Reconstruction of big sensor data

With ubiquitous sensors continuously monitoring and collecting large amounts of information, there is no doubt that this is an era of big data. One of the important source for scientific big data is the datasets collected by Internet of things (IoT). For an IoT application to analyze big sensor data, it is necessary that the data are clean and lossless. However, data loss in IoT is common due to unreliable wireless link or hardware failure in the nodes. To reconstruct the big sensor data, this paper first presents an algorithm based on matrix completion method. Then for multi-attributes sensor data, a tensor-based method is provided to estimate missing values. Moreover, an effective solution is proposed using the alternating direction method of multipliers. Finally, the experiments with real-word sensor data show the effectiveness of the proposed methods.