A Method for Physical Monitoring of Wireline Communications Based on Compressed Sensing and Spread-Spectrum Signals

This paper presents a method based on Compressed Sensing (CS) and spread-spectrum signals for estimating the frequency response of electrical communication cables, with applications in cable monitoring for diagnostics and physical-layer security. Compared with traditional systems in the literature, the use of CS relaxes the data-acquisition requirements regarding sampling frequency, thereby reducing the cost of electronic components and enabling integration of the method into the network transceivers for real-time, continuous cable monitoring. Moreover, the use of spread-spectrum signals excites the cable’s high-frequency response, allowing for the detection of defects as short as 0.5 cm. The experimental evaluation demonstrates the method’s capability to accurately estimate the cable’s frequency response in the band of interest, up to the 40th Nyquist band with an undersampling ratio of 10, and to detect defects, specifically the lack of cable insulation.