Missing Data Reconstruction Using Adaptively Updated Dictionary in Wireless Sensor Networks

Abstract

Due to external interference or fault, the collected sensor data is often missed or abnormal. It's significant to reconstruct the missing data, especially the large-scale missing data. In this paper, a missing sensor data reconstruction method based on the adaptively updated dictionary is presented. The K-SVD algorithm is used to train the historical data frames which are collected at different time to generate the original dictionary atoms. Moreover, in order to meet the realtime, continuous characteristics of sensor data, an adaptive dictionary update algorithm is studiedwhich. It calculates the correlation between the current reconstructed data frame and the largest weight frame in the training dictionary to update the dictionary. The experimental results are fully analyzed by the open data set. The results show that the proposed method has higher reconstructed precision especially the interval of data frames which is more than 60 minutes compared with other commonly used methods .

Reconstructing Missing and Anomalous Data Collected from High-Frequency In-Situ Sensors in Fresh Waters

“…More recently, but also based on daily data, various infilling techniques, such as regression, scaling, and equi-percentile approaches [ 24 ], along with dynamic regression models [ 25 ], have been used to reconstruct missing streamflow data. Methods developed in other domains, such as computer science, have reconstructed missing sensor data based on temporal or spatial correlation, interpolation, and sparse theory [ 26 ]. In sensor networks, linear and non-linear regression methods have been developed that use the non-missing data adjacent to the missing data [ 27 ], along with algorithms based on combining K-means algorithms and neural networks with particle swarm optimization [ 28 ].…”

IoT Devices Signals Processing Based on Shepard Local Approximation Operators Defined in Riesz MV-Algebras

“…Missing data can be caused by many things, but most of the times it is due to a malfunction of an IoT device or a communication problem between the IoT device and the processing applications. There is a continuous focus on finding new methods to fill in the missing data using various mathematical methods (Zhao and Zheng, 2017;Ruan et al, 2017;Leturiondo et al, 2017;Xu et al, 2017), methods that can be used to develop software modules that act as input validators for industrial automated control systems. In reality, the signal collected by IoT devices creates a discrete-time signal from a continuous process, called sample (Rajeshwari and Rao, 2008).…”