一种基于指数损失函数的多类分类AdaBoost算法及其应用

 提出一种新的多类分类AdaBoost算法——使用多类分类指数损失函数的前向逐步叠加模型FSAMME（forward stagewise additive modeling using a multi class exponential loss function）。该算法是基于原始的两类分类AdaBoost算法归结为使用两类分类指数损失函数的前向逐步叠加模型的统计学观点,将两类分类的前向逐步叠加模型自然扩展到多类分类情况下得到的,并采用多类指数损失函数和前向逐步叠加模型对FSAMME进行了详细的理论证明。该算法大大降低对弱分类器的精度要求,只需每个弱分类器的精度比随机猜测好;算法简单明了,不用把多类问题转化为多个两类问题,而是直接求解多类分类问题,大大减小计算复杂度和计算量。通过对基准数据库的测试分类及航空发动机故障样本的诊断,结果表明:FSAMME算法一方面可达到较高的分类诊断准确率,其准确率明显高于AdaBoost.M1,略高于AdaBoost.MH;另一方面可大大减小计算成本,满足在线快速分类诊断的要求。     

A suboptimal algorithm for modulation classification

In this paper we present a suboptimal algorithm for modulation classification to classify the general M-ary phase-shifted keying (MPSK) signal buried in additive white Gaussian noise (AWGN). We first derive the phase density functions of MPSK signals, then develop the required statistics for modulation classification and demonstrate a classifier for CW, binary phase-shifted keying (BPSK), quadrature phase-shifted keying (QPSK), and 8PSK. The structure of the proposed classifier is flexible and is easy to expand. The performance of classifier is evaluated in terms of the probability of successful classification. An example (BPSK/QPSK case) is provided to demonstrate the capabilities of the proposed classifier. The performance is evaluated through the theoretical approach and the Monte Carlo computer simulations and is compared with that previously published in 1992. It is shown that the performance of the proposed classifier is better. Further improvement in performance can be obtained by increasing the length of observation interval.     

Adaptive boosting for SAR automatic target recognition

The paper proposed a novel automatic target recognition (ATR) system for classification of three types of ground vehicles in the moving and stationary target acquisition and recognition (MSTAR) public release database. First MSTAR image chips are represented as fine and raw feature vectors, where raw features compensate for the target pose estimation error that corrupts fine image features. Then, the chips are classified by using the adaptive boosting (AdaBoost) algorithm with the radial basis function (RBF) network as the base learner. Since the RBF network is a binary classifier, the multiclass problem was decomposed into a set of binary ones through the error-correcting output codes (ECOC) method, specifying a dictionary of code words for the set of three possible classes. AdaBoost combines the classification results of the RBF network for each binary problem into a code word, which is then "decoded" as one of the code words (i.e., ground-vehicle classes) in the specified dictionary. Along with classification, within the AdaBoost framework, we also conduct efficient fusion of the fine and raw image-feature vectors. The results of large-scale experiments demonstrate that our ATR scheme outperforms the state-of-the-art systems reported in the literature     

基于支持向量机的组合分类方法及应用

为了解决采用神经网络、决策树作为弱分类器的AdaBoost组合分类存在的不足,进一步改善组合分类效果,提出采用支持向量机(SVM)作为弱分类器的一种新的组合分类诊断方法——AdaBoost-SVM。该方法没有采用一个固定的SVM的核参数,而是自适应调整SVM中的核参数,从而得到一组有效的SVM弱分类器。通过对基准数据库的测试及航空发动机故障样本的诊断,结果表明,所提AdaBoost-SVM方法较好地解决了现有的Ada-Boost组合分类方法中存在的弱分类器本身参数选取困难问题及训练轮数的合理选取问题,并具有更好的泛化性能,更适合对分散程度较大、聚类性较差的航空发动机故障样本进行分类。     

Hybrid discriminative/class-specific classifiers for narrowband signals

The class-specific (CS) method of signal classification operates by computing low-dimensional feature sets defined for each signal class of interest. By computing separate feature sets tailored to each class, i.e., CS features, the CS method avoids estimating probability distributions in a high-dimension feature space common to all classes. Building a CS classifier amounts to designing feature extraction modules for each class of interest. In this paper we present the design of three CS modules used to form a CS classifier for narrowband signals of finite duration. A general module for narrowband signals based on a narrowband tracker is described. The only assumptions this module makes regarding the time evolution of the signal spectrum are: (1) one or more narrowband lines are present, and (2) the lines wandered either not at all, e.g., CW signal, or with a purpose, e.g., swept FM signal. The other two modules are suited for specific classes of waveforms and assume some a priori knowledge of the signal is available from training data. For in situ training, the tracker-based module can be used to detect as yet unobserved waveforms and classify them into general categories, for example short CW, long CW, fast FM, slow FM, etc. Waveform-specific class-models can then be designed using these waveforms for training. Classification results are presented comparing the performance of a probabilistic conventional classifier with that of a CS classifier built from general modules and a CS classifier built from waveform-specific modules. Results are also presented for hybrid discriminative/generative versions of the classifiers to illustrate the performance gains attainable in using a hybrid over a generative classifier alone.     

神经网络集成在发动机故障诊断中的应用研究

提出一种基于A daBoost的集成神经网络故障诊断方法,利用多层前向神经网络作为故障弱分类器,通过简单地训练若干个单一神经网络并将其预测结果进行合成,实现了对航空发动机多类故障的诊断。针对一个涡轮喷气发动机气路部件的仿真实验表明,这种方法提高了最终故障分类器的泛化能力,便于工程应用。      

基于DSP的三自由度肌电假手实时控制方法

 利用安置在拇长屈肌,指深屈肌和指伸肌上的3个电极所测得的肌电信号,采用所提出的新的模式分类器,可以实现基于DSP的三自由度假手手指运动的实时控制。该分类器采用自回归(AR)参数模型和样本熵的方法构造特征矢量,经过由弹性反向传播(RP)算法构建的3层前馈神经网络的分类,能够成功地分辨出拇指、食指和中指的弯曲与伸展运动,平均识别率可以达到91%以上。实验结果表明,该分类器具有很高的辨识能力,同时由于其较小的计算量,也为嵌入式的多自由度肌电假手控制提供了一种新的控制方法。     

Automatic target classification of slow moving ground targets in clutter

A new approach is proposed which will allow air-to-ground target classification of slow moving vehicles in clutter. A wideband space-time adaptive (STAP) filter architecture, based on subbanding, is developed and coupled with a one dimensional template-based minimum mean squared error (MMSE) classifier. The performance of this STAP/ATC (automatic target classification) algorithm is quantified using an extensive simulation. The level of residual clutter afforded by various filter configurations and the associated incremental improvement in ATC performance is quantified, revealing the potential for realizable hardware and software implementations to achieve acceptable ATC performance.     

多重信号分类算法测向精度的仿真分析

作为空间谱估计理论体系中的标志性算法,多重信号分类算法从1979年提出后就一直是阵列信号处理领域的研究热点之一。针对复杂电磁环境中阵列测向的实际应用问题,从多信源、含噪宽带信号、入射角度、阵列间距这4个因素对多重信号分类算法的测向精度作了仿真分析,得出了一些结论。     

Classification of radar targets using synthetic neural networks

Radar target classification performance of neural networks is evaluated. Time-domain and frequency-domain target features are considered. The sensitivity of the neural network algorithm to changes in network topology and training noise level is examined. The problem of classifying radar targets at unknown aspect angles is considered. The performance of the neural network algorithms is compared with that of decision-theoretic classifiers. Neural networks can be effectively used as radar target classification algorithms with an expected performance within 10 dB (worst case) of the optimum classifier     

基于SVM算法的虚假航迹识别

针对云雨杂波和主被动干扰导致多雷达传感器产生虚假目标航迹的问题,利用支持向量机(SVM)算法的自主学习能力,通过构建基于数据驱动的判别模型进行虚假航迹识别。针对航迹起始得到的目标潜在航迹,利用人工智能数据驱动、自学习的特点,设计了SVM算法。通过对已标记真假的目标航迹样本进行离线学习,形成虚假航迹识别的SVM分类器,实现了基于数据驱动的判别模型代替先验知识规则约束的固定模型,并在工程应用中,利用SVM分类器在线识别虚假航迹,完成实时剔除。通过实测雷达数据实验验证,该算法的目标虚假航迹准确率高达95%以上,完全满足实际的工程应用需求。相比基于阈值或规则进行硬性判断的传统虚假航迹识别方法,所提出的算法不仅提高了准确率,还具有较高的实时性,能够适应复杂多变的杂波环境,在实际应用中具有更强的适应性和实用性。因此,提出的基于SVM算法的虚假航迹识别方法对于密集杂波场景下的虚假航迹剔除问题具有显著的实际应用价值。     

Diverse AdaBoost-SVM分类方法及其在航空发动机故障诊断中的应用

 提出采用考虑到精度/差异权衡的SVM作为弱分类器的一种新的组合分类诊断方法——Diverse AdaBoost-SVM。该方法通过在一组具有适当精度的弱分类器中进一步选择具有较大差异性的弱分类器,对这些具有较大差异性的弱分类器进行组合,从而较好解决AdaBoost算法中存在的精度/差异权衡的难题;同时该方法也较好地解决了现有的AdaBoost方法存在的弱分类器本身参数选取困难问题及训练轮数T的合理选取问题。通过对基准数据库的测试及航空发动机故障样本的诊断,结果表明和其他方法相比,Diverse AdaBoost -SVM方法具有更好的泛化性能,更适合对分散程度较大、聚类性较差的航空发动机故障样本进行分类,也更适合对非对称故障样本集进行分类。     

A multiresolution approach to discrimination in SAR imagery

We develop and test a new algorithm for discriminating man-made objects from natural clutter in synthetic-aperture radar (SAR) imagery. This algorithm exploits characteristic variations in speckle pattern as image resolution is varied from course to fine. We model these variations as an autoregression in scale, and then use the autoregressive model to define a multiresolution log-likelihood ratio discriminant. We incorporate this discriminant into the existing Lincoln Laboratory SAR system for automatic target recognition (ATR), and test the augmented system by applying it to millimeter-wave SAR imagery having 0.3 m resolution and representing 56 square kilometers of terrain. At a probability of detection of 0.95, the addition of the multiresolution discriminant reduces the number of natural-clutter false alarms by a factor of six.     

Implementing digital terrain data in knowledge-aided space-time adaptive processing

Many practical problems arise when implementing digital terrain data in airborne knowledge-aided (KA) space-time adaptive processing (STAP). This paper addresses these issues and presents solutions with numerical implementations. In particular, using digital land classification data and digital elevation data, techniques are developed for registering these data with radar return signals, correcting for Doppler and spatial misalignments, adjusting for antenna gain, characterizing clutter patches for secondary data selection, and ensuring independent secondary data samples. These techniques are applied to select secondary data for a single-bin post-Doppler STAP algorithm using multi-channel airborne radar measurement (MCARM) program data. Results with the KA approach are compared with those obtained using the standard sliding window method for choosing secondary data. These results illustrate the benefits of using terrain information, a priori data about the radar, and the importance of statistical independence when selecting secondary data for improving STAP performance     

一种基于证据距离的多分类器差异性度量

 多分类器系统因其能够显著提升分类精度而引发了广泛关注。多分类器系统中各子分类器间的差异性是提升融合分类精度的先决条件。提出了一种基于证据距离的分类器系统差异性度量,同时基于该度量提出一种多分类器系统构造方法。综合了既有差异性度量、所提新差异性度量以及在训练样本集上的分类性能等多个指标,实现了多分类器系统的有效构造。实验结果表明,所提差异性度量及多分类器系统构造方法是合理的,能有效提升融合分类精度。     

一种基于协方差矩阵重构的相干信源DOA估计算法

针对空间相干信源的波达方向估计问题,提出了一种基于协方差矩阵重构的TSVD-ESPRIT算法。它利用包含所有信源信息的特征向量构造Toeplitz协方差矩阵,避免了阵列有效孔径的损失,分辨率高且稳定性好;并且利用ESPRIT算法代替MUSIC算法进行DOA估计,避免了谱峰搜索,大大降低了计算量。数据仿真和分析证明了该算法的正确性和有效性。     

Optimal sensor selection strategy for discrete-time stateestimators

A novel sensor selection strategy is introduced, which can be implemented on-line in time-varying discrete-time system. We consider a case in which several measurement subsystem are available, each of which may be used to drive a state estimation algorithm. However, due to practical implementation constraints (such as the ability of the on-board computer to process the acquired data), only one of these subsystems can actually by utilized at a measurement update. An algorithm is needed, by which the optimal measurement subsystem to be used is selected at each sensor selection epoch. The approach described is based on using the square root V-Lambda information filter as the underlying state estimation algorithm. This algorithm continuously provides its user with the spectral factors of the estimation error covariance matrix, which are used in this work as the basis for an on-line decision procedure by which the optimal measurement strategy is derived. At each sensor selection epoch, a measurement subsystem is selected, which contributes the largest amount of information along the principal state space direction associated with the largest current estimation error. A numerical example is presented, which demonstrates the performance of the new algorithm. The state estimation problem is solved for a third-order time-varying system equipped with three measurement subsystem, only one of which can be used at a measurement update. It is shown that the optimal measurement strategy algorithm enhances the estimator by substantially reducing the maximal estimation error     

An online ensemble semi-supervised classification framework for air combat target maneuver recognition

Online target maneuver recognition is an important prerequisite for air combat situation recognition and maneuver decision-making. Conventional target maneuver recognition methods adopt mainly supervised learning methods and assume that many sample labels are available. However, in real-world applications, manual sample labeling is often time-consuming and laborious. In addition, airborne sensors collecting target maneuver trajectory information in data streams often cannot process information in real time. To solve these problems, in this paper, an air combat target maneuver recognition model based on an online ensemble semi-supervised classification framework based on online learning, ensemble learning, semi-supervised learning, and Tri-training algorithm, abbreviated as Online Ensemble Semi-supervised Classification Framework (OESCF), is proposed. The framework is divided into four parts: basic classifier offline training stage, online recognition model initialization stage, target maneuver online recognition stage, and online model update stage. Firstly, based on the improved Tri-training algorithm and the fusion decision filtering strategy combined with disagreement, basic classifiers are trained offline by making full use of labeled and unlabeled sample data. Secondly, the dynamic density clustering algorithm of the target maneuver is performed, statistical information of each cluster is calculated, and a set of micro-clusters is obtained to initialize the online recognition model. Thirdly, the ensemble K-Nearest Neighbor (KNN)-based learning method is used to recognize the incoming target maneuver trajectory instances. Finally, to further improve the accuracy and adaptability of the model under the condition of high dynamic air combat, the parameters of the model are updated online using error-driven representation learning, exponential decay function and basic classifier obtained in the offline training stage. The experimental results on several University of California Irvine (UCI) datasets and real air combat target maneuver trajectory data validate the effectiveness of the proposed method in comparison with other semi-supervised models and supervised models, and the results show that the proposed model achieves higher classification accuracy.     

基于小波包分析与多核学习的滚动轴承故障诊断

为了更准确地诊断滚动轴承故障,提出了一种基于小波包分析与多核学习的滚动轴承故障诊断方法.该方法首先对振动信号进行3层小波包分解,将振动信号分解为不同频带的信号,提取各频带的相对能量特征,构建特征向量;然后采用多核学习算法从训练样本集中学习核函数与分类器;最后使用训练出的分类器识别滚动轴承故障类型.为了验证方法的有效性,进行了滚动轴承故障诊断实验,实验结果表明该方法的故障诊断准确率达到98.25%,与传统的基于小波包与支持向量机的滚动轴承故障诊断方法相比,其故障诊断准确率更高,同时由于避免了核函数的选择问题,该方法更便于实际应用.      

Few-shot electromagnetic signal classification: A data union augmentation method

Deep learning has been fully verified and accepted in the field of electromagnetic signal classification. However, in many specific scenarios, such as radio resource management for aircraft communications, labeled data are difficult to obtain, which makes the best deep learning methods at present seem almost powerless, because these methods need a large amount of labeled data for training. When the training dataset is small, it is highly possible to fall into overfitting, which causes performance degradation of the deep neural network. For few-shot electromagnetic signal classification, data augmentation is one of the most intuitive countermeasures. In this work, a generative adversarial network based on the data augmentation method is proposed to achieve better classification performance for electromagnetic signals. Based on the similarity principle, a screening mechanism is established to obtain high-quality generated signals. Then, a data union augmentation algorithm is designed by introducing spatiotemporally flipped shapes of the signal. To verify the effectiveness of the proposed data augmentation algorithm, experiments are conducted on the RADIOML 2016.04C dataset and real-world ACARS dataset. The experimental results show that the proposed method significantly improves the performance of few-shot electromagnetic signal classification.