Research And Implementation Of Grid Technology In MUSER Imaging

Along with the implementation of the SKA plan, radio interference array imaging is currently a hot research topic. As a new generation of solar radio interference arrays in our country, the process of data observation and processing of MUSER is similar to the SKA's. How to handle massive amounts of data generated by MUSER becomes a central issue for astronomy and technical workers.Aiming at the characteristics of MUSER mass observation data real-time processing, overall structure of MUSER observation data acquisition, and sampling and imaging process are analyzed in detail.This paper mainly discussed a research on the Gridding technology and its implementation in MUSER imaging, the concrete contends is as follows:1. According to the principle of synthetic aperture, we analyzed the process of telescope image reconstruction, and carried on extensive research on data acquisition, processing, and construction of MUSER observation and data processing system.2. Gridding alogrithm for MUSER imaging processing is studied in-depth, and the properties of the Fourier transform are analyzed. Especially, we analyzed the process of convolution algorithm of grid sampling function and weighted interpolation function, and the choice of optimal results are obtained.3. On the basis of the above research, the paper designed and implemented the high-speed real-time Gridding process for the imaging of MUSER. By using GPU parallel computing, the performance of Gridding has been greatly increased, at 4X?10X.In general, this paper studies and realizes MUSER imaging high performance Gridding, data processing has laid a good foundation for the construction of MUSER, this research realize the core algorithm based on CUDA parallel and successfully applied to MUSER data processing pipeline, which have achieved a very good performance.