Application Of SSA Based On Improved Cao Algorithm In GNSS-MR Sea Level Height Retrieval

Real-time and accurate monitoring of sea level height is of great significance for human production and life,as well as research on ocean climate.Using traditional tide gauges to monitor sea level height has low fault tolerance and cannot obtain absolute sea level height.Using the Global Navigation Satellite System Multipath Reflectometry(GNSS-MR)technology to retrieve the signal noise to ratio(SNR)of GNSS station data for sea level height,making up for the shortcomings of tidal stations,has become a new method for sea level height monitoring.GNSS-MR technology uses quadratic polynomial fitting to separate direct and reflected signals,and uses Lomb-Scargle spectral analysis to process the residual sequence to ultimately obtain sea level height inversion values.However,when using quadratic polynomial fitting to obtain the residual sequence,there will be noise that affects the inversion accuracy.In this paper,singular spectrum analysis based on the improved Cao algorithm is used to decompose and reconstruct the original signal-to-noise ratio time series data to obtain the residual sequence after noise removal,thereby improving the accuracy of sea level height inversion.The main research contents and conclusions of this article are as follows:1.The research status of GNSS-MR technology for retrieving sea level height and singular spectrum analysis based on improved Cao algorithm at home and abroad is introduced,and the relationship between signal to noise ratio and multipath effect is described in detail.2.The basic principles of singular spectrum analysis based on the improved Cao algorithm are studied,including the decomposition and reconstruction of time series data by singular spectrum analysis and how the improved Cao algorithm determines the optimal embedding dimension.3.Using two methods,singular spectrum analysis and singular spectrum analysis based on the improved Cao algorithm,to retrieve sea level heights from GPS L1,L2,and L5 signal data,the results show that the sea level heights obtained by singular spectrum analysis and singular spectrum analysis based on the improved Cao algorithm are closer to the measured values of tidal stations compared to the quadratic polynomial method,Moreover,the inversion results of singular spectrum analysis based on the improved Cao algorithm are better than those of singular spectrum analysis,verifying the feasibility and superiority of singular spectrum analysis based on the improved Cao algorithm for sea level height inversion.4.The singular spectrum analysis based on the improved Cao algorithm is used to retrieve sea level height from GNSS multimode and multi frequency data(GPS and GLONASS),and the results are compared and analyzed.Then,an experimental analysis was conducted on the influencing factors of GNSS-MR sea level height retrieval,and an experimental comparison was conducted from three aspects: satellite altitude angle,satellite signal frequency,and satellite data sampling rate.The results showed that selecting the appropriate altitude angle range,appropriate satellite signal frequency,and high frequency satellite data sampling rate,the accuracy of sea level height retrieval values was higher.