Federated Unfolding Learning for CSI Feedback in Distributed Edge Networks
Bibliographic record
Abstract
In distributed edge networks employing frequency division duplex, the feedback of channel state information (CSI) from the edge devices to the edge server always consumes a lot of spectrum resources, resulting in a serious communication burden. In this paper, we first propose an end-to-end unfolding neural network framework inspired by the soft threshold iterative algorithm (U-ISTANet). The proposed U-ISTANet integrates the advantages of compression awareness and neural networks. Especially, the compression matrix and sparse transformation of channel matrix can be learned for accurate CSI compression and recovery. And a lightweight version of U-ISTANet, called U-ISTANet-L, is proposed to reduce the training parameters. To reduce the data transmission overhead in the centralized learning framework, we extend the proposed U-ISTANet-L to a federated U-ISTANet-L (FU-ISTANet-L), which can train a more generalizable model by increasing the number of edge devices to enlarge the data set in a distributed learning manner. The proposed FU-ISTANet-L reduces the transmission overhead and increases the training speed while achieving a performance close to that of centralized learning. Furthermore, we propose a personalized FU-ISTANet-L (P-FU-ISTANet-L) to solve the heterogeneous data training problem in different communication environments. Specifically, we first obtain a pre-trained model by federation unfolding learning, and then each edge device fine-tunes the model using only a small amount of train data to obtain a personalized model for local channel environment. Extensive experimental results are provided to show that the proposed networks achieve a significant performance over the benchmarking schemes in terms of the normalized mean square error.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".