Research Article
A novel data quality assessment framework for vehicular network testbeds
@INPROCEEDINGS{10.4108/eai.28-9-2017.2273211, author={Daxin Tian and Yukai Zhu and Jianshan Zhou and Xuting Duan and Yunpeng Wang and Jeungeun Song and Hui Rong and Peng Guo}, title={A novel data quality assessment framework for vehicular network testbeds}, proceedings={12th EAI International Conference on Testbeds and Research Infrastructures for the Development of Networks \& Communities}, publisher={EAI}, proceedings_a={TRIDENTCOM}, year={2018}, month={1}, keywords={connected vehicles data quality assessment data fusion machine learning vehicular network}, doi={10.4108/eai.28-9-2017.2273211} }- Daxin Tian
Yukai Zhu
Jianshan Zhou
Xuting Duan
Yunpeng Wang
Jeungeun Song
Hui Rong
Peng Guo
Year: 2018
A novel data quality assessment framework for vehicular network testbeds
TRIDENTCOM
EAI
DOI: 10.4108/eai.28-9-2017.2273211
Abstract
Big data technique is considered as a powerful tool to exploit all the potential of the Internet of Things and the smart cities. The development of internet of Vehicles (IoV) and wireless communication technologies have boosted diverse applications related to smart cities and Cyber-Physical Systems, but the data quality of vehicular sensors is an important issue due to the high-speed mobile wireless communication environment and physical sensor noise. This paper presents our experiences for big data analytics based on a vehicular network testbed, in terms of sensors data management, multi-dimension data fusion and data quality assessment for the vehicular sensor data. The proposed data quality assessment framework consist of feature extraction based on multi-sensor data fusion and multi-level wavelet transform, as well as a semi-supervised learning based classification algorithm. The comparison experiment shows that the proposed framework and approaches can extract feasible features and solve the unbalanced label problems, which achieve a better assessment effect.
