基于增强型模糊神经网络的盲均衡算法研究

提出一种利用增强型模糊神经网络进行盲均衡的新算法.增强型模糊神经网络具有很好的非线性逼近能力和映射能力,符合非线性通信技术处理的特点.给出增强型神经网络的结构和状态方程,提出代价函数,推导出均衡参数的迭代公式.仿真表明,本算法收敛后误码率减减小,收敛效果较好.     

基于分数间隔的四二阶归一化累积量盲均衡算法

针对目前分数间隔盲均衡算法存在的缺陷,提出了基于分数间隔的四二阶归一化累积量盲均衡算法.先对接收信号进行分数间隔采样,然后利用四二阶归一化累积量将盲均衡算法归结为无约束的极值问题,从而简化了算法,加快了收敛速度,降低了误码率.仿真验证了算法的有效性.     

模糊神经网络在多源数据融合中的应用研究

在T-S模糊神经网络数据融合的基础上,改进了标准T-S模糊融合算法中的模糊算子,并利用聚类算法对网络结构中模糊隶属度个数进行选取.通过仿真实验,验证了改进的算法在融合过程中的合理性、稳定性和准确性.以及聚类算法在T-S模糊神经网络数据融合算法中运用的合理性和有效性.     

粒子群优化模糊神经网络在语音识别中的应用

针对模糊神经网络训练采用BP算法比较依赖于网络的初始条件,训练时间较长,容易陷入局部极值的缺点,利用粒子群优化算法(PSO)的全局搜索性能,将PSO用于模糊神经网络的训练过程.由于基本PSO算法存在一定的早熟收敛问题,引入一种自适应动态改变惯性因子的PSO算法,使算法具有较强的全局搜索能力.将此算法训练的模糊神经网络应用于语音识别中,结果表明,与BP算法相比,粒子群优化的模糊神经网络具有较高的收敛速度和识别率.     

Lagrange神经网络在盲多用户检测中的应用

利用Lagrange神经网络的基本原理,在线性约束恒模算法(LCCMA)基础上,通过增加约束条件,提出了一种多约束Lagrange神经网络恒模"盲多用户检测"算法.通过仿真实验表明,算法比传统最陡下降恒模算法(SDCMA)在误码率等方面有所改善.     

基于模糊神经网络建模的RFID室内定位算法

将模糊神经网络FNN应用于基于RFID技术的室内定位系统IPS,提出一种基于模糊神经网络的RFID室内定位算法,算法将参考标签数据作为神经网络的训练样本,建立"标签接收信号强度与标签读写器间距离RSSI-DIST"的映射模型,然后利用最小二乘解确定目标的位置坐标.同时,对比了传统BP神经网络和FNN网络在建模和定位中的性能.在仿真和硬件平台测试中,模糊神经网络都要比BP表现出更优异的性能,表明基于模糊神经网络的算法更适合于IPS系统.     

一种新的双模式盲均衡算法研究

针对恒模算法(CMA)收敛速度较慢、收敛后均方误差较大的缺点,提出一种新的双模式盲均衡算法.在算法初期,利用能快速收敛的归一化恒模算法(NCMA)进行冷启动,在算法收敛后切换到判决引导(DD-LMS)算法,减少误码率.计算机仿真表明,提出的新算法有较快的收敛速度和较低的误码率.     

GIHS-TSFNN并行学习算法及其应用研究

针对传统T-S模糊神经网络的随机初始网络参数导致网络学习速度慢、易陷入局部解以及运算精度低等缺陷,提出了一种应用佳点集的改进和声搜索算法(GIHS)优化T-S模糊神经网络的并行学习算法.首先应用佳点集择优构造更加高质量的初始和声库,然后搜索过程中进行参数动态调整,并且每次迭代产生多个新解,充分利用和声记忆库的信息,以提高算法的全局搜索能力和收敛速度.其次,将GIHS算法与T-S神经网络相结合构建并行学习算法,实现两种算法的并行交互集成,得到了最优参数配置以提高T-S模糊神经网络的泛化能力.最后将该算法应用到农业干旱等级预测中以解决旱情评估问题.仿真实验表明,GIHS算法性能优于基本HS和IHS算法,且与T-S模糊神经网络、HS算法优化的T-S模糊神经网络和IHS算法优化的T-S模糊神经网络相比,具有更高的预测准确度.     

基于GA-BP的模糊神经网络控制器与Elman辨识器的系统设计

提出了一种基于神经网络的模糊控制系统 ,该系统由模糊神经网络控制器和模型辨识网络组成 .文中介绍了模糊神经网络控制器采用遗传算法离线优化与 BP算法在线调整 ,给出了具体控制算法 ,推导了变形 Elmam网络的系统辨识算法 .仿真结果表明了此法的可行性和有效性 .     

基于改进型模糊神经网络的信用卡客户违约预测

本文在传统模糊神经网络基础上,采用灰狼优化算法计算神经网络的初始权值和阈值,提出了一种改进型模糊神经网络算法,并建立了信用卡客户违约预测模型。改进型模糊神经网络具有很好的非线性拟合能力和很好的全局搜索能力,解决了传统模糊神经网络算法收敛速度慢,容易陷入局部最优的问题。最后,通过预测信用卡客户违约问题,与支持向量机算法、传统模糊神经网络算法和卡方自动交互诊断器算法相比较,验证了改进型模糊神经网络算法的准确性、高效性和鲁棒性,平均准确率达到了94.1%。     

基于一维搜索的ICA自适应算法及其在股票分析中的应用

在固定步长的ICA极大似然估计自适应算法的基础上,通过一维搜索引入了步长修正方案,使新算法可在收敛速度和稳定状态时的失调误差这两个性能指标上达到最佳结合点,具有较好的时变系统跟踪能力。仿真结果证实了本文所提出的算法可以有效地提高ICA的自适应性,能够更准确地完成盲源分离。在此基础上将算法用在时变性很强的股票数据上,以验证该算法的有效性和可行性。     

Further Developments in an Iterative Projection and Contraction Method for Linear Programming

A linear programming problem can be translated into an equivalent general linear complementarity problem, which can be solved by an iterative projection and contraction (PC) method [6]. The PC method requires only two matrix-vector multiplications at each iteration and the efficiency in practice usually depends on the sparsity of the constraint-matrix. The prime PC algorithm in [6] is globally convergent; however, no statement can be made about the rate of convergence. Although a variant of the PC algorithm with constant step-size for linear programming [7] has a linear speed of convergence, it converges much slower in practice than the prime method [6]. In this paper, we develop a new step-size rule for the PC algorithm for linear programming such that the resulting algorithm is globally linearly convergent. We present some numerical experiments to indicate that it also works better in practice than the prime algorithm.     

基于模拟退火算法的盲均衡器设计

盲均衡技术是不需要训练序列的自适应均衡技术.该项技术可以消除由于通信信道失真造成的码间干扰.本文的目的是对盲均衡技术作一介绍,包括对广泛应用的最大归一化累积量判据的介绍和讨论,并在此基础上提出了基于模拟退火技术的设计方法,最后,对这种盲均衡器骑行了仿真,并得出了相应的结论.     

A gradient-related algorithm with inexact line searches

In this paper, a new gradient-related algorithm for solving large-scale unconstrained optimization problems is proposed. The new algorithm is a kind of line search method. The basic idea is to choose a combination of the current gradient and some previous search directions as a new search direction and to find a step-size by using various inexact line searches. Using more information at the current iterative step may improve the performance of the algorithm. This motivates us to find some new gradient algorithms which may be more effective than standard conjugate gradient methods. Uniformly gradient-related conception is useful and it can be used to analyze global convergence of the new algorithm. The global convergence and linear convergence rate of the new algorithm are investigated under diverse weak conditions. Numerical experiments show that the new algorithm seems to converge more stably and is superior to other similar methods in many situations.     

Chaotic convergence of the decision-directed blind equalization algorithm

Classically, adaptive equalization algorithms are analyzed in terms of two possible steady state behaviors: convergence to a fixed point and divergence to infinity. This twofold scenario suits well the modus operandi of linear supervised algorithms, but can be rather restrictive when unsupervised methods are considered, as their intrinsic use of higher-order statistics gives rise to nonlinear update expressions. In this work, we show, using different analytical tools belonging to dynamic system theory, that one of the most emblematic and studied unsupervised approaches – the decision-directed algorithm – is potentially capable of presenting behaviors, like convergence to limit-cycles and chaos, that transcend the aforementioned dichotomy. These results also indicate theoretical possibilities concerning step-size selection and initialization.     

A superlinearly and quadratically convergent SQP type feasible method for constrained optimization

A new SQP type feasible method for inequality constrained optimization is presented, it is a combination of a master algorithm and an auxiliary algorithm which is taken only in finite iterations. The directions of the master algorithm are generated by only one quadratic programming, and its step-size is always one, the directions of the auxiliary algorithm are new “secondorder“ feasible descent. Under suitable assumptions, the algorithm is proved to possess global and strong convergence, superlinear and quadratic convergence.     

New higher-order numerical one-step methods based on the Adomian and the modified decomposition methods

We develop new, higher-order numerical one-step methods and apply them to several examples to investigate approximate discrete solutions of nonlinear differential equations. These new algorithms are derived from the Adomian decomposition method (ADM) and the Rach-Adomian-Meyers modified decomposition method (MDM) to present an alternative to such classic schemes as the explicit Runge-Kutta methods for engineering models, which require high accuracy with low computational costs for repetitive simulations in contrast to a one-size-fits-all philosophy. This new approach incorporates the notion of analytic continuation, which extends the region of convergence without resort to other techniques that are also used to accelerate the rate of convergence such as the diagonal Padé approximants or the iterated Shanks transforms. Hence global solutions instead of only local solutions are directly realized albeit in a discretized representation. We observe that one of the difficulties in applying explicit Runge-Kutta one-step methods is that there is no general procedure to generate higher-order numeric methods. It becomes a time-consuming, tedious endeavor to generate higher-order explicit Runge-Kutta formulas, because it is constrained by the traditional Picard formalism as used to represent nonlinear differential equations. The ADM and the MDM rely instead upon Adomian’s representation and the Adomian polynomials to permit a straightforward universal procedure to generate higher-order numeric methods at will such as a 12th-order or 24th-order one-step method, if need be. Another key advantage is that we can easily estimate the maximum step-size prior to computing data sets representing the discretized solution, because we can approximate the radius of convergence from the solution approximants unlike the Runge-Kutta approach with its intrinsic linearization between computed data points. We propose new variable step-size, variable order and variable step-size, variable order algorithms for automatic step-size control to increase the computational efficiency and reduce the computational costs even further for critical engineering models.