针对不同长度信号的调制识别难题，提出统一骨干+专家接口框架。
A Unified Backbone--Expert Framework with Relation-Token and Residual--Classifier Interfaces for Automatic Modulation Recognition
- 共享卷积状态空间骨干，搭配关系令牌与残差分类器双接口
- 短序列用复平面描述符补信息，长序列用多尺度残差修正特征
- 在RML2016.10b和HisarMod2019上分别达67.28%和87.19%准确率
自动调制识别（AMR）在不同观测长度下面临表征瓶颈，单一模型架构难以兼顾。为此，我们提出统一骨干-专家框架，采用通用卷积状态空间骨干和两个专用接口。对于短序列，在编码前注入显式的滞后感知复平面描述符作为关系令牌，以补偿信息损失；对于长序列，设计门控多尺度残差精修模块以修正特征图，并结合固定平均分类器协作，挖掘互补证据。该框架在RML2016.10b上实现67.28 ± 0.14%的平均准确率，在HisarMod2019上达87.19 ± 0.77%（三次运行均值±样本标准差）。通过三种子实验、原长跨配置测试及受控窗口研究验证，确认专家接口解耦优于通用架构。
原文摘要 · Abstract (English)
Automatic modulation recognition (AMR) faces distinct representation bottlenecks under varying observation lengths, where a single model architecture often fails to excel. To address this, we propose a unified backbone-expert framework with a common convolutional state-space backbone and two specialized interfaces. For short sequences, we inject explicit lag-aware complex-plane descriptors as relation tokens before encoding to compensate for information loss. For long sequences, we design a gated multi-scale residual refinement module to correct the feature map, combined with a fixed-averaging classifier collaboration to harness complementary evidence. Our framework achieves overall average accuracies of 67.28 \pm 0.14% on RML2016.10b and 87.19 \pm 0.77% on HisarMod2019 (mean \pm sample standard deviation over three runs), respectively. The framework's efficacy is further validated through three-seed ablations, native-length cross-configuration tests, and controlled window studies, confirming the benefit of expert-interface decoupling over one-size-fits-all architectures.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。