College of Geo?exploration Science and Technology, Jilin University, Changchun 130026, China

 With respects to the conventional surface seismic profile (SSP) data, the vertical seismic profile (VSP) data usually have higher exploration accuracy, but the imaging range of VSP data is limited due to the limited observation system. In order to solve the problem of limited imaging range of VSP data, the authors have combined Radon transform with cross?correlation seismic interferometry and convolution seismic interferometry. The VSP data are transformed into virtual SSP data by virtual source method in the Radon domain. The characteristics of random noise is suppressed and the signal / noise ratio of virtual SSP records is improved by using the Radon transform meth? od. The effectiveness of virtual source method in the Radon domain is verified by numerical simulation, which is fur? ther applied to the original VSP data with the added random noise. The results show that compared with the tradition? al virtual source method, the virtual source method in the Radon domain can effectively suppress random noise and improve the signal / noise ratio of virtual source records. With respects to the original VSP data, the newly formed vir? tual SSP records after migration imaging can effectively expand the image range and improve the resolution.