基于卷积神经网络的通信信号调制识别研究

针对传统人工提取专家特征来进行通信信号识别的方法存在局限性大、低信噪比下准确率低的问题,提出一种复基带信号与卷积神经网络自动调制识别相结合的新方法。该方法将接收到的信号进行预处理,得到包含同相分量和正交分量的复基带信号,该信号作为输入卷积神经网络模型的数据集,通过多次训练调整模型结构以及卷积核、步长、特征图和激活函数等超参数,利用训练好的模型对通信信号进行特征提取和识别。实现了对2FSK、4FSK、BPSK、8PSK、QPSK、QAM16和QAM64 七种数字通信信号类型的识别分类。实验结果表明,当信噪比为0dB时,七种信号的平均识别准确率已达94.61%,验证了算法是有效的且在低信噪比条件下有较高的准确率。     

基于深度卷积神经网络的数字调制方式识别

针对非协作通信条件下信号调制方式识别问题,提出了一种基于深度神经网络的调制方式自动识别新方法。该方法对接收到的信号进行预处理,生成星座图,并将星座图形状作为深度卷积神经网络的输入,根据训练好的网络模型对调制信号进行分类识别。与以往的识别方法相比,该方法利用卷积神经网络自动学习各种数字调制信号的星座图特征,克服了特征提取困难,通用性不强,抗噪声性能差等缺点,处理流程简单,并对星座图的形变具有不敏感性。针对4QAM、16QAM和64QAM三种典型的数字调制方式,进行了仿真实验,当信噪比大于4时,调制方式的识别正确率大于95％,实验结果表明,基于深度卷积神经网络的信号调制方式识别方法是有效的。     

基于高阶累积量和DNN模型的井下信号识别方法

针对矿井复杂异构的无线环境,提出一种基于高阶累积量和DNN模型的井下信号识别方法,实现了井下BPSK,QPSK,8PSK,2FSK,4FSK,8FSK,32QAM,64QAM,OFDM等数字信号的自动调制识别。分析得到9种数字信号的高阶累积量理论值,并通过傅里叶变换提高信号辨识度;分析井下小尺度衰落信道对高阶累积量的影响,推导出经过井下衰落信道后信号的高阶累积量计算表达式,根据高阶累积量理论值构造特征参数并训练DNN模型,实现信号识别。仿真分析结果表明,该方法在矿井Nakagami-m衰落信道下有出色的调制识别性能,信噪比为-5 dB时平均正确识别率为89.2%以上,信噪比为5 dB以上时平均正确识别率为100%。该方法为在特殊复杂环境下的信号识别检测提供了新思路。     

基于小波和高阶累积量的数字调制识别算法

利用基于高阶累积量的数字调制识别算法对数字调制信号进行分类识别时,六阶及六阶以上累积量的计算过于复杂,且多进制频移键控(MFSK)与8PSK信号各阶累积量的值均相等,直接计算无法识别。针对此问题,提出了一种基于小波和高阶累积量相结合的分类算法,先对MFSK与8PSK信号进行小波变换,再利用四阶累积量进行识别。实验证明,利用该算法所提取的特征参数能有效抑制高斯白噪声,除了识别2ASK/BPSK,4ASK,2FSK,4FSK,QPSK,8PSK信号外,还可识别16QAM,并且计算量小,易于实现。当信噪比大于等于3dB时,总体识别率达到96%。与已有算法相比,仿真结果证明了该算法的优越性。     

基于高阶累积量的数字调制信号识别方法

通信信号调制方式自动识别在信号检测、威胁分析、频谱监测等领域有着重要的地位,是非合作通信关注的关键技术.针对单一累积量调制信号识别有限且识别率低等问题,利用信号的二、四、六阶累积量特征所构造的矢量集,实现了MASK、MPSK、MFSK、MQAM四类信号的类间识别,以及2ASK、4ASK、8ASK,2PSK、4PSK、8PSK,2FSK、4FSK、8FSK,4QAM、16QAM、64QAM的类内识别.在Matlab环境下进行了仿真实验,实验结果表明,该方法在信噪比大于5 dB时可以达到90％以上的识别率.     

Using fuzzy clustering and TTSAS algorithm for modulation classification based on constellation diagram

The automatic recognition of the modulation format of a detected signal, the intermediate step between signal detection and demodulation, is a major task of an intelligent receiver, with various civilian and military applications. Obviously, with no knowledge of the transmitted data and many unknown parameters at the receiver, such as the signal power, carrier frequency and phase offsets, timing information, etc., blind identification of the modulation is a difficult task. This becomes even more challenging in real world.In this paper I develop a novel algorithm using Two Threshold Sequential Algorithmic Scheme (TTSAS) algorithm and pattern recognition to identify the modulation types of the communication signals automatically. I have proposed and implemented a technique that casts modulation recognition into shape recognition. Constellation diagram is a traditional and powerful tool for design and evaluation of digital modulations. In this paper, modulation classification is performed using constellation of the received signal by fuzzy clustering and consequently hierarchical clustering algorithms are used for classification of Quadrature–Amplitude Modulation (QAM) and Phase Shift Keying (PSK) modulations and also modulated signal symbols constellation utilizing TTSAS clustering algorithm, and matching with standard templates, is used for classification of QAM modulation. TTSAS algorithm used here is implemented by the Hamming neural network. The simulation results show the capability of this method for modulation classification with high accuracy and appropriate convergence in the presence of noise.     

QAM调制技术及其MATLAB仿真

文章首先概述了QAM调制技术的原理,然后重点介绍了QAM调制技术在有线电视系统中的作用和意义,并且对QAM的频道设置和输出进行了说明。文章还利用MATLAB对基于QAM的数字通信系统进行仿真,得到了良好效果的接收星座图和补偿后的星座图、信道估计图、系统BER分析图。仿真结果表明这种通过MATLAB实现的数字通信系统具有较强的可实现性,为实际应用和科学合理地设计正交幅度调制系统,提供了高效的仿真平台。     

多功能数字信号调制器的研究

本文介绍以８０３１单片机为核心的数字信号调制器的实现原理，由于采用了倍频技术和复合波形合成技术，因而这种数字信号调制器的传输速率在１００－９６００ｂｉｔ／ｓ范围内可调，并且在ＡＳＫ、ＦＳＫ、ＰＳＫ、ＤＰＳＫ、ＱＡＭ等信号调制方式可供选择，以便适用于各种特性的信道。     

基于轻量型网络的单载波信号调制识别

针对当前应用深度学习实现数字信号调制识别过程中网络复杂、计算量高、硬件平台要求高的问题, 本文提出了在改进的MobileNetV3轻量级神经网络中使用信号星座图调制识别的方法. 首先, 将接收到的MPSK和MQAM信号转换成星座图像, 将其进行灰度图像提取, 灰度图像增强, 构建星座图的图像数据集, 然后将ResNet中的跨层结构引入MobileNetV3网络, 解决了随着网络层数的增加, 权重减小而导致的梯度消失现象. 最后将星座图数据集用于训练MobileNetV3的轻量型神经网络权重, 对星座图像进行识别. MobileNetV3基于深度卷积可分离和神经架构搜索(network architecture search, NAS)技术在保证识别精度的前提下, 大大降低了参数量和训练时间, 将对于简单信号的调制识别, 轻量型神经网络可以有效简化网络结构, 降低对硬件设备的要求. 仿真结果表明, 针对的调制信号(BPSK、QPSK、8PSK、16QAM、64QAM), 能实现识别率为99.76%的调制识别, 相较于传统应用深度学习实现调制识别的网络, 网络参数量和计算量明显减小.     

基于DSP软件无线电数字调制解调技术的应用研究

该系统主要是研究基于软件无线电思想的调制解调技术。在以TI公司的TMS320C6711数字信号处理器(DSP)为核心的软件无线电平台上实现FSK,QPSK,QAM等多种制式的调制解调功能,通过对主机中虚拟平台的操作,实现对调制解调制式的实时选择和数据测量的实时显示,并以QPSK为例进行了观察和分析。     

低信噪比下数字调制信号脉内特征提取

为了改善小波变换法在低信噪比下提取数字调制信号脉内特征时的性能,首先利用WVD变换在零频处既可以保持原始信号调制特性不变,又具有很好的噪声抑制作用的特性,对信号进行降噪处理,然后从理论上分析了PSK信号和FSK信号小波脊线的性质,根据不同调制信号小波脊线的差异可以得出其调制类型,进而可以得到信号的频率和码速等脉内调制信息.通过对低信噪比下PSK信号和FSK信号的计算机仿真,取得了较好的效果,也验证了方法的可行性.     

Classification of modulation signals using statistical signal characterization and artificial neural networks

Modulation recognition systems have to be able to correctly classify the incoming signal modulation scheme in the presence of noise. A new method for classification of analogue and digital modulated signals at low signal-to-noise ratio (SNR) is introduced in this paper. This method uses the statistical signal characterization (SSC) to extract parameters to classify the different modulation signals. The SSC technique produces a set of four numerical parameters for a specific modulated signal. Subsequent comparison of these parameters to those of other waveforms provides the basis for our classification system. The results of SSC technique are applied to an artificial neural network (ANN) to have a robust classification system in the presence of noise down to SNR of 3 dB. No a priori information is required by this technique about the set of input waveforms. The input to the classification system can be analogue or digital signals or a combination of both. The proposed technique shows a 100% efficiency of classification of analogue signals or digital signals at SNR of 7 dB. This classification efficiency reduces to 83% and 86% for analogue or digital signals at SNR of 3 dB. The SSC technique shows better classification results in comparison with other techniques with an important advantage over other methods, which is the simplicity of the neural network needed with this technique due to the small number of features used in the classification.     

Blind signal modulation recognition through clustering analysis of constellation signature

Automatic recognition of digital modulation schemes is becoming an active research area in many covert operations. It has many military applications where surveillance and electronic warfare requires a type of modulation in intercepted signal to prepare jamming signals. Most of the approaches are based on modulated signal's component, but the modulation type can be best identified with the use of constellation diagram. The proposed technique is able to recognize M-QAM, M-ASK, and M-PSK modulation scheme in Additive White Gaussian Noise (AWGN) environment. As the constellation points form clusters in the I-Q plane, the order of the modulation can be obtained by estimating the correct number of clusters, which is calculated by OPTICS algorithm. The least square error has been calculated using linear regression from the obtained constellation points, to identify either ASK or PSK and QAM. The error is least for ASK which differentiates ASK from PSK and QAM. To identify between the PSK and QAM, k-means clustering is employed to find the number of centroids equal to order of modulation estimated by OPTICS. With the difference in maximum and minimum absolute value of the centroids, PSK or QAM is recognized. The proposed method shows an improvement in the classification accuracy which reaches 100% using 1024 symbols at 20 dB compared to 98.89%, 98.05%, and 98% when using more complex classifiers like Support Vector Machine, Naive Bayes Classifier, KNN respectively. The method used is unsupervised whereas most of the methods in the literature require training phase to set the thresholds or weights for final model to detect modulation type. This algorithm is also implemented in LabVIEW, and tested on real-time signals. An intelligent system is made which does not require any knowledge of symbol rate, carrier frequency, and any training phase to set thresholds, and detects the type of modulation blindly in real time. Modulated RF signals are generated by NI PXIe-5673 (RF transmitter). NI PXI 5600 is used to downconvert RF signal and NI PXI-5142 (100 MS/s OSP digitizer) is used to sample the downverted signal.     

基于软件无线电和神经网络的RFID调制制式测试研究

RFID测试技术的发展对RFID技术的成熟和广泛应用具有重要的理论意义和实用价值;RFID调制制式测试(识别)则是RFID测试技术重要研究内容之一;文中提出了一种新的基于软件无线电的RFID调制方式测试方法;`在软件无线电的基础上,采用4个特征参数,用BP神经网络实现对2ASK、4ASK、2FSK、4FSK和BPSK等RFID系统典型调制方式进行识别;讨论了方案设计,给出了仿真实验结果;仿真结果表明该方法具有较好的准确性和稳健性。     

Automatic determination of digital modulation types with different noises using Convolutional Neural Network based on time–frequency information

In this study, a novel digital modulation classification model has been proposed for automatically recognizing six different modulation types including amplitude shift keying (ASK), frequency shift keying (FSK), phase-shift keying (PSK), quadrate amplitude shift keying (QASK), quadrate frequency shift keying (QFSK), and quadrate phase-shift keying (QPSK). The determination of modulation type is significant in military communication, satellite communication systems, and submarine communication. To classify the modulation types, we have proposed a two-stage hybrid method combining short-time Fourier transform (STFT) and convolutional neural network (CNN). In the first stage, as the data source, the time–frequency information from these modulation signals have been extracted with STFT. This information has been obtained as 2D images to feed the input of the CNN deep learning method. In the second stage, the obtained 2D time–frequency information has been given to the input of the CNN algorithm to classify the modulation types. In this work, noises at various SNR values from 0 dB to 25 dB were created and added to the modulated signals. Even in the presence of noise, the proposed hybrid deep learning model achieved excellent results in the noised-modulation signals.     

基于高效自适应聚类算法的调制识别研究

提出了一种基于星座聚类的通信信号调制识别新方法.该方法将星座图形状作为调制识别的特征,运用聚类算法EAFCM(efficient adaptive fuzzy C-means)重建接受信号的星座图.基于模糊C-均值(FCM)聚类算法的自适应高效聚类算法EAFCM不仅克服了模糊C-均值聚类算法需要预先确定聚类参数c、对初始中心敏感等不足,而且具有良好的抗噪声性能.将该方法应用到对PSK/QAM信号的调制识别,实验结果表明该方法是实际有效的.     

Classification of communications signals using an advanced technique

Because of rapid growing of radio communication technology of late years, importance of automatic classification of digital signal type is rising increasingly. This paper presents an advanced technique that identifies a variety of digital signal types. This method is a hybrid heuristic formed by a radial basis function neural networks (as a classifier) and particle swarm optimization technique. A suitable combination of higher order statistics up to eighth are proposed as the prominent characteristics of the considered signals. In conjunction with neural network we have used a cross-validation technique to improve the generalization ability. Experimental results indicate that the proposed technique has high percentage of correct classification to discriminate different types of digital signal even at low SNRs.     

基于FPGA的数字信号发生器及其在QPSK调制中的应用

在现代通信中利用载波携带数字信息的传输应用广泛,如PSK、QAM等调制技术均要用到正弦形式的载波。传统的载波产生采用模拟电路实现,不利于集成且性能受限。文章讨论使用FPGA芯片、利用直接数字频率合成技术制作成数字信号发生器,该信号发生器具有体积小、功能强的优点,且结合该信号发生器在QPSK调制器中的实际应用验证了该方案的合理性。     

基于自适应级联模糊推理的调制识别算法

针对非平稳的数字调制信号,构造新的高阶交又累量特征;利用神经网络的学习机制实现自适应模糊推理调制识别器的非线性动态建模;采取分层决策的级联结构,提高了特征与识别器的契合度,最大程度上减少了隶属度函数和模糊规则的冗余;根据特征样本的大致分布建立蕴涵初始经验的级联模糊神经网络系统,使知识推理结构明确可控;通过样本训练实现结构参数自适应调整和优化,完成其逼近求精.仿真实验证明,该系统在信噪比等环境参数变化较大的情况下具有更好的稳健性,其算法识别率和效率相对于神经网络识别器和模糊识别器有明显提高.     

Learn class hierarchy using convolutional neural networks

A large amount of research on Convolutional Neural Networks (CNN) has focused on flat Classification in the multi-class domain. In the real world, many problems are naturally expressed as hierarchical classification problems, in which the classes to be predicted are organized in a hierarchy of classes. In this paper, we propose a new architecture for hierarchical classification, introducing a stack of deep linear layers using cross-entropy loss functions combined to a center loss function. The proposed architecture can extend any neural network model and simultaneously optimizes loss functions to discover local hierarchical class relationships and a loss function to discover global information from the whole class hierarchy while penalizing class hierarchy violations. We experimentally show that our hierarchical classifier presents advantages to the traditional classification approaches finding application in computer vision tasks. The same approach can also be applied to some CNN for text classification.