Multi-variable data Imputation Method for Fog Computing enabled Industrial Internet of Things

This paper proposes a recovery method for missing multivariate time-varying data in Fog Computing-based Industrial Internet of Things. With the rapid development of Internet, handling high-dimensional non-Euclidean data has become a significant challenge. To address this issue, this paper employs Graph Convolutional Networks (GCN) to capture the underlying structure of the data and analyzes the relationships between datasets through convolutional networks. This method is capable of estimating true signals from partially missing or corrupted observations, effectively restoring missing sensor data. Through simulation experiments, the effectiveness of this method has been demonstrated in restoring missing data that follows Pearson correlation coefficients, identical distributions, and linear coefficients.