Search

GNSS reflectometry (GNSS-R) is a method to derive sea level using Signal to Noise Ratio (SNR) from the Global Navigation Satellite Systems (GNSS). SNR data consist of the direct signal from the satellite (multipath) and of the signals reflected by the sea surface, and hence separating the multipath is necessary to extract the signal from the sea surface. The process of separating multipath may affect the number of data and may eventually affect the quality of the derived sea level values. There are two multipath separation techniques that are mostly used: polynomial fitting and wavelet decomposition. This study investigates the performance of both techniques by applying them to analyze three months of the L1 SNR data of Global Positioning System (GPS) and Globalnaya Navigatsionnaya Sputnikovaya Sistema (GLONASS) as observed from two stations, Barus (CBRS) at North Sumatera from January 1 to March 31, 2022, and Morotai (CMOR) at North Maluku, Indonesia using data from February 1 to May 1, 2022. Comparison with sea level from tide gauge observations shows a high correlation for both techniques, with correlation coefficients of approximately 0.90 and 0.97 for CBRS and CMOR, respectively. The Root Mean Square Error (RMSE) of polynomial fitting for CBRS and CMOR have the same value, 11.5 cm, whereas those of wavelet are 11.4 cm and 11.5 cm. Since polynomial fitting and wavelet decomposition show similar performance, we conclude that both techniques give comparable accuracy of multipath SNR data for GNSS-R in Indonesia with appropriate quality control parameters. [ABSTRACT FROM AUTHOR]