Real-time Processing for Remote Sensing Satellite Data Based on Stream Computing

With the development of data achieving ability of high resolution remote sensing satellites and the enhancement of data receiving ability on the ground, the data processing workload of the existing ground application system for remote sensing satellites is growing, and the demand of real-time data processing is increasingly higher. In recent years, stream computing has become a research hotspot due to the high performance for real-time concurrent processing and distributed computing. In this paper, a new real-time processing method of remote sensing satellite data is proposed by the framework of stream computing. Firstly, according to the characteristics of remote sensing data processing, the stream computing is modeled, the processing time of the instance is abstracted, and the multi-task optimization and scheduling methods are given. Then the real-time data processing system of remote sensing satellite is implemented by using this method, and the data tuples, processing components and task topology are redesigned. Finally, the data processing time and throughput rate of the system are tested and analyzed. Experimental results show that the real-time performance of the system is greatly improved.