Imputing the missing data in IoT based on the spatial and temporal correlation

Missing data in IoT occurs due to a variety of reasons such as unstable network communication, synchronization problems, unreliable sensor devices, environmental factors and other device malfunctions which often resulted in data incompleteness. Missing data imputation is the most common preprocessing task to dealing with incomplete data. Though missing data is common in IoT but missing data imputation is hardly seen in the IoT environment. As a result, when analytics is performed on IoT data with missing values, it leads to the decline in accuracy and reliability of the data analysis results. In this paper, a novel ST-correlated proximate missing data imputation model is proposed to deal with missing data problem in IoT. Experimental results have proved that the proposed method has outperformed the existing single imputation and multiple imputation statistical methods in terms of accuracy.