cs.LGApr 11, 2026

Cross-Validated Cross-Channel Self-Attention and Denoising for Automatic Modulation Classification

Authors: Prakash Suman, Yanzhen Qu

Organizations: Computer Science, Engineering, and Technology, Colorado Technical University, 1575 Garden of the Gods Road, Colorado Springs, 80907, CO, USA.

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

This study addresses a key limitation in deep learning Automatic Modulation Classification (AMC) models, which perform well at high signal-to-noise ratios (SNRs) but degrade under noisy conditions due to conventional feature extraction suppressing both discriminative structure and interference. The goal was to develop a feature-preserving denoising method that mitigates the loss of modulation class separation. A deep learning AMC model was proposed, incorporating a cross-channel self-attention block to capture dependencies between in-phase and quadrature components, along with dual-path deep residual shrinkage denoising blocks to suppress noise. Experiments using the RML2018.01a dataset employed stratified sampling across 24 modulation types and 26 SNR levels. Results showed that denoising depth strongly influences robustness at low and moderate SNRs. Compared to benchmark models PET-CGDNN, MCLDNN, and DAE, the proposed model achieved notable accuracy improvements across -8 dB to +2 dB SNR, with increases of 3%, 2.3%, and 14%, respectively. Cross-validation confirmed the model's robustness, yielding a mean accuracy of 62.6%, macro precision of 65.8%, macro-recall of 62.6%, and macro-F1 score of 62.9%. The architecture advances interference-aware AMC by formalizing baseband modeling as orthogonal subproblems and introducing cross-channel attention as a generalized complex interaction operator, with ablations confirming the critical role of feature-preserving denoising for robustness at low-to-medium SNR.

Figures & tables

Figure 1: Proposed model architecture
Figure 2: Cross-channel self-attention block
Figure 3: Denoising block A and B
Figure 4: Garrote thresholding and its derivative
Figure 5: Proposed model architecture with tensor shapes
Figure 6: Normalized modulation class distribution for train, validation, and test samples
Figure 7: Normalized SNR level distribution for train, validation, and test samples
Figure 8: Proposed model training and validation loss
Figure 9: Proposed model training and validation accuracy
Figure 10: Proposed model modulation classification accuracy at each SNR level
FoldOverall Accuracy (%)Macro-Precision (%)Macro-Recall (%)Macro-F1 (%)
063.2167.1363.7363.73
162.6164.3862.6562.65
262.3865.4862.6762.67
362.8866.7063.3263.32
461.8865.1562.2662.26
Mean ±\pm Std62.6±0.562.6\pm 0.565.8±1.265.8\pm 1.262.6±0.562.6\pm 0.562.9±0.562.9\pm 0.5
Table 1: Cross-validation results
Figure 11: Proposed model vs benchmark models modulation classification accuracy at each SNR level
ModelLSTMddHeadsCCSADen.Den.Thres.ParamsFLOPsAvg. Acc.Max Acc.Inference time/
No.Units  BlocksBlock ABlock B  (M)(%)(%)sample (ms)
116322133Garrote165,58023462.7997.880.53
216322233Garrote178,28427063.5198.000.68
316322333Garrote190,98830663.4798.130.79
416322222Garrote85,35423262.7797.640.63
516322211Garrote40,87816157.9692.740.58
Table 2: Ablation study of the proposed model configurations.
Figure 12: Ablation study: Average modulation classification accuracy vs. FLOP; color by inference time/sample (ms)
Figure 13: Ablation study: average modulation classification accuracy per SNR level
AMCAutomatic Modulation Classification
SNRSignal-to-Noise Ratio
I/QIn-phase / Quadrature
LSTMLong Short-Term Memory
DP-DRSNDual-Path Deep Residual Shrinkage Network
CNNConvolutional Neural Network
RNNRecurrent Neural Network
Table 16
Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

AuthorModel NameInput Signal/Pre-processingDen.DL ArchitectureTrainable ParametersDatasetAvg AccuracyMax Accuracy
CNN Models
Shaik and Kirthiga Shaik and Kirthiga (2021)DenseNet––CNN19kC55%–
Shen et al. Shen et al. (2023)MSSAI/Q–CNN36k–218kC55.25%–60.90%–
Li et al. Li et al. (2023)LightMFFSI/Q, A/P–CNN95kA63.44%–
     95kB65.44%–
Shi et al. Shi et al. (2022)–I/Q–CNN, SE block113kC–98.70%
Appendix
Table 4: Summary of existing CNN, RNN, Transformer, SNN, Hybrid deep learning AMC models; Dataset A: RML2016.10a, B: RML2016.10b, C: RML2018.01a, and NS: Non-standard dataset
Figure 14: Proposed model confusion matrices at -4dB, 0dB, +4dB, and +30dB SNR
More in this topic

Automatic Modulation Classification

Automatic modulation classification identifies radio signal modulation schemes from received waveforms, often under noise, fading, and limited labeled data.

See all 17 papers

Explore similar work

CardsList
May 12, 2026

Modulation Consistency-based Contrastive Learning for Self-Supervised Automatic Modulation Classification

Deep learning-based AMC methods have achieved remarkable performance, but their practical deployment remains constrained by the high cost of labeled data. Although self-supervised learning (SSL) reduces the reliance on labels, existing SSL-based AMC methods often rely on task-agnostic pretext objectives misaligned with modulation classification, leading to representations entangled with nuisance factors such as symbol, channel, and noise. In this paper, we identify intra-instance modulation consistency as a task-aware structural prior, whereby different temporal segments of the same signal may differ in waveform while preserving the same modulation type, thus providing a principled cue for task-aligned self-supervision. Based on this prior, we propose Mod-CL, a Modulation consistency-based Contrastive Learning framework that constructs positive pairs from different temporal segments of the same signal instance, to encourage the model to learn shared modulation information while suppressing nuisance variations. We further develop a contrastive objective tailored to Mod-CL, which jointly exploits temporal segmentation and data augmentation to pull together views sharing the same modulation semantics while avoiding supervisory conflicts within each signal instance. Extensive experiments on RadioML datasets show that Mod-CL consistently outperforms strong baselines, especially in low-label regimes, achieving substantial improvements in linear probing accuracy.

Chenxu Wang, Shuang Wang, Lirong Han +4

School of Artificial Intelligence, Xidian University, Xi’an 710071, China · Unmanned System Research Institute, Northwestern Polytechnical University, Xi’an 710072, China · School of Electronics And Information, Northwestern Polytechnical University, Xi’an 710072, China

Contrastive LearningSelf-Supervised LearningAutomatic Modulation Classification
AskPDF
Aug 25, 2026

HFEMCNet: A Compact Hybrid Frequency Enriched Multi Channel Network for Automatic Modulation Classification

Automatic modulation classification (AMC) of received radio signals is prudent for further signal processing tasks such as communication monitoring, cognitive radio operation, and interference mitigation in the electromagnetic spectrum. Traditional methods often rely on handcrafted features and struggle under complex channel conditions, whereas deep learning (DL) architectures can learn discriminative representations directly from raw signals. In this work, we propose HFEMCNet, a novel compact hybrid frequency-enriched multi-channel network that jointly exploits spatiotemporal dependencies and frequency domain information derived via Fast Fourier Transform (FFT). HFEMCNet integrates raw IQ samples with hierarchical spectral features to produce a rich signal representation, which is processed by convolutional layers for spatial feature extraction and Long Short-Term Memory (LSTM) units for temporal modeling. Experimental evaluation on benchmark datasets RML2016.10a, RML2016.10b, and over-the-air RML2018.01a demonstrates that HFEMCNet significantly outperform contemporary state-of-theart DL models in classification accuracy while achieving a reduced parameter count, smaller memory footprint, and lower tail-latency, making it suitable for real-time deployment on resource-constrained platforms. These results highlight the effectiveness of combining hybrid architectures with frequencydomain enrichment for robust and efficient AMC.

Qamar Ijaz, Nayyer Aafaq

Automatic Modulation ClassificationConvolutional RNNsFrequency-Domain Feature LearningEfficient Neural Network Inference
AskPDF
Jul 9, 2026

DKDNet: Dual Knowledge and Data-Driven Network for Cross-Domain Automatic Modulation Classification

The dynamics of communication environments induce significant distribution shifts across domains, challenging the generalization of deep learning-based automatic modulation classification (AMC) models. While existing UDA methods alleviate this problem by aligning source and target features, they give limited consideration to modulation-specific structures that remain informative across domain conditions. In this paper, we consider signal prior knowledge, grounded in communication protocols and physical principles, as a potential way to enhance cross-domain representation learning. Given that different priors may vary in modulation discriminability, domain stability, and complementarity, this paper first analyzes five commonly adopted signal representations that instantiate different signal priors. From them, in-phase/quadrature (IQ), amplitude--phase (AP), and autocorrelation function (ACF) are selected as compact prior-guided inputs. Based on that, a dual knowledge and data-driven network (DKDNet) is proposed for cross-domain AMC. The multi-representation feature encoder (MRFE) and dynamic lightweight fusion unit (DLFU) are designed to achieve unified representation learning and adaptive feature fusion, and the resulting fused features are optimized with modulation classification and adversarial domain alignment objectives. Experiments on both simulated and public datasets validate the rationality of the prior selection and demonstrate the superiority of the proposed method.

Shuang Wang, Chenxu Wang, Hantong Xing +3

School of Artificial Intelligence, Xidian University, Xi’an 710071, China · School of Electronics And Information, Northwestern Polytechnical University, Xi’an 710072, China · Unmanned System Research Institute, Northwestern Polytechnical University, Xi’an 710072, China

Unsupervised Domain AdaptationAutomatic Modulation ClassificationWireless Communications
AskPDF