粒子种群优化（PSO）算法的性能研究

近年来，一种新的基于种群优化的算法——粒子种群优化（PSO）算法，正受到人们的普遍关注。首先介绍了PSO原理及具体实现步骤，接着对各种常见PSO算法，例如原始算法、惯性权值算法、限制因子算法等进行了解释。在此基础上，对PSO算法典型模型的参数选择，如惯性权值、加权系数、最大速度等，进行了详细研究，并给出了实验结果，得出了相关结论．为今后参数的选择提供了参考。接着讨论了PSO在神经网络、模糊逻辑系统和进化计算等计算智能领域及其它工程领域的应用，最后给出了进一步的研究方向。     

基于PSO-BP算法的无线传感器网络定位优化

在研究现有定位算法的基础上,针对基于接收信号强度指示(RSSI)定位模型中的参数易受环境影响等问题,提出了一种新型的粒子群优化(PSO)算法与后向传播(BP)神经网络相结合的算法.BP网络算法权值的修正依赖于非线性梯度值,易形成局部极值,同时学习次数较多,需先通过粒子群算法进行优化.为了提高定位精度,首先采用速度常量法滤波处理,然后通过改进的混合优化算法对BP神经网络初始权值和阈值进行优化,并分析算法的性能.试验中隐层节点个数采用试错法,从12到19变化,以确定合适数目.实验结果表明,与一般加权算法和传统BP算法相比,改进的混合优化算法可大幅改善测距误差对定位误差的影响,同时可使25 m内最小定位误差小于0.27 m.     

遗传二进制多粒子群优化算法及其在子阵STAP中的应用

抑制杂波和干扰的子阵空时自适应处理的性能会受子阵方向图栅瓣的影响而严重下降,非均匀子阵划分和子阵权值设计可有效抑制天线方向图的栅瓣,消除STAP性能曲线上的栅凹口.由于子阵划分和子阵权值分别对应于整数和实数变量,使得该问题为一种特殊的混合优化问题.针对该特殊问题本文提出了一种新的遗传二进制多粒子群优化算法-GBMPSO.该算法不仅可处理二进制编码粒子,而且采取了多粒子群共同进化且在同一种群内部进行交叉和变异的机制,有效避免了算法早熟.针对四种典型测试函数的仿真实验表明,GBMPSO算法与二进制PSO算法和遗传算法相比在收敛速度和算法精度上均有明显的优势.最后,给出了该算法在子阵STAP中应用的实例.     

环境自适应的拓扑联合感知虚拟网映射算法

针对虚拟网映射算法环境适应度低、拓扑关联性较差且映射开销较大的问题,该文提出一种环境自适应的拓扑联合感知虚拟网映射算法。首先提出一种加权相对熵排序方法对具有多指标的节点进行量化处理,依环境变化赋予节点指标不同的权值;在虚拟节点排序阶段采用加权相对熵和广度优先搜索算法双重排序,物理节点排序中引入就近度与加权相对熵算法配合使用,实现了对虚拟拓扑和物理拓扑的联合感知;最后利用k-最短路径算法完成虚拟链路映射。仿真结果表明,该算法依据环境变化自适应调整指标权值,提高了虚拟网映射成功率和收益开销比。     

LAT模型中基于次用户权值优化的频谱感知

全双工认知无线电LAT(Listen-and-Talk)模型因自干扰消除不完全而使得频谱感知能力下降.为提升频谱感知性能,提出了LAT模型下基于次用户加权的协作式频谱感知方法,分别研究了决策融合和数据融合下次用户最优权值的算法.求解最优权值时利用了凸优化方法,在处理非凸问题时利用了序列规划的方法,得到了较好的优化结果.仿真实验表明,在加权融合机制下的LAT模型得到了趋近于传统LBT(Listen-before-Talk)模型的频谱感知能力.     

基于项权值变化的完全加权正负关联规则挖掘

本文提出一种基于项权值变化的完全加权正负关联规则挖掘算法,解决了基于项权值变化的负模式挖掘问题.该算法考虑项权值依赖于事务记录的特点,采用新的项集剪枝方法和模式评价框架,通过项集的项内权值比和维数比的简单计算和比较,挖掘有效的完全加权正负关联规则.实验结果表明,与现有无加权正负关联规则挖掘算法比较,本文算法能避免无效的模式出现,其挖掘时间和候选项集数量明显减少,减幅最大分别可达94.09%和88.16%.     

一种对任意线阵天线的主波束赋形方法

基于自适应天线阵理论,给出了一种用于任意线阵天线的主波束赋形算法.该方法通过迭代获得一组最优权值,用来减小加权赋形方向图和期望方向图在主波束上的差别,同时将旁瓣电平降低到期望值.一些算例充分说明了这种算法的有效性.该算法可以有效的控制主波束方向、宽度、形状和旁瓣电平.     

传感器网络中航迹关联的加权模糊聚类算法

提出一种基于特征加权的模糊聚类算法用于传感器网络中的航迹关联.该方法首先将分布式多传感器获得的航迹进行数据同化,然后通过同步采样将所有航迹映射为高维空间中的一组点集,最后通过特征加权的模糊c均值(FCM)聚类算法实现航迹的分类关联和信息融合.在聚类算法中通过ReliefF算法实现了特征权值的自动确定,自适应地考虑了不同时刻航迹位置对数据关联的不同影响.实验结果表明本文提出的航迹关联算法不仅具有良好的关联效果,而且通过特征的自适应加权提高了信息融合质量.     

基于加权K近邻算法的缺失数据填补研究

针对数据缺失问题,本文在完全随机缺失的前提下,对完整数据集进行不同比例的挖空处理,并使用K近邻算法进行缺失值填补;采用交叉验证法优化K值;最后借用高斯函数,对传统K近邻算法进行加权处理,提出加权K近邻算法.实验结果表明,不论K取值多大,加权K近邻算法填补效果均优于传统K近邻算法;且K=2时,两种算法填补效果达到最佳.     

基于PSO-LBG算法的最优码本设计

矢量量化（VQ）是一种极其重要的信号压缩方法，广泛地应用于图像信号压缩、语音信号压缩领域。它I的主要问题是码本设计，在码本设计过程中，有许多算法被提出。本文提出了PSO和LBG算法相结合的1PSO—LBG算法采设计码本，改善了码本质量，提高了收敛速度。     

Be/spl acute/zier representations for the multiobjective optimization of conformal array amplitude weights

For conformal phased arrays, generally the excitation amplitude of the array elements must be adjusted in order to maintain low sidelobes as the array is scanned. While the desired phase weights for maximum gain are deterministically set by the array geometry and scan angle, the representation of optimum low sidelobe amplitude weights remains an open problem. Following up on prior work using the efficiency-constrained optimization of a modified Bernstein polynomial for low sidelobe conformal array synthesis, a Be/spl acute/zier surface is shown to provide a good representation of the optimized amplitude weights with a reduced number of parameters, while demonstrating /spl epsiv/-constraint multi-objective optimization of conformal aperture efficiency versus sidelobe level. These results are extended to include a Be/spl acute/zier volume representation for the multiobjective optimization of conformal aperture efficiency versus both sidelobe level and scan angle.     

自适应动态Meta粒子群优化算法综合多方向图共形阵列

基于全波仿真得到的广义阵元有源方向图,该文提出一种用于综合多方向图共形阵列的新方法：自适应动态Meta粒子群优化(ADMPSO)算法。在传统Meta粒子群优化(MPSO)算法基础上,定义了优势子群和非优子群的概念,并通过植入非优子群裁减、优势子群规模膨胀以及惯性权重自适应更新等机制,实现了优化过程中多子群的自适应动态调整,全面提高了算法性能。ADMPSO成功用于12元微带锥面共形阵列非赤道面的多方向图综合,综合过程考虑了由共形载体导致的阵元极化指向各异特征,在公共激励存在约束情况下,使阵列同时实现了笔形、平顶,以及余割平方波束总功率方向图,其与该阵列全波数值仿真完全吻合,优化结果和收敛速度相比于其他算法均有显著改善。     

Analysis of a sign algorithm with delayed data

An adaptive filter whose weights are adapted using a sign algorithm with a delayed error signal is analyzed. For stationary environments it is proved that the excess average absolute estimation error is bounded for all values of the error signal delay and the algorithm step size. For the nonstationary case when the optimal filter weights are time varying, the optimum step size which minimizes the excess average absolute error is derived. It is shown that the optimum step size does not depend on the additive noise power. The analytical results are supported by computer simulations     

Generation of short optical pulses in semiconductor lasers by combined dc and microwave current injection

We report picosecond pulse generation in low threshold buried optical guide lasers using combined direct and microwave current excitation. The pulse widths were obtained as a function of direct current for several levels of RF excitation using lasers 125 and 380 μm long. The pulses have a Gaussian shape with full widths at half maximum intensity ranging from 19 to 57 ps. The pulse widths were obtained from the second harmonic autocorrelation. The experimental results are in reasonable agreement with the theory of short pulse formation in sinusoidally modulated lasers. The pulse width decreases with increasing dc and microwave current, and decreasing laser length. The shortest pulses were obtained with a 125μm long laser using 0.25 W RF at 1 GHz and 35 mA de bias. Multiple pulses are emitted at high dc excitation levels. Using the above laser at the indicated current levels the emission consists of a burst of ∼10 pulses which are separated by 30 ps, and has an approximately exponentially decaying amplitude. Saturable absorption was introduced in the lasers by degradation and results in shortening the pulses. A comparison is made of the pulse widths obtained for sinusoidal microwave current modulation and for pulsed excitation where the excitation is obtained from a step recovery diode. Reduced pulse widths are obtained for short current pulse excitation.     

基于Fourier-Mellin变换和相干系数法的重复轨道干涉SAR图像配准新方法

相对于单轨双天线InSAR系统,重复轨道InSAR系统由于轨道不能完全平行,导致InSAR图像对间存在夹角,使其配准更为困难。该文针对非平行重复轨道的InSAR问题,提出了一种结合Fourier-Mellin变换和相干系数法的InSAR图像配准算法,对存在夹角的SAR复图像对进行精确配准。该方法用于粗配准时,通过引入非线性图像幅度变换预处理,克服了Fourier-Mellin变换无法直接处理SAR图像对幅度分布动态范围大且不均匀的问题,而用于精配准时,通过引入相干系数法,精确估计了复图像对之间存在的微小角度。以SIR-C/X重复轨道的X波段SAR数据为例,详细比较了传统相干系数配准算法和该文算法在配准精度和计算时间上的差异,从而验证了算法的有效性。     

Partially decoupled Volterra filters: formulation and LMSadaptation

The adaptation of Volterra filters by one particular method-the method of least mean squares (LMS)-while easily implemented, is complicated by the fact that upper hounds for the values of step sizes employed by a parallel update LMS scheme are difficult to obtain. In this paper, we propose a modification of the Volterra filter in which the filter weights of a given order are optimized independently of those weights of higher order. Using this approach, we then solve the minimum mean square error (MMSE) filtering problem as a series of constrained optimization problems, which produce a partially decoupled normal equation for the Volterra filter. From this normal equation, we are able to develop an adaptation routine that uses the principles of partial decoupling that is similar in form to the Volterra LMS algorithm but with important structural differences that allow a straightforward derivation of bounds on the algorithm's step sizes; these bounds can be shown to depend on the respective diagonal blocks of the Volterra autocorrelation matrix. This produces a reliable set of design guidelines that allow more rapid convergence of the lower order weight sets     

Subcarrier weighting: a method for sidelobe suppression in OFDM systems

In this letter, a method for sidelobe suppression in OFDM systems is proposed and investigated. The proposed method is based on the multiplication of the used subcarriers with subcarrier weights. The subcarrier weights are determined in such a way that the sidelobes of the transmission signal are minimized according to an optimization algorithm which allows several optimization constraints. As a result, sidelobe suppression by subcarrier weighting reduces OFDM sidelobes by more than 10 dB in the average without requiring the transmission of any side information.     

A parallel binary structured LMS algorithm for transversal adaptive filters

The adaptation process in digital filters requires extensive calculation. This computation makes adaptation a slow and time consuming process. Simple, but versatile, parallel algorithms for adaptive filters, suitable for VLSI implementation, are in demand. In this paper a regular and modular parallel algorithm for an adaptive filter is presented. This parallel structure is based on the Gradient Vector Estimation Algorithm, which minimizes the Mean Square Error. In the parallel method, the tap weights of the adaptive filter are updated everys steps, whereas in the recursive algorithms, the tap weights are updated at each step. Fors step update, bit strings of lengths are used to derive the terms with which the tap weights of the adaptive filter are calculated. The algorithm presented computes the tap weights at timen+s as a function of the tap weights at timen, the inputs from timen+1 ton+s−1, and the desired output from timen+1 ton+s−1. The algorithm also can be mapped to a VLSI architecture that is both regular and modular and allows either expansion of the order of the filter or the degree of parallelism obtainable. A comparison between the performance of the sequential LMS algorithm, Fast Exact LMS algorithm, and the parallel binary structured LMS algorithm is presented.     

Coverage path planning for UAVs based on enhanced exact cellular decomposition method

In this paper, an enhanced exact cellular decomposition method to plan the coverage path of UAVs in a polygon area is proposed. To be more specific, the contributions of the paper are: firstly, the turning motion of UAVs is shown to be less efficient from the viewpoints of route length, duration and energy. Secondly, the problem of coverage Path Planning (CPP) in a convex polygon area is transformed to width calculation of the convex polygon, and a novel algorithm to calculate the widths of convex polygons with time complexity of O(n) is developed. The path of the least number of turns for an UAV based on the widths of convex polygons is devised. Thirdly, a convex decomposition algorithm for minimum width sum based on the greedy recursive method which revolves around decomposing the concave area into convex subregions is developed. It is proved that the algorithm is a polynomial time algorithm. To avoid unnecessary back and forth motion, some entirely adjacent subregions are combined. Finally, comparing different weights of two joint-points, a subregion connection algorithm based on minimum traversal of weighted undirected graph is proposed to connect the coverage paths of the subregions. Simulation results show that the proposed method is feasible and effective.     

遗传算法在天线阵自适应算法中的应用

本文采用一种新的基于遗传算法的天线阵列的自适应算法。根据实际的硬件限制,不改变阵列的幅度加权,而通过遗传算法对相位加权作全局优化。为减小对天线方向图的破坏,并希望在干扰源来波方向实时置零,因而采用只优化搜索数字移相器的最后几位。计算机模拟结果证实了这种方法的有效性。