Title: Imputation of missing sensor data values using in-exact replicas

Authors: Michael W. Bigrigg, H. Scott Matthews, James H. Garrett

Addresses: Department of Civil and Environmental Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA. ' Department of Civil and Environmental Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA. ' Department of Civil and Environmental Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA

Abstract: Sensor network systems attempt to exploit redundancy which is assumed to be inherent in a pervasive sensor network due to the density of the sensor nodes. However, there is a lack of sensor data stream correlation between seemingly redundant sensors. The approach presented the uses and prediction techniques that have been used to anticipate a sensor value using its own history, to allow one sensor to be able to anticipate the data from another different sensor, given a historical stream of data from the sensor to be anticipated. The specific mechanisms explored are artificial neural networks, pattern recognition and multivariate polynomial regression.

Keywords: imputation; infrastructure management; missing data; redundancy; replicas; sensor data; sensor networks; sensor values; artificial neural networks; ANNs; pattern recognition; multivariate polynomial regression; data stream correlation.

DOI: 10.1504/IJISTA.2009.025103

International Journal of Intelligent Systems Technologies and Applications, 2009 Vol.7 No.1, pp.4 - 23

Available online: 12 May 2009 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article