A data reconstruction model addressing loss and faults in medical body sensor networks

Due to limited resource, noise and unreliable link, data loss and sensor faults are common in medical body sensor networks (BSN). Most available works used data reconstruction to improve data quality in traditional wireless sensor networks (WSN). However, existing data reconstruction schemes using redundant information of WSN can not provide a satisfactory accuracy for BSN. In light of this, a Bayesian network based data reconstruction scheme is formalized in this paper, which rebuilds data using conditional probabilities of body sensor readings to recover missing data and sensor faults, rather than the redundant information collected from a large number of sensors. Experiments on extensive online data set show that the performance of our scheme outperforms all available data reconstruction schemes.