It is a future trend to use wireless sensor network to detect operation state of large machinery and carry out fault diagnosis.A rolling bearing fault diagnosis model based on wireless sensor network was proposed

and principal component analysis and neural network technology were used to carry out wireless network data fusion in the network

with which original vibration signals were converted into fault type

in this way

correspondence and network energy consumption were reduced.In the experiment

the test subject was a deep groove ball bearing which is commonly used in engineering.Vibration data in normal and fault state collected by sensors were analyzed and calculated

data fusion of feature level was carried out on cluster nodes to extract fault feature in order to find out fault type.Experiment results prove that the model and the data fusion algorithm are reliable and effective.