Efficient Residual Shrinkage CNN Denoiser Design for Intelligent Signal Processing: Modulation Recognition, Detection, and Decoding
Bibliographic record
Abstract
The noises embedded in signals will degrade the signal processing quality. Traditional denoising algorithms might not work in practical systems since the statistical characteristics of noises might not be learned. To address this issue, we propose an efficient residual shrinkage convolutional neural network (RSCNN) aided denoiser based on the principle of the domain transformation, shrinking and inverse transforming operations conducted by the traditional denoiser. The proposed RSCNN is composed by the batch normalization layer, domain transformation layers, the shrinkage module and inverse transformation layers, wherein transformation layers consist of convolutional layers and the nonlinear activation function. Moreover, we propose a thresholds learning subnetwork to automatically determine the thresholds, so as to enhance noise suppressing performances. Furthermore, we compose the data set by preprocessing the received signals, and design the loss function according to different denoising requirements. To validate the efficiency and universality of the RSCNN aided denoiser, we apply the proposed RSCNN denoiser to three different application scenarios, including the modulation recognition, detection and decoding. After the offline training, at the online deployment stage, we utilize the RSCNN denoiser to reduce the noise power and improve the signal to noise ratios. Simulation results demonstrate that the proposed intelligent denoiser can efficiently improve the signal processing capabilities to achieve higher modulation recognition accuracy, better detection and decoding performances with lower complexity than benchmark schemes.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".