强背景噪声环境下的线性调频信号频率估计

估计信号的瞬时频率常用Cohen类时频分布的时间一阶条件矩，但这些方法在更低信噪比下存在较大的估计方差。基于线性调频信号的自相关和互相关可以实现噪声压缩的结论，提出了改进的基于互Wigner-Ville分布线性调频信号的频率估计方法。该方法具有更强的抗噪声能力，在多元信号频率估计时，抑制了交叉项的影响。仿真实验的结果证明了算法的有效性。     

部分样本雷达信号的参数估计

线性调频信号参数检测是SAR对抗的一个重要问题 ,其估计参数的方法有多种 ,但对于分析部分样本LFM信号而言 ,绝大多数估计方法的误差很大。在用快速相关解线调估计和分析部分样本LFM信号存有误差的基础上 ,提出了一种改进的快速相关解线调算法运用于部分样本LFM信号的起始频率和调频斜率参数的估计 ,减小了误差。最后对改进的算法进行了仿真 ,证实了改进后的有效性。     

基于直接判决和导频跟踪的OFDM系统快时变信道估计

提出了一种基于直接判决的OFDM系统的快时变信道估计方法。采用了直接判决算法进行信道估计，并从中选择有效的估计结果，联合导频信号进行信道跟踪。将基于训练序列的信道估计结果作为直接判决算法的初始值，利用传输信号直接判决的统计特性进行了信道估计，并利用改进的导频算法进一步地跟踪信道在时间上的变化。Simulink仿真结果表明，该估计算法适用于时变信道，比基于导频的信道估计方法和基于训练序列的信道估计方法效果都要好。     

基于一种改进型阵列结构联合估计二维方向角

针对一种基于ESPRIT的算法进行二维DOA估计时，在一定条件下会出现角度估计模糊的现象，分析指出了其中方法性能缺陷的理论原因，优化了阵列结构，在此基础上对矩阵束求解方法进行改进，针对类L型阵列结构采用改进算法能有效解决DOA联合估计中的角度配对问题，实现了信号方向参数的自动配对。仿真实验证实本文提出的改进算法的有效性。     

利用小波对信号进行去噪及参数估计

在以往的小波软阈值去噪的基础上对阈值进行了一定的改进，利用小波系数对信号进行了参数的估计，与其它方法相比较，该估计参数的方法能在较低的信噪比，下比较准确的计算出信号的到达时间等重要参数，满足准确性和实时性的要求。     

基于改进GA的周期长码直扩信号PN码估计

针对周期长码直接扩频信号的PN码盲估计问题,提出了一种基于改进GA的扩频码估计方法.首先,通过观测信号来初始化种群,在种群更新时保留最优个体并加入一部分随机产生的个体防止局部最优的方法来改进GA;然后,根据信号协方差矩阵的谱范数来估计PN码与信息码的同步点;最后,截取一周期的观测信号,利用改进的GA来估计PN码.仿真结...     

SCM-OFDM系统中一种新型的盲频偏估计方法

该文提出了一种用于重叠码调制-正交频分复用(SCM-OFDM)系统中基于信号检测的频偏估计方法。这种估计方法首先利用SCM迭代检测器输出的判决符号重构接收信号,再利用最小均方误差准则对频偏进行估计。该方法不需要额外的训练序列或导频符号,是一种高效的盲频偏估计方法。此外,该文还提出了一种改进的频偏估计方法。在改进的方法中,接收机将频偏估计与信号检测进行联合迭代;这样在每一次迭代过程中,信号检测和频偏估计都需要执行一次,并且接收机在信号检测时可以利用上一次得到的频偏估计值进行频偏补偿。仿真结果表明,该文提出的基于信号检测的频偏估计方法能够高效并准确地估计出SCM-OFDM系统的频偏。     

低信噪比情况下调频信号的瞬时频率估计

提出一种基于互Wigner-Ville分布(XWVD)的瞬时频率迭代估计方法.理论分析了该方法的收敛性,通过仿真比较了各种瞬时频率估计方法在噪声下的估计方差,证明此方法在低信噪比情况下对估计线性调频信号的瞬时频率有较好的效果.并采用加窗的方法改进了此算法,仿真结果证明,改进的方法对非线性调频信号的瞬时频率进行了有效估计.     

有色噪声下不相关和相干混合信号的DOA估计

针对有色噪声背景下的不相关和相干混合入射信号，本文提出了一种新的波达角度（Direction Of Arrival,DOA）估计方法 .首先对混合信号协方差矩阵进行分析和处理以消除其中的有色噪声部分.在此基础上，先利用多重信号分类（Multiple Signal Classification,MUSIC）方法或旋转不变信号参数估计（Estimation of Signal Parameters via Rotational Invariance Techniques,ESPRIT）法估计出不相关信号的DOA；然后利用改进的空间差分方法构造出一个新的只含有相干信号的协方差矩阵，且无秩亏损；最后利用MUSIC算法或ESPRIT算法从中估计出相干信号的DOA.和文献报道的方法相比，新方法具有更优的混合信号DOA估计性能，尤其对于相干信号.仿真结果验证了该算法的有效性.     

信噪比时变情况下的自适应时延估计方法

本文介绍了一种自适应时延估计的改进方法，此方法用于在信噪比时变情况下，对空间上分离的两个传感器所接收的带限信号的时延估计。它的优点在于：当信噪比变化或未知时，它能对时延的变化随时做出估计。文中给出了此方法的估计精度和收敛特性。     

基于流形学习的新高光谱图像降维算法

提出了一种新的基于图像块距离的邻域选择方法,并将其应用于流形学习中,得到一类新的高光谱图像非线性降维算法。该类算法利用高光谱图像物理特性,结合图像的光谱信息和空间信息,在最大限度减小图像信息冗余的基础之上,很好地保持了原始数据集的特性。与其它高光谱图像的降维算法相比,改进的流形学习算法不仅考虑到高光谱图像本身的空间关系,而且利用图像块距离更好地保持了数据点之间的局部特性,从而有效地去除原始数据集光谱维和空间维的冗余信息。实际高光谱数据的实验结果表明,所提出的算法在应用于高光谱图像分类时,与其它方法相比具有更高的分类精度。     

High resolution and low distortion digital look-up-table sinusoidal oscillators

We propose two new methods to reduce the spurious harmonic distortion of a digital sinusoidal signal generated using the look-up-table (LUT) method. Spurious harmonic distortion arises when we try to interpolate the value of a sample that is not present in the LUT. The first proposed method depends on a real time evaluation of the missing sample value without the need for any additional LUT length. The second proposed method requires an additional table and some real time processing to calculate the value of the interpolated sample. The two proposed methods are simulated and their performance is compared with that of the existing methods; namely direct LUT, linear interpolation and trigonometric interpolation. The simulation results show that our proposed methods are superior to both direct LUT and linear interpolation methods. Moreover, for practical table lengths, the performance of our proposed methods is of the same order as that of the trigonometric interpolation method.     

Evaluating color vision deficiency daltonization methods using a behavioral visual-search method

Daltonization methods are used to automatically improve color images for color-deficient people. A comparison of different daltonization methods, however, is still left undone. We propose a visual-search method to evaluate daltonization methods by assessing behavioral performances of the attentional mechanism through the analysis of accuracy and response time data. Firstly, we show that the visual-search methodology can indeed be used to evaluate daltonization methods. Secondly, we argue that a combination of natural images and Ishihara images is needed to highlight differences between the daltonization methods. Our results indicate that the investigated daltonization methods can be ranked from highest to lowest as following: Firstly, the method proposed by Kotera; secondly, the method proposed by Fidaner et al.; thirdly, the method proposed by Huang et al.; and lastly the method proposed by Kuhn et al.     

Universal Context Based Decoding with Low-Density Parity-Check Codes

Universal estimation strategies are proposed to improve channel decoding of sequences that contain context based redundancy. The new methods combine techniques from universal compression, such as the Burrows-Wheeler Transform (BWT) and segmentation of piecewise stationary memoryless sources (PSMS's) with recently proposed methods of discrete denoising. Simulation results with systematic low density parity check (LDPC) codes show significant improvements of the proposed methods on standard decoding, even when the actual sequence context model is unknown in advance. The combined methods inherit advantages of each of the separate methods.     

Cooperative Learning Algorithms for Data Fusion Using Novel $L_{1}$ Estimation

Two novel L₁ estimation methods for multisensor data fusion are developed, respectively in the case of known and unknown scaling coefficients. Two discrete-time cooperative learning (CL) algorithms are proposed to implement the two proposed methods. Compared with the high-order statistical method and the entropy estimation method, the two proposed estimation methods can minimize a convex cost function of the linearly fused information. Furthermore, the proposed estimation method can be effectively used in the blind fusion case. Compared with the minimum variance estimation method and linearly constrained least square estimation method, the two proposed estimation methods are suitable for non-Gaussian noise environments. The two proposed CL algorithms are guaranteed to converge globally to the optimal fusion solution under a fixed step length. Unlike existing CL algorithms, the proposed two CL algorithms can solve a more complex L₁ estimation problem and are more suitable for weight learning. Illustrative examples show that the proposed CL algorithms can obtain more accurate solutions than several related algorithms.     

A new class of zero-correlation zone sequences

In this correspondence, two new methods for constructing sets of zero-correlation zone (ZCZ) sequences are proposed. These methods are based on perfect sequences and unitary matrices. Compared with previous methods, the proposed methods can generate sets of nonbinary ZCZ sequences having a longer zero-correlation zone. The sequences obtained by these methods are suitable for approximately synchronized code-division multiple-access (AS-CDMA) systems.     

On Blind Timing Acquisition and Channel Estimation for Wideband Multiuser DS-CDMA Systems

The problems of blind timing acquisition and channel estimation for DS-CDMA signals in multipath fading channels are investigated. Using only the spreading code of the desired user, methods based on QR decompositions are proposed. These methods perform comparably or even better than subspace based methods with an order lower complexity. Furthermore, the methods exhibit significantly more robustness to channel order mismatch. Based on the acquired timing information, channel estimation algorithms are also developed which are competitive with the previously proposed subspace based channel estimation algorithms. In addition, a channel order estimation algorithm is proposed for the scenario where the order is unknown. Performance of the proposed algorithms is evaluated through simulation and comparison to asymptotic Cramér-Rao bounds.     

斜视SAR运动补偿研究

该文建立了斜视SAR的运动误差模型,分析了不同斜视角度和波束宽度情况下运动误差造成的影响,提出了小斜视和大斜视情况下各自的运动补偿方法,分析表明该方法较常规运动补偿方法适用范围更广。仿真和实测数据的处理结果表明了该方法的有效性。     

仿射不变的中心投影

 考虑将区域和轮廓类不变特征提取算法结合:利用中心投影将目标转化为闭曲线,对闭曲线运用轮廓类方法提取不变特征.证明了图像经中心投影所得闭曲线保持仿射变换的结论.介绍了一种利用小波变换进行仿射不变特征提取的算法.对一些汉字、26个大写英文字母等的实验结果表明结论的有效性,与传统方法相比该算法对噪声具有较强的鲁棒性.     

Improved plot fusion method for dynamic programming based track before detect algorithm

The detection of randomly distributed extended multi-targets under various circumstances is of special interest. An improved plot fusion method is proposed in this work. The proposed plot fusion method is based on a multi-contour tracking and region growing algorithm. The proposed method is applied to the DP-TBD algorithm for extended targets in the K-distributed sea clutter. Models and shortcomings of existing methods are presented. The principle of the DP-TBD algorithm for extended targets and implementation of the improved plot fusion method are explained in detail. In the simulation, the methods are applied to almost all situations, close trajectories, crossing trajectories, small targets, weak targets and targets in a high clutter region. Comparison with existing methods demonstrates that the proposed method is practical and superior to the existing methods.