Abstract | ||
---|---|---|
We propose a Bayesian network based method for the fault diagnosis problem of medical body sensor networks used to collect physiological signs to monitor the health of patients. We formalize a Bayesian network to describe the body sensor network considering both the spatial and temporal correlation in measurements at different sensors. Then we give the theoretical analysis of the fault detection, false alarm of this method, and the error probability after executing the fault diagnosis algorithm. Finally, Experiments carried out on synthetic medical datasets by injecting faults into real medical datasets show that the simulation performance matches the theoretical analysis closely, and the proposed approach possesses a good detection accuracy with a low false alarm rate. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/MSN.2015.21 | MSN |
Keywords | Field | DocType |
fault diagnosis, body sensor networks, Bayesian network | Data mining,False alarm,Pattern recognition,Computer science,Fault detection and isolation,Bayesian network,Artificial intelligence,Constant false alarm rate,Probability of error,Wireless sensor network | Conference |
Citations | PageRank | References |
3 | 0.39 | 13 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Haibin Zhang | 1 | 118 | 18.58 |
Jiajia Liu | 2 | 1372 | 94.60 |
Rong Li | 3 | 3 | 0.73 |