信号处理2026，Vol.42Issue(4)：544-556,13.DOI:10.12466/xhcl.2026.04.008
融合归因分析与注意力机制的自动调制分类可信优化方法
A Trustworthy Optimization Method for Automatic Modulation Classification Integrating Attribution Analysis and Attention Mechanism
摘要
Abstract
In complex channel environments,automatic modulation classification in wireless communication places high demand on model performance and reliability.Although deep learning methods have achieved significant progress in this task,their decision-making processes remain"black boxes,"limiting their applicability in critical scenarios.Moreover,existing automatic modulation classification(AMC)model optimization largely relies on human expertise or"black-box"exploration,lacking reliable evidence to guide model decisions and enhancements.To address this,this pa-per proposes a trustworthy optimization method for integrating AMC attribution analysis and an attention mechanism.Specifically,feature attributions were first obtained using integrated gradients and DeepLIFT,quantifying the contribu-tions of input signal amplitude/phase and in-phase/quadrature channel features to the model output.This allows identifi-cation of feature regions that are decisive for classification results.The credibility of these attributions is verified against the physical characteristics of modulated signals.Subsequently,trustworthy attributed features are used as attention weights to optimize the original signals and enhance the model's responsiveness to decisive feature regions,thereby si-multaneously improving classification performance and interpretability.This optimization process relies on trustworthy attributions to ensure consistency between physical signal characteristics and model decisions,thereby enabling credible model optimization.The designed optimization module does not alter the original network structure,ensuring method generality and engineering feasibility.Finally,a signal constellation-based interpretability metric is proposed,to quan-tify the model's interpretability through an alignment analysis between attribution and physical features.The experimen-tal results demonstrate that,without changing the network structure,this method improves the classification accuracy of convolutional neural network and long short-term memory models by approximately 12%and 7%,respectively,while achieving a quantifiable correspondence between key features and model decisions.关键词
自动调制分类/归因分析/注意力机制/可解释人工智能Key words
automatic modulation classification/attribution analysis/attention mechanism/explainable artificial intelligence分类
信息技术与安全科学引用本文复制引用
王硕,张骁议,唐浩,徐博..融合归因分析与注意力机制的自动调制分类可信优化方法[J].信号处理,2026,42(4):544-556,13.基金项目
国家自然科学基金(62563007,62563008)The National Natural Science Foundation of China(62563007,62563008) （62563007,62563008）
