用于最大似然2-DOA估计的裂变粒子群算法

最大似然2-DOA估计是一种抗干扰能力强、精确度高的测向算法。对其目标函数求解时容易收敛到局部最优解,最终影响最大似然2-DOA估计的性能。考虑到粒子群（PSO）算法具有全局化寻优能力及计算量巨大的特点,本文提出了一种裂变粒子群(FPSO)算法,在保证全局化寻优能力的同时大大降低计算量。但当2-DOA估计环境恶劣时FPSO算法仍然需要较大的迭代次数才能达到一定的2-DOA估计精度。为了进一步降低计算量,提出了改进的FPSO算法。其采用AP算法初始化全局最优粒子,精确了初始化精度从而降低精确2-DOA估计所需的迭代次数。仿真证明,FPSO算法和改进的FPSO算法能够分别在AWGN信道和短波信道下以较低的计算量实现精确的2-DOA估计,其均方根误差逼近其CRB。      

基于PSO算法的系统可靠性优化分析

提出了将粒子群优化算法(Patticle Swarm Optimization Algorithm,PSO)用于求解系统可靠性优化问题,建立了系统的可靠性模型,分别采用遗传算法(Genetic algonthm,GA)和PSO算法进行了优化仿真,结果表明采用PSO算法和GA算法都能实现系统可靠性优化,但是相比之下PSO算法的计算精度和求出最优解的概率更高,需较少的迭代次数,能更稳定的求解最优解,而且没有求解早熟的弱点,因此PSO算法更适合于系统可靠性优化。     

混合粒子对优化算法在说话人识别中的应用

在粒子群优化(Particle Swarm Optimization, PSO)和混合蛙跳算法(Shuffled Frog-Leaping Algorithm, SFLA)的基础上,该文提出了一种新的混合粒子对优化(Shuffled Particle-Pair Optimizer, SPPO)算法,应用于矢量量化的说话人识别。该算法将全局信息交换和局部深度搜索相结合寻求最佳的说话人码本。群体按适应值分为3个粒子对,每个粒子对由两个粒子构成,按先后顺序执行PSO算法中的速度位置更新和LBG算法以实现局部细致搜索,间隔一定的迭代次数通过SFLA混合策略实现粒子对间的信息交换,从而使群体向全局最优解靠近。实验结果表明,本算法始终稳定地取得显著优于LBG,FCM,FRLVQ-FVQ和PSO算法的说话人识别性能,较好地解决了初始码本影响的识别性能的问题,且在计算时间和收敛速度方面有相当的优势。     

基于函数复杂度的自适应模拟退火和禁忌搜索新算法

在求解多峰复杂函数的过程中,传统的模拟退火算法和禁忌搜索算法经常出现算法快速收敛于局部最优解、后期收敛速度变慢和搜索能力变差等问题.为解决这些问题,本文给出函数复杂度的定义,并提出基于函数复杂度的自适应模拟退火和禁忌搜索算法.该算法首先根据函数复杂度自适应调整步长控制参数,然后根据调整后步长求得函数的粗糙解,在此基础上再使用初始步长求得全局最优解.实验表明,该算法不仅可以跳出局部最优解的限制,并且减少了迭代次数,有效地提高了全局和局部搜索能力.     

基于混合优化算法的纳米薄膜参数表征

为了在椭圆偏振测量过程中得到精确的纳米薄膜参数,提出了一种求解纳米薄膜参数的混合优化算法。结合人工神经网络算法反向传播和粒子群算法快速寻优的特点,建立了改进粒子群-神经网络(Improved Particle Swarm Optimization-Neural Network,IPSO-NN)混合优化算法。该算法在较少的迭代次数下具有快速跳出局部最优解的能力,从而快速寻找椭偏方程最优解。文中使用该算法对标称值为(26.7±0.4)nm的硅上二氧化硅纳米薄膜厚度标准样片进行薄膜参数计算。结果表明:采用IPSO-NN混合优化算法计算薄膜厚度时相对误差小于2%,折射率误差小于0.1。同时,文中通过实验对比了传统粒子群算法与IPSO-NN算法,验证了IPSO-NN算法计算薄膜参数时能有效优化迭代次数和寻找最优解的过程,实现快速收敛,提高计算效率。     

基于TOFS的宽带信号波束空间DOA估计

频域子空间正交性测试(TOFS)方法是宽带信号DOA估计的一类重要算法。TOFS无需进行聚焦,避免了角度的预估计,在中等信噪比时性能较优,但该方法信噪比门限较高。文中利用均匀线阵模型,将波束空间方法应用于TOFS算法,在降低分辨信噪比门限的同时减少了TOFS算法的运算复杂度。通过仿真验证了新方法性能优于TOFS算法。     

一种新的基于粒子群密度的聚类算法

提出了一种新的粒子群密度聚类算法和对粒子群的初始化方法。该算法具有传统粒子群算法寻找最优解的特点,同时从密度的角度考虑了数据总体的分布,增强了寻找局部最优解的能力,并通过对粒子群的初始化加快了粒子群的收敛速度,得到了更好的聚类效果。对仿真数据和IRIS真实数据的实验结果证明,该算法聚类效果优于传统粒子群聚类算法和K均值算法。     

均匀搜索粒子群算法

针对基本粒子群优化算法容易陷入局部最优解的问题,本文定义了PSO粒子搜索中心的概念,并对其随机状态下粒子搜索中心在全局最优解与局部最优解之间的概率密度进行了计算,在此基础上提出了粒子搜索中心在两个最优解之间均匀分布的均匀搜索粒子群算法,并通过7个Benchmark函数与基本PSO算法进行了对比实验及算法分析,实验分析结...     

一种新的自聚焦相干宽带DOA算法

传统的基于相干信号空间聚焦的宽带DOA算法具有良好的空间分辨能力、较低的信噪比门限,但必须通过角度预估计来构造相应的聚焦矩阵.本文提出了一种不需要角度预估的相干宽带DOA算法,直接利用阵列接收数据构造聚焦变换矩阵,避免了预估计角度对聚焦变换的影响,同时降低了算法的运算量.计算机仿真表明该算法是切实可行的.     

基于混合粒子群优化算法的医学图像配准

提出了一种基于粒子群算法PSO和Powell的混合优化算法,将PSO算法的全局搜索能力与Powell算法的局部寻优能力有机地结合起来.即在PSO算法每步迭代后对当前的局部最优解增加一步Powell局部寻优算法,进而保证了解的精确性的同时提高了求解的速度.将其应用于配准过程中对目标函数的最优化,取得了比较好的效果.     

Low-Complexity DOA Estimation Based on Constraint Solution Space

The Weighted Subspace Fitting (WSF) algorithm is one of the universal algorithms in Direction-Of-Arrival (DOA) estimation, which is of high accuracy. However, it involves the multi-dimensional nonlinear optimization problem, and the computational complexity is usually high. In this paper, we propose a low-complexity DOA estimation algorithm based on constraint solution space. Firstly, we use ESPRIT algorithm to limit the solution space around the best solution and reduce the computational range. Then, we find the best solution in a smaller solution space constraint by Cramr-Rao Bound (CRB), and seek repeatedly until reaching the global optimal solution of WSF algorithm by using the space of the best solution. By limiting the searching process in smaller solution space, this strategy controls the direction of convergence and reduces computational complexity. The experimental results show that this algorithm needs less iterations when the same DOA accuracy is required, and the computational complexity is apparently reduced.

     

Computationally efficient DOA estimation for coprime linear array: a successive signal subspace fitting algorithm

ABSTRACT

In this paper, direction of arrival (DOA) estimation of multiple signals with coprime array is investigated and signal subspace fitting (SSF) method is linked to the coprime array, which achieves a better DOA estimation performance than the traditional uniform array. While the SSF method requires expensive computational cost in the case of multiple signals due to the multidimensional global angular searching, we propose a successive SSF (S-SSF) algorithm from a computationally efficient perspective. In the proposed algorithm, we employ rotational invariance and coprime property to obtain the initial estimates. Then, via a successive scheme, we transform the traditional multidimensional global angular searching problem into one-dimensional partial angular searching one. Consequently, the computational complexity has been significantly reduced. Specifically, the proposed S-SSF algorithm can obtain almost the same DOA estimation performance as SSF but with remarkably lower complexity. Finally, Cramer-Rao Bound (CRB) is provided and numerical simulations demonstrate the effectiveness of the proposed algorithm.     

恒模特性在DOA中的新应用及其克拉美-罗界研究

研究了恒模特性和DOA估计之间的联系，基于最小二乘恒模和权向量多项式求根提出一种简单有效的LSCM-DOA估计新算法，同时为分析此DOA估计算法的性能，详细推导了恒模信号在任意阵型下的克拉美-罗界。理论分析和实际仿真证明该算法几乎达到其克拉美-罗界，而且性能相对于未应用恒模特性的MUSIC和ESPRIT等算法较优越，尤其在小快拍数和低信噪比情况下取得了良好的效果，并且因为未进行谱峰搜索而将计算复杂度控制在较低范围内。另外，该方法还解决了强信号源未移除时对弱信号源的DOA估计问题。计算机仿真证明了算法理论是正确有效的。     

基于阵列流型盲辨识的MIMO雷达二维DOD和二维DOA联合估计

MIMO（Multiple Input Multiple Output）雷达基于分集增益理念,使其相对于相控阵雷达,在目标探测、参数测量、多目标分辨及干扰识别和抑制等方面具有明显优势.目标角度估计是雷达目标参数测量的核心内容,也是雷达对空域目标进行定位和跟踪的前提.本文基于双L型阵列,提出了一种高精度低复杂度的双基地MIMO雷达二维离开角和二维到达角联合估计的新算法.通过对匹配滤波后的阵列接收数据进行子空间分解,实现了阵列流形矩阵的盲辨识,进而获得目标二维到达角和二维离开角的闭式解.所提算法估得的收发四维角（二维离开角和二维到达角）能够自动配对,与2-D ESPRIT（Two Dimensional Estimating Signal Parameters via Rotational Invariance Techniques）算法相比,运算复杂度约是其三分之一,角估计性能相当.仿真实验证明了所提算法以较低的运算复杂度,实现了对目标收发四维角的高精度联合估计.     

共形阵列天线信源方位与极化状态的联合估计算法

 针对共形天线阵列流形的特点,提出了共形阵列天线信源方位与极化状态的联合估计算法.算法的方位估计不需要信源极化状态的任何信息,估计精度高、分辨力强,在完成信源方位估计的同时,还可以精确的估计出信源的极化状态,不需要参数配对,实现了共形阵列天线信源方位与极化状态的联合估计.算法仅需要二维参数搜索,计算量比已有的联合估计算法小,且适用于任意共形载体,具有广泛的应用范围.对算法参数估计的理论性能进行了分析推导,给出了参数估计的CRB(Cramer-Rao Bound),并通过Monte-Carlo仿真实验验证了理论分析的正确性与算法的有效性.     

Fast algorithms for single frequency estimation

In this paper, some new estimators of the frequency of a single complex sinusoid are presented. The rotate-add-decimate (RAD) method of Crozier is first refined to more closely approach the Cramer-Rao Bound (CRB). An additional modification yields an unbiased estimator (ERAD) that essentially achieves the CRB above a signal-to-noise ratio (SNR) threshold comparable to that of RAD. In addition, this estimator is proven to achieve the CRB for high SNR. The ERAD method requires approximately 2N complex multiply-adds and log/sub 2/N arctangents. A modified ERAD (MERAD) is proposed that matches the SNR threshold and computational complexity of the RAD method (approximately 3N complex multiply-adds and log/sub 2/N arctangents) but achieves the CRB for high SNR.     

A new DOA estimation approach for QPSK Alamouti STBC using ICA technique‐based PSO algorithm

Alamouti space time block code (STBC) has been a revolutionary technology in multiple‐input multiple‐output (MIMO) wireless communication since it provides full transmission diversity. To reduce a multi‐path effect and a consumed power, the dynamic beam‐forming technique should be used to enable antennas focusing on a particular area. The aim of this paper is how to reduce the computational complexities of independent component analysis (ICA) and speed up the algorithm used in estimating the direction of arrival (DOA) angles. First, we derive a simple formula to reduce the number of unknown DOA to be one only. Then, real‐imaginary (Re‐Im) decomposition for MIMO system is used to reduce the computational complexities of ICA algorithm. The novel criteria used in this paper is that the kurtosis measuring for the extracted source will be minimum at one of the unknown values of DOA angles. Finally, particle swarm optimization (PSO) will be used as an effective tool to locate the DOA angle positions that minimize the kurtosis measuring. Performance analysis of the proposed approach with QPSK Alamouti STBC in MIMO channel was implemented using MATLAB. The validated criterions for the new approach were first examined. Then, root‐mean‐square‐error (RMSE) was employed to test the proposed approach at different SNR levels. The main parameters that influence on this approach were evaluated. It was found that superior performance could be obtained at ?DOA > 10⁰ when antenna spacing set to be λ/2 using at least 10³ snapshots. The important point observed during simulations was computational complexity (and latency) of the proposed approach was reduced to the minimum by employing Re‐Im decomposition model and PSO algorithm. Consequently, this approach is very efficient for hardware implementations.     

Efficient 2-D DOA and frequency estimation for L-shaped array via RD-PM

This paper addresses the issue of joint two-dimensional direction of arrival (2-D DOA) and frequency estimation via reduced-dimensional propagator method (RD-PM) with L-shaped array. The proposed algorithm has no need for eigenvalue decomposition of the sample covariance matrix and simplifies three-dimensional global spectral search within the three-dimensional propagator method (3-D PM) to one-dimensional local search, which greatly reduces computational complexity. Furthermore, the proposed algorithm can work under both uniform and non-uniform L-shaped array and can achieve paired 2-D DOA and frequency estimates automatically. In addition, the 2-D DOA and frequency estimation performance for the proposed method is approximate 3-D PM algorithm and parallel factor (PARAFAC) method but exceeds the estimating signal parameters via rotational invariance techniques (ESPRIT) algorithm and improved PM algorithm. The detailed derivation of Cram´er-Rao bound (CRB) is provided and the simulation results demonstrate the effectiveness and superiority of the proposed approach.     

柱面共形阵列高性能低复杂度DOA估计算法

针对柱面共形阵列的波达方向(DOA)估计问题,从信号子空间的角度分析了在阵元遮挡下应用多重信号分类(MUSIC)算法的性能缺陷。在此基础上提出通过偏置常数的方法克服经典MUSIC算法的阵元遮挡问题。进一步提出一种基于数据自适应子阵分割的快速DOA估计算法,该方法先利用稀疏采样的偏置MUSIC算法进行DOA预估,依此确定所需要的子阵及二维搜索区域,确定MUSIC算法的搜索范围,进而得到高精确度的DOA估计。利用子阵分割的方法进行DOA估计,避免了经典MUSIC算法因阵元遮挡导致运算量大、精确度低等问题。仿真结果表明,该方法能大幅度降低运算复杂度,同时提高DOA估计精确度。     

交替投影算法的二维拓展

交替投影(AP)的极大似然算法由于其最接近CR下界的良好似然估计性能和适当的计算量受到人们的重视.本文通过采用直角交叉的两个直线阵实现了AP算法的二维拓展,计算量约与两个一维DOA估计计算量相当.