基于改进CEEMDAN分解的GNSS-RTK监测信号处理方法

A GNSS-RTK Monitoring Signal Processing Method Based on Improved CEEMDAN Decomposition

GNSS-RTK监测精度受多路径效应和仪器内部白噪声的影响,为此,提出基于有效系数筛分本征模式函数(IMF)分量的改进CEEMDAN算法分离信号和噪声,以提升GNSS-RTK监测精度.首先,通过仿真信号验证该算法的白噪声分离效果.然后,进行GNSS-RTK连续多天定点观测试验,验证采用改进CEEMDAN算法提取多路径误差的可行性.接着,在开阔的平坦地面和水面分别进行GNSS-RTK稳定性试验,分析不同环境下的多路径效应.最后,以天津海河大桥为对象,对GNSS-RTK监测信号进行滤波降噪,进而提取桥梁在环境激励下的动力响应.结果表明:改进CEEMDAN算法能够实现更精确的IMF分量分解与重构,从而有效分离高频白噪声分量和低频多路径误差;在开阔的平坦地面和水面环境下多路径效应影响均较小,多路径误差序列频域分布为地面(0.05 Hz)<水面(0.10 Hz).结合结构自振频率及现场环境确定带通滤波频带,对GNSS-RTK实测信号进行滤波降噪处理,成功提取了海河大桥的动态位移,结果与拾振器监测结果基本一致.

The monitoring accuracy of GNSS-RTK is affected by the multipath effect and white noise inside the instrument.Therefore,an improved CEEMDAN algorithm based on an effective coefficient to sift the IMF components is proposed to separate the signal and noise and improve the monitoring accuracy.First,the proposed algorithm was verified by signal simulation.Second,a consecutive and fixed-point monitoring experiment was conducted using GNSS-RTK to verify the reliability of the improved CEEMDAN in extracting the multipath errors.Next,GNSS-RTK stability tests were performed in an open flat ground and water area to analyze the multipath effect in different environments.Finally,taking Tianjin Haihe Bridge as an example,the GNSS-RTK monitored signal was filtered and the dynamic response of the bridge under environmental excitation was obtained.Results showed that the improved CEEMDAN algorithm can achieve a more accurate decomposition and reconstruction of the IMF components,thus effectively separating the high-frequency white noise components and low-frequency multipath errors.In the open flat ground and water area,the multipath effect exhibited a small influence.The frequency domain distribution of the multipath error sequence was as following:Ground(0.05 Hz)<water surface(0.10 Hz).The frequency band for filtering was determined considering the natural structural frequency and monitoring environment,based on which the GNSS-RTK signal was filtered and the dynamic displacement of the Haihe Bridge was successfully extracted. Themonitored results of GNSS-RTK were consistent with the results acquired from the vibration sensors.