Machine Learning, Deep Learning and Computational intelligence for CSI Compression and Semantic Communication
The explosive growth of data and demand for ultra-reliable, low-latency communication necessitates a paradigm shift from traditional bit-level transmission to Semantic Communication. This workshop aims to provide a focused platform for researchers from the Machine Learning (ML) / Deep Learning (DL) / Computational Intelligence (CI) and Wireless Communication communities to collaborate.
The primary objective is to explore and leverage existing, well-established ML/DL/CI algorithms to address crucial challenges in Channel State Information (CSI) Compression and the implementation of robust and efficient Semantic Communication systems. By bridging these two critical fields, the workshop seeks to foster novel, promising solutions that optimize resource utilization and elevate the intelligence of future wireless networks.
We invite submissions detailing original, unpublished research that utilizes advanced computational intelligence techniques to enhance the semantic and efficiency aspects of wireless communication.
We welcome theoretical and experimental contributions across various facets of this emerging field. Specific areas of interest include, but are not limited to, the application of Dimensionality Reduction techniques for Semantic Communication classification tasks, such as PCA, LDA, KLDA, and ICA. We are keenly interested in exploring Semantic Communication systems utilizing Large Language Models (LLMs), as well as solutions for Semantic Communication for regression tasks. Further critical areas involve the design of Robust precoders for Semantic Communication under various channel models, and the investigation of complex communication scenarios like Multi-source Semantic Communication and Multi- task Semantic communication. The workshop also welcomes papers focusing on the integration of semantic principles with advanced physical layer technologies, specifically MIMO-OFDM based semantic communication, OTFS based semantic communication, NOMA based semantic communication, and IRS based semantic communication. Finally, a core focus remains on foundational efficiency problems, specifically CSI Compression and Estimation techniques.
Topics:
- Dimensionality reduction techniques for Semantic Communication for classification that include PCA, LDA, KLDA, ICA etc.
- Semantic communication using Large Language Model
- Robust precoder design for Semantic Communication with various channel model
- Semantic communication for regression
- Multi-source Semantic Communication
- Multi-task Semantic communication
- MIMO-OFDM based semantic communication
- OTFS based semantic communication
- NOMA based semantic communication
- IRS based semantic communication
- CSI Compression and Estimation
Paper Submission Guidelines
Papers submitted to ICIN 2026 Workshops will be assessed based on originality, technical soundness, clarity and interest to a wide audience. All submissions must be written in English and must use standard IEEE two-column conference template, available for download from the IEEE website: https://www.ieee.org/conferences/publishing/templates.html
Workshop papers can be up to 8 pages (Full papers) and 5 pages (Short papers), including tables, figures and references.
Only PDF files will be accepted for the review process and all submissions must be done electronically through EDAS at: https://edas.info/N34512
Important Deadlines:
- Workshop paper submission deadline: January 4, 2026
- Acceptance notification: January 18, 2026
- Camera-ready paper submission: January 25, 2026
Contact:
Dr. E.S. Gopi,
E-Mail: esgopi@nitt.edu
IEEE senior member of the Computational Intelligence Society and the Signal Processing Society.
Co-ordinator and Head, Pattern recognition and Computational intelligence Group,