Abstract: At present, the sea surface wind field observation methods are limited. The space-based observation in view of global navigation satellite system-reflection (GNSS-R) provides a new approach for global wind field sensing. The GNSS-R sea surface wind speed sensing technology has the characteristics of allweather, low power consumption, wide coverage, multiple signal sources and low cost, which has gained widespread attention gradually. Since the designed GNSS system primitively aimed to provide global users with passive navigation services based on radio pseudo-range measurement, and not designed as a space-based remote sensing radiation source, the GNSS-R reflected signal had the cheracteristics such as scattered observation points, weak signal power and signal processing algorithm with large amount of calculations. Taking the BF-1 satellite as the research object, the GNSS-R signal processing algorithm and system were studied. In view of the characteristics of GNSS-R signal, the requirements and constraints of space-borne real-time processing, it was proposed to combine the dynamic phase compensation technology on the short-term cycle correlation and rotation transform to reduce the GNSS-R signal processing burden without distortion. The computational burden of the proposed algorithm was 1/8.99 of the one for the traditional circular correlation method, meanwhile a FPGA based GNSS-R signal processing system was designed. The pay load was tested on the ground and in orbit. The results show that the GNSS-R signal processing system and the proposed signal processing algorithm meet the satellite’s requirements (5-frame per second). The DDM results have be used in the wind field inversion effectively.

Key words: space-borne GNSS-R, sea wind observation, payload, key technologies, design and verification