Hits:

Indexed by:会议论文

Date of Publication:2015-01-01

Included Journals:CPCI-S

Volume:37

Page Number:673-678

Key Words:Wireless Sensor Network; Missing Data Imputation; Modified Frequent Itemsets Mining; Nearest Neighbor Estimation

Abstract:Due to causes such as unreliability of the transport protocol, energy exhaustion and noise disturbance etc in wireless sensor network, the uploaded data on the sensor node usually tend to be incomplete, which brings about a series of inconvenience for analysis and operation in the subsequent. Therefore, it is necessary for us to make compensation for the missing data. In this paper, we put forward one kind of method of combing with modified Frequent Itemsets mining and NN search(FINN) to make estimation for the missing data in the wireless sensor network and use estimated data to replace missing value. Because the final operation only uses similar data, it is unnecessary to use all the data, so it can reduce unnecessary error and enhance precision of estimated value.