带有距离走动和多普勒扩散的高速运动目标检测

针对运动目标在长时间积累过程中的距离走动和多普勒扩散问题,提出了一种基于多种方法相结合的高速多目标检测和参数估计方法.该方法利用楔形变换校正目标包络的距离走动,运用循环自相关函数法和"解线频调"法联合估计并补偿目标的二次相位,最后通过相参积累聚集能量来检测目标.所提方法有效地补偿了目标的距离走动和时变多普勒频率,延长了信号相干积累时间并提高了检测性能.该算法运算量小,适合于硬件实现.对算法的性能做了分析并进行了仿真,仿真结果验证了算法的有效性.     

ISAR成像中散射点越分辨单元走动校正算法

从目标回波的精确模型入手,探讨了在逆合成孔径雷达成像中散射点越分辨单元走动校正算法与常规距离多普勒算法、极坐标插值算法在处理散射点越分辨单元走动方面的差异,并从最大成像范围及算法对模型的依赖等方面对这些算法进行比较．结果表明,散射点越分辨单元走动校正算法适宜于逆合成孔径雷达成像中散射点走动的校正．     

基于Keystone-FRFT的机动弱目标检测算法

针对机动弱目标检测中出现的距离走动和多普勒频率走动问题,在分析回波信号走动的基础上,提出一种基于楔形变换和分数阶傅里叶变换的弱目标检测算法。算法使用楔形变换补偿距离走动,利用分数阶傅里叶变换补偿多普勒频率走动,在低信噪比情况下能有效地检测机动弱目标。仿真分析表明算法有效。     

距离保护在小电流接地系统中的应用新判据

在小电流接地系统当发生异地两点接地时,目前的电流保护无法实现100%有选择性动作,若在此中情况下采用距离保护,在并列运行线路中通过采用自适应的逻辑判别方法实现先切除不重要用户及只切除一条线路;串联线路中通过在电源线、供电线(串联线路)上,用相间距离Ⅰ段动作后辅以接地距离元件的测量判别,或上、下线路间进行数据通信的方式可以实现只切除供电线路,从面实现100%有选择性动作。     

一种低信噪比信号的调制盲识别方法

提出了一种利用包络高阶特征J进行调制方式识别的方法。该方法具有计算简单、无需知道信号的先验信息,同时具有较好的噪声抑制等特点,能在低信噪比情况下快速、有效地进行盲调制识别,易于实时应用和工程的实现。其仿真结果表明:该方法在0 dB信噪比条件下能够识别信号的调制方式,性能优于传统的包络调制识别方法。     

一种新的图像距离在人脸识别中的应用

提出了一种新的图像距离-- 图像匹配距离(Image Matching Distance,IMMD)．IMMD考虑图像中每个像素与对应图像特定区域的关系,在特定区域寻找与该像素匹配的点,从而将图像的灰度值及其坐标位置引入到图像的相似性度量中．这样使得IMMD对人脸姿态、表情、角度变化具有较好的鲁棒性．用基于图像匹配距离的最近邻分类器进行人脸识别．实验结果表明,基于IMMD的方法优于基于传统欧氏距离和图像欧氏距离的同类型方法．     

运用BP神经网络的FSK信号智能检测

针对强噪声背景下FSK信号的智能检测，提出了一种新的方法．运用接收端FSK信号与发射信号的互相关信息，作为特征量，利用BP神经网络实现模式分类，达到智能检测FSK信号的目的．仿真结果表明，该方法能够有效地抑制噪声，实现FSK信号的智能检测．     

动态目标图象的信号检测方法研究

在信号处理中，用普通相关方法对信噪比信号的检测经常得到不很好的结果，本人出了用双相关图象去除杂光背景干扰的算法，该算法与普通相关检测方法上比有１０ｄＢ左右的信噪比增益。     

一种不同距离的雷达检测概率计算模型

基于雷达探测距离与信噪比，信噪比与检测概率、虚警概率之间的关系，推导并提出了一种不同距离的雷达检测概率计算模型，并对该模型的适用范围进行了充分的讨论．通过与文献[1]和[2]中提出模型的比较，验证了该模型的正确性．     

BP神经网络在小信号有效值非线性误差校正中的应用

系统传输的非线性是产生测量误差的主要原因之一。为保证测量精度,应对非线性误差进行校正。描述了BP神经网络的基本模型及算法,阐述了神经网络校正非线性误差的基本原理,提出了用BP神经网络来校正交变小信号真有效值非线性误差的方法。最后给出了校正AD637非线性误差的实验数据。结果表明：运用BP神经网络校正非线性误差,提高了交变小信号真有效值的测量精度。     

Research of Velocity Compensation Method Based on Range Profile Feature

In this paper, based on the feature of high resolution one-dimensional range profile, two effective motion compensation methods are presented. Firstly, the processing method of stepped-frequency and the response of target moving to range profile are analyzed. Secondly, the function of range profile entropy and range profile contrast are presented for velocity compensation, and then the theory analysis, math model, and solution method are analyzed in detail. Finally, several simulation experiments are designed to prove the accuracy and effectiveness of these two methods. From the final theory analysis and simulation experiments, the conclusion can be drawn that these two methods are effective, which can get much higher velocity estimation accuracy, well real-time and easy to be used in project application. After motion compens-ation, the high resolution one-dimensional range profile will be much better than that used to be, and is used for the detection, recognition and ranging of moving targets.     

毫米波PSF雷达运动目标距离成像的补偿

利用毫米波脉冲阶梯跳频雷达波形的特点，解决了在运动目标高分辨距离像中存在的距离－速度模糊问题，实现了运动目标距离成像的补偿，计算机仿真实验验证了算法的有效性。     

Channel Estimation and Compensation for Underwater Acoustic Pipeline Communication

An estimation and compensation algorithm for underwater acoustic pipeline channel is investigated. A joint time-frequency adaptive signal-to-noise ratio (SNR) estimation based on the maximum likelihood method is introduced firstly, and the Cramer-Rao lower bound (CRLB) is proposed so as to evaluate the performance of the SNR estimation algorithm. For frequency-selective fading channel part, estimation and compensation are made to improve the robustness of the system on the basis of the LMS algorithm. Furthermore,real-time update iteration algorithm in the frequency domain is investigated to realize synchronous receiving and estimation. For verification, simulations and actual data tests were made, and the results show that the algorithm possesses great robustness, efficiency and accuracy inrealization of SNR estimation, signal detection and frequency impulse compensation for the channel.     

小波阈值降噪在岸桥小车轨道振动信号处理中的应用

针对测量时振动信号易受噪声干扰的特点,采用小波降噪法对原始信号进行降噪处理。小波具有“变焦距”、较好的时频局部分析能力等特性,适合非平稳信号振动噪声处理。分别通过小波阈值降噪法对仿真信号并口岸桥小车轨道铰点振动信号进行降噪处理。试验表明,小波阈值降噪法能够较好的消除噪声,能有效提高信噪比、降低均方误差,为进一步分析振动信号奠定了基础。     

改进的格拉布斯准则在信号检测门限估值中的应用

观测数据中野值（信号或干扰等）的存在将会影响信号检测门限估计的准确度和可信度,直接应用格拉布斯准则仅能有效地处理观测数据中野值占有率小于１０％时的情况,本文根据门限估计的需求,对该准则作了必要的改进,使之能够适应野值可能较多时的情况,改进后的准则将野值占有率提高了近１６％,,即达到２７％,此外,在野值的除过程中,野值的位置信息也可同时获得,为门限检测提供了信号所处的位置信息。     

近程雷达目标多分量信号处理

研究近程雷达FM多分量信号条件下目标距离测量 .利用目标多点散射模型、TLS ESPRIT和数理统计方法确定平均距离 .计算单分量信号瞬时频率的方法不适合于多分量信号 .多分量信号瞬时频率可以利用TLS ESPRIT结合数理统计的方法获得 .计算机模拟证明这种方法具有很高的测距精度     

A Novel Long-Time Coherent Integration Method for Moving Target Detection

Long-time coherent integration (LTCI) can remarkably improve the detection ability of radar for moving target. To increase the processing efficiency, this paper proposes a novel LTCI method based on segment time reversing transform (STRT) and chirp z-transform (CZT). In this method, STRT operation is first presented to estimate the Doppler ambiguity factor, and keystone transform (KT) is used to correct the whole range migration (RM). Then, CZT and non-uniform fast Fourier transform (NUFFT) are used to estimate the parameters as well as correct the second and third order Doppler frequency migration (DFM). Compared with the existing methods, the proposed method can achieve RM correction and DFM correction without repetitive operation. The effectiveness of the proposed method is validated by both simulated and real data.     

小波分析在突变信号检测中的应用

本文探讨基于小波变换模最大值沿尺度演变的信号突变检测的基本原理与方法,在不同尺度上分析和处理信号的各种频率成分,使信号的奇点、突变点放大,提高信号的分辨率、信噪比.提出通过二进小波变换检测信号奇异点的实用技术,有效地检测出滚动轴承故障发生的起始点,为在线故障诊断做出了有益的探索.     

几种SAR距离弯曲校正算法比较分析

针对SAR成像中的距离弯曲校正问题，给出了频谱分段搬移和插值校正方法的一般步骤，最后对这些距离弯曲校正方法进行了仿真，并将距离弯曲校正后的成像结果和相应的质量指标进行了比较，得到了相关结论．     

Tracking algorithm of BPSK signal in low bit SNR and high dynamic scenarios

A new tracking algorithm is proposed aiming at the tracking problem in low bit signal-to-noise ratio( i. e.,Eb /N0) scenarios,in which the bit clock regenerated by bit synchronization loop decides loop update moment. The double frequency processing and non-coherent accumulation technologies are applied to eliminate the impact of data polarity inversion,and then long time accumulation improves the input signal-to-noise ratio of discriminator. The frequency locked loop and phase locked loop constitute a carrier loop in parallel,which can meet the high dynamic demands. The effectiveness of this algorithm has been corroborated by theoretical analysis,simulation and measurements,and the new tracking algorithm has been used in an aerospace engineering project successfully.