DCSCS: A Novel Approach to Improve Data Accuracy for Low Cost Meteorological Sensor Networks

摘要

The technology of Wireless Sensor Network (WSN) has facilitated the meteorological parameter monitoring. To reduce the cost, most monitoring systems collect desired data through cheap, unprofessional meteorological equipments, so do our system. However, due to the low cost of equipments and the poor package, the data collected by unprofessional equipments is inaccurate. In this study, we propose a Data Correction Scheme for Cheap Sensor (DCSCS) to improve data accuracy. The Discrete Wavelet Transform algorithm is used to filter the collected data to clear the noise and find the data pattern. Then we use Back Propagation neural network to establish a data correction model. Finally, we use the model to correct the other data. Our experiment shows that the meteorological data corrected by our method performs a very good result both in maximum error, mean error and correlation coefficient.