Test Problem Generator by Neural Network for Algorithms that Try Solving Nonlinear Programming Problems Globally

A test problem generator, by means of neural networks nonlinear function approximation capability, is given in this paper which provides test problems, with many predetermined local minima and a global minimum, to evaluate nonlinear programming algorithms that are designed to solve the problem globally.     

Evaluation of fuzzy regression models by fuzzy neural network

In this paper, a novel hybrid method based on fuzzy neural network for approximate fuzzy coefficients (parameters) of fuzzy linear and nonlinear regression models with fuzzy output and crisp inputs, is presented. Here a neural network is considered as a part of a large field called neural computing or soft computing. Moreover, in order to find the approximate parameters, a simple algorithm from the cost function of the fuzzy neural network is proposed. Finally, we illustrate our approach by some numerical examples.     

射频功率放大器的线性化技术

本文阐述了射频功率放大器非线性产生的原因 ,介绍了几种射频功放的线性化技术 ,以便于射频功放设计者参考     

Design and analysis of a flat gain and linear low noise amplifier using modified current reused structure with feedforward structure

In this paper, a Low Noise Amplifier (LNA) with the current reused topology is proposed for wideband applications. To increase input impedance matching common source with inductive degeneration and RC shunt feedback structure is used. To extend the bandwidth, inductive series peaking technique is utilized. In the next stage, two parallel structure is hired to have a high voltage gain with low power consumption in addition to improve linearity. Also, by using the self-forward-body-bias (SFBB) technique, supply voltage is reduced and as a result power consumption is decreased further. The proposed LNA exhibits the high and flat gain of 14.7–15.4 ​dB, input return loss of less than −11 ​dB and noise figure range of 2.3–4.4 ​dB from 1 ​GHz up to 8 ​GHz. It consumes 5.4 ​mW from a 1.2 ​V power supply. The achieved IIP3 range for the proposed LNA is 0 ​dBm up to +2.7 ​dBm. The proposed LNA occupies 0.45 ​mm² in 0.18-μm CMOS technology.     

Anti-swing control of a hydraulic loader crane with a hanging load

In this paper, anti-swing control for a hydraulic loader crane is presented. The difference between hydraulic and electric cranes are discussed to show the challenges associated with hydraulic actuation. The hanging load dynamics and relevant kinematics of the crane are derived to create the 2-DOF anti-swing controller. The anti-swing controller is added to the electro-hydraulic motion controller via feedforward. A dynamic simulation model of the crane is made, and the control system is evaluated in simulations with a path controller in actuator space. Simulation results show significant reduction in the load swing angle during motion. Experiments are carried out to verify the performance of the anti-swing controller, showing good suppression of the payload angle in practice.     

微波前馈功率放大技术

技术的发展带动了对高线性、高功率微波放大器的需求。传统的功率回退应用方法存在器件及效率方面的限制。文中分析了前馈技术提高功放线性度的原理,并给出Ｃ波段前馈功率放大器实例。     

一类平衡前馈序列的构造与性质

该文采用m序列选控传统前馈密码系统的方法,在继承了Bent函数谱值均匀优点的同时,使得产生的序列具有理想的平衡性、大的周期和高的线性复杂度,且能抵抗现有的线性逼近攻击和相关攻击。     

数字自适应前馈功放线性化研究

本文研究了一种新的前馈线性化功放的结构,省去了导频信号,在经典的两环前馈法基础上引入第三环,通过第三环直接提取交调信号,同时在线性化功放系统中引入了DSP控制器,用最优化算法进行控制,实现了对线性化功放的自适应控制。此方案被用于cdma2000基站功放的设计,使三载波CDMA信号输出的ACPR低于-51dB。     

混沌与神经网络相结合的预测算法及其应用

探讨了混沌理论与前馈人工神网络相结合进行非线性时间序列预测的基本原理,并对影响短波通信的两个参数;年太阳黑子数和天１０ｃｍ辐射强度,在ＳＶＤ滤波的基础上进行了预测。     

二元Bent序列的前馈生成＊

通过对[1]中二元Bent序列的等价定义,本文给出了使用前馈网络生成二元Bent序列的快速方法,使二元Bent序列的产生比[2]中的方法简单,迅速。