MambaJSCC: Adaptive Deep Joint Source–Channel Coding With Generalized State Space Model
- 22 December 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25 (15361276) , 9264-9279
Abstract
Lightweight and efficient neural network models for deep joint source-channel coding (JSCC) are crucial for semantic communications. In this paper, we propose a novel JSCC architecture, named MambaJSCC, that achieves great performance with low computational and parameter overhead. MambaJSCC utilizes the visual state space model with channel adaptation (VSSM-CA) blocks as its backbone for transmitting images over wireless channels, where the VSSM-CA primarily consists of the generalized state space models (GSSM) and the zero-parameter, zero-computational channel adaptation method (CSI-ReST). We design the GSSM module, leveraging reversible matrix transformations to express generalized scan expanding operations, and theoretically prove that two GSSM modules can effectively capture global information. We discover that GSSM inherently possesses the ability to adapt to channels, a form of endogenous intelligence. Based on this, we design the CSI-ReST method, which injects channel state information (CSI) into the initial state of GSSM to utilize its native response, and into the residual state to mitigate CSI forgetting, enabling effective channel adaptation without introducing additional computational and parameter overhead. Experimental results on different devices, including IoT device JETSON AGX ORIN, show that MambaJSCC not only outperforms existing JSCC methods (e.g., SwinJSCC) across various scenarios but also significantly reduces parameter size, computational overhead, and inference delay. We have released our code and pre-trained models at https://github.com/Wireless3C-SJTU/MambaJSCC, allowing full reproduction of our results.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (62222111)
- National Natural Science Foundation of China (62431015)
- National Natural Science Foundation of China (62125108)
- National Natural Science Foundation of China (625B2112)
- National Natural Science Foundation of China (U24B20131)
- National Natural Science Foundation of China (62301471)
- National Key Laboratory for Positioning, Navigation and Timing Technology; and in part by the National Science and Technology Major Project—Mobile Information Networks (2024ZD1300700)
This publication has 29 references indexed in Scilit:
- MambaJSCC: Deep Joint Source-Channel Coding with Visual State Space Model2024
- ZigMa: A DiT-style Zigzag Mamba Diffusion Model2024
- Deep Joint Source-Channel Coding for CSI Feedback: An End-to-End ApproachIEEE Journal on Selected Areas in Communications, 2022
- Nonlinear Transform Source-Channel Coding for Semantic CommunicationsIEEE Journal on Selected Areas in Communications, 2022
- Swin Transformer: Hierarchical Vision Transformer using Shifted Windows2021
- The Open Images Dataset V4International Journal of Computer Vision, 2020
- Sigmoid-weighted linear units for neural network function approximation in reinforcement learningNeural Networks, 2018
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric2018
- Deep Residual Learning for Image Recognition2016
- Multiscale structural similarity for image quality assessment2004