Font Size: a A A

Satellite Radar Altimeter Waveform Simulation And Retracking Based On Singular Spectrum Analysis

Posted on:2020-10-13Degree:MasterType:Thesis
Country:ChinaCandidate:W Q DiFull Text:PDF
GTID:2480306308452134Subject:Geodesy and Survey Engineering
Abstract/Summary:PDF Full Text Request
Satellite altimetry was applied to marine geodesy,geophysics and oceanography with its unique advantages.With the improvement of satellite orbit determination precision and observation error correction theory,Waveform retracking became the key to improve the quality of satellite altimetry data.In order to better improve the altimeter data quality,this paper proposed a new waveform retracking method based on singular spectrum analysis(SSA).The main research contents of this paper were as follows:(1)Radar altimetry waveform simulation.According to the formation mechanism of radar altimeter echo,the theoretical echo was simulated based on the Brown-Hayne model,and the effect factors of altimeter echo shape were analyzed.Based on the Brown-Hayne model,the theoretical waveform was simulated.The Gaussian white noise sequence with specific mean and amplitude was added to the trailing edge of the theoretical waveform to represent the actual waveform trailing edge oscillation.The Gaussian peak function was used to generate the waveform peaks to represent the actual waveform peaks.A total of six types of waveforms,such as ocean waveform,ocean-like waveform,double-peak waveform,ladder waveform,single-peak waveform,and other waveform,were simulated.The simulation waveforms were arranged in a certain order to construct waveform series,including from land to ocean and ocean to land modes.(2)Waveform retracking based on SSA.Based on the principle of SSA algorithm,the denoising and reconstruction process of SSA were used to filter the waveform.The OCOG,threshold algorithm and 5-? algorithm were analyzed.The original waveforms series was decomposed,denoised,and reconstructed by SSA firstly,and the reconstructed waveform series was retracked by OCOG,threshold and 5-? algorithm secondly.(3)Waveform retracking method verification.The new method proposed in this paper was verified by the simulated waveform series and the 20 Hz waveform data of 19.6°N?23.8°N in the South China Sea region of Jason-1 satellite cycle110 pass088.The simulation waveform series were processed by SSA to discuss the influence of different window lengths on the reconstructed waveforms,and the result showed that the best selection of window length was the total number of single waveform gates.The reconstructed waveforms were retracked.It was found that the retracking gates of reconstructed waveforms were closer to the tracking gate 32.5,which proved that the denoising effect of the SSA was obvious.The retracking results of two simulation waveform series of from sea to land and from land to sea modes were identical,indicating that the SSA was not affected by the waveform series modes.The original measured waveforms were denoised and reconstructed by SSA to obtain reconstructed waveforms.The original waveforms and the reconstructed waveforms were retracked by OCOG,threshold algorithm and 5-?algorithm respectively.Using the improvement percentage(IMP)to evaluate the sea surface height(SSH),it was found that the IMP of threshold algorithm based on SSA was largest,whici was 41.8%with highest precision.The crossover point residual of SSH in South China Sea region was obtained,and the crossover point residual of threshold algorithm based on SSA is 0.1483m,which was less than the 0.1563m corresponding to the threshold algorithm.(4)The mean sea surface height model constrcution in South China Sea.The SGDR data of Jason-1 satellite of all available cycles from 2002 to 2013 in South China Sea was retracked by threshold algorithm based on SSA to correct the SSH data.After collinear adjustment,crossover adjustment and grid process,a new mean sea surface height(MSSH)model with resolution of 2'×2' in the South China Sea was established.Comparing the new MSSH model with the WHU2013,CLS15 and DTU15 global MSSH models in the South China Sea region,it had found that the established MSSH model in this paper was closest to the DTU15 model.The RMS of residual between new MSSH model and DTU15 model was 6 cm,and the MEAN of the residual was controlled within 5 cm.The result showed that the MSSH model in the South China Sea established in this paper met the accuracy requirements.
Keywords/Search Tags:Satellite altimetry, Singular spectrum analysis(SSA), Waveform retracking, Waveform simulation, Mean sea surface height model(MSSH)
PDF Full Text Request
Related items