基于卡尔曼和最小二乘的抗野值降噪方法研究

Study on Outliers Suppression Denoising Method Based on Kalman Filter and Least Square Fitting

为降低信号中野值对滤波的影响,提高卡尔曼滤波精度,提出了一种基于卡尔曼和最小二乘的抗野值降噪方法。方法在进行卡尔曼滤波过程中,首先利用莱特准则对野值进行判别,而后采用最小二乘拟合方法对野值进行修正,并通过引入调节因子降低连续型野值拟合误差的影响。实验结果表明,上述方法不仅能有效降低孤立型野值的影响,而且具有较强的抗连续型野值干扰能力,同时还具有较强的鲁棒性,是一种有效可行的抗野值降噪方法。

In order to eliminate the influence of outliers on filter and improve the precision of kalman filter,an outliers suppression denoising method based on kalman filter and least square fitting is proposed.When the Kalman filter was used to reduce the interference caused by noise,the outliers were identified by Letts criterion at first.Then the outliers were amended by Least Square algorithm,and the modulation factor was proposed to reduce fitting error of continuous outliers finally.The experiment results indicate that the prosed approach can reduce not only the interference caused by solitary outliers but also the interference caused by continuous outliers,and the robustness of the approach is high,thus it is an effective outliers suppression denoising approach.