Project Description
Functional Data Analysis-aided Learning Methods for RobustWireless Measurements
Many existing data-driven solutions are black-box approaches, may not be robust and adaptive, and work only for low-dimensional and discrete data. In fact, wireless data belong to the class of functional data, which can be represented by curves or functions. Advancements in modern Internet technology have enabled the collection of sophisticated, high-dimensional wireless datasets, which triggered the study of functional data for wireless measurements. Our focus on functional data analysis (FDA) in this project is also motivated by: (i) in many wireless measurements research, the data generating process naturally follows a stochastic function; (ii) many wireless data-driven problems can be better approached if the data are considered as functions; (iii) FDA can overcome the curse of dimensionality problem faced by traditional approaches; and (iv) FDA can provide explainable results such as performance guarantees. Recognizing the significance of the aforementioned problems and observations, a team of three investigators from Auburn University (AU) and Florida International University (FIU), with complementary expertise will work on developing flexible and practical methodologies and methods for dealing with large and complex wireless measurement data by applying FDA.
Oct. 1, 2023 ~ Sept. 30, 2026
Project Team
Related Publications (journal & magazine)
-
Yang Yang, Mingzhe Chen, Yufei Blankenship, Jemin Lee, Zabih Ghassemlooy, Julian Chen, and Shiwen Mao, “Positioning using wireless networks: Applications, recent progress, and future challenges,” IEEE Journal on Selected Areas in Communications, to appear. DOI: 10.1109/JSAC.2024.3423629.
-
S. Wang, Z. Shang, and G. Cao, “Optimal classification for functional data,” Statistica Sinica, vol.34, pp.1545-1564, 2024.
-
S. Wang and G. Cao, “Multiclass classification for multidimensional functional data through deep neural networks,” Electronic Journal of Statistics, vol.18, no.1, pp.1248-1292, Mar. 2024. DOI: 10.1214/24-EJS2229
-
S. Wang, G. Cao, Z. Shang, Scandinavian Journal of Statistics, “Deep neural network classifier for multidimensional functional data,” vol.50, no.4, pp.1667-1686, Dec. 2023. https://doi.org/10.1111/sjos.12660
-
S. Wang, G. Cao, and Y. Huang, “Review on functional data classification,” WIREs Computational Statistics, e1638, Nov. 2023. https://doi.org/10.1002/wics.1638.
-
X. Wang, Z. Yu, S. Mao, J. Zhang, S. C.G. Periaswamy, and J. Patton, “MapLoc: LSTM-based location estimation using uncertainty radio maps,” IEEE Internet of Things Journal, vol.10, no.15, pp.13474-13488, Aug. 2023. DOI: 10.1109/JIOT.2023.3262619.
Related Publications (conference)
-
Ziqi Wang and Shiwen Mao, “AIGC for wireless data: The case of RFID-based human activity recognition,” in Proc. IEEE ICC 2024, Denver, CO, June 2024, pp.4060-4065.
-
Tianya Zhao, Ningning Wang, Guanqun Cao, Shiwen Mao, Xuyu Wang, “Functional data analysis assisted cross-domain Wi-Fi sensing using few-shot learning,” in Proc. IEEE ICC 2024, Denver, CO, June 2024, pp.4780-4785.
Related Publications (others)
-
T. Zhao, X. Wang, S. Mao, S. Vucetic, and J. Wu, “Adversarial machine learning for wireless localization,” Chapter 8 in Network Security Empowered by Artificial Intelligence, Y. Chen, J. Wu, P. L. Yu, and C. Wang (Editors), Berlin, Germany: Springer Nature, 2024. DOI: 10.1007/978-3-031-53510-9_8.
-
G. Cao, “FDASTATAUBURN/FDADNN: Functional regression via deep neural network,” [online] Available: https://github.com/FDASTATAUBURN/FDADNN.
We acknowledge the generous support from our sponsor
This project is supported in part by the National Science Foundation under Grants CNS-2319342 and CNS-2319343. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the foundation.