A Multiparameter Spoofing Detection Method Based on Parallel CNN-Transformer Neural Network with Gating Mechanism
-
摘要: 在日益复杂的电磁环境中,卫星导航信号极易受到欺骗干扰,因此有必要对欺骗干扰进行检测。针对多种欺骗干扰场景下,单一参数欺骗检测方法不通用、不够鲁棒的问题,该文提出了一种基于门控机制的并行CNN-Transformer神经网络(PCTN)的多参数欺骗干扰信号检测算法,用于检测接收机是否受到欺骗信号的干扰。该算法考虑了欺骗干扰在跟踪阶段对接收机各参数的影响,并从中提取出多个特征信息组成多维时间序列。对序列进行预处理后,利用PCTN网络提取序列的时间前后联系和多个维度上的特征信息。实验结果表明,所提算法在TEXBAT数据集以及实采数据集上分别取得了99%和94.5%以上的总体检测率,能够有效提升多场景下欺骗干扰检测算法的通用性和泛化性。Abstract:
Objective Global Navigation Satellite Systems (GNSS) provide location, velocity, and timing services globally and are widely used. However, their signals are highly susceptible to interference from natural environments or human factors, and existing single-parameter and multi-parameter detection methods have limitations. In an increasingly complex electromagnetic environment, satellite navigation systems face a growing risk of deception and interference. Therefore, it is essential to refine deception interference detection techniques to enhance the generality and adaptability of detection algorithms. This study proposes a multi-parameter deception interference detection algorithm that addresses the limitations of existing methods, ensures the secure and reliable operation of GNSS receivers, and contributes to the safety and stability of satellite navigation systems. Methods Extract key information from the receiver tracking phase. Select five observation metrics: code rate, discriminator result, Doppler shift, carrier-to-noise ratio, and SQM index. Due to the large fluctuations in the original values, apply sliding window processing using Moving Variance (MV) and Moving Mean (MA) to obtain nine feature parameters, forming a multidimensional time series sample. This approach better captures signal feature trends, reduces the effect of data fluctuations, and provides a stable and reliable data foundation for subsequent detection. Construct a Parallel CNN-Transformer Neural network (PCTN) based on a gating mechanism. The network consists of three convolutional neural network modules, eight Transformer encoder modules, and one gating module. The gating mechanism learns the weights of the two branches, fuses their outputs, and detects deception interference signals. Evaluate the model using the TEXBAT dataset and an actual dataset, comparing its performance with five existing algorithms. Results and Discussions The PCTN algorithm performs well on the TEXBAT dataset. As shown in Fig. 6 , its classification accuracy for real signals reaches 99.222%, exceeding that of the five comparison algorithms. The ROC curve (Fig. 8 ) and evaluation metrics (Table 3 ) indicate that the PCTN algorithm achieves the highest AUC value and outperforms others in accuracy, precision, recall, and F1 score, demonstrating stable classification performance across various deception scenarios and effectively distinguishing deception signals from real signals. A deception interference collection platform collects actual data, and after fine-tuning, the model is tested. The PCTN algorithm maintains significant advantages, achieving the highest AUC value in the ROC curve (Fig. 10 ). As shown inTable 4 , its detection accuracy remains above 94.5%, exceeding other algorithms. Compared with its performance on the TEXBAT dataset, the PCTN algorithm exhibits only a 5% decrease on the actual dataset, significantly lower than other algorithms. This demonstrates its robustness, strong generalization capability, and effectiveness in detecting deception interference in new scenarios.Conclusions This study proposes a multi-parameter deception interference detection algorithm based on Deep Learning (DL). The method extracts multiple parameter features from the receiver tracking stage, forms multidimensional time series samples, and employs the PCTN model for detection. Experimental results demonstrate that, compared with five existing algorithms, the proposed method offers significant advantages. On the TEXBAT dataset, it achieves high accuracy across various deception scenarios. On the actual dataset, it exhibits better generalization performance and effectively differentiates deceptive signals from real signals, even with new datasets. Future research can focus on deploying the algorithm on hardware platforms to enable real-time and accurate deception interference detection in practical satellite navigation scenarios. This will further enhance the security of satellite navigation systems and support the reliable application of satellite navigation technology in complex electromagnetic environments. -
表 1 9个特征参数的总结
符号 特征参数 单位(未滑窗处理前) 观测量 f1 码率-MV MHz 码率 f2 载波相位差-MV cycle 鉴别器结果 f3 载波频率差-MV Hz f4 码相位差-MV chip f5 多普勒频移-MV Hz 多普勒频移 f6 载噪比-MA dB 载噪比 f7 载噪比-MV dB f8 Ratio-MA \ SQM指标 f9 Ratio-MV \ 表 2 模型主要参数表
参数名称 参数值 特征向量维度(d_model) 512 前馈网络隐藏层维度(d_hidden) 1 024 编码器层数 8 CNN输入通道数 9 CNN 1维卷积和的大小 4 批量大小(batch_size) 32 表 3 6种算法在不同欺骗场景下的各项评估指标结果(%)
算法 评估指标 欺骗场景 ds2 ds3 ds4 ds8 SVM 精确度 90.343 94.481 94.157 92.343 召回率 90.333 94.111 93.833 92.222 F1得分 90.333 94.099 93.822 92.217 准确率 90.333 94.111 93.833 92.222 Bagging 精确度 89.240 91.835 91.167 90.656 召回率 88.444 90.444 89.944 89.556 F1得分 88.387 90.365 89.870 89.485 准确率 88.444 90.444 89.944 89.556 MLP 精确度 92.890 96.278 95.970 93.559 召回率 92.889 96.111 95.833 93.556 F1得分 92.889 96.108 95.830 93.555 准确率 92.889 96.111 95.833 93.556 Transformer 精确度 98.165 98.283 98.544 95.732 召回率 98.111 98.222 98.5 95.722 F1得分 98.111 98.222 98.5 95.722 准确率 98.111 98.222 98.5 95.722 PCLN 精确度 99.001 98.889 99.223 96.192 召回率 99.000 98.889 99.222 96.056 F1得分 99.000 98.889 99.222 96.053 准确率 99.000 98.889 99.222 96.056 PCTN 精确度 99.779 99.667 99.668 97.214 召回率 99.778 99.667 99.667 97.111 F1得分 99.778 99.667 99.667 97.110 准确率 99.778 99.667 99.667 97.111 表 4 6种算法在实采数据集上的各项评估指标结果(%)
评估指标 SVM Bagging MLP Transformer PCLN PCTN 精确度 82.248 76.249 72.404 80.852 87.66 94.579 召回率 81.439 73.404 71.509 80.649 87.491 94.526 F1得分 81.325 72.676 71.231 80.617 87.477 94.525 准确率 81.439 73.404 71.509 80.649 87.491 94.526 -
[1] HEIN G W. Status, perspectives and trends of satellite navigation[J]. Satellite Navigation, 2020, 1(1): 22. doi: 10.1186/s43020-020-00023-x. [2] 王芝应, 聂俊伟, 李峥嵘, 等. BOC接收机捕获阶段转发式欺骗信号检测算法[J]. 全球定位系统, 2016, 41(5): 13–17. doi: 10.13442/j.gnss.1008-9268.2016.05.003.WANG Zhiying, NIE Junwei, LI Zhengrong, et al. A repeater spoofing signal detection algorithm in GNSS acquisition period of BOC modulation receiver[J]. GNSS World of China, 2016, 41(5): 13–17. doi: 10.13442/j.gnss.1008-9268.2016.05.003. [3] 王文益, 王沛菡. 基于捕获结果的GNSS欺骗式干扰检测[J]. 信号处理, 2021, 37(8): 1460–1469. doi: 10.16798/j.issn.1003-0530.2021.08.013.WANG Wenyi and WANG Peihan. GNSS spoofing interference detection based on acquisition results[J]. Journal of Signal Processing, 2021, 37(8): 1460–1469. doi: 10.16798/j.issn.1003-0530.2021.08.013. [4] SHANG Shunshun, LI Hong, WEI Yimin, et al. GNSS spoofing detection and identification based on clock drift monitoring using only one signal[C]. 2020 International Technical Meeting of the Institute of Navigation, San Diego, USA, 2020: 331–340. doi: 10.33012/2020.17147. [5] YUAN Dingbo, LI Hong, WANG Fei, et al. A GNSS acquisition method with the capability of spoofing detection and mitigation[J]. Chinese Journal of Electronics, 2018, 27(1): 213–222. doi: 10.1049/cje.2017.11.001. [6] FANG Jingxiaotao, YUE Jiang, XU Bing, et al. A post-correlation graphical way for continuous GNSS spoofing detection[J]. Measurement, 2023, 216: 112974. doi: 10.1016/j.measurement.2023.112974. [7] 王璐, 张林杰, 吴仁彪. 功率监测与SQM融合的GNSS欺骗干扰检测[J]. 信号处理, 2023, 39(3): 505–515. doi: 10.16798/j.issn.1003-0530.2023.03.013.WANG Lu, ZHANG Linjie, and WU Renbiao. GNSS spoofing detection based on power monitoring combined with SQM[J]. Journal of Signal Processing, 2023, 39(3): 505–515. doi: 10.16798/j.issn.1003-0530.2023.03.013. [8] 许睿, 岳帅, 唐瑞琪, 等. 欺骗环境下GNSS信号估计与定位修正技术[J]. 航空学报, 2020, 41(10): 323930. doi: 10.7527/S1000-6893.2020.23930.XU Rui, YUE Shuai, TANG Ruiqi, et al. GNSS signal estimation and position correction algorithm under spoofing attacks[J]. Acta Aeronautica et Astronautica Sinica, 2020, 41(10): 323930. doi: 10.7527/S1000-6893.2020.23930. [9] 贾琼琼, 朱传胜. 基于多通道SQM指标联合的矢量接收机多径干扰检测方法[J]. 全球定位系统, 2023, 48(3): 110–119. doi: 10.12265/j.gnss.2023037.JIA Qiongqiong and ZHU Chuansheng. Multipath interference detection method for vector receivers based on joint multi-channel SQM metrics[J]. GNSS World of China, 2023, 48(3): 110–119. doi: 10.12265/j.gnss.2023037. [10] ROTHMAIER F, CHEN Y H, LO S, et al. A framework for GNSS spoofing detection through combinations of metrics[J]. IEEE Transactions on Aerospace and Electronic Systems, 2021, 57(6): 3633–3647. doi: 10.1109/TAES.2021.3082673. [11] LI Jing, CHEN Zhengkun, RAN Zixuan, et al. The GNSS spoofing detection method based on AdaBoost[C]. The 2023 6th International Symposium on Autonomous Systems, Nanjing, China, 2023: 1–6. doi: 10.1109/ISAS59543.2023.10164411. [12] REN Yingying, RESTIVO R D, TAN Wenkai, et al. Knowledge distillation-based GPS spoofing detection for small UAV[J]. Future Internet, 2023, 15(12): 389. doi: 10.3390/fi15120389. [13] LI Lixuan, SUN Chao, ZHAO Hongbo, et al. GNSS spoofing detection using moving variance of signal quality monitoring metrics and signal power[C]. The 14th EAI International Conference on Communications and Networking in China, Shanghai, China, 2020: 537–548. doi: 10.1007/978-3-030-41114-5_40. [14] HU Jie, LI Shen, and SUN Gang. Squeeze-and-excitation networks[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 7132–7141. doi: 10.1109/CVPR.2018.00745. [15] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]. The 31st International Conference on Neural Information Processing Systems, Long Beach, USA, 2017: 6000–6010. [16] LEMMENES A, CORBELL P, and GUNAWARDENA S. Detailed analysis of the TEXBAT datasets using a high fidelity software GPS receiver[C]. The 29th International Technical Meeting of the Satellite Division of the Institute of Navigation, Portland, USA, 2016: 3027–3032. doi: 10.33012/2016.14668. [17] HUMPHREYS T E, BHATTI J A, SHEPARD D, et al. The Texas spoofing test battery: Toward a standard for evaluating GPS signal authentication techniques[R]. Austin: The University of Texas at Austin, 2012. -