Diagnostic Methodology for Surface Defects in Curved Automotive Radomes Using a Phase Correlation Matching Process

In this article, a new class of diagnostic methodology based on reconstruction phase deviation is proposed. Using Fourier optics (FOs) and the Tikhonov method, the original wavefront distribution fields are reconstructed on the surface of real-life automotive radomes. During the phase deviation reconstruction, wrapped phase errors significantly degrade the quality of the reconstructed field. To mitigate the effect of wrapped phase errors, the phase correlation matching (PCM) process is proposed. The simulation results of the diagnostic methodology demonstrate that the PCM process effectively identifies dielectric defect regions on curved surfaces. In addition, both simulation and experimental diagnostics scenarios are conducted to validate the proposed methodology. The scenarios are structured based on various parameters, such as dielectric size, source directivity, and the radius of curvature for both homogeneous and inhomogeneous radome covers. As a result, the proposed diagnostic methodology achieves an accuracy of over 90% in localized defects such as dielectric patches and protrusions on the radome surface. Finally, the proposed method exhibits higher computational efficiency compared with previously reported diagnostic methods; furthermore, the proposed algorithm presents a potential solution for quality control procedures on automotive assembly lines and real-life radome applications.