共查询到17条相似文献,搜索用时 109 毫秒
1.
为了充分挖掘高维特征空间中辐射源的细微特征, 提出一种基于全局潜在低秩表示(Global Latent Low Rank Representation, GLat-LRR)的通信辐射源潜在细微特征提取方法.首先, 提取通信辐射源信号的瞬时频率, 通过傅里叶变换将信号投影到高维特征空间; 挖掘特征样本之间全局的低秩结构和维度之间全局的潜在低秩关系, 将特征样本集作为整体应用到潜在低秩表示模型中, 利用维度之间低秩关系得到特征样本集的潜在部分矩阵, 每个列向量即为每个通信辐射源信号的潜在细微特征向量.在实际采集的同厂家同型号FM电台数据集上, 该方法提取的潜在细微特征能够显著提高通信辐射源个体识别的性能. 相似文献
2.
3.
4.
个体特征选择和提取是辐射源个体识别的关键,直接决定分类识别性能的好坏。由于在实际工程应用中,利用暂态特征进行通信辐射源个体识别难以实现,本文从稳态特征出发,对通信辐射源个体特征提取技术进行了综述,对特征的产生机理、在信号传播过程中所受到的污染以及在实际工程应用中的可行性做了归纳与分析。最后,指出了目前通信辐射源个体特征提取技术存在的问题,展望了个体识别技术未来可能的研究方面。 相似文献
5.
针对复杂电磁环境下通信辐射源个体识别问题,提出了一种小样本条件下基于深度置信网络的通信辐射源个体识别方法.首先分析通信辐射源信号频带内互调干扰信号的幅度和相位特性,建立基于互调干扰信号的通信辐射源个体特征;然后对辐射源信号进行预处理得到通信辐射源信号的矩形积分双谱,再采取对比散度的方法,利用高阶谱自底向上训练每个受限玻尔兹曼机,通过多次迭代得到合适的权重、隐藏层的偏差和可见层的偏差,从而提取出辐射信号的互调干扰信号特征;最后使用softmax分类器对训练模型进行微调,获得面向通信辐射源细微特征识别的深度学习网络.通过计算机的仿真得到了超过80%的识别率,进一步验证了该方法的有效性. 相似文献
6.
7.
针对低信噪比条件下通信辐射源个体识别率低的问题,提出了一种基于Welch功率谱和卷积神经网络的通信辐射源个体识别方法.构建了由20个基于ZigBee协议的物联网设备组成的测试平台,将ZigBee信号前同步码部分的Welch功率谱数据作为辐射源指纹特征送入卷积神经网络进行分类.该方法在低信噪比条件下很好地保留了辐射源的指纹特征,结合卷积神经网络强大的微特征提取能力,对辐射源进行了有效分类.实验结果证明,在瑞利信道及低信噪比条件下,所提方法的识别效果明显优于其他方法. 相似文献
8.
基于模糊函数的雷达辐射源个体识别 总被引:1,自引:0,他引:1
雷达辐射源个体识别是现代电子情报和电子支援系统的重要研究内容。本文在辐射源个体特征分析的基础上,提出了一种基于模糊函数的辐射源个体识别算法。针对模糊函数的冗余性,分别利用模糊函数切片和局部模糊函数切片表征雷达的细微差异与个体特征,并推导了局部模糊函数切片的快速算法。仿真实验验证了提出算法对于辐射源个体特征提取和识别的可行性和有效性。 相似文献
9.
10.
提出了一种基于经验模态分解(EMD,empirical mode decomposition)的通信辐射源个体识别方法。首先采用EMD方法将稳态信号的主要信号成分与杂散成分分离开来;然后提取杂散成分的频域特征作为信号的细微特征;最后运用支持向量机(SVM,support vector machine)分类器对多个通信辐射源个体进行分类识别。无线网卡个体识别的实验结果表明,对同型号的辐射源个体,该方法可以取得较好的识别效果;尤其在较低信噪比下,与其他方法相比,该方法获得更高的识别率。 相似文献
11.
12.
通信电台发射的信号通常表现出一定的细微特征差异,针对这种细微特征差异,本文在论证跳频信号跳变瞬时包络可以作为电台个体细微特征的基础上,提出了一种基于跳频信号瞬时包络特征的跳频电台个体识别方法。首先基于一种改进的基于小波变换的包络提取算法,提取出了样本信号跳变时刻的瞬时包络,并减轻了噪声等因素的影响。其次,分离并定量计算其盒维数和信息维数等瞬时特征,将得到的跳频信号的瞬时细微特征变换为一个特征向量,之后采用基于构造型神经网络的分类方法实现不同跳频电台的个体识别。最后对实际工作状态下3种型号电台进行个体识别,实际数据的实验结果验证了算法的有效性。 相似文献
13.
一种新的混沌同步及保密通信方式 总被引:20,自引:3,他引:17
本文提出一种新的混沌同步及保密通信方式。在发送端,将信息信号与混沌载波的和经取模运算后再嵌入混沌动力系统的迭代之中以实现调制;在接收端,用一个相应的非动力学系统从接收信号中提取混沌载波并进而恢复信息信号。研究表明,该混沌同步方式具有较强的抗干扰能力并且实现容易;该保密通信系统保密性能良好。 相似文献
14.
To measure the similarity between the fine features of communication transmitters, a method of individual communication transmitter identification based on maximum correntropy was put forward. Firstly, the square integral bispectra was extracted to characterise the communication transmitters, and then optimization function based on maximum correntropy criterion was constructed. Secondly, the half-quadratic technique transformed the nonlinear op-timization problem to a weighted linear least squares problem. Finally, for identifying the communication transmitters, the sparse coefficient computed by active set algorithm was utilized to construct the classifier. Using the FM radios with same manufacturer and model, the actually collected data sets verified the feasibility and effectiveness of the method. 相似文献
15.
利用信号围线积分双谱分形特征实现电台识别 总被引:1,自引:0,他引:1
信号的双谱能反映信号的细微特征,可用于电台识别中,但将它直接应用于电台识别需要计算复杂的匹配模板,增加分类器的复杂度,影响识别效率。针对此问题,提出了一种将信号围线积分双谱的分形特征作为电台特征参数的识别方法。首先由信号双谱估计值求出围线积分双谱,然后利用盒维数和信息维数定量描述围线积分双谱波形的复杂度,并将这两种分形维数作为特征向量,最后利用支持向量机(SVM)实现电台分类识别。对两部实际电台所发射的2FSK信号利用所提方法进行分析,结果表明在信噪比为7 dB及以上时,电台正确识别率能达到94.29%以上,验证了所提方法的可行性。 相似文献
16.
Sathyan T. Kirubarajan T. 《IEEE transactions on circuits and systems. I, Regular papers》2006,53(7):1597-1609
In this paper, a new Markov-jump-system (MJS)-based secure chaotic communication technique is proposed. An MJS evolves by switching from one state evolution model to another according to a finite state Markov chain. The transmitter in the proposed communication system is an MJS consisting of multiple transmission maps, that is, the transmitter switches from one chaotic map to another during the transmission of data. This switching feature makes it difficult to identify and follow the transmission without knowing the transmitter parameters, i.e., to eavesdrop, thereby increasing the security offered by the inherently secure chaotic communication system. If the chaotic maps used at the transmitter, and the corresponding Markov transition probability matrix of the MJS are known to the (authorized) receiver, then a multiple model estimator can be used to track the MJS transmitter. In this paper, the use of the interacting multiple model (IMM) estimator is proposed as part of the receiver to follow the switching transmitter. The effectiveness of the IMM-estimator-based receiver to follow the switching transmitter is evaluated by means of simulations. A new modulation technique that uses the MJS transmitter is also introduced. Further, it is shown that the same receiver framework, when used as a receiver for chaotic parameter modulation, provides significant performance improvement in terms of bit-error rate compared to a receiver that uses extended Kalman filter. In addition, the seemingly more complex IMM-estimator-based receiver is shown to significantly reduce the computational complexity per transmitted bit, thus resulting in increased data rate. 相似文献
17.
Experiment and analysis of the effect of fine tracking system on the unstable platform in laser communication 总被引:3,自引:0,他引:3
Atmospheric turbulence and platform vibration in the space optical communication can cause the offset and jitter of beam, which further result in the fluctuations of received optical power. To resist this effect, a communication system with fine tracking systems in the receiver and transmitter is designed. The system is used in the experiment of laser communication between high-rise buildings over a distance of 3.5 km. After adding a vibration source to the transmitter, the centroids of spots captured by the camera of the transmitter and the optical power of receiver are recorded for the purpose of analysis. When the vibration source works at the designated frequency, a peak appears at the corresponding frequency in the spectrum of the spot centroids and the optical power of receiver. Then the peak disappears once the fine tracking system begins to work. Compared with the condition without the fine tracking system, the minimum value of the optical power of receiver is increased by 5 dB, and the standard deviation is decreased by 30%. 相似文献