Sparse inversion SVD method for dispersion extraction of ultrasonic guided waves in cortical bone

In the study, we propose a sparse inverse singular value decomposition (SI-SVD) technique, which combines SVD and a least-squares sparse inversion method to improve the resolution of the ultrasonic (US) guided waves (GW) spectrum of long cortical bones. The method was assessed using synthetic data and experimental signals recorded on a 4-mm thick bone-mimicking plate. The results illustrated that the SI-SVD technique can achieve a more accurate determination of the dispersion relationships with enhanced noise filtering, which should be helpful to the USGW-based assessment of long cortical bone.