基于分布式小型前端组阵的长波信号智能检测方法  

An Intelligent Detection Method of Long-wave Signal Based on Distributed Small Front Arrays

作　　者：陈怀 胡赟鹏[1] 沈智翔[1] 李睿[2] CHEN Huai;HU Yunpeng;SHEN Zhixiang;LI Rui(Institute of Information Systems Engineering,Information Engineering University,Zhengzhou 450001,China;Unit 72221 of PLA,Jinan 250000,China)

机构地区：[1]信息工程大学信息系统工程学院,郑州450001 [2]中国人民解放军72221部队,济南250000

出　　处：《电讯技术》2025年第3期354-362,共9页Telecommunication Engineering

基　　金：国家科学自然基金资助项目(62171468)。

摘　　要：面向分布式小型化磁感前端阵列接收长波信号场景,针对长波频段噪声复杂的特性,提出了一种宽带多信号智能联合检测方法。该方法基于已知参考信号样本预训练神经网络,通过神经网络在预训练阶段学习分布式接收样本矢量在各维度上的潜在复杂关联性规律,进而部署网络后可输出基于输入样本矢量联合概率的置信度量,用于判断当前样本是否存在目标信号从而得到检测结果。基于宽带信号仿真数据集进行实验,结果表明算法可直接对宽带数据进行处理,并能有效完成频谱感知,能够在低信噪比和相关噪声条件下获得接近理论处理增益的检测性能,达到80%以上的检测率。在此基础上,采用中科院空天信息创新研究院布设于内蒙古的超短基线电磁探测阵列(Mini-array by Chinese Academy of Sciences,CASMA)采集的实际长波信号数据进行性能验证,算法性能测试结果同样验证了其有效性。该方法不限于长波信号,也适用于其他具备参考信号条件下的信号盲检测的场景,实现在信道参数未知、信号微弱等盲环境条件下获得更优的目标信号检测性能。In consideration of the complex noise characteristics of long-wave,an intelligent joint detection method for wideband multi-signal is proposed for the scenario in which a distributed miniaturized magnetic sensing front array is used to receive long-wave signals.The method leverages pre-trained neural networks using known reference signal samples to learn the potential correlation laws of the distributed received sample vectors in each dimension during the pre-training phase.After deployment,the network can output confidence metrics based on the joint probability of input sample vectors,enabling the determination of whether the current sample contains a target signal and thus yielding detection results.The algorithm can directly process wideband data and effectively perform spectrum sensing.It achieves near-theoretical processing gain in detection performance under low signal-to-noise ratio(SNR)and correlated noise conditions,with a detection rate exceeding 80%.Furthermore,the method is validated using actual long-wave signal data collected by the ultra-short baseline electromagnetic detection array(Mini-array by Chinese Academy of Sciences,CASMA),which is deployed in Inner Mongolia Autonomous Region by the Institute of Space Information Innovation,Chinese Academy of Sciences.The performance test results further confirm the effectiveness of the proposed method.This method is not only applicable to long-wave signals,but also to other signal blind detections with reference signal,for achieving better detection performance under such blind environmental conditions as channel parameter unknown or weak signals.

关 键 词：分布式小型前端阵列 长波信号检测 相关噪声 神经网络 

分 类 号：TN911.7[电子电信—通信与信息系统]