Comparative Analysis of Different Sensor Designs for SAR-Based Pipeline Inspection in Near and Far Field Imaging

Abstract

One of the essential parts of any synthetic aperture radar (SAR) based microwave imaging system for pipeline inspection is the antenna, which determines the system ability to detect defects on an object reconstructed image. As the standoff distance increases, the antenna’s ability to detect defects on a material becomes influenced by the field of operation making it less effective, thereby affecting the defect information used to characterise the reconstructed pipe sample. In this paper, we conduct a comparative analysis of various antennas performances for pipeline inspection in near and far field operations using a SAR-based microwave system. An antipodal Vivaldi antenna (AVA) and 2-AVA are designed, and their performances are compared to an open-ended waveguide (OEW). The defective pipe image is reconstructed using the autofocus range-Doppler algorithm (RDA) for the OEW and AVA and improved multiple signal classification least square estimation (MUSIC-LSE) for the 2-AVA. Simulation and experimental results showed that the three antennas exhibit different performances depending on the region of operations. For the near field inspection, the OEW has a higher signal-to-noise-ratio (SNR) of 19 dB, fewer errors squared error (SE), and COV of 15% compared to the AVA with 16 dB, and 2-AVA with 14 dB, respectively. For the far field inspection, the 2-AVA gives a better SNR of 2 dB compared to the OEW and AVA with SNRs of −5 dB and −3 dB respectively. Thus, the choice of the antenna for the inspection depends on the region of operation.

Keywords

• Autofocus RDA
• coefficient of variance (COV)
• cross-range profile reconstruction (CPR)
• image reconstruction
• multiple signal classification least square estimation (MUSIC-LSE)
• open-ended waveguide