一种基于误差放大的快速BP学习算法

针对目前使用梯度下降原则的BP学习算法，受饱和区域影响容易出现收敛速度趋缓的问题，提出一种新的基于误差放大的快速BP学习算法以消除饱和区域对后期训练的影响．该算法通过对权值修正函数中误差项的自适应放大，使权值的修正过程不会因饱和区域的影响而趋于停滞，从而使BP学习算法能很快地收敛到期望的精度值．对3-parity问题和Soybean分类问题的仿真实验表明，与目前常用的Delta-bar-Delta方法、加入动量项方法、Prime Offset等方法相比，该方法在不增加算法的复杂度和额外的CPU机时的情况下能更快地收敛到目标精度值．     

具有混沌激励函数的BP网络算法

该文讨论了BP网络学习过程中的假饱和现象，并给出了一种改进的算法，有效地解决了假饱和的问题。仿真结果表明，该方法不但可以提高网络学习的快速性，而且具有一定的避免权值落入局部极小点的能力，从而提高了网络的收敛精度，同时，该算法还能提高网络的泛化能力。     

噪声环境下IEEE 802.11 DCF的性能仿真及算法改进

在对噪声环境下IEEE 802．11 DCF性能进行仿真分析的基础上，提出了一种适应于无线噪声环境的DCF改进算法，该算法通过接收端反馈消息来区分因冲突和误码产生的数据包传输失败，并在发送端针对不同的传输失败原因分别采取不同的退避算法。仿真结果表明改进算法的性能优于DCF，饱和吞吐量比DCF最多能够提高103％。     

带有饱和像素的CCD图像拖尾修正算法研究

该文分析了CCD像元饱和对帧转移CCD拖尾修正算法的影响,分析表明残余拖影误差与饱和像元位置的关系。并提出了一种利用CCD的虚拟参考行可计算出残余拖影误差的修正算法,该算法可有效修正带有饱和像素的图像拖影影响。     

输入饱和情况下农用拖拉机的直线导航控制

本文以农用拖拉机为研究对象, 提出了一种基于饱和方法的直线路径跟踪导航算法. 该算法可以为农用拖拉机导航中普遍存在的输入饱和控制问题提供一种有效的解决办法. 首先, 基于直线导航系统运动学模型, 将其线性化得到易于控制设计的线性系统模型. 其次, 设计非饱和控制器, 使得系统在该非饱和控制器作用下全局镇定. 最后, 结合饱和函数和非饱和控制器, 构造出饱和控制器, 并证明在该饱和控制器作用下系统全局渐近稳定. 该控制方案的优点在于: 控制器结构较为简单, 且饱和控制器中从外到内的饱和度并不需要按照递减顺序选取, 从而提高了闭环系统的收敛性能. 仿真结果验证了上述方法的有效性.     

一种基于放大误差信号的自适应BP算法

针对标准BP算法受饱和区域影响而导致收敛速度慢的问题,提出一种基于放大误差信号的自适应BP算法,该算法通过修改激励函数的导数,使权值的修正过程不会因饱和区域而趋于停滞,并分析了改进算法的收敛性。仿真结果表明,改进的BP算法加快了收敛速度,并在一定程度上提高了全局收敛能力。     

BP网络模型中输出神经元过早饱和的机理研究

在数学理论上分析了BP神经网络输出神经元出现过早饱和现象的内在机制，提出了相关定理并加以证明。阐释了动量项对出现该现象起着十分重要作用，分析了其它一些文献提出的改进算法与该文所提出的定理之间的联系。因此，该定理对改进BP算法在理论上能够提供很好的指导。     

数字PID控制中饱和问题的研究

研究了数字PID算法中的饱和效应,提出克服饱和现象的各种算法。还通过仿真,研究了比例饱和、积分饱和、微分饱和三种情况下各种算法的性能。最后指出微分项带滤波的PID算法具有良好的抗饱和能力。     

溢出并行检测低功耗饱和乘加单元的研究与实现

提出了一种溢出并行检测的低功耗饱和乘加单元的新结构，并采用0．35μm CMOS工艺实现了该结构．采用这种结构的MAC单元，不再需要额外地执行特殊的溢出检测指令，减少了取指译码的功耗开销．该结构通过专门的溢出检测单元来并行检测计算溢出的情况，一旦检测到结果溢出，就立即停止最后一级的串行加法运算过程，采取饱和操作，把结果置成正的最大值，保证以后无用的计算不要继续进行，降低了计算功耗．实验结果表明，同传统的MAC单元相比，计算溢出之后不但不用执行专门的饱和溢出指令，提高了MAC单元的运算速度，而且可以节省10％～20％的功耗．     

共轴八旋翼无人飞行器的偏航静态抗饱和补偿控制

设计了一种共轴八旋翼无人飞行器,与四旋翼飞行器相比,其具有更大的驱动能力、更强的带载能力和一定的冗余能力．首先,建立了飞行器的动力学模型．针对共轴八旋翼飞行器偏航运动能力比俯仰、滚转运动能力弱,偏航容易出现执行器饱和现象的问题,从实际工程出发提出了基于线性自抗扰控制器的静态抗饱和补偿器．线性自抗扰算法易于工程调节,能够实时估计与补偿外界扰动．静态抗饱和补偿器不增加系统阶次,有效抑制偏航执行器饱和．利用李亚普诺夫稳定理论证明了基于线性自抗扰控制器的静态抗饱和偏航控制系统的稳定性．最后,通过共轴八旋翼飞行器的仿真实验与原型机比较实验验证了算法的有效性与鲁棒性．原型机实验结果表明：在室内固定干扰下,执行器退出饱和的最长时间约为4 s,偏航角误差收敛到±0.085 rad;在室外变干扰下,执行器退出饱和的最长时间约为9 s,偏航角误差收敛到±0.127 rad．基于线性自抗扰控制器的静态抗饱和补偿器在外界干扰情况下能够有效地抑制执行器饱和,具有良好的偏航控制性能与强鲁棒性．     

Distributed model predictive control for polytopic uncertain systems subject to actuator saturation

In this paper, we present a distributed model predictive control (MPC) algorithm for polytopic uncertain systems subject to actuator saturation. The global system is decomposed into several subsystems. A set invariance condition for polytopic uncertain system with input saturation is identified and a min–max distributed MPC strategy is proposed. The distributed MPC controller is designed by solving a linear matrix inequalities (LMIs) optimization problem. An iterative algorithm is developed for making coordination among subsystems. Case studies are carried out to illustrate the effectiveness of the proposed algorithm.     

Stability analysis for delta operator systems subject to state saturation

In this paper, we investigate the problem of stability analysis for linear delta operator systems subject to state saturation. Both full state saturation and partial state saturation are investigated for the delta operator systems. Two equivalent necessary and sufficient conditions are identified such that the system with full state saturation is globally asymptotically stable. Based on the sufficient conditions, an iterative algorithm is proposed for testing global asymptotic stability of the system with full state saturation. A new globally asymptotically stable condition is also proposed for the partial state saturation system. Two numerical examples on a ball and beam model are given to show the effectiveness of the proposed method.     

Inertial motion tracking using sensor saturation compensation with <Emphasis Type="Italic">l</Emphasis><Subscript>1</Subscript> norm regularization

A smoother is proposed for inertial sensor-based motion tracking, where the sensor saturation is compensated. The sensor saturation estimation term is added using l ₁ norm regularization techniques to the standard smoothing problem. The proposed algorithm is in the form of a quadratic optimization problem. Three slightly different methods are proposed, where the geometric models on sensor saturation are slightly different. Through simulation and experiments, it is shown that the proposed method effectively compensates the sensor saturation.     

Set-membership filtering for systems with sensor saturation

This paper addresses the set-membership filtering problem for a class of discrete time-varying systems with sensor saturation in the presence of unknown-but-bounded process and measurement noises. A sufficient condition for the existence of set-membership filter is derived. A convex optimisation method is proposed to determine a state estimation ellipsoid that is a set of states compatible with sensor saturation and unknown-but-bounded process and measurement noises. A recursive algorithm is developed for computing the ellipsoid that guarantees to contain the true state by solving a time-varying linear matrix inequality. Simulation results are provided to demonstrate the effectiveness of the proposed method.     

Nonlinear Thf‐Fxlms Algorithm For Active Noise Control With Loudspeaker Nonlinearity

Adaptive algorithms are prevalently applied in the design of nonlinear active noise control (ANC) systems. The most important nonlinearity in ANC is the saturation effect produced by the electro‐acoustical sensors and transducers. The dominant saturation nonlinearity is in the transducers, which can be represented by a Wiener model. An effective solution to mitigate such nonlinear distortion is to employ the Nonlinear Filtered‐X Least Mean Square (NLFXLMS) algorithm. The controller compensates the nonlinearity using a model of the saturation effect represented by the scaled error function (SEF). However, the NLFXLMS is limited by two practical issues such that the degree of nonlinearity has to be known in advance and the SEF cannot be evaluated in real time. In this work, the NLFXLMS algorithm is modified by incorporating the tangential hyperbolic function (THF) to model the saturation effect of the loudspeaker. The proposed THF‐NLFXLMS algorithm, models the nonlinear secondary path and applies the estimated degree of nonlinearity in the control algorithm design. The results show that the Wiener secondary path, with saturation nonlinearity represented by SEF, can be modelled by THF with a certain degree of accuracy and can yield a reasonable estimate of the degree of nonlinearity. The performance of the proposed algorithm is comparable with the benchmark NLFXLMS and is superior to the conventional FXLMS as well as the second order Volterra algorithm of similar computational complexity with the proposed algorithm.     

Particle swarm optimization based nonlinear active noise control under saturation nonlinearity

In this paper, a method is proposed to overcome the saturation non-linearity linked to the microphones and loudspeakers of active noise control (ANC) system. The reference microphone gets saturated when the acoustic noise at the source increases beyond the dynamic limits of the microphone. When the controller tries to drive the loudspeaker system beyond its dynamic limits, the saturation nonlinearity is also introduced into the system. The secondary path which is generally estimated with a low level auxiliary noise by a linear transfer function does not model such saturation nonlinearity. Therefore, the filtered-x least mean square (FXLMS) algorithm fails to perform when the noise level is increased. For alleviating the saturation nonlinearity effect a nonlinear functional expansion based ANC algorithm is proposed where the particle swarm optimization (PSO) algorithm is suitably applied to tune the parameters of a filter bank based functional link artificial neural network (FLANN) structure, named as PSO based nonlinear structure (PSO-NLS) algorithm. The proposed algorithm does not require any computation of secondary path estimate filtering unlike other conventional gradient based algorithms and hence has got computational advantage. The computer simulation experiments show its superior performance compared to the FXLMS, filtered-s LMS and genetic algorithms under saturation present at both at secondary and reference paths. The paper also includes a sensitivity analysis to study the effect of different parameters on ANC performance.     

基于放大误差信号的BP算法

提出了一种基于放大误差信号的改进BP算法,构造了一个新的函数修改激励函数的导数,通过放大误差信号,加大网络权值调整力度,避免BP算法易于陷入饱和区域来加快收敛速度。仿真结果表明,该算法在收敛速度等方面大大优于其他一些算法。     

基于多小波变换和分块SVD的彩色图像水印算法

提出了一种基于多小波变换和分块SVD的非盲水印算法。选择彩色图像的饱和度分量作为水印的嵌入域,水印为有意义的灰度图像,水印经Arnold置乱后被嵌入到饱和度分量多小波变换的不同中频区域中。实验证明,该算法不仅对噪声、JPEG压缩、剪切及Photoshop处理具有很好的鲁棒性,而且能抵抗旋转、放缩和平移等几何攻击。