to perform continuous monitoring was identified. Although reflectometry is an efficient method to diagnose simple topologies, it remains limited in the case of complex branched networks due to multipath fading of the test signal during its propagation. In practice, a distributed diagnosis strategry is necessary when multiple reflectometers (referred as diagnosis sensors) at several points of a complex network at the same time, thus allowing to solve ambiguities due to the existence of multiple paths and to guarantee a reliable and continuous diagnosis in complex wiring system. In this paper, the use of chaotic signals for distributed diagnosis and the detection of defects in the complex wiring network is presented experimentally. Chaos time-domain reflectometry (chaos-TDR) using low auto-correlation sidelobes signals, shows high performances in very noisy environment and is a suitable candidate for on-line diagnosis. Low cross-correlation chaotic signals enable several reflectometers (sensors) to simultaneously inject their signals in the cables, without any interference. Experimental results show that the proposed method performs well in a branched wiring network as it permits to detect and locate faults accurately even when the target system is operating. Also, the inference of wired network topology is realized by using sensor measurements at multiple cable ends.