High-Resolution Radar Imaging Through a Pipe Via MUSIC and Compressed Sensing

Abstract

Techniques for high-resolution radar imaging through a pipe are explored. To improve the angular resolution imposed by the pipe diameter, two high-resolution processing methods, MUSIC and compressed sensing (CS), are adapted for through-pipe radar imaging. First, the signal model, for propagation through the pipe, is formulated such that MUSIC and CS can be used. Then, methods for forming radar images from scattering collected through a pipe are derived. The methods are then demonstrated through electromagnetic simulation and verified with measurement data. It is shown that both methods provide high angular resolution in comparison with matched filter processing. The primary difference between the methods is that MUSIC requires multistatic data collection, while monostatic data collection is sufficient for CS. The performance of the two methods in terms of the number of targets and the signal-to-noise ratio is also evaluated. It is found that both methods can be used to image a complex scene composed of multiple point scatterers. CS, however, requires fewer number of data points than MUSIC. On the other hand, MUSIC is found to be less susceptible to noise than CS.