基于Kullback-Leiber距离的迁移仿射聚类算法

针对迁移聚类问题,该文提出一种新的基于Kullback-Leiber距离的迁移仿射聚类算法(TAP_KL)。该算法从概率角度重新解释AP算法的目标函数,并借助于信息论中最常见的一种距离度量,即Kullback-Leiber距离,测量源域与目标域代表点的相似性。另外,通过详细分析TAP_KL算法与AP算法的目标函数,得出一个重要结论,即可以将源域与目标域的相似性嵌入到目标域数据集相似性矩阵的计算中,从而直接利用AP算法的优化算法优化TAP_KL算法的目标函数,解决基于代表点的迁移聚类问题。最后,通过基于4个数据集的仿真实验,进一步验证了TAP_KL算法在解决迁移聚类问题时的有效性。     

基于直接数据域方法的自适应多波束形成算法

提出一种基于直接数据域最小二乘方法的自适应多波束形成算法,包括前向计算、后向计算和前一后向计算.利用天线阵元输出复电压的单快拍数据构建矩阵方程,采用共轭梯度法求解得到阵列的自适应权值向量,从而在所有期望信号方向形成接收波束,同时在各干扰方向形成深零陷,使信干噪比显著提高.由于只需对单快拍数据进行处理,并且避免了样本协方差矩阵的构造及矩阵求逆运算,故计算复杂度较传统算法低.仿真结果表明,算法具有良好的期望信号恢复和干扰置零性能.因此,可适用于卫星通信和移动通信等领域的实时性处理.     

基于余弦相似性的自适应权重的改进FCM算法

模糊C-均值聚类算法(FCM)是一种经典的聚类算法,主要通过迭代更新隶属度和聚类中心来提高聚类的有效性.FCM算法的性能主要通过类内紧性和类间分离性来评价,但其既依赖于初始聚类中心,也对噪声非常敏感.考虑到每个数据点和每个聚类中心对目标函数的不同重要性,本文提出了一种具有自适应权重的改进FCM聚类算法(Hybrid FCM).主要贡献:将2个具有自适应指数p和q的自适应权向量ψ和φ引入FCM的目标函数,以体现不同数据点和聚类中心的重要性;为提高聚类性能,自适应指数p、q和模糊因子m采用粒子群优化算法(PSO)优化,新提出的聚类评价指标AWCVI作为PSO算法的适应度函数;迭代过程中利用余弦相似性对隶属度函数进行修正,提高算法的鲁棒性.实验表明,本文提出的算法能够有效地提高聚类效果.     

机载雷达的预滤波直接数据域算法

空时自适应处理(STAP)作为一种地杂波环境下机载雷达自适应检测小目标的方法已得到广泛研究,干扰非一致性是当前的一个重要方面。由杂波脊外的强干扰(或目标)引起的分立非一致性是市区和尖峰状杂波环境下STAP性能下降的主要原因。利用级联算法得到的自适应权可用于STAP干扰的分立非一致性问题。该方法由一个直接数据域算法和一个统计自适应算法级联构成。该文提出一种带杂波衰减预滤波器的直接数据域算法来增强算法对分立非一致干扰的抑制,从而提高了STAP对付分立非一致干扰的性能。     

一种全最小二乘算法及其在非线性滤波器中的应用

针对输入输出观测数据均含有噪声的滤波问题,提出了一种稳定的总体最小二乘自适应算法。该算法以系统的增广权向量的瑞利商(RQ)与对增广权向量的最后元素的约束的和作为总损失函数,利用梯度最陡下降原理导出权向量的自适应迭代算法,并将该算法应用于非线性Volterra滤波器。研究了算法的稳定性能,提出的算法不仅有良好的收敛性能,而且在权向量的自适应迭代时不需要标准化处理,使得算法的实施更为简单。仿真实验表明,无论在线性系统或非线性系统,本文算法的收敛性能,鲁棒抗噪性能和稳态收敛精度明显高于其它同类总体最小二乘算法。     

基于AWDO算法的主动悬架LQR控制优化

车辆主动悬架LQR控制研究的关键是性能指标加权系数的选择,针对这个问题,文中提出了一种基于改进风驱动优化算法的LQR控制器加权系数优化方法。利用协方差矩阵自适应评价策略算法(CMAES)确定传统风驱动优化算法(WDO)的固有参数,得到自适应风驱动优化算法(AWDO)。将LQR控制器性能指标的加权系数作为优化目标,在解空间中迭代搜索,寻找目标函数值最小的位置。通过实验仿真,并与粒子群算法(PSO)和遗传算法(GA)比较,该算法的收敛速度、收敛精度和鲁棒性更好,优化后的主动悬架性能比相应被动悬架的性能更优。     

一种利用结构特点实现复数域联合对角化解盲源分离新算法研究及应用

联合对角化方法是求解盲源分离问题的有力工具.但是现存的联合对角化算法大都只能求解实数域盲源分离问题,且对目标矩阵有诸多限制.为了求解更具一般性的复数域盲源分离问题,提出了一种基于结构特点的联合对角化（Structural Traits Based Joint Diagonalization,STBJD）算法,既取消了预白化操作解除了对目标矩阵的正定性限制,又允许目标矩阵组为复值,具有极广的适用性.首先,引入矩阵变换,将待联合对角化的复数域目标矩阵组转化为新的具有鲜明结构特点的实对称目标矩阵组.随后,构建联合对角化最小二乘代价函数,引入交替最小二乘迭代算法求解代价函数,并在优化过程中充分挖掘所涉参量的结构特点加以利用.最终,求得混迭矩阵的估计并据此恢复源信号.仿真实验证明与现存的有代表性的对目标矩阵无特殊限制的复数域联合对角化算法FAJD算法及CVFFDIAG算法相比,STBJD算法具有更高的收敛精度,能有效地解决盲源分离问题.     

核加权RX高光谱图像异常检测算法

提出了一种新的基于混合核函数的加权RX算法,用于高光谱图像异常检测.在将原始高光谱数据非线性映射到高维特征空间以挖掘高光谱图像波段间蕴含的非线性信息后,自适应地赋予特征空间RX算子中采样协方差矩阵各光谱向量相应的权值.权值的大小与光谱向量到质心的距离成反比,从而削减了协方差矩阵中异常数据比重,使加权协方差矩阵更好地表征背景数据分布.最后利用核函数性质将高维特征空间的内积运算转化为低维输入空间的核函数计算,并根据高光谱数据特点线性组合新型光谱核函数和径向基核函数以改善算法性能.为验证算法的有效性,利用真实的高光谱数据进行了仿真实验,结果表明该算法优于特征空间的RX算法,能检测到更多的异常目标.     

相控阵部分自适应波束形成收发算法

为减少大型相控天线阵的优化唯相权值的计算量,本文提出一种新的相控阵部分自适应波束形成的收发算法.它采用自适应部分地选择控制零点的控制阵元(PA),按照最大信干噪比原则,用部分唯相共轭梯度法计算各个阵元的优化唯相权矢量来减轻干扰的影响.模拟结果表明,当干扰数目小于阵元半数时计算性能有很大提高,同时天线远场模式的主瓣损失、波形变化及干扰方向零点深度与全阵自适应相比几乎没有变化.     

一种基于迭代信息熵权的改进LFCM算法

LFCM算法作为一种划分式聚类算法,它根据新旧聚类中心之差是否小于设定阈值决定是否停止迭代,但存在聚类准确率偏低的缺陷.熵是信息论中有效度量不确定性的一项指标.为此,文章提出了一种基于迭代信息熵权的改进LFCM算法,该算法将熵权法与LFCM算法进行了有效的融合.算法的改进思路为:(1)输入样本结合隶属度函数,求得数据矩阵,根据该数据矩阵求取新的信息熵和权重;(2)目标函数采用权重函数,算法通过判断新旧权重的差值是否小于设定阈值作为停止迭代条件.仿真实验结果表明,基于迭代信息熵权的改进LFCM算法可以有效提高聚类准确率及紧凑率.     

A ROBUST PHASE-ONLY DIRECT DATA DOMAIN ALGORITHM BASED ON GENERALIZED RAYLEIGH QUOTIENT OPTIMIZATION USING HYBRID GENETIC ALGORITHM

A robust phase-only Direct Data Domain Least Squares （D3LS） algorithm based on generalized Rayleigh quotient optimization using hybrid Genetic Algorithm （GA） is presented in this letter. The optimization efficiency and computational speed are improved via the hybrid GA composed of standard GA and Nelder-Mead simplex algorithms. First, the objective function, with a form of generalized Rayleigh quotient, is derived via the standard D3LS algorithm. It is then taken as a fitness function and the unknown phases of all adaptive weights are taken as decision variables Then, the nonlinear optimization is performed via the hybrid GA to obtain the optimized solution of phase-only adaptive weights. As a phase-only adaptive algorithm, the proposed algorithm is simpler than conventional algorithms when it comes to hardware implementation. Moreover, it proc- esses only a single snapshot data as opposed to forming sample covariance matrix and operating matrix inversion. Simulation results show that the proposed algorithm has a good signal recovery and interferences nulling performance, which are superior to that of the phase-only D3LS algorithm based on standard GA.     

实数遗传算法的改进及性能研究

提出一种粒子群优化方法(PSO)与实数编码遗传算法(GA)相结合的混合改进遗传算法(HIGAPSO).该方法采用混沌序列产生初始种群、非线性排序选择、多个交叉后代竞争择优和变异尺度自适应变化等改进遗传操作;并通过精英个体保留、粒子群优化及改进遗传算法(IGA)三种策略共同作用产生种群新个体,来克服常规算法中收敛速度慢、早熟及局部收敛等缺陷.通过四个高维典型函数测试结果表明该方法不但显著提高了算法的全局搜索能力,加快了收敛速度;而且也改善了求解的质量及其优化结果的可靠性,是求解优化问题的一种有潜力的算法.     

Power allocation in cognitive radio networks over Rayleigh-fading channels with hybrid intelligent algorithms

In this paper, we introduce a hybrid strategy which combines pattern search (PS) optimization and genetic algorithm (GA) to address the problem of power allocation in cognitive radio networks. Considering the fluctuating interference thresholds in cognitive networks, an approach for promoting the coexistence of licensed users and cognitive users is designed. Secondly, based on the analysis of transmission outage probability, a corresponding objective function with regard to the power allocation over Rayleigh fading channels is obtained. It is a difficult task to obtain this objective function directly by using traditional methods, such as common mathematical deduction or linear programming, due to the nonlinearity and complexity of the underlying optimization problem. Inspired by the concept of intelligent algorithms, we employ the scheme of combining PS optimization and GA method, which are both efficient intelligent algorithms to address this challenge. The advantage of this hybrid strategy is that it can overcome the instability problem of GA as well as the local convergency problem of PS method. Thus, the hybrid intelligent method can attain a global and steady outcome. We improve the performance of power allocation strategy with an acceptable increase in computation overhead. The numerical results are encouraging and show that the proposed approach is worthy of consideration in achieving complicated power optimization. Hence, we achieve steady and rational outcomes by applying the proposed hybrid strategy when traditional method is to be ineffective in addressing the nonlinear objective.     

基于自适应遗传算法的星载SAR天线方向图与模糊综合

该文针对改善星载合成孔径雷达(SAR)的模糊特性,提出了一种自适应遗传算法。该算法同时对模糊和方向图进行优化。首先确定模糊区域,然后以天线方向图的主瓣宽度和副瓣电平(包括星载SAR模糊区域的副瓣电平)为目标函数,应用自适应遗传算法对天线方向图进行综合。为了避免早熟的现象,在该算法中,交叉概率、变异概率和变异范围同时进行了自适应的变化。和非自适应遗传算法相比较,该算法迭代步骤少,收敛速度快。仿真结果表明,模糊度得到了很好的抑制,对星载SAR系统设计具有实际意义。     

一种基于实数编码遗传算法的常模盲均衡

盲均衡可以看作代价函数优化问题。为了改进经典常模算法的性能,研究了利用实数编码遗传算法的常模盲均衡,把均衡器系数向量作为遗传算法的决策变量,采用轮盘赌选择和精英保留策略相结合的混合选择算子、算术交叉算子和非均匀变异方式,经过一系列的遗传操作,搜索到适应度值最高的个体,即均衡器的最优系数。计算机仿真结果证明了算法具有收敛速率快、能够搜索到全局最优解等特点。     

基于迭代二阶锥的唯相位波束形成

唯相位自适应波束形成技术对普通相控阵雷达自适应干扰抑制是非常重要的,在数字阵列雷达中,唯相位技术则可以充分利用各阵元发射模块微波功率从而提高雷达的威力。该文提出了一种新的唯相位波束形成方法,假定初始唯相位权重矢量有个较小的相位扰动,将唯相位模型的目标函数和约束函数在该相位扰动区域内分别用泰勒一阶展开式来近似,则可以将原来的非凸问题转化为凸优化问题,通过二阶锥规划方法(SOCP)求得使当前目标函数最小的扰动矢量,然后更新得到新的权重矢量并代替原来的权重矢量,重复上述迭代过程直到满足停止条件,可以得到满足要求的唯相位权重。计算机仿真结果验证了该方法的正确性和有效性。     

Cognitive radio adaptation using particle swarm optimization

One of the basic capabilities of cognitive radio is to adapt the radio parameters according to the changing environment and user needs. This paper proposes a new adaptation method which uses particle swarm optimization (PSO) to optimize cognitive radio parameters given a set of objectives. The procedure of the proposed method is presented and multicarrier system is used for simulation analysis. Experimental results show that the proposed method performs far better than genetic algorithm (GA)‐based adaptation method in terms of convergence speed, converged fitness values, and stability. The proposed method can also provide the tradeoffs of the objective functions, and the resulting parameter configuration is consistent with the weights of the objective functions. Copyright © 2008 John Wiley & Sons, Ltd.     

Dynamic adaptive discrete particle swarm optimization algorithm based method on low-power mapping in network-on-chip

Compared to 2D NoC, 3D NoC has better integrated density and system performance, which was a reliable method to solve the problem about low-power mapping. On the basis of the traditional particle swarm optimization algo-rithm (PSOA), a dynamic adaptive discrete particle swarm optimization algorithm (DADPSOA) was proposed . Parame-ter in this algorithm was adjusted dynamically based on the degree of early convergence and the charge of individual adap-tive value to approach the optimal solution. At the same time, the reasonable structure of the particles was made aiming at reducing the time complexity of this algorithm. Experimental results show that comparing with the random mapping, genetic algorithm (GA), PSOA and dynamic ant colony algorithm (DACA), DADPSOA can save the execution time, reduce the communication power consumption of mapping results. The power consumption of the task graph is reduced.     

GA and Lyapunov theory-based hybrid adaptive fuzzy controller for non-linear systems

In this present article, a new hybrid methodology for designing stable adaptive fuzzy logic controllers (AFLCs) for a class of non-linear system is proposed. The proposed design strategy exploits the features of genetic algorithm (GA)-based stochastic evolutionary global search technique and Lyapunov theory-based local adaptation scheme. The objective is to develop a methodology for designing AFLCs with optimised free parameters and guaranteed closed-loop stability. Simultaneously, the proposed method introduces automation in the design process. The stand-alone Lyapunov theory-based design, GA-based design and proposed hybrid GA–Lyapunov design methodologies are implemented for two benchmark non-linear plants in simulation case studies with different reference signals and one experimental case study. The results demonstrate that the hybrid design methodology outperforms the other control strategies on the whole.     

基于说话人识别的GMM/GA算法

在实时平台上,高斯混合模型(GMM)具有计算有效性和易于实现的优点。最大似然规则中,模型参数不断更新,但由于爬山特征,任意的原始模型参数估计通常将导致局部最优;遗传算法(GA)适于求解复杂组合优化问题及非线性函数优化。提出了基于说话人识别的可以解决GMM局部最优问题的GMM/GA新算法,实验结果表明,提出的GMM/GA新算法比纯粹的GMM算法能获得更优的效果。