arXiv：cs.LG（机器学习，全量分类）· Youngsun Kong, Ki H. Chon·· 14 小时前AI 评分31
PI-AMFM：面向生物医学信号变基数 AM-FM 模态分解的置换不变学习框架
PI-AMFM: Permutation-Invariant Learning for Variable-Cardinality AM-FM Mode Decomposition in Biomedical Signal Analysis
AI 导读
研究者提出置换不变神经网络框架 PI-AMFM，用于变基数 AM-FM 模态分解，结合多尺度时序编码器、Mamba 主干与分量存在性估计，训练时采用置换不变的匈牙利匹配。在合成 AM-FM 信号上，其分解误差、瞬时频率误差、重建误差和模态数误差均低于对比方法，并在交叉 chirp 示例中保持整体轨迹模式。该模型仅在合成信号上训练，却在光电容积脉搏波记录中恢复了心脏与呼吸动态。
正文
Abstract:Physiological recordings often contain nonstationary oscillatory components whose number and dynamics vary across signals. Amplitude- and frequency-modulated (AM-FM) representations are well suited to characterizing such dynamics and have shown broad utility in biomedical signal analysis. Recent approaches have incorporated neural networks to learn mode decomposition patterns from data, but component cardinality is often predefined or determined through separate stopping or selection mechanisms. We propose a permutation-invariant neural framework for variable-cardinality AM-FM mode decomposition (PI-AMFM). PI-AMFM combines a multiscale temporal encoder, Mamba backbone, and component-presence estimation, with permutation-invariant Hungarian matching during training. On synthetic AM-FM signals, PI-AMFM achieved lower decomposition, instantaneous-frequency, reconstruction, and mode-count errors than the compared methods while preserving the overall trajectory pattern in a crossing-chirp example. On photoplethysmographic recordings, recovered modes captured cardiac and respiratory dynamics despite training only on synthetic signals. These results support the feasibility of PI-AMFM for variable-cardinality decomposition of nonstationary biomedical signals.
| Comments: | 5 pages, 3 figures |
| Subjects: | Signal Processing (eess.SP); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.00819 [eess.SP] |
| (or arXiv:2610.00819v1 [eess.SP] for this version) | |
| https://doi.org/10.48550/arXiv.2610.00819 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Youngsun Kong [view email]
[v1]
Wed, 30 Sep 2026 23:24:38 UTC (2,734 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org