PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 29, 2026Chinese Optics0 citationsOpen Access

面向空间激光干涉的相位计自动重锁技术设计与实验验证

View Full Paper
XWXin-yu WANGRYRun YANGHLHe-Shan Liu

Key Points

  • This research aims to develop an automatic re-locking technology for phase meters used in space laser interferometry, addressing current technical limitations.
  • Proposes a loss detection strategy combining instantaneous frequency values and frequency change rates.
  • Utilizes both raw data from loop filters and down-sampled data from CIC filters for loss determination.
  • Implements dual frequency prediction algorithms for periodic and non-periodic signals alongside interpolation techniques.
  • Achieved an average re-lock time of 32 μs and a maximum of 60 μs, significantly faster than the FFT method by two orders of magnitude.
  • Re-lock speed remained effective regardless of loss duration, maintaining microsecond-level speeds even after 10 seconds of lock loss.
  • Frequency estimation error stabilized below 10 Hz across a signal-to-noise ratio range from −10 to 10 dB.

Abstract

面向空间激光干涉的相位计,当锁相环路发生失锁时,现阶段普遍采用FFT(Fast Fourier Transform,快速傅里叶变换)测频法重新完成信号捕获,该方法存在测频精度偏低(100 Hz量级)、重锁耗时较长(约7 ms)等技术问题。本文提出一种与FFT协同部署的自动重锁技术,该技术采用瞬时频率值与频率变化率相结合的失锁检测策略,同时选取环路滤波器原始数据与CIC(Cascaded Integrator-Comb,级联积分器梳状滤波器)降采样数据两类数据源完成失锁判断,失锁发生后通过复位操作清除积分误差,并接收频率预测算法输出的预测值。该频率预测算法针对周期信号采用波形生成算法,针对非周期信号采用二阶多项式预测算法,同时结合插值技术生成对应的频率预测值。该自动重锁技术与FFT采用并行部署的方式且形成明确的功能分工,其中该技术依托信号自身的规律性开展频率预测,负责处理所有规律信号的失锁场景(无论失锁时长)以及短时(<1 s)非规律信号的快速重锁,FFT则负责处理非规律信号以及长时复杂失锁场景下的信号重新捕获,二者形成优势互补的工作模式。实验验证结果表明,在规律信号失锁的场景下,本研究提出的算法平均重锁时间为32 μs,最大重锁时间为60 μs,相较于FFT方法提升两个数量级,且重锁速度与失锁时长无关联,即在失锁时长达到10 s时仍能保持数十微秒量级的重锁速度,同时在−10~10 dB的信噪比范围内,频率估计误差稳定在10 Hz以下,即使信噪比低至−10 dB时仍可实现稳定锁定。这种与FFT协同部署的架构在保留FFT宽频捕获能力的基础上,显著提升了规律信号场景下的快速重锁能力,为空间引力波探测任务提供了高精度、快响应、强稳定性的相位测量技术支撑。

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

WANG et al. (2026) studied this question.

synapsesocial.com/papers/6a192df7fab5b468c4416f50https://doi.org/10.37188/co.2026-0033
Ask AI
Helpful
Bookmark
Share
View Full Paper