arXiv cs.AI· Zhiyi Yin, Zhizhao Wen, Ruoxin Zhang, Chao Wang, Puyang Xu, Yifan Jiang, Yusi Gong·· 12 小时前AI 评分18
FuseAMC：融合CNN与累积量分类器的CPU高效调制识别决策融合模型
Deep Meets Classical: A CPU-Efficient Decision-Fusion Model for Automatic Modulation Classification
AI 导读
FuseAMC 将约 198k 参数的残差 CNN 与梯度提升树累积量分类器的类别后验拼接，用逻辑回归元组合器做决策融合，在 RadioML 2016.10a 的 44k 帧留出测试集上达到 55.8% 总体准确率，比最佳累积量特征基线高 5.1 个百分点、比单独 CNN 高 3.2 个百分点。
正文 · 原文
Abstract:Automatic modulation classification (AMC) underpins spectrum sensing and non-cooperative signal identification in cognitive radio, yet the two standard CPU-feasible approaches to it, a compact convolutional neural network (CNN) on raw in-phase and quadrature (I/Q) samples and a classical higher-order-cumulant feature classifier, are rarely compared against each other conditioned on signal-to-noise ratio (SNR). We show that they are, in fact, complementary: the CNN is markedly more robust at low SNR, where hand-crafted statistics collapse toward chance, while the cumulant classifier is more accurate at high SNR, where higher-order statistics cleanly separate dense constellations. We exploit this complementarity with FuseAMC, a lightweight decision-fusion model that trains a logistic-regression meta-combiner over the concatenated class-posteriors of an approximately 198k-parameter residual CNN and a gradient-boosted-tree cumulant classifier, on a validation split disjoint from test. On RadioML 2016.10a (5 seeds, 44k-frame held-out test set), FuseAMC reaches 55.8% overall accuracy, improving on the best cumulant-feature baseline by 5.1 points and on the CNN alone by 3.2 points (paired-bootstrap 95% confidence-interval lower bounds of 5.0 and 3.1 points, both positive), while matching the best classical baseline in its own strong high-SNR regime and the CNN in its strong low-SNR regime. With about 0.2 million parameters, a model footprint of a few megabytes, and roughly 3 MFLOPs per frame, the whole pipeline trains and runs on one CPU core, targeting general-purpose and embedded processors without hardware acceleration.
| Comments: | 5 pages, 2 figures, 5 tables. Accepted at the 2026 6th International Conference on Artificial Intelligence, Automation and High Performance Computing (AIAHPC) |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.05937 [cs.LG] |
| (or arXiv:2611.05937v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.05937 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhiyi Yin [view email]
[v1]
Wed, 4 Nov 2026 04:25:29 UTC (107 KB)
来源：arXiv cs.AI · arxiv.org