Underdetermined DOA estimation and blind separation of non-disjoint sources in time-frequency domain based on sparse representation method

This paper deals with the blind separation of nonstation-ary sources and direction-of-arrival （DOA） estimation in the under-determined case, when there are more sources than sensors. We assume the sources to be time-frequency （TF） disjoint to a certain extent. In particular, the number of sources presented at any TF neighborhood is strictly less than that of sensors. We can identify the real number of active sources and achieve separation in any TF neighborhood by the sparse representation method. Compared with the subspace-based algorithm under the same sparseness assumption, which suffers from the extra noise effect since it can-not estimate the true number of active sources, the proposed algorithm can estimate the number of active sources and their cor-responding TF values in any TF neighborhood simultaneously. An-other contribution of this paper is a new estimation procedure for the DOA of sources in the underdetermined case, which combines the TF sparseness of sources and the clustering technique. Sim-ulation results demonstrate the validity and high performance of the proposed algorithm in both blind source separation （BSS） and DOA estimation.

Modeling of unsupervised knowledge graph of events based on mutual information among neighbor domains and sparse representation

Text event mining,as an indispensable method of text mining processing,has attracted the extensive attention of researchers.A modeling method for knowledge graph of events based on mutual information among neighbor domains and sparse representation is proposed in this paper,i.e.UKGE-MS.Specifically,UKGE-MS can improve the existing text mining technology's ability of understanding and discovering high-dimensional unmarked information,and solves the problems of traditional unsupervised feature selection methods,which only focus on selecting features from a global perspective and ignoring the impact of local connection of samples.Firstly,considering the influence of local information of samples in feature correlation evaluation,a feature clustering algorithm based on average neighborhood mutual information is proposed,and the feature clusters with certain event correlation are obtained;Secondly,an unsupervised feature selection method based on the high-order correlation of multi-dimensional statistical data is designed by combining the dimension reduction advantage of local linear embedding algorithm and the feature selection ability of sparse representation,so as to enhance the generalization ability of the selected feature items.Finally,the events knowledge graph is constructed by means of sparse representation and l1 norm.Extensive experiments are carried out on five real datasets and synthetic datasets,and the UKGE-MS are compared with five corresponding algorithms.The experimental results show that UKGE-MS is better than the traditional method in event clustering and feature selection,and has some advantages over other methods in text event recognition and discovery.

Representation learning via an integrated autoencoder for unsupervised domain adaptation

The purpose of unsupervised domain adaptation is to use the knowledge of the source domain whose data distribution is different from that of the target domain for promoting the learning task in the target domain.The key bottleneck in unsupervised domain adaptation is how to obtain higher-level and more abstract feature representations between source and target domains which can bridge the chasm of domain discrepancy.Recently,deep learning methods based on autoencoder have achieved sound performance in representation learning,and many dual or serial autoencoderbased methods take different characteristics of data into consideration for improving the effectiveness of unsupervised domain adaptation.However,most existing methods of autoencoders just serially connect the features generated by different autoencoders,which pose challenges for the discriminative representation learning and fail to find the real cross-domain features.To address this problem,we propose a novel representation learning method based on an integrated autoencoders for unsupervised domain adaptation,called IAUDA.To capture the inter-and inner-domain features of the raw data,two different autoencoders,which are the marginalized autoencoder with maximum mean discrepancy(mAE)and convolutional autoencoder(CAE)respectively,are proposed to learn different feature representations.After higher-level features are obtained by these two different autoencoders,a sparse autoencoder is introduced to compact these inter-and inner-domain representations.In addition,a whitening layer is embedded for features processed before the mAE to reduce redundant features inside a local area.Experimental results demonstrate the effectiveness of our proposed method compared with several state-of-the-art baseline methods.

Multiple Emitters Localization by UAV with Nested Linear Array:System Scheme and 2D-DOA Estimation Algorithm

Unmanned Aerial Vehicle(UAV)equipped with uniform linear array has been applied to multiple emitters localization.Meanwhile,nested linear array enables to enhance localization resolution and achieve under-determined Direction of Arrival(DOA)estimation.In this paper,we propose a new system structure for emitters localization that combines the UAV with nested linear array,which is capable of significantly increasing the positioning accuracy of interested targets.Specifically,a localization scheme is designed to obtain the paired two-dimensional DOA(2D-DOA,i.e.azimuth and elevation angles)estimates of emitters by nested linear array with UAV.Furthermore,we propose an improved DOA estimation algorithm for emitters localization that utilizes Discrete Fourier Transform(DFT)method to obtain coarse DOA estimates,subsequently,achieve the fine DOA estimates by sparse representation.The proposed algorithm has lower computational complexity because the coarse DOA estimates enable to shrink the range of over-complete dictionary of sparse representation.In addition,compared to traditional uniform linear array,improved 2D-DOA estimation performance of emitters can be obtained with a nested linear array.Extensive simulation results testify the effectiveness of the proposed method.

Multiple transformation analysis for interference separation in TDCS

Various types of interference signals limit the practical application of transform domain communication systems(TDCSs)in the severe electromagnetic field,an orthogonal basis learning method of transformation analysis(OBL-TA)is proposed to effectively address the problem of obtaining an optimal transform domain based on sparse representation.Then,the sparse availability is utilized to obtain the optimal transformation analysis by the iterative methods,which yields the sparse representation for transform domain(SRTD)in unrestricted form.In addition,the iterative version of SRTD(I-SRTD)in unrestricted form is obtained by decomposing the SRTD problem into three sub-problems and each sub-problem is iteratively solved by learning the best orthogonal basis.Furthermore,orthogonal basis learning via cost function minimization process is conducted by stochastic descent,which is assured to converge to a local minimum at least.Finally,the optimal transformation analysis is developed by the effectiveness of different transform domains according to the accuracy of the sparse representation and an optimal transformation analysis separately(OPTAS)is applied to the synthesized signal forms with conic alternatives,dualization,and smoothing.Simulation results demonstrate that the superiorities of the proposed methods achieve the optimal recovery and separation more rapidly and accurately than conventional methods.

基于梯度阈值的太赫兹时域信号自适应稀疏算法

胶接结构广泛应用于航空航天等国防领域,但在工艺制作及使用过程可能会产生胶接界面脱粘缺陷和损伤,由于太赫兹无损检测技术对非金属材料良好的穿透性能,已被广泛应用于复合材料的无损检测中,太赫兹无损检测技术在多层胶接结构样件胶层内部缺陷的无损检测方面具有较大优势。利用反射式太赫兹时域光谱系统检测多层胶接结构样件,得到的具有样件内部材料信息的太赫兹时域信号,但信号中还包含了大量的冗余特征和噪声等无效信息,这些无效信息大大降低了信号处理和分析效率。针对这一问题,文中提出了基于二阶梯度法提取太赫兹时域信号有效特征,以飞行时间误差为限制条件基于信号的时域特征自适应确定阈值,稀疏太赫兹时域信号,减少信号中冗余无效信息,实现太赫兹时域信号的有效压缩。然后,通过二值化图像分割识别多高斯恢复信号和太赫兹时域光谱系统检测信号的太赫兹图像缺陷区域。最后,制备具有脱粘缺陷的多层胶接结构样件,开展太赫兹无损检测实验。结果表明:文中算法的数据压缩率达到了81%,相比传统压缩算法离散余弦变换提高了59%,相比主成分分析算法提高了75%,相比K-SVD字典学习算法提高了26%,缩短了约80%的数据计算时间,减小了约95%数据存储空间占用,且缺陷识别偏差不超过0.05。文中算法极大地提高了数据处理和分析效率,保证了缺陷识别的精度。

迭代重加权时域平面波叠加法的非稳态声场重建

提出了迭代重加权时域平面波叠加法,并应用于非稳态声场重建。该方法首先通过时域离散卷积得到全息面时域声压与虚源面时域波数谱离散状态公式;然后根据稀疏表示和迭代重加权,得到残差范数与加权解范数最小化问题表达式;最后将每一对波数下虚源面时域波数谱与时域传播核卷积叠加得到重建面上的时域声压。为了验证该方法的有效性,给出了激励板声源的仿真验证,并在半消声室内进行了实验验证。讨论了一些参数对该方法的影响,并通过与基于Tikhonov正则化的时域平面波叠加法比较,突出该方法的重建优势。仿真和实验结果表明,该方法对非稳态声场重建具有很好的效果,并且重建精度高。

基于多源域领域适应字典学习和稀疏表示的脑电情感识别方法

脑电信号容易记录且不易伪装,基于脑电信号的情感识别越来越受到人们的关注.然而,人类情感具有多样性和个体可变性,基于脑电信号的情感识别仍是情感计算领域的难题.本文提出一种多源域领域适应字典学习和稀疏表示方法.为减少源领域和目标领域数据分布的差异,将所有领域的数据投影到共享子空间,并在共享子空间中学习一个共有字典.根据稀疏重建的最小化类内误差和最大化类间误差准则,稀疏表示具有类别的分辨能力.另外,每个源域自适应学习领域权重,可以避免负迁移的发生.模型参数的求解通过参数交替优化方法,所有参数可同时达到最优解.DEAP数据集的实验结果显示本文方法在所有对比方法中是最优的.

基于稀疏字典表示的无监督域适应学习算法

现有迁移学习方法通常仅关注如何减小源域和目标域之间的分布差异,忽略了源域和目标域样本自身的偏差。为此提出一种基于稀疏字典表示的无监督域适应学习方法(SRDA)。利用字典学习重构源域和目标域的样本,学习源域和目标域的字典及样本的稀疏表示,在其稀疏表示空间添加约束,使目标域可以在稀疏表示空间和源域共享同一个分类器,通过在源域稀疏表示空间学习的分类器对目标域进行分类,完成目标域分类任务,实现域适应学习。实验结果表明,所提出的SRDA算法能够有效完成源域与目标域间的知识迁移,且性能优于传统基于特征迁移的域适应方法。

基于压缩采样的电力系统负荷频率自动控制研究

为降低新能源接入对电力系统的影响,研究了基于压缩采样的电力系统负荷频率自动控制方法。利用匹配追踪算法对采集到的新能源发电输出功率信号进行稀疏表示,并将稀疏表示后的信号投射至测量矩阵得到低维的测量值。在测量值范围内,运用L1范数优化方法得到重构后新能源发电输出功率。将该功率与期望功率的偏差作为扰动项,接入基于比例积分(PI)控制器的负荷频率控制模型。利用变论域方法调节PI控制器输入与输出论域。根据模糊推理定义伸缩因子,构建变论域模糊PI控制器使之作用于负荷频率控制模型,完成电力系统负荷频率的自动控制。试验结果证明:该方法控制后的输出功率稳定,负荷扰动下区域控制偏差和频率偏差小,电力系统负荷频率自动控制的精度最高可达99%,具有实用性。

基于认知驱动的变换域通信智能抗干扰方法

为满足未来无人系统通信智能抗干扰的实际需要,针对传统变换域通信系统(transform domain communication system,TDCS)自身开放性有限、干扰应对能力不足等问题,设计了基于认知引擎驱动的智能系统架构,并针对各认知引擎驱动子模块提出了3种改进方法,包括基于稀疏逼近的未知干扰处理、基于稀疏表示的变换学习干扰识别以及针对性的干扰变换稀疏分析方法。实验结果表明,识别子模块与传统的分类器相比,整体的干扰识别率提高了5.2%,并且可实现无监督的学习;同时,针对典型干扰的重构精度在90%以上,实现了不同干扰类型的最优变换处理,显著提高了系统的抗干扰性能,传输误码率逼近理想水平。

复杂场景下声频传感器网络核稀疏表示车辆识别

提出一种基于核稀疏表示的声频传感器网络车辆识别方法.首先利用Mel频率倒谱系数提取车辆声音特征;然后采用核方法将其投影到高维特征空间以实现线性可分,将线性稀疏表示扩展到核域空间,构建过完备字典,求解核稀疏最优化问题对目标车辆进行分类.实验验证了该算法在声频数据集结构复杂的情况下,能有效地识别目标车辆,与传统的声频目标分类算法相比,提高了识别的准确率.

一种快速的基于压缩感知的多普勒高分辨方法

利用雷达目标在多普勒域的稀疏性,基于压缩感知的目标多普勒估计方法,能够在有限的相干积累时间内实现多普勒的高分辨.然而,即使采用压缩感知中的一种高效算法——正交匹配追踪算法,其运算复杂度也相对较高.为了进一步降低运算复杂度,对接收脉冲进行分组,将一维的多普勒估计问题转化为一个二维的稀疏信号重构问题,进而利用一种针对二维稀疏信号优化的低复杂度正交匹配追踪算法对其进行估计.仿真表明,该方法具有较高的运算效率,并能够获得接近直接应用传统的正交匹配追踪算法的多普勒分辨率.

基于DTCWT和稀疏表示的红外偏振与光强图像融合

针对红外偏振与光强图像彼此包含共同信息和特有信息的特点,提出了一种基于双树复小波变换和稀疏表示的图像融合方法.首先,利用双树复小波变换获取源图像的高频和低频成分,并用绝对值最大值法获得融合的高频成分;然后,用低频成分组成联合矩阵,并使用K-奇异值分解法训练该矩阵的冗余字典,根据该字典求出各个低频成分的稀疏系数,通过稀疏系数中非零值的位置信息判断共有信息和特有信息,并分别使用相应的规则进行融合;最后,将融合的高低频系数经过双树复小波反变换得到融合图像.实验结果表明,本文提出的融合算法不仅能较好地凸显源图像的共有信息,而且能很好地保留它们的特有信息,同时,融合图像具有较高的对比度和细节信息.

基于多尺度几何分析的复杂网络压缩策略

自然界许多系统都具有多尺度效应,人们对系统的观察或测量也往往是在不同尺度上进行的。复杂网络可视为描述复杂系统的抽象模型,多尺度几何分析提供一种有效刻画复杂系统结构由粗到细的显微镜策略。该文从多尺度几何分析的角度对复杂网络压缩问题进行了有益的尝试,提出一种有效的网络压缩策略。策略分析及通过对不同拓扑类型的网络仿真表明基于多尺度几何分析的复杂网络压缩策略是可行的,能有效地帮助人们通过尽可能少的信息来分析、刻画大型复杂网络,实现网络数据及结构的稀疏表示。

稀疏标签传播:一种鲁棒的领域适应学习方法

稀疏表示因其所具有的鲁棒性,在模式分类领域逐渐得到关注.研究了一种基于稀疏保留模型的新颖领域适应学习方法,并提出一种鲁棒的稀疏标签传播领域适应学习(sparse label propagation domain adaptation learning,简称SLPDAL)算法.SLPDAL通过将目标领域数据进行稀疏重构,以实现源领域数据标签向目标领域平滑传播.具体来讲,SLPDAL算法分为3步:首先,基于领域间数据分布均值差最小化准则寻求一个优化的核空间,并将领域数据嵌入到该核空间;然后,在该嵌入核空间,基于l1-范最小化准则计算各领域数据的核稀疏重构系数;最后,通过保留领域数据间核稀疏重构系数约束,实现源领域数据标签向目标领域的传播.最后,将SLPDAL算法推广到多核学习框架,提出一个SLPDAL多核学习模型.在鲁棒人脸识别、视频概念检测和文本分类等领域适应学习任务上进行比较实验,所提出的方法取得了优于或可比较的学习性能.

复数图像去噪算法的稀疏编码实现

针对复数图像系统所获取的图像具有噪声的问题,将稀疏编码理论扩展到复数域中,并且运用该理论提出一种复数据的降噪算法.该算法采用复数域的正交匹配追赶算法对图像进行稀疏编码,利用自适应方式从待训练含噪图像样本中获取表示词典,并指出最终图像的降噪效果与表示的稀疏度有关.实验结果表明,文中算法对于复数图像取得了很好的降噪效果,尤其是可以有效地提高复数相位的降噪信噪比.

基于多尺度字典学习的图像融合方法

将小波域多尺度分析的思想和图像域单尺度稀疏表示的思想有效结合,提出基于多尺度字典学习的图像融合方法。首先将训练图像变换到小波域,分别对各个子带系数训练字典;根据训练的字典求解并融合源图像各个子带的稀疏表示系数;经过逆小波变换重构融合图像。提出的方法综合了学习字典的稀疏特性和小波分析的多分辨率特性。实验结果表明较现有基于图像域字典学习的融合方法和基于小波域多尺度分析的融合方法均具有更优的融合效果。

基于双树复数小波和波原子稀疏图像表示的压缩传感图像重构

目前用于压缩传感(CS)图像重构的大部分算法都是基于图像在单一基上的稀疏性来实现。但很多图像在两种或多种基上具有稀疏表示,当用一种基进行稀疏图像表示时,往往不能有效捕捉图像的结构特性,导致图像重构质量不高。为此本文提出了一种基于波原子和双树复数小波混合基的图像稀疏表示方法,利用线性Bregman迭代来进行重构的压缩传感算法。该算法在每一次迭代更新后用梯度下降法进行全变差调整,再分别在两种基上执行软阈值处理来减小图像的l_1范数。实验结果表明本文算法有效提高了重构图像的质量。