Determining missing values in dimension incomplete databases using spatial-temporal correlation techniques

In recent years, wireless sensor networks (WSNs) are extensively used in environment monitoring applications. It is paramount that data from these sensors be reliable since it could be used for critical decision making. However the data acquired is typically not usable directly as it suffers from noise, missing data and incompleteness. When the dimensionality of the collected data is lower than its actual dimensionality, the correspondence relationship between dimensions and their associated values is lost resulting in dimension incomplete problem. Querying incomplete databases has gained substantial research interests. Many techniques are being proposed to deal with incomplete databases by estimating and replacing missing sensor values using a well-suited statistical imputation technique. Some of the methods which are applicable to impute missing data in sensor readings are WARM (Window Association Rule Mining), AKE (Applying K-nearest Neighbor Estimation) however these methods are used as avoidance methods, which detect the presence of incomplete data and impute the value for the missing value before storing the data into the database, to further avoid querying dimension incomplete databases. No substantial research has been focused to deal with missing values present in the existing databases. Querying such dimension incomplete databases could lead to obtaining incomplete results. Considering this limitation this paper proposes to incorporate the above avoidance methods as a part of searching dimension incomplete databases and also proposes newer version to the existing WARM method. The advantage of the proposed approach is that the result of the user query will always have complete and accurate data.