Department of Electrical & Computer Engineering UC Davis

NSF SpecEES Project

PI: Z. Ding and L. Lai (UC Davis)

Visit PI Institutional Research Pages for more related works: Prof. Z. Ding and Prof. L. Lai

Project Overview

The overarching goal of this project is to develop a robust and secure wireless function computing framework to specifically address spectrum efficiency, energy efficiency, and security issues in wireless networks for IoT applications. The underlying principle of our framework is that in many IoT applications, instead of recovering the full data observed by various devices, the role of wireless networks is to assist certain decision making based on function computation of distributed data. Such IoT devices can send all their raw data to the decision maker, which can then compute the function of interest. However, this approach will incur a tremendous amount of unnecessary overheads since most of the raw data are discarded after the functions of interest are computed, and require more costy spectrum/energy. The main idea of this project is to develop secure and spectrum/energy efficient protocols that enable the decision maker to compute functions of interest without first recovering the full data from sensing devices. Thus, instead of being treated as mere data pipes, wireless links become an integral part of the smart decision process in IoT applications. This approach will undoubtedly improve spectrum and energy efficiency of wireless IoT systems and reduce decision delays.

Research Thrust Synopsis
  • Robust distributed optimization algorithms for IoT systems. Data in IoT systems are often collected at IoT devices at distributed locations. In this setup, distributed optimization algorithms can suitably compute parameters of interest, as these methods do not require the transfer of all data to a centralized location. However, existing distributed optimization methods are not robust to networking errors or adversarial attacks. Existing algorithms may fail even if a single IoT device is compromised. In this thrust, we aim to design optimization algorithms that are robust against link errors and adversarial attacks.
  • Over-the-air computation for fast data fusion from distributed IoT devices. Sensor data from distributed low power IoT devices are often sent to fusion centers for joint processing and analysis. IoT systems can achieve high energy and spectrum efficiency by enabling grant-free access and over-the-air computations. In this thrust, we focus on blind over-the-air fusion and computation without requiring orthogonal access or pilot resources for a priori estimation of channel state information.
  • End-to-end training and learning of networked artificial intelligence systems. Multimedia based detection and classification in a networked environment is among the top artificial intelligence (AI) applications. Traditional media encoders developed for human subjects are not optimized for AI tasks. In this thrust, we study the end-to-end joint optimization of media encoder and deep learning architecture to customize media encoders for specific machine learning objectives.
  • Exploring graph signal processing for data processing. Graph signal processing (GSP) has demonstrated strong promises in modeling and capturing signal features from multiple (e.g. distributed) sources. However, GSP is limited in processing pair-wise signal relationships. In this thrust, we generalize GSP by investigating the efficacy of higher-graph signal processing (HGSP) to model multi-lateral interactions of practical signals from sensor clusters.
Papers
  1. S. Zhang, S. Cui and Z. Ding, ``Hypergraph Spectral Analysis and Processing in 3D Point Cloud," in IEEE Transactions on Image Processing, vol. 30, pp. 1193-1206, 2021, doi: 10.1109/TIP.2020.3042088.

  2. C. J. Feres and Z. Ding, ``Wirtinger Flow Meets Constant Modulus Algorithm: Revisiting Signal Recovery for Grant-Free Access,'' in IEEE Transactions on Signal Processing, vol 69, pp. 6515-6529, Aug. 2021. doi: 10.1109/TSP.2021.3103038.

  3. Q. Deng, S. Zhang and Z. Ding, ``An Efficient Hypergraph Approach to Robust Point Cloud Resampling,'' IEEE Transactions on Image Processing, vol. 31, pp. 1924-1937, 2022, doi: 10.1109/TIP.2022.3149225.

  4. C. Feres, ``Geometrical Frameworks for Wireless Access in Large Scale Multi-Antenna Networks", PhD Dissertation, University of California Davis, Jan. 2022

  5. Y. Jin and L. Lai, ``Privacy Protection In Learning Fair Representations”, Proc. IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 2964-2968, Singapore, 2022.

  6. S. Qi, L. D. Chamain, and Z. Ding, ``Hierarchical Training for Distributed Deep Learning based on Multimedia Data over Band-limited Networks’’, Accepted. IEEE International Conference on Image Processing, Bordeaux, France, 2022.

  7. C. Feres, B. C, Levy and Z. Ding, ``Over-the-Air Collaborative Learning in Joint Decision Making" IEEE Global Communications Conference, Accepted, Rio de Janeiro, Brazil, Dec. 2022`

  8. P. Zhao and L. Lai, ``Minimax Optimal Estimation of KL Divergence for Continuous Distributions," IEEE Trans. Information Theory, vol. 66, no. 12, pp. 7787-7811, Dec. 2020.

  9. Y. Jin and L. Lai, ``On the Adversarial Robustness of Hypothesis Testing," IEEE Trans. Signal Processing, vol. 69, pp. 515-530, 2021.

  10. G. Liu and L. Lai, ``Action-Manipulation Attacks Against Stochastic Bandits: Attacks and Defense," IEEE Trans. Signal Processing, vol. 68, pp. 5152-5165, 2020.

  11. C. J. Feres and Z. Ding, ``Wirtinger Flow Meets Constant Modulus Algorithm: Revisiting Signal Recovery for Grant-Free Access," IEEE Trans. Signal Processing, 2021.

  12. Y. Jin and L. Lai, ``Privacy-Accuracy Trade-Off of Inference as Service," IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2021, pp. 2645-2649.

  13. F. Li, L. Lai and S. Cui, ``On the Adversarial Robustness of Linear Regression," IEEE 30th Intl. Workshop on Machine Learning for Signal Processing (MLSP), 2020.

  14. F. Li, L. Lai and S. Cui, ``On the Adversarial Robustness of Feature Selection Using LASSO," IEEE 30th Intl. Workshop on Machine Learning for Signal Processing (MLSP), 2020.

  15. S. Zhang, S. Cui and Z. Ding, ``Hypergraph-Based Image Processing," 2020 IEEE International Conference on Image Processing (ICIP), 2020, pp. 216-220, doi: 10.1109/ICIP40778.2020.9190874.

  16. S. Zhang, S. Cui and Z. Ding, ``Hypergraph Spectral Clustering for Point Cloud Segmentation," IEEE Signal Processing Letters, vol. 27, pp. 1655-1659, 2020, doi: 10.1109/LSP.2020.3023587.

  17. S. Zhang, S. Cui and Z. Ding, ``Hypergraph Spectral Analysis and Processing in 3D Point Cloud," IEEE Transactions on Image Processing, vol. 30, pp. 1193-1206, 2021, doi: 10.1109/TIP.2020.3042088.

  18. Lai, Lifeng, and Bayraktar, Erhan, ``On the Adversarial Robustness of Robust Estimators,'' IEEE Transactions on Information Theory, vol. 66, no. 8, pp. 5097-5109, August, 2020.

  19. L. D. Chamain and Z. Ding, ``Improving Deep Learning Classification of JPEG2000 Images Over Bandlimited Networks,'' IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4062-4066, Barcelona, Spain, 2020.

  20. Liu, Guanlin, and Lai, Lifeng, ``Action-Manipulation Attacks on Stochastic Bandits,'' Proc. of 45th International Conference on Acoustics, Speech, and Signal Processing (ICASSP) pp. 3112-3116, Barcelona, Spain, 2020.

  21. J. Dong, Y. Shi, and Z. Ding, ``Blind Over-the-Air Computation and Data Fusion via Provable Wirtinger Flow,'' IEEE Trans. on Signal Processing, vol. 68, pp. 1136-1151, 2020.

  22. Jalali, Amin, and Ding, Zhi, ``Joint Detection and Decoding of Polar Coded 5G Control Channels,'' IEEE Transactions on Wireless Communications, vol. 19, no.3, pp. 2066-2078, March 2020.

  23. X. Ma, L. Lai and S. Cui, `` Optimal Two-Stage Bayesian Sequential Change Diagnosis," IEEE International Symposium on Information Theory (ISIT), 2020, pp. 1130-1135, doi: 10.1109/ISIT44484.2020.9173938.

  24. Li, Fuwei, Lai, Lifeng, and Cui, Shuguang, ``On the Adversarial Robustness of Subspace Learning'', IEEE Transactions on Signal Processing, vol. 68, pp. 1470-1483, 2020.

  25. S. Zhang, Z. Ding and S. Cui ``Introducing Hypergraph Signal Processing: Theoretical Foundation and Practical Applications,'' IEEE Internet of Things Journal, vol. 7, pp. 639-660, Jan. 2020.

  26. Jin, Yulu, and Lai, Lifeng, ``Adversarially Robust Hypothesis Testing'', Proc. of Asilomar Conference on Signals, Systems, and Computers, pp. 1806-1810, Pacific Grove, CA, USA, 2019.

  27. Li, Fuwei, Lai, Lifeng, and Cui, Shuguang, ``On the Adversarial Robustness of Subspace Learning,'' Proc. 44th International Conference on Acoustics, Speech, and Signal Processing (ICASSP) pp. 2477-2481, Brighton, United Kingdom, 2019.

  28. Tu, Wenwen and Lai, Lifeng, ``On Function Computation With Privacy and Secrecy Constraints,'' IEEE Transactions on Information Theory, 65 (10), pp. 6716-6733, 2019

  29. Cho, Myung and Lai, Lifeng and Xu, Weiyu, ``Generalized Distributed Dual Coordinate Ascent in a Tree Network for Machine Learning,'' 2019 IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 3512-3516, Brighton, UK, May 2019.

  30. Wu, Wenhao and Ding, Zhi, ``A Markovian Design of Bi-Directional Robust Header Compression for Efficient Packet Delivery in Wireless Networks'', IEEE Transactions on Wireless Communications, 18 (1), pp. 20-33, 2019.

  31. Bouchoucha, Taha and Chuah, Chen-Nee and Ding, Zhi, ``Topology Inference of Unknown Networks Based on Robust Virtual Coordinate Systems'', IEEE/ACM Transactions on Networking, 27 (1), pp. 405-418, 2019

  32. X. Cao and L. Lai, ``Distributed Approximate Newton's Method Robust to Byzantine Attackers,'' IEEE Trans. on Signal Processing, , vol. 67, pp. 6011-6025, 2020.

  33. X. Cao and L. Lai, ``Distributed Gradient Descent Algorithm Robust to an Arbitrary Number of Byzantine Attackers,'' IEEE Trans. on Signal Processing, , vol. 67, pp. 5850-5864, 2019.

  34. Zhao, Puning, and Lai, Lifeng, ``Analysis of KNN Information Estimators for Smooth Distributions,'' IEEE Transactions on Information Theory, vol. 66, No. 6, pp. 3798-3826, 2019.

  35. Zhang, Han, Ai, Bo, Xu, Wenjun, Xu, Li, and Cui, Shuguang, ``Multi-Antenna Channel Interpolation via Tucker Decomposed Extreme Learning Machine,'' IEEE Transactions on Vehicular Technology vol. 68, no.7, pp. 7160-7163, 2019.

  36. L. D. Chamain and Z. Ding, US Patent Application Serial No. 64/048,582: ``Quannet: Joint Image Compression and Classification over Channels with Limited Bandwidth'', filed on 06 July 2020.


    37. Return to PI Institutional Research Pages for more related works: Prof. Z. Ding and Prof. L. Lai

Questions or comments? Email (zding(a)ucdavis.edu).