A Starlink Downlink Signal Detection Algorithm Based on Segmented Cross-Correlation Spectrograms and SD-YOLOv9

Liu Yongjun; ShenLei; HanYu; ShenLin

doi:10.11959/j.issn.1000-0801.DXKX260050

您当前的位置：

首页 >

文章列表页 >

A Starlink Downlink Signal Detection Algorithm Based on Segmented Cross-Correlation Spectrograms and SD-YOLOv9

更新时间：2026-07-23

- A Starlink Downlink Signal Detection Algorithm Based on Segmented Cross-Correlation Spectrograms and SD-YOLOv9
- Telecommunications Science (2026)
- 作者机构：
  
  1.杭州电子科技大学通信工程学院，浙江杭州 310018
  2.中国电子科技集团有限公司第三十六研究所，浙江嘉兴 314000
  3.杭州第一技师学院，浙江杭州 310023
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.DXKX260050
  CLC： TP393
- Received：16 January 2026，
  
  Revised：2026-04-01，
  
  Accepted：14 July 2026，
- 稿件说明：
移动端阅览
Liu Yongjun, ShenLei, HanYu, et al. A Starlink Downlink Signal Detection Algorithm Based on Segmented Cross-Correlation Spectrograms and SD-YOLOv9[J/OL]. Telecommunications Science, 2026.
DOI：

Liu Yongjun, ShenLei, HanYu, et al. A Starlink Downlink Signal Detection Algorithm Based on Segmented Cross-Correlation Spectrograms and SD-YOLOv9[J/OL]. Telecommunications Science, 2026. DOI： 10.11959/j.issn.1000-0801.DXKX260050.

摘要

针对Starlink低轨卫星下行通信中因卫星高速运动引起的显著多普勒频偏以及低信噪比环境，导致传统正交频分复用（orthogonal frequency division multiplexing，OFDM）信号检测方法可靠性显著下降的问题，本文提出一种基于分段互相关谱图和Starlink-Detect YOLOv9（SD-YOLOv9）网络相结合的Starlink下行信号检测新算法。所提算法首先利用Starlink信号主同步序列（Primary Synchronization Signal

PSS）的重复结构，通过分段互相关与非相干累加生成具有显著谱线结构的二维特征图，将传统一维信号峰值检测问题转化为二维图像特征识别任务；此外，设计了一种名为SD-YOLOv9的增强型目标检测网络，在Backbone末尾引入STCSPA (Swin Transformer-Enhanced CSP with Attention)模块，将Swin Transformer的局部窗口注意力机制嵌入Cross Stage Partial（CSP）框架，以增强特征建模与上下文表达能力；同时将Backbone和Neck结构中传统的下采样卷积模块替换为SD-ADown (Adaptive Downsampling) 模块，实现自适应多分支特征融合，减少特征损失并提升弱信号检测性能。实验结果表明，在AWGN信道的-13 dB极低信噪比条件下，本文方法的检测准确率高达100%，相较于基准YOLOv9模型、基于互相关和CapNN算法分别提升了7%和48%，且相较于传统单一的差分互相关、直接互相关及分段互相关算法分别提升了99%、63%和41%。同时，消融实验进一步验证了STCSPA与SD-ADown模块的有效性。本研究在解决高动态无线通信检测难题方面的有效性与优越性，为下一代卫星通信系统提供了可靠的技术思路。

Abstract

To address the problem that significant Doppler frequency offsets caused by high-speed satellite movement and low signal-to-noise ratio (SNR) environments severely degrade the reliability of traditional orthogonal frequency division multiplexing (OFDM) signal detection methods in Starlink low Earth orbit (LEO) satellite downlink communications

this paper proposes a novel Starlink downlink signal detection algorithm integrating segmented cross-correlation spectrograms with the Starlink-Detect YOLOv9 (SD-YOLOv9) network. The proposed algorithm first utilizes the repetitive structure of the Starlink Primary Synchronization Signal (PSS) to generate two-dimensional (2D) feature maps with distinct spectral line structures via segmented cross-correlation and non-coherent accumulation

effectively converting the traditional one-dimensional (1D) signal peak detection problem into a 2D image feature recognition task. Furthermore

an enhanced object detection network named SD-YOLOv9 is designed. Specifically

an STCSPA (Swin Transformer-Enhanced CSP with Attention) module is introduced at the end of the Backbone to embed the local window attention mechanism of the Swin Transformer into the Cross Stage Partial（CSP）framework

thereby enhancing feature modeling and context representation capabilities. Simultaneously

the conventional downsampling convolution modules in the Backbone and Neck structures are replaced with SD-ADown (Adaptive Downsampling) modules to achieve adaptive multi-branch feature fusion

minimizing feature loss and improving detection performance for weak signals. Experimental results demonstrate that in the AWGN channel at an extremely low SNR of -13 dB

the detection accuracy of the proposed method reaches 100%. This represents an improvement of 7% and 48% compared to the baseline YOLOv9 model and the cross-correlation and CapNN-based algorithm

respectively. Additionally

it provides performance gains of 99%

63%

and 41% compared to traditional singular differential cross-correlation

direct cross-correlation

and segmented cross-correlation algorithms

respectively. Ablation studies further validate the effectiveness of the STCSPA and SD-ADown modules. The effectiveness and superiority of this study in resolving detection challenges in high-dynamic wireless communications provide reliable technical insights for next-generation satellite communication systems.

关键词

Keywords

references

汤媛媛 , 王珂 , 亢衡 , 等 . 面向空天地一体化网络的低轨星座跳波束调度技术研究 [J ] . 移动通信 , 2025 , 49 ( 10 ): 62 - 73,84 .

TANG Y Y , WANG K , KANG H , et al . Research on beam-hopping scheduling technology for LEO constellations in space-air-ground integrated networks [J ] . Mobile Communications , 2025 , 49 ( 10 ): 62 - 73,84 .

陈全 , 杨磊 , 郭剑鸣 , 等 . 低轨巨型星座网络：组网技术与研究现状 [J ] . 通信学报 , 2022 , 43 ( 5 ): 177 - 189 .

CHEN Q , YANG L , GUO J M , et al . LEO mega-constellation networks: Networking technologies and research status [J ] . Journal on Communications , 2022 , 43 ( 5 ): 177 - 189 .

DENG R , DI B , ZHANG H , et al . Ultra-dense LEO satellite constellations: How many LEO satellites do we need? [J ] . IEEE Transactions on Wireless Communications , 2021 , 20 ( 8 ): 4843 - 4857 .

NEINAVAIE M , KHALIFE J , KASSAS Z M . Acquisition, Doppler tracking, and positioning with Starlink LEO satellites: First results [J ] . IEEE Transactions on Aerospace and Electronic Systems , 2021 , 58 ( 3 ): 2606 - 2610 .

HUANG M , CHEN J , FENG S . Synchronization for OFDM-based satellite communication system [J ] . IEEE Transactions on Vehicular Technology , 2021 , 70 ( 6 ): 5693 - 5702 .

CHO Y S , KIM J , YANG W Y , et al . MIMO-OFDM wireless communications with MATLAB [M ] . Hoboken : John Wiley & Sons , 2010 .

HUMPHREYS T E , IANNUCCI P A , KOMODROMOS Z M , et al . Signal structure of the Starlink Ku-band downlink [J ] . IEEE Transactions on Aerospace and Electronic Systems , 2023 , 59 ( 5 ): 6016 - 6030 .

刘慧 , 刘祖深 , 许虎 , 等 . 5G NR小区搜索算法研究及FPGA实现 [J ] . 国外电子测量技术 , 2021 , 40 ( 6 ): 152 - 158 .

LIU H , LIU Z S , XU H , et al . Research on 5G NR cell search algorithm and FPGA implementation [J ] . Foreign Electronic Measurement Technology , 2021 , 40 ( 6 ): 152 - 158 .

YANG X , XIONG Y , JIA G , et al . PSS based time synchronization for 3GPP LTE downlink receivers [C ] // Proceedings of 2011 IEEE International Conference on Communication Technology . Piscataway : IEEE Press , 2011 : 930 - 933 .

JOO J S . A study on partial correlation based PSS detection schemes for initial cell search in LTE systems [J ] . Journal of the Institute of Electronics and Information Engineers , 2014 , 51 ( 9 ): 3 - 10 .

张德民 , 李秀 , 陈嘉田 . 5G系统中主同步信号定时同步算法的研究 [J ] . 光通信研究 , 2019 ( 3 ): 59 - 64 .

ZHANG D M , LI X , CHEN J T . Research on timing synchronization algorithm for primary synchronization signal in 5G systems [J ] . Study on Optical Communications , 2019 ( 3 ): 59 - 64 .

孙倩楠 , 张治中 . 5G系统小区搜索PSS定时同步方法 [J ] . 电讯技术 , 2020 , 60 ( 6 ): 678 - 683 .

SUN Q N , ZHANG Z Z . PSS timing synchronization method for cell search in 5G systems [J ] . Telecommunication Engineering , 2020 , 60 ( 6 ): 678 - 683 .

董宝江 , 彭琛 , 卢贺 . 5G NR的小区搜索改进算法 [J ] . 电信科学 , 2021 , 37 ( 2 ): 48 - 54 .

DONG B J , PENG C , LU H . Improved cell search algorithm for 5G NR [J ] . Telecommunications Science , 2021 , 37 ( 2 ): 48 - 54 .

李文星 , 刘祖深 , 铁奎 , 等 . 基于分段互相关的5G新空口PSS同步算法 [J ] . 现代电子技术 , 2024 , 47 ( 21 ): 7 - 12 .

LI W X , LIU Z S , TIE K , et al . Segmented cross-correlation-based PSS synchronization algorithm for 5G new radio [J ] . Modern Electronics Technique , 2024 , 47 ( 21 ): 7 - 12 .

SAARNISAARI H , DE LIMA C M . 5G NR over satellite links: Evaluation of synchronization and random access processes [C ] // Proceedings of 2019 21st International Conference on Transparent Optical Networks (ICTON) . Piscataway : IEEE Press , 2019 : 1 - 4 .

卿朝进 , 杨娜 , 唐书海 , 等 . 基于1-D CNN的二阶段OFDM系统定时同步方法 [J ] . 计算机应用研究 , 2023 , 40 ( 2 ): 565 - 570 .

QING C J , YANG N , TANG S H , et al . Two-stage timing synchronization method for OFDM systems based on 1-D CNN [J ] . Application Research of Computers , 2023 , 40 ( 2 ): 565 - 570 .

KAUR G . Neural network (ConvNN and CapNN) based joint synchronization of timing and frequency for CO-OFDM system [J ] . Optical and Quantum Electronics , 2024 , 56 ( 6 ): 923 .

WANG C Y , YEH I H , LIAO H Y M . YOLOv9: Learning what you want to learn using programmable gradient information [C ] // Proceedings of the European Conference on Computer Vision . Cham : Springer Nature Switzerland , 2024 : 1 - 21 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on human action recognition based on mobile communication network signals

Heterogeneous graph neural network algorithm based on topology information enhancement

Cross-view contrastive learning and multi-channel hypergraph convolution for session-based recommendation

Text classification model based on GNN and attention mechanism

Multiscale residual temporal convolutional networks-based intrusion detection model in Internet of things

Related Author

Qian Neng

Lu Chengling

Wan Yuxuan

Wu Zhefu

YAO Bowu

DENG Kun

WEI Zhenhua

WU Tong

Related Institution

Institute of Cyberspace Security, Zhejiang University of Technology

School of Electronics and Communication Engineering, Zhejiang Post and Telecommunication College

College of Artificial Intelligence, Jiaxing University

College of Information Science and Engineering, Jiaxing University

Provincial Key Laboratory of Multimodal Perceiving and Intelligent Systems, Jiaxing University

⁰