Towards Robust and Efficient Dynamic Spectrum Sharing with Knowledge Transfer
NSF Award CNS-2245918 · Principal Investigator: Dr. Lusi Li · Old Dominion University
Learning transferable knowledge for wireless systems that are robust, efficient, secure, and privacy-aware. Explore selected results in semantic communications, adversarial robustness, and incomplete multiview learning.
IEEE CNS 2026
Channel-aware universal adversarial perturbations expose vulnerabilities in cross-modal retrieval across varying wireless channel conditions.
IEEE GLOBECOM 2026
Text-triggered backdoor attacks manipulate image generation and reconstruction in diffusion-driven semantic communications while preserving normal behavior on benign inputs.
Paper · Coming soon
IEEE GLOBECOM 2026
Lightweight residual adaptation and channel-aware training improve cross-modal retrieval over noisy wireless channels while keeping pretrained encoders frozen.
Paper · Coming soon
IEEE MILCOM 2025
Energy-based calibration preserves semantic content during image transmission over noisy wireless channels.
ACM Multimedia 2025
Energy-guided imputation, attention fusion, and contrastive learning improve clustering with missing views.
AAAI 2026
Hierarchical imputation and dual alignment learn robust representations when some views are missing.
IEEE TNNLS · DOI: 10.1109/TNNLS.2025.3617888
Prototype-guided graph learning captures multilevel relationships and recovers missing views in the embedding space.
Implementations and reproduction instructions
ECSC — Energy-calibrated semantic communication · MILCOM 2025
SemCom-UAP — Channel-aware universal perturbations · CNS 2026
SemCom-DE-UAP — Data-efficient universal perturbations · MILCOM 2026
ReT-UAP — Transferable attacks on vision-language models · Information Fusion 2026
DIMVC-HIA — Hierarchical imputation and alignment · AAAI 2026
PGFormer — Prototype-graph transformer · IEEE TNNLS
ERL-Net (ESI-net repository) — Energy-guided selective imputation · IEEE TNNLS 2026
EnergyIMVC — Energy-based incomplete multiview clustering · ACM MM 2025
MICA — Multilevel imputation and contrastive alignment · Neural Networks 2025
SEMI — Structure exploration and missing-view inference · Information Fusion 2024
(IEEE MILCOM Workshop) Patrick Martin, Yiming DU, Chunjiang Zhu, and Lusi Li*. "Graph-Enhanced Cross-Modal Semantic Communication, " In IEEE MILCOM Workshop on LEAF: Lightweight Edge Adaptive Foundation Models for Edge Intelligence and Autonomous Reasoning, 2026. (Accepted)
(NIPS Workshop) Zemin Chen, Yao Wang, Junrui Zhang, Lusi Li, Daniel Takabi, and Rui Ning. "CRAE-GUARD: Clean-Referenced Absolute Evidence for Sampling-Robust Quantum Backdoor Detection," In NIPS 2026 Workshop on SaTQuML: Secure and Trustworthy Quantum Machine Learning, 2026. (Accepted)
(IEEE MILCOM) Ziyu Wang, Rui Ning, Shuai Hao, Mohammad GhasemiGol, Daniel Takabi, and Lusi Li*. " Data-Efficient Universal Adversarial Perturbations for Cross-Modal Retrieval in Semantic Communications," In MILCOM 2026-2026 IEEE Military Communications Conference (MILCOM). (Accepted). [PDF]
(IEEE CNS) Ziyu Wang, Yiming Du, Jian Li, Rui Ning, Shuai Hao, Daniel Takabi, and Lusi Li*. " SemCom-UAP: Universal Adversarial Perturbations Against Cross-Modal Retrieval in Semantic Communications," in 2026 IEEE Conference on Communications and Network Security (CNS). (Accepted). [PDF]
(IEEE GLOBECOM) Yiming Du, Jian Li, Ziyu Wang, Mustafa Ibrahim, Rui Ning, Daniel Takabi, Shuai Hao, and Lusi Li*. "RoSC: Robust Semantic Communication for Cross-Modal Retrieval." in GLOBECOM 2026-2026 IEEE Global Communications Conference. (Accepted). [PDF]
(IEEE GLOBECOM) Jian Li, Ziyu Wang, Yiming Du, Zemin Chen, Rui Ning, Daniel Takabi, and Lusi Li*. "DiffSem-BD: Text-Triggered Backdoor Attacks on Diffusion-Driven Semantic Communications," in GLOBECOM 2026-2026 IEEE Global Communications Conference. (Accepted). [PDF]
(IEEE QCE) Junrui Zhang, Zemin Chen, Lusi Li, Mohammad Ghasemigol, Daniel Takabi, and Rui Ning. "Input-Aware Dynamic Backdoor Attack Against Quantum Neural Networks, " in 2026 IEEE International Conference on Quantum Computing & Engineering (QCE). (Accepted). [PDF]
(IEEE ICDCS) Jiawei Chen, Lusi Li, Daniel Takabi, Masha Sosonkina and Rui Ning. "Heterogeneous Graph Backdoor Attack." In 2026 IEEE 46th International Conference on Distributed Computing Systems (ICDCS), pp. 872-882. IEEE, 2026. DOI: 10.1109/2575-8411.2026.00088.
(INFFUS) Ziyu Wang, Jian Li, Yiming Du, Rui Ning, Jiang Li, Daniel Takabi, and Lusi Li*. "ReT-UAP: Rethinking the Multi-Surrogate Universal Adversarial Perturbations on Vision-Language Pre-trained Models," Information Fusion, 2026.
(TMLR) Miao Lin, Feng Yu, Rui Ning, Lusi Li, Jiawei Chen, Qian Lou, Mengxin Zheng, Chunsheng Xin, and Hongyi Wu. "RPP: A Certified Poisoned-Sample Detection Framework for Backdoor Attacks under Dataset Imbalance." Transactions on Machine Learning Research (TMLR), 2026.
(IEEE TNNLS) Ziyu Wang, Jian Li, Yiming Du, Rui Ning, Shuai Hao, Daniel Takabi, and Lusi Li*. "Not All, but the Right Ones: Energy-Guided Representation Learning for Incomplete Multi-View Clustering." IEEE Transactions on Neural Networks and Learning Systems (2026).
(IEEE INFOCOM) Miao Lin, Junrui Zhang, Jian Li, Feng Yu, Lusi Li, Chunsheng Xin, Hongyi Wu, and Rui Ning. “ACS-Boot: Efficient Randomized Smoothing for Robustness Certification on Resource-Constrained Edge Devices.” IEEE INFOCOM 2026. (in press)
(IEEE INFOCOM) Zemin Chen, Jian Li, Miao Lin, Austin Mao, Lusi Li, Rui Ning, ChunSheng Xin, and Hongyi Wu. “TrojanEdge: Mutual Information-Enhanced Robust and Persistent Backdoor Attacks for Edge and On-Device Deployments.” IEEE INFOCOM 2026. (in press)
(IEEE INFOCOM) Yiming Du, Ziyu Wang, Jian Li, Rui Ning, and Lusi Li*. " Deep Incomplete Multi-View Clustering via Hierarchical Imputation and Alignment." AAAI 2026 (in press). [paper][source code]
(IEEE MILCOM) Ziyu Wang, Yiming Du, Rui Ning, Daniel Takabi, and Lusi Li*. “ECSC: Energy-Calibrated Semantic Communication.” IEEE MILCOM 2025. [paper][source code]
(IEEE MILCOM) Yao Wang, Chunyu Hu, Jian Li, Rui Ning, Lusi Li*, and Daniel Takabi. “Contrastive Multi-Hop Semantic Communication.” IEEE MILCOM 2025. [paper]
(IEEE TNNLS) Yiming Du, Yao Wang, Ziyu Wang, Rui Ning, Lusi Li*. “PGFormer: A Prototype-Graph Transformer for Incomplete Multiview Clustering.” IEEE Transactions on Neural Networks and Learning Systems (2025). [paper]
(ACM MM) Ziyu Wang, Yiming Du, Rui Ning, and Lusi Li*. "Energy-based Deep Incomplete Multi-View Clustering." Proceedings of the 33rd ACM International Conference on Multimedia 2025 (in press). [source code]
(NEUNET) Ziyu Wang, Yiming Du, Yao Wang, Rui Ning, and Lusi Li*. "Deep Incomplete Multi-view Clustering via Multi-level Imputation and Contrastive Alignment." Neural Networks 181 (2025): 106851. [paper] [source code]
(NEUNET) Qian Tao, Chenghao Liu, Yuhan Xia, Yong Xu, and Lusi Li*. "Adaptive multi-graph contrastive learning for bundle recommendation." Neural Networks 181 (2025): 106832. [paper]
(IEEE SMC) Leizhen Zhang, Lusi Li, Di Wu, Sheng Chen, and Yi He. “Fairness-Aware Streaming Feature Selection with Causal Graphs.” In 2024 IEEE International Conference on Systems, Man, and Cybernetics (SMC). [paper] [source code]
(ICNC) Lusi Li*, Rui Ning, Ziyu Wang, and Shuai Hao. "A Dual Transfer Framework for Cooperative Spectrum Sensing in Cognitive Radio." In 2024 International Conference on Computing, Networking and Communications (ICNC), pp. 1065-1070. IEEE, 2024. [paper]
(NEUNET) Qian Tao, Jianpeng Liao, Enze Zhang, Lusi Li*. "A Dual Robust Graph Neural Network Against Graph Adversarial Attacks." Neural Networks, 175 (2024): 106276. [paper]
(INFFUS) Ziyu Wang, Lusi Li*, Xin Ning, Wenkai Tan, Yongxin Liu, and Houbing Song. "Incomplete Multi-view Clustering via Structure Exploration and Missing-view Inference." Information Fusion (2024): 102123. [paper] [source code]
Principal Investigator: Dr. Lusi Li (CS at ODU)
Student Investigator: Ziyu Wang (CS at ODU)
Student Investigator: Yiming Du (CS at ODU)
Student Investigator: Yao Wang (CS at ODU)
Student Investigator: Jian Li (CS at ODU)
Academic Collaborator: Dr. Rui Ning (CS at ODU)
Academic Collaborator: Dr. Shuai Hao (CS at ODU)
Academic Collaborator: Dr. Jiang Li (ECE at ODU)
Academic Collaborator: Dr. Daniel Takabi (Cybersecurity at ODU)
Academic Collaborator: Dr. Yi He (DS at W&M)
Academic Collaborator: Dr. Hongyi Wu (The University of Arizona)
Academic Collaborator: Dr. Chunsheng Xin (Iowa State University)
The proliferation of wireless devices and the increasing demand for wireless services, coupled with inefficient spectrum allocation and limited spectrum availability, exacerbate the scarcity of wireless spectrum resources in next-generation (NextG) networks. This project focuses on the development of novel transfer learning (TL) frameworks for dynamic spectrum sharing (DSS) to enable knowledge transfer across users, environments, and wireless systems, offering viable approaches to intelligently utilize underutilized licensed spectrum more effectively. DSS wireless systems exhibit characteristics such as dynamic environments, heterogeneous networks, massive connections, interference, high communication overhead, limited computing, and storage capacity, as well as security and privacy concerns, making it challenging to learn and leverage transferable knowledge. Moreover, achieving the desired performance of knowledge transfer often requires substantial amounts of high-quality training data, while transferring data knowledge may raise security and privacy issues, limiting adaptation and generalization to other tasks. Therefore, this project aims to explore novel TL strategies for learning transferable knowledge and addressing concerns related to robustness, efficiency, security, and privacy in DSS systems.
A key thrust of the project involves a systematic investigation into the characteristics and parameters of target DSS wireless systems, alongside an exploration of the fundamental principles, theories, and unique challenges associated with knowledge transfer. These studies aim to bridge the gap between system characteristics and algorithm development. Based on the findings, the research team will (1) design an ensemble evaluation scheme to assess the robustness, efficiency, security, and privacy of TL-based DSS frameworks, (2) develop efficient TL-based DSS frameworks for adaptive spectrum sensing, selection, access, and handoff, and (3) create robust security and privacy TL strategies for monitoring, detecting, mitigating, and preventing various malicious attacks, while also protecting sensitive data. Concurrently, the research team will develop a Wireless Knowledge Transfer testbed that incorporates transferable knowledge, evaluation schemes, pre-trained TL models, attack knowledge databases, and security and privacy strategies. This testbed is to facilitate and standardize research on knowledge reuse in wireless communication systems. The integration of research and education plans will equip the NextG workforce in the fields of DSS, artificial intelligence, transfer learning, and cybersecurity. Outreach activities will establish connections between the DSS research, and minority groups, K-12 students, and college students through various learning approaches.
Conference Leadership: Serving as Technical Program Committee (TPC) Co-Chair for CAPWIC 2027, collaborating with the organizing committee to support the conference. [CAPWIC 2027]
Research Engagement: Participated in ODU Computer Science Ph.D. Research Gatherings (2026).
Graduate Program Leadership: Served as Graduate Program Director for the M.S. program in Computer Science at ODU (December 2025–present).
Undergraduate Research Mentoring: Mentored three undergraduate students through the Summer 2026 NSF Research Experiences for Undergraduates (REU) program, “Graph Learning and Network Analysis: From Foundations to Applications,” and one additional undergraduate student through the NSF CISE REU supplement.
Research Outreach: Served as faculty lead for CS Trick or Research, helping undergraduate and graduate students explore research topics and opportunities (2025).
K–12 STEM Engagement: Served as a judge for the AI track of the Great Computer Challenge (GCC) for K–12 students (2024, 2025, 2026).
Student Innovation: Served as a judge for the Computer Science Graduate Society (CSGS) Hackathon 2024-2026.
AI Outreach: Served as a panelist for an AI Q&A panel hosted by the ACM student organization.
Student Organization Advising: Served as faculty advisor for ODU’s Association for Computing Machinery Women (ACM-W) chapter (2022–present).