The Study Of Least-square Migration With Frequency-division Encoding

With the development of seismic exploration target area increasingly complex, for the sake of efficient lighting for the complicated target area, wide-azimuth even full-azimuth acquisition is usually used to collect field data. This method can improve the imaging quality in the target zone, but multiples the data size. At the same time, as the precision of the migration image is remand to increase again and once again, a lot of high precision migration imaging method arises. All of those have brought serious challenges to data processing. How to improve the efficiency of the collection and processing under the premise of ensuring the imaging quality, become the focus of many scholars study.In recent years, blended data is more and more get the attention of many scholars in the field of seismic exploration. blended data acquisition increases the shot point sampling rate and the reflection information of per unit volume, enhances the efficiency of field data acquisition. Blended data imaging using linear characteristics of seismic wave field propagation, synthesizes one shot data to super-gathers before migration. this method can immensely improve the efficiency of migration by reducing the number of migration, However, because of the crosstalk generated by irrelevant souce-field and receiver-field when imaging, the quality of the images is greatly reduced. In order to reduce or remove the crosstalk and obtain high-quality migration image, many scholars have done research and found that encoding is a effective method. Based on deep understanding of the previous research of other experts, this article will introduce the principle of encoding method in detail, list some familiar coding method,compare those methods to frequency-division code and demostrte the effectiveness of the method in 2D SEG/EAGE model. Some other study such as code scheme and supergather number will also be studied in this article.LSM(Least-Square Migration) is also a recent hot topics in the study of experts and scholars. Conventional migration operator is just the conjugate transpose of the seismic wave modeling operator, and not the inverse of the modeling operator, so its imaging ability is limited. The LSM take the imaging as a inversion problem, can make the images get higher resolution and higher fidelity. In this paper, model matching data of the least-squares sense will be expounded, the general steps of solving the optimization problem will be given, and the one-way wave continuation will be deduced. We will combine the split-step LSM and coding method to migrate the blended data.