LEO constellation-based electromagnetic monitoring intelligent processing framework and a review of key technologies
-
摘要: 依托低軌星座構建電磁頻譜監測系統成為實現全球電磁頻譜管理的有效途徑與當前的研究熱點。傳統低軌電磁監測系統架構采用“星上采集與處理”的模式，即衛星對信號進行采集并處理后，將處理的結果回傳到地面。這導致系統性能受限于單星載荷。針對此問題提出采集與處理分離的低軌電磁監測系統智能處理框架，衛星作為數據的轉發節點，僅負責采集信號，地面數據中心對數據進行下一步處理。同時，針對傳統技術方法難以高效處理該架構下地面數據中心海量數據的問題，將深度學習與傳統架構下的關鍵技術進行了有機融合，為實現全球時空連續電磁頻譜監測提供了新的選擇。梳理了基于深度學習的頻譜感知、盲源分離和無源定位三大關鍵技術及其近幾年研究進展；重點討論了各關鍵技術向星座系統遷移的適用性問題與技術核心突破問題，給出了低軌電磁監測系統智能處理框架中關鍵技術的下一步研究建議。Abstract: The development of an electromagnetic spectrum monitoring (ESM) system based on a low-earth orbit (LEO) constellation has shown to be an effective method of achieving global ESM and is now a research hotspot in several fields. In the classic LEO-based ESM system, the “on-satellite acquisition and processing” architecture is used in which the satellite gathers and analyzes electromagnetic signal data before transmitting the processed results back to the data center on the ground. Although this framework can reduce the transmission pressure on the satellite-ground link, it yields a limited system performance of the single satellite payload. This paper proposes an intelligent processing framework for the LEO-based ESM system with separate acquisition and processing. In this framework, the satellites serve as forwarding nodes for electromagnetic signal data. The satellites are only responsible for acquiring electromagnetic signal data, which is then processed by a data center on the ground. Unlike the traditional framework, this framework delivers massive amounts of raw electromagnetic data to the ground. To address the problem that the massive data in this framework are difficult to process using traditional technical methods, deep learning is organically integrated with the key technologies of the traditional framework. Deep learning provides a new option for realizing global space–time continuous ESM. The three key technologies involved in the proposed framework are spectrum sensing, blind source separation, and passive positioning based on deep learning, and their research progress in recent years has been sorted out. Compared with ground-based systems, constellation-based systems have the following characteristics: (1) the satellites are far away from the radiation source; (2) the satellites are fast; (3) the satellites show long-distance distribution among them; (4) the topology of the constellation is always in high-speed dynamic change. These characteristics cause a significant divergence between their essential technologies and the research of ground-based systems for these technologies. However, the present efforts relating to essential technologies are based on research conducted on ground-based platforms. There is an issue of applicability to consider when immediately transitioning them to the constellation-based system. Thus, the suitability of each important technology for the migration of constellation-based systems is thoroughly examined. The future trajectory of each major technological breakthrough is then investigated. Finally, recommendations for further studies are made based on the leading technologies of the intelligent processing framework for LEO-based ESM systems.
-
圖 1 低軌電磁監測系統的具體實現. (a) 中國航天科工集團[1]; (b) 美國的HawkEye 360[5]; (c) 盧森堡的Kleos Space[5]; (d) 法國的UnseenLabs[5]
Figure 1. Concrete implementation of a LEO-based ESM system: (a) China Aerospace Science and Industry Corporation[1]; (b) HawkEye 360 of the United States[5]; (c) Kleos Space of Luxembourg[5]; (d) Unseenlabs of France[5]
圖 2 低軌電磁監測系統關鍵技術
Figure 2. Key technologies of LEO constellation electromagnetic spectrum monitoring systems
圖 3 低軌電磁監測系統智能處理框架
Figure 3. Intelligent processing framework of LEO constellation electromagnetic spectrum monitoring system
圖 4 神經網絡基本結構. (a)全連接神經網絡; (b)卷積神經網絡; (c)循環神經網絡; (d) transformer
Figure 4. Basic structure of neural network: (a) fully connected neural network; (b) convolutional neural network; (c) recurrent neural network; (d) transformer
圖 5 低軌電磁監測系統智能處理關鍵技術
Figure 5. Key technologies of LEO constellation electromagnetic spectrum monitoring intelligent processing framework
圖 6 協作感知類型. (a) 分布式; (b) 中繼式; (c) 分簇式; (d) 集中式
Figure 6. Types of collaborative sensing: (a) distributed; (b) relayed; (c) clustered; (d) centralized
圖 7 盲源分離類型. (a) 超定盲源分離; (b) 正定盲源分離; (c) 欠定盲源分離
Figure 7. Blind source separation type: (a) over-determined blind source separation; (b) positive-determined blind source separation; (c) under-determined blind source separation
久色视频
圖 8 無源定位方法. (a) 兩步法; (b) 直接法
Figure 8. Passive positioning methods: (a) two-step positioning; (b) direct position determination
-
參考文獻
[1] Su K, Ma Q, Zhu W Q, et al. Micro-satellite based electromagnetic spectrum detection system technologies. Aerosp Electron Warf, 2018, 34(6): 6 doi: 10.3969/j.issn.1673-2421.2018.06.002蘇抗, 馬琴, 朱偉強, 等. 基于微納衛星的電磁頻譜監測系統技術. 航天電子對抗, 2018, 34(6):6 doi: 10.3969/j.issn.1673-2421.2018.06.002 [2] Xia Z Q. Enlightenment of iridium’s comeback. Mod Enterp, 2001(6): 42 [3] Martínez P F O, Uribe G G A, Mosquera P F L. OneWeb: web content adaptation platform based on W3C Mobile Web Initiative guidelines. Ingeniería E Investig, 2011, 31(1): 117 [4] Sun Z. A LM-11 carrier rocket successfully sends the first satellite in Hongyun project. Aerosp China, 2019(1): 42 doi: 10.3969/j.issn.1002-7742.2019.01.014孫喆. “長征”十一號火箭成功發射“虹云”工程首顆衛星. 中國航天, 2019(1):42 doi: 10.3969/j.issn.1002-7742.2019.01.014 [5] Xia R, Deng B Y, Wang J C. A review of the progress of low-orbit electromagnetic monitoring system // 2021 National Conference on Electromagnetic Spectrum Security and Control. Nanjing, 2021: 262夏瑞, 鄧博于, 王敬超. 低軌電磁監測系統進展綜述//2021全國電磁頻譜安全與控制大會. 南京, 2021: 262 [6] Jewett R. HawkEye 360 Prepares to Expand RF Tracking Capabilities With Second Cluster [EB/OL]. (2021-1-28) [2022-3-23]. https://www.satellitetoday.com/imagery-and-sensing/2021/01/28/hawkeye-360-prepares-to-expand-rf-tracking-capabilities-with-second-cluster/ [7] Liu X, Sun Q Q, Lu W D, et al. Big-data-based intelligent spectrum sensing for heterogeneous spectrum communications in 5G. IEEE Wirel Commun, 2020, 27(5): 67 doi: 10.1109/MWC.001.1900493 [8] Oquab M, Bottou L, Laptev I, et al. Learning and transferring mid-level image representations using convolutional neural networks // 2014 IEEE Conference on Computer Vision & Pattern Recognition. Columbus, 2014: 1717 [9] Graves A, Mohamed A R, Hinton G. Speech recognition with deep recurrent neural networks // IEEE International Conference on Acoustics, Speech and Signal Processing. Vancouver, 2013: 6645 [10] Otter D W, Medina J R, Kalita J K. A survey of the usages of deep learning for natural language processing. IEEE Trans Neural Netw Learn Syst, 2021, 32(2): 604 doi: 10.1109/TNNLS.2020.2979670 [11] Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks. Science, 2006, 313(5786): 504 doi: 10.1126/science.1127647 [12] Rumelhart D E, Hinton G E, Williams R J. Learning representations by back propagating errors. Nature, 1986, 323(6088): 533 doi: 10.1038/323533a0 [13] Lecun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition. Proc IEEE, 1998, 86(11): 2278 doi: 10.1109/5.726791 [14] Williams R J, Zipser D. A learning algorithm for continually running fully recurrent neural networks. Neural Comput, 1989, 1(2): 270 doi: 10.1162/neco.1989.1.2.270 [15] Vaswani A, Shazeer N, Parmar N, et al. Attention is All you Need // 31st Conference on Neural Information Processing Systems (NIPS 2017). Long Beach, 2017 [16] Zheng S, Chen S, Yang L, et al. Big data processing architecture for radio signals empowered by deep learning: Concept, experiment, applications and challenges. IEEE Access, 2018, 6: 55907 doi: 10.1109/ACCESS.2018.2872769 [17] Lee D J, Jang M S. Optimal spectrum sensing time considering spectrum handoff due to false alarm in cognitive radio networks. IEEE Commun Lett, 2009, 13(12): 899 doi: 10.1109/LCOMM.2009.12.091448 [18] Taherpour A, Nasiri-Kenari M, Gazor S. Multiple antenna spectrum sensing in cognitive radios. IEEE Trans Wirel Commun, 2010, 9(2): 814 doi: 10.1109/TWC.2009.02.090385 [19] Zeng Y H, Liang Y C. Robustness of the cyclostationary detection to cyclic frequency mismatch // 21st Annual IEEE International Symposium on Personal, Indoor and Mobile Radio Communications. Istanbul, 2010: 2704 [20] Chen H S, Gao W, Daut D G. Signature based spectrum sensing algorithms for IEEE 802.22 WRAN // 2007 IEEE International Conference on Communications. Glasgow, 2007: 6487 [21] Liu C, Wang J, Liu X M, et al. Maximum eigenvalue-based goodness-of-fit detection for spectrum sensing in cognitive radio. IEEE Trans Veh Technol, 2019, 68(8): 7747 doi: 10.1109/TVT.2019.2923648 [22] Zhou M. SSDF Attack and Defense Strategies in Cooperative Spectrum Sensing of Cognitive Radio Networks [Dissertation]. Hangzhou: Zhejiang University, 2016周明. 認知無線電網絡合作頻譜感知中的SSDF攻擊及其防御機制[學位論文]. 杭州: 浙江大學, 2016 [23] Hu L N, Cao N, Mao M H, et al. Dynamic adaptive double-threshold cooperative spectrum sensing with multi-level quantization // 2019 IEEE 3rd Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC). Chongqing, 2019: 1381 [24] Hamza D, A?ssa S, Aniba G. Equal gain combining for cooperative spectrum sensing in cognitive radio networks. IEEE Trans Wirel Commun, 2014, 13(8): 4334 doi: 10.1109/TWC.2014.2317788 [25] Ali A, Hamouda W. Advances on spectrum sensing for cognitive radio networks: Theory and applications. IEEE Commun Surv Tutor, 2017, 19(2): 1277 doi: 10.1109/COMST.2016.2631080 [26] Pan G L, Li J, Lin F. A cognitive radio spectrum sensing method for an OFDM signal based on deep learning and cycle spectrum. Int J Digit Multimed Broadcast, 2020, 2020: 5069021 [27] Liu C, Wang J, Liu X M, et al. Deep CM-CNN for spectrum sensing in cognitive radio. IEEE J Sel Areas Commun, 2019, 37(10): 2306 doi: 10.1109/JSAC.2019.2933892 [28] Zhao Y X, Zhu X M, Duan G X, et al. Spectrum sensing for cognitive radio using FLOM and CNN in alpha noise // 2021 IEEE Wireless Communications and Networking Conference (WCNC). Nanjing, 2021: 1 [29] Uvaydov D, D’Oro S, Restuccia F, et al. DeepSense: fast wideband spectrum sensing through real-time In-the-loop deep learning // IEEE INFOCOM 2021?IEEE Conference on Computer Communications. Vancouver, 2021: 1 [30] Chen Z B, Xu Y Q, Wang H B, et al. Deep STFT-CNN for spectrum sensing in cognitive radio. IEEE Commun Lett, 2021, 25(3): 864 doi: 10.1109/LCOMM.2020.3037273 [31] Xie J, Fang J, Liu C, et al. Unsupervised deep spectrum sensing: A variational auto-encoder based approach. IEEE Trans Veh Technol, 2020, 69(5): 5307 doi: 10.1109/TVT.2020.2982203 [32] Lee W, Kim M, Cho D H. Deep cooperative sensing: Cooperative spectrum sensing based on convolutional neural networks. IEEE Trans Veh Technol, 2019, 68(3): 3005 doi: 10.1109/TVT.2019.2891291 [33] Cai P X, Zhang Y, Pan C Y. Coordination graph-based deep reinforcement learning for cooperative spectrum sensing under correlated fading. IEEE Wirel Commun Lett, 2020, 9(10): 1778 doi: 10.1109/LWC.2020.3004687 [34] Chen Z B, Xu Y Q, Wang H B, et al. Federated learning-based cooperative spectrum sensing in cognitive radio. IEEE Commun Lett, 2022, 26(2): 330 doi: 10.1109/LCOMM.2021.3114742 [35] Sarikhani R, Keynia F. Cooperative spectrum sensing meets machine learning: Deep reinforcement learning approach. IEEE Commun Lett, 2020, 24(7): 1459 doi: 10.1109/LCOMM.2020.2984430 [36] Wang C C, Zeng Y H. Research status and prospects of underdetermined blind source separation algorithms. J Beijing Univ Posts Telecommun, 2018, 41(6): 103 doi: 10.13190/j.jbupt.2018-004王川川, 曾勇虎. 欠定盲源分離算法的研究現狀及展望. 北京郵電大學學報, 2018, 41(6):103 doi: 10.13190/j.jbupt.2018-004 [37] Bofill P, Zibulevsky M. Underdetermined blind source separation using sparse representations. Signal Process, 2001, 81(11): 2353 doi: 10.1016/S0165-1684(01)00120-7 [38] Carabias-Orti J J, Nikunen J, Virtanen T, et al. Multichannel blind sound source separation using spatial covariance model with level and time differences and nonnegative matrix factorization. IEEE/ACM Trans Audio Speech Lang Process, 2018, 26(9): 1512 doi: 10.1109/TASLP.2018.2830105 [39] Lee D D, Seung H S. Learning the parts of objects by non-negative matrix factorization. Nature, 1999, 401(6755): 788 doi: 10.1038/44565 [40] Wu W F, Chen X H, Su X J, et al. Wavelet decomposition algorithm for source number estimation of mechanical vibration. Mech Sci Technol Aerosp Eng, 2011, 30(10): 1679 doi: 10.13433/j.cnki.1003-8728.2011.10.018毋文峰, 陳小虎, 蘇勛家, 等. 機械振動源數估計的小波方法. 機械科學與技術, 2011, 30(10):1679 doi: 10.13433/j.cnki.1003-8728.2011.10.018 [41] Li Z N, Liu W B, Yi X B. Underdetermined blind source separation method of machine faults based on local Mean decomposition. J Mech Eng, 2011, 47(7): 97 doi: 10.3901/JME.2011.07.097李志農, 劉衛兵, 易小兵. 基于局域均值分解的機械故障欠定盲源分離方法研究. 機械工程學報, 2011, 47(7):97 doi: 10.3901/JME.2011.07.097 [42] Hershey J R, Chen Z, Roux J L, et al. Deep clustering: discriminative embeddings for segmentation and separation // 2016 IEEE International Conference on Acoustics, Speech and Signal Processing. Shanghai, 2016: 31 [43] Wang S S, Naithani G, Virtanen T. Low-latency Deep Clustering for Speech Separation//ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Brighton, 2019: 76 [44] Chen Z, Luo Y, Mesgarani N. Deep attractor network for single-microphone speaker separation // 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). New Orleans, 2017: 246 [45] Han C, Luo Y, Mesgarani N. Online deep attractor network for real-time single-channel speech separation // ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing. Brighton, 2019: 361 [46] Drude L, Higuchi T, Kinoshita K, et al. Dual frequency- and block-permutation alignment for deep learning based block-online blind source separation // 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Calgary, 2018: 691 [47] Kinoshita K, Drude L, Delcroix M, et al. Listening to each speaker one by one with recurrent selective hearing networks // 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Calgary, 2018: 5064 [48] Brunner G, Naas N, Palsson S, et al. Monaural music source separation using a ResNet latent separator network // 2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI). Portland, 2019: 1124 [49] Luo Y, Mesgarani N. Conv-TasNet: Surpassing ideal time–frequency magnitude masking for speech separation. IEEE/ACM Trans Audio Speech Lang Process, 2019, 27(8): 1256 doi: 10.1109/TASLP.2019.2915167 [50] Zhao M C, Yao X J, Wang J, et al. Single-channel blind source separation method of spatial aliasing signal based on Stacked-TCN. Syst Eng Electron, 2021, 43(9): 2628 doi: 10.12305/j.issn.1001-506X.2021.09.32趙孟晨, 姚秀娟, 王靜, 等. 基于Stacked-TCN的空間混疊信號單通道盲源分離方法. 系統工程與電子技術, 2021, 43(9):2628 doi: 10.12305/j.issn.1001-506X.2021.09.32 [51] Togami M. Deep Multi-channel speech source separation with time-frequency masking for spatially filtered microphone input signal // 2020 28th European Signal Processing Conference (EUSIPCO). Amsterdam, 2021: 266 [52] Watanabe R, Kitamura D, Saruwatari H, et al. DNN-Based frequency component prediction for frequency-domain audio source separation // 2020 28th European Signal Processing Conference (EUSIPCO). Amsterdam, 2021: 805 [53] Drude L, Hasenklever D, Haeb-Umbach R. Unsupervised training of a deep clustering model for multichannel blind source separation // ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Brighton, 2019: 695 [54] Schmidt R. Multiple emitter location and signal parameter estimation. IEEE Trans Antennas Propag, 1986, 34(3): 276 doi: 10.1109/TAP.1986.1143830 [55] Weiss A J. Direct position determination of narrowband radio frequency transmitters. IEEE Signal Process Lett, 2004, 11(5): 513 doi: 10.1109/LSP.2004.826501 [56] Carter G C. Coherence and time delay estimation. Proc IEEE, 1987, 75(2): 236 doi: 10.1109/PROC.1987.13723 [57] So H, Ching P C, Chan Y T. A new algorithm for explicit adaptation of time delay. IEEE Trans Signal Process, 1994, 42(7): 1816 doi: 10.1109/78.298289 [58] Ho K C. Bias reduction for an explicit solution of source localization using TDOA. IEEE Trans Signal Process, 2012, 60(5): 2101 doi: 10.1109/TSP.2012.2187283 [59] Ulman R, Geraniotis E. Wideband TDOA/FDOA processing using summation of short-time CAF’s. IEEE Trans Signal Process, 1999, 47(12): 3193 doi: 10.1109/78.806065 [60] Ho K C, Lu X N, Kovavisaruch L. Source localization using TDOA and FDOA measurements in the presence of receiver location errors: Analysis and solution. IEEE Trans Signal Process, 2007, 55: 684 doi: 10.1109/TSP.2006.885744 [61] Zhang S X, Xing M D. A novel Doppler chirp rate and baseline estimation approach in the time domain based on weighted local maximum-likelihood for an MC-HRWS SAR system. IEEE Geosci Remote Sens Lett, 2017, 14(3): 299 doi: 10.1109/LGRS.2016.2633359 [62] Zhang W Q. Fast Doppler rate estimation based on fourth-order moment spectrum. Electron Lett, 2015, 51(23): 1926 doi: 10.1049/el.2015.2182 [63] Wu H, Yao F Q, Chen Y, et al. Multibit-quantization-based collaborative spectrum sensing scheme for cognitive sensor networks. IEEE Access, 2017, 5: 25207 doi: 10.1109/ACCESS.2017.2767101 [64] Weiss A J. On the accuracy of a cellular location system based on RSS measurements. IEEE Trans Veh Technol, 2003, 52(6): 1508 doi: 10.1109/TVT.2003.819613 [65] Weiss A J, Amar A. Direct geolocation of stationary wideband radio signal based on time delays and Doppler shifts // 2009 IEEE/SP 15th Workshop on Statistical Signal Processing. Cardiff, 2009: 101 [66] Chen X, Wang D, Liu Z P, et al. A fast direct position determination for multiple sources based on radial basis function neural network // 2018 13th APCA International Conference on Automatic Control and Soft Computing (CONTROLO). Chengdu, 2018: 381 [67] Zhao C, Zhao Y J. One recurrent neural networks solution for passive localization. Neural Process Lett, 2019, 49(2): 787 doi: 10.1007/s11063-018-9856-y [68] Pak S, Chalise B K, Himed B. Target localization in multi-static passive radar systems with artificial neural networks // 2019 International Radar Conference (RADAR). Toulon, 2019: 1 [69] Liu G N, Wu H C, Xiang W D, et al. Indoor object localization and tracking using deep learning over received signal strength // 2020 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB). Paris, 2020: 1 [70] Wang Z W, Hu D X, Zhao Y J, et al. Real-time passive localization of TDOA via neural networks. IEEE Commun Lett, 2021, 25(10): 3320 doi: 10.1109/LCOMM.2021.3097065 [71] Gotsis K A, Kaifas T N, Siakavara K, et al. Direction of Arrival (DoA) estimation for a Switched-Beam DS-CDMA System using Neural Networks // 2007 19th International Conference on Applied Electromagnetics and Communications. Dubrovnik, 2007: 1 -
下載:
點擊查看大圖
圖(8)
計量
- 文章訪問數: 561
- HTML全文瀏覽量: 241
- PDF下載量: 105
- 被引次數: 0
