GNSS坐标时间序列的小波去噪

Wavelet Denoising to the GNSS Position Time Series

针对影响小波去噪质量的各种因素进行分析,确定最佳分解层次,利用软阈值法对GNSS坐标时间序列进行小波去噪,使用信噪比和均方根误差来衡量去噪效果。分析表明,阈值估计准则是小波去噪的关键,不同阈值的去噪结果存在明显差异,利用启发式Stein无偏风险阈值,对水平方向的坐标时间序列可以得到较为理想的去噪效果。

In this paper the various factors for the influence on the denoising quality are analysed to confirm the best decomposition level,and the soft threshold method is used in the wavelet denoising to the GNSS position time series with SNR(Signal Noise Ratio)and RMSE(Root Mean Squard Error)for evaluating the equality of denoising.The analysis shows that there are obvious differences between the different criterions of threshold estimation which are the key to the wavelet denoising,and the better effect for the horizontal position time series can be obtained by the heuristic threshold method of Stein unbiased risk.