Diagnosis Method for Automotive Radar Cover using Fourier Optics and Tikhonov Method

The paper presents a radar cover defect identification method applicable to curved surfaces. The method localizes the defects due to fabrication errors over the radar cover surface. Fourier Optics and Tikhonov regularization are utilized for field reconstruction. Next, phase correlation matching is introduced to visualize the local phasic deviations on the surface without encountering issues owing to the periodic nature of the phase. The presented method is validated using field measurement at 76.5 GHz. The radar cover topology under test is selected as a sector of a cylindrical surface. The dielectric square patch with a length of 0.7 λ0 is detected at a maximum 6.06 λ0 away from the reconstruction region center. The method confirms the accuracy of a maximum of 9.63% deviation for sparse defect localization.