Data Aggregation in Wireless Sensor Network Based on Deep Learning Model
Qiu Lid · Chuangan jishu xuebao · 2014
In order to improve the performance of data fusion in wireless sensor network,a data aggregation algorithm SAEMDA( stacked autoencoder model data fusion algorithm) based on deep learning model was proposed,which combined stacked autoencoder( SAE) and wireless sensor network clustering routing protocol. Feature extraction and classification model( SAEM) is designed by SAEMDA to extract and classify the data features of nodes in each cluster,and then SAEMDA sends the features fused in the same class to Sink node. Either offline supervised learning algorithm or online unsupervised learning algorithm can be used to train the SAEM. Simulation results show that compared with BPFDA and SOFMDA,SAEMDA can improve the data fusion accuracy by 7.5 percentage points at most in similar situations of energy consumption.