Abstract
We propose an improved scheme based on a rank reduction algorithm to separate the simultaneous source data generated by the simultaneous source acquisition. Obviously, the interferences from other sources will increase the Hankel matrix rank to full rank. As a consequence, the rank reduction algorithm can be used to suppress the interferences effectively. Unfortunately, it is difficult to set a suitable rank due to the complexity of seismic data, resulting in residual noise or damaged signals. To address the shortage, we use a flattening operator to simplify the simultaneous source data, and then apply the rank reduction algorithm to process the data iteratively. In the proposed iterative framework, the flattening operator and rank reduction operator can be written into a coherence-operator to shape the deblending inversion problem. Examples of simulated synthetic and field data have been used to test the effectiveness of the proposed method.
Hongling Chen
Phd candidate student of Information and Communication Engineering
My research interests include seismic data high-resolution and parameters inversion, seismic proccesing.