一种基于HMM的雷达目标识别方法

根据雷达目标散射点的一般模型和特定条件的简化模型；提出用RELAX方法从高分辨雷达回波提取目标散射点分布的位置信息作为目标HRRR的特征向量；利用HRRR对雷达视角敏感这一特点，用隐马尔可夫过程表征多视角雷达回波序列，获得目标距离-方位两维信息，用若干HMM子过程构成的模式链表征一个飞行目标的飞行姿态变化，从而采用基于隐马尔可夫模型的分类器实现目标类属和方位的自动识别和分类，实测数据的计算机仿真结果表明，这一方法的平均识率为99.80%和81.2%。     

层数对线性神经元感知机性能的影响

用神经元激活函数的线性模型，研究层数对感知器网络性的影响。结论是，层数Ｌ的选取应不大于神经元最多层神经元的个数；权初值的设影响算法步长的选取范围。     

高分辨雷达目标的RELAX与MRELAX参数特征

用RELAX算法和改进的MRELAX算法对实测雷达数据分别进行处理,从高分辨雷达回波距离像中提取目标特征参数并进行比较分析.仿真实验证明,用MRELAX算法对高分辨雷达目标数据进行处理与用RELAX算法处理的结果比较,MRELAX算法得到了更加清晰的目标特征.     

A Target Recognition Approach to Projecting HRR Profiles onto Subspace

Abstract: A array of the azimuthally averaged range-profile vectors and the inter-class and intra-class divergence matrixesare constructed iwth many frames of the high resolution range profiles which result from radar echoes of airplanes. Takingthe methods of whitening transformation and SVD produces a system of subspace vectors for target recognition. Where-upon, a template library for target recognition is built by the projection of a class-mean vector made from the radar dataonto the subspace for recognition. By Euclidean distance, a comparison is made between the above projection and eachtemplate in the library, to decide which class the target belongs to. Finally, simulations with the experimental radar dataarte given to show that the proposed method is robust to variation in azimuth and immune to additive Gaussian noisewhen SNR≥5dB.     

区间值模糊集的分解定理及相似度的计算

针对区间值模糊集相似度公式只是经验公式并没有理论依据问题.建立了区间值模糊集新的截集概念,研究了相似度计算方法.结果表明:所建立的新截集概念是以往截集概念的推广.在新截集理论的基础上,进一步给出了区间值模糊集的分解定理;根据该截集发展出一种推导区间值模糊集相似度的方法,建立了模糊集相似度与区间值模糊集相似度之间的联系,可进一步应用到模糊风险分析研究中.     

基于Bayesian正则化算法的非线性函数拟合

为克服常规BP算法在解决非线性函数拟合时泛化能力不强的问题,本文研究了用贝叶斯正则化算法来提高网络泛化能力的问题,结果表明在相同网络规模或误差条件下,Bayesian正则化算法泛化能力明显优于基本BP算法及其它改进的BP算法,且收敛速度较快,拟合效果好。     

第三代移动通信中的联合时空信号处理（英文）

使用空间和时间联合方法解决第三代移动通信中的信号处理；根据基站的空一时信号模型，考虑了具体算法的结构和复杂性；通过对直扩-码分多路接入信号、多用户、多径瑞利衰落信道下自适应天线阵接收端信号的数值分析表明理论可行性；研究表明当SNR达9dB时，误码率下降100倍左右。     

Real Time Simulation of Large Scale Ocean Surface near the Seashore

A new wave modeling method and a level of detail (LOD) scheme are proposed for ocean surface simulation in this paper.The modeling method describes ocean wave by modifying the sine wave,and gets wave direction at any position of ocean surface under any ocean floor conditions using wave num decomposition.LOD scheme is proposed based on quad-tree block,which simplifies the ocean surface regular mesh and realizes the real-time rendering of large-scale ocean surface.Experimental results show that these methods can get realistic effect and fast rendering speed,which are appropriated to the applications of 3D games and battlefield simulation.