Source-independent Hessian-free Gauss-Newton Full-waveform Inversion

Summary The estimation of soruce wavelet is very important for implementing full-waveform inversion (FWI) successfully. Many FWI algorithms estimate the soruce signature iteratively in the inversion process. In this paper, a source-independent method is adpoted by normalizing the wavefield. Furthermore, the gradientbased methods for FWI suffer from slow local convergence rate. A Hessian-free (HF) Gauss-Newton method is implemented in this research by solving the Newton system with a conjugate-gradient (CG) method. The CG method only needs Hessian-vector products instead of constructing the Hessian matrix explicitly. In this paper, the Hessian operator in HF Gauss-Newton method is corrected combining with the source-independent method. We demonstrate with numerical examples that the proposed strategies can improve the convergence rate and reduce the computational burden.