Filtering and Outlier Detection

In addition to noise, process data may also contain outliers (gross er- 

---

In section Structural Parametric Identification by Extended Kalman Filter, online structural parametric identification using the EKF will be briefly reviewed. In section Online Identification of Noise Parameters, an online identification algorithm for the noise parameters in the EKF is introduced. Then, in section Outlier-Resistant Extended Kalman Filter, an online outlier detection algorithm is presented, and it is embedded into the EKF. This algorithm allows for robust structural identification in the presence of possible outliers. In section Online Bayesian Model Class Selection, a recursive Bayesian model class section method is presented for non-parametric identification problems. 

In this section, the outlier-resistant extended Kalman filter (OR-EKF) is introduced for online outlier detection and robust structural parametric identification using dynamic response data. In this algorithm, an online outlier detection algorithm is embedded into the EKF. Section Online Outlier Detection by Outlier Probability introduces the concept of outlier probability, and it will be utilized for online outlier detection. This outlier detection algorithm is embedded in the EKF for online structural identification. Section Procedure of the Outlier-Resistant Extended Kalman Filter summarizes the procedure of the OR-EKF algorithm. 

Filtering refers to exclusion of bad probes on the array or the whole array. Visual inspection of array images and color-coded plot images are the first tools that should be used to detect outlier probes or arrays [23]. Clustering (described below) is another common way to identify outlier array among the whole set of arrays in a study. 

---