Uplink Assisted Joint Channel Estimation and CSI Feedback: An Approach Based on Deep Joint Source-Channel Coding

2025-04-15

AI Summary

In FDD MIMO systems, conventional modular AI-based CSI feedback—where channel estimation and quantization are decoupled—suffers from suboptimal performance, reconstruction cliffs, and distribution shift. To address these issues, this paper proposes an uplink-assisted end-to-end framework jointly optimizing channel estimation and CSI feedback. Leveraging the partial reciprocity between uplink and downlink channels, we integrate overhead-free uplink CSI into a deep joint source-channel coding (DJSCC) architecture, enabling joint optimization and end-to-end co-training of all modules. Experimental results demonstrate substantial improvements in CSI reconstruction accuracy and robustness, completely eliminating the “performance cliff.” Ablation studies confirm the efficacy of each design component, while scalability and generalization analyses validate strong cross-scenario adaptability. The proposed framework achieves state-of-the-art performance under practical constraints, offering a principled solution to CSI feedback in FDD massive MIMO.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Multimodal Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In frequency division duplex (FDD) multiple-input multiple-output (MIMO) wireless communication systems, the acquisition of downlink channel state information (CSI) is essential for maximizing spatial resource utilization and improving system spectral efficiency. The separate design of modules in AI-based CSI feedback architectures under traditional modular communication frameworks, including channel estimation (CE), CSI compression and feedback, leads to sub-optimal performance. In this paper, we propose an uplink assisted joint CE and and CSI feedback approach via deep learning for downlink CSI acquisition, which mitigates performance degradation caused by distribution bias across separately trained modules in traditional modular communication frameworks. The proposed network adopts a deep joint source-channel coding (DJSCC) architecture to mitigate the cliff effect encountered in the conventional separate source-channel coding. Furthermore, we exploit the uplink CSI as auxiliary information to enhance CSI reconstruction accuracy by leveraging the partial reciprocity between the uplink and downlink channels in FDD systems, without introducing additional overhead. The effectiveness of uplink CSI as assisted information and the necessity of an end-toend multi-module joint training architecture is validated through comprehensive ablation and scalability experiments.

Problem

• Improves downlink CSI acquisition in FDD MIMO systems
• Addresses sub-optimal performance in modular CSI feedback designs
• Enhances CSI reconstruction using uplink-downlink channel reciprocity

Innovation

• Deep joint source-channel coding architecture
• Uplink CSI as auxiliary information
• End-to-end multi-module joint training