光通信技术2026，Vol.50Issue(3)：29-35,7.DOI:10.13921/j.cnki.issn1002-5561.2026.03.005
基于Q学习超启发算法的FBG混叠光谱解调方法
FBG aliasing spectrum demodulation method based on Q-learning hyper-heuristic algorithm
摘要
Abstract
To address the reflection spectrum aliasing problem in fiber Bragg grating(FBG)multiplexing systems,a Q-learning hyper-heuristic differential evolution(QHDE)algorithm is proposed in this paper.Based on the differential evolution(DE)framework,the algorithm constructs an operator library containing four heterogeneous mutation strategies,and introduces the Q-learning mechanism to achieve the adaptive selection of mutation strategies.By evaluating the performance of the strategy in real time,computational resource scheduling is dynamically optimized.The demodulation results show that the QHDE algo-rithm achieves picometer-level accuracy for wavelength demodulation of FBG aliasing spectrum in the test range,and its de-modulation performance is significantly better than the DE algorithm with single mutation strategy,exhibiting excellent robust-ness and demodulation capability.关键词
光纤布喇格光栅/波长解调/超启发算法/Q学习/光谱混叠/差分进化Key words
fiber Bragg grating/wavelength demodulation/hyper-heuristic algorithm/Q-learning/spectral aliasing/differential evolution分类
信息技术与安全科学引用本文复制引用
夏雨,刘亚文,孙宾..基于Q学习超启发算法的FBG混叠光谱解调方法[J].光通信技术,2026,50(3):29-35,7.基金项目
江苏省特种设备安全监督检验研究院直属分院科技计划项目(2024JC01-KT-01)资助. （2024JC01-KT-01）
