Special Session 8  会议特别专题 8

AI-aided Signal Processing

Description: The field of Signal Processing (SP), traditionally anchored in mathematical models and statistical methods like the Fourier Transform and Kalman filtering, is experiencing a fundamental transformation fueled by Artificial Intelligence (AI), especially deep learning. While classical SP provides interpretability and optimality under defined conditions, its effectiveness diminishes when confronted with the high dimensionality, non-linearity, and complexity characteristic of many modern, real-world signals and systems where accurate underlying models are elusive.
This paradigm shift necessitates innovative approaches that transcend the limitations of purely model-based or purely data-driven methods. Below, we highlight the critical motivations driving the integration of AI into signal processing:

1. Overcoming Classical Limitations: Traditional SP techniques often struggle with complex, non-stationary signals or when system dynamics are poorly understood. AI, particularly deep learning, excels at learning intricate patterns directly from data, offering performance breakthroughs in scenarios challenging for classical methods.
2. The Rise of Hybrid Intelligence: The most impactful advancements stem not from replacing SP blocks with AI, but from synergistic hybridization. Combining the principled structure and domain knowledge embedded in classical algorithms with the adaptive learning and complex pattern recognition capabilities of AI yields systems more powerful than either approach in isolation. This fusion is key to tackling problems with partially known dynamics or intricate data structures.
3. Beyond Block Replacement: The field has matured beyond viewing AI as a simple substitute. There's a growing recognition of the enduring value of SP principles as structural priors or complementary components within AI-enhanced systems, leading to a pragmatic synthesis of analytical rigor and empirical learning.
4. Automated Feature Discovery: Deep learning offers the ability to automatically learn hierarchical features from raw signals, reducing the need for manual, expert-driven feature engineering and potentially uncovering non-intuitive signal characteristics.
5. Transforming SP Research & Education: AI's influence extends to the methodology of the field itself. AI tools are now employed not just as subjects of study but as instruments to analyze pedagogical effectiveness and accelerate knowledge discovery within the SP community.

This special session provides a vital forum to explore the dynamic intersection of AI and Signal Processing. By uniting researchers and practitioners, we aim to dissect the challenges and opportunities presented by this convergence, examining novel hybrid architectures, integration strategies, and the impact across diverse applications. Contributions will shape the understanding of how AI is fundamentally redefining signal analysis, interpretation, and manipulation for future intelligent systems. We invite contributions on theoretical foundations, algorithmic innovations, and practical implementations that showcase the power and potential of AI-aided Signal Processing.

Session organizers
Assoc. Prof. Yimao Sun, Sichuan University, China
Prof. Yue Ivan Wu, Sichuan University, China
Assoc. Prof. Weiliang Zuo, Xi’an Jiaotong University, China

The topics of interest include, but are not limited to:
▪ AI-augmented Classical Signal Processing
▪ AI-aided Kalman Filtering and Adaptive Filtering
▪ Compressive Sensing and Sparse Signal Processing with AI
▪ AI-augmented Optimization and Inverse Problems
▪ AI-aided Signal Enhancement and Denoising
▪ AI-driven Time-Frequency Analysis
▪ Intelligent Beamforming and Precoding in MIMO System
▪ AI in Near-Field Signal Processing
▪ AI-driven Channel Estimation, Prediction, and Modeling in Wireless communications
▪ AI for Semantic Communication
▪ AI-based Audio and Speech Processing
▪ Generative AI for Signal Synthesis, Augmentation, and Analysis
▪ Biomedical Signal Processing and Medical Image Analysis With AI
▪ AI for Sensor Data Processing and Sensor Fusion
▪ Explainable AI for interpreting AI-aided signal processing systems

Submission method
Submit your Full Paper (no less than 4 pages with two colums) or your paper abstract-without publication (200-400 words) via Online Submission System, then choose Special Session 8 (AI-aided Signal Processing)
Template Download

Introduction of Session organizers

Assoc. Prof. Yimao Sun, Sichuan University, China

Y. Sun (Member, IEEE) received the B.Eng. degree in electronic information engineering in 2013, and the M.Sc. and Ph.D. degrees in information and communication engineering in 2015 and 2019, respectively, all from the University of Electronic Science and Technology of China (UESTC), Chengdu, China. From September 2017 to September 2018, he was a Visiting Ph.D. Student sponsored by the China Scholarship Council (CSC) at the University of Missouri, Columbia, MO, USA, where he collaborated with Prof. K. C. Ho (Fellow, IEEE). Since March 2021, he has been an Associate Research Fellow with the College of Computer Science, Sichuan University, Chengdu, China. His research interests include statistical signal processing, parameter estimation, passive source localization and tracking, intelligent signal processing, array signal processing, and direct position determination. He was recognized as an Outstanding IoT Teacher of Sichuan Province in 2023. He was the Principal Investigator of projects funded by the National Natural Science Foundation of China (NSFC) Fund. He has also served as a core member in several NSFC General Programs and a sub-project of the National Key R&D Program related to 5G-based indoor localization and visible light communication.
Dr. Sun serves as a Guest Editor for Frontiers in Signal Processing and Applied Sciences. He has served as an Award Chair for ICSPS 2025 and a Session Chair for multiple international conferences, and has been an invited speaker at events such as ICSPS 2024, the Academic Event Series on "Target Detection and Localization in Complex Environments" (2023), and the Frontier Technology Forum on "Intelligent Spectrum Management and Utilization" (2022). He has also served on the committees of multiple international conferences and is an active reviewer for numerous journals, including the IEEE Transactions on Signal Processing, IEEE Transactions on Wireless Communications, IEEE Journal of Selected Topics in Signal Processing, and IEEE Internet of Things Journal.



Prof. Yue Ivan Wu, Sichuan University, China

Yue Ivan WU (Senior Member, IEEE) obtained a Ph.D. in electronic & information engineering in 2010 from the Hong Kong Polytechnic University. He joined Sichuan University (Chengdu, Sichuan, China) in July 2013, and is currently a Professor in the College of Electronics and Information Engineering, Sichuan University (Chengdu, Sichuan, China).
Y. I. Wu is/was on the editorial boards of The Journal of the Acoustical Society of America, IEEE Transactions on Intelligent Transportation Systems, IET Signal Processing, Electronics Letters, IEEE Access, and Telecommunication Systems.
He has some experience on the research areas of space-time signal processing and acoustic vector sensors.



Assoc. Prof. Weiliang Zuo, Xi’an Jiaotong University, China

Weiliang Zuo received the B.E. degree in electrical engineering and the Ph.D. degree in control science and engineering from Xi'an Jiaotong University, Xi'an, China, in 2010 and 2018, respectively. During 2016 to 2017, he was a visiting Ph.D student in the School of Electrical and Computer Engineering at Georgia Institute of Technology (Georgia Tech). He is currently an associate professor at the College of Artificial Intelligence, Xi'an Jiaotong University. His current research interests include array and statistical signal processing, pattern recognition, and medical imaging processing. He has published scientific articles in refereed journals, such as IEEE Journal of Selected Topics on Signal Processing, IEEE Transactions on Signal Processing, Pattern Recognition, etc.