自适应复信道均衡的一种新的神经网络方法

近年来,神经网络已经广泛地应用到许多信号处理问题中.自适应信道均衡是数字通信系统中的一个重要问题.在本文中我们提出一种基于复数赫布类型算法的自适应信道均衡器.计算机模拟表明,无论在线性还是非线性信道中,所提出的均衡器均表现出良好的性能,这为自适应复信道均衡提供了一种新的方法.     

一种新的复值递归神经网络训练方法及其应用

一种新的基于数字滤波器理论的全互连复值递归神经网络训练方法被提出.每个递归神经元均具有复数ⅡR滤波器结构.通过优化ⅡR滤波器的系数来更新神经网络的权值,而优化过程则采用逐层优化(LBLO)技术和递归最小平方(RLS)方法.该算法的性能通过将其应用于复信道均衡来加以说明.计算机仿真结果表明,该算法具有较快的收敛速度.这为快速训练复值递归神经网络提供了一条新的途径.     

基于自适应神经网络的二维线性相位FIR滤波器优化设计

提出一种基于自适应三角函数基神经网络的二维线性相位FIR滤波器优化设计方法.该方法根据二维线性相位FIR滤波器幅频响应特性,采用三角函数基神经网络优化算法计算滤波器系数,同时在神经网络训练过程引入自适应学习率算法,提高神经网络的学习效率和收敛速度.通过训练神经网络的权值,使二维线性相位FIR滤波器幅频响应与理想幅频响应...     

基于Takagi-Sugeno型自适应模糊神经网络的模拟电路故障诊断

该文提出了一种基于Takagi-Sugeno型自适应模糊神经网络故障诊断方法。首先通过电路仿真获得故障样本,其次利用主成分分析对故障样本进行降维处理,减少自适应模糊神经网络的输入,降低训练时间,然后采用BP算法与最小二乘法相结合的混合学习算法训练自适应模糊神经网络的连接权值和隶属度函数。仿真结果表明,此方法能够快速有效地对模拟电路的故障进行诊断和定位,表现出了很好的应用潜力,在容差模拟电路故障诊断领域具有较好的应用前景。     

CPSN在信道均衡中的应用

信道均衡是用来消除码间串扰、对信道畸变进行补偿,从而在接收端正确的重建发送信号的滤波方法。由于信道均衡可以看作一个模式分类问题,恰恰神经网络具有良好的模式分类特征,因此针对复信道可采用CPSN进行二进制自适应信道均衡。仿真结果表明,该算法对于严重码间串扰及适度非线性畸变的复信道效果良好。     

基于非监督权值预学习加速网络的训练

本文提出一种基于非监督学习的Ｏｊａ规则和有监督学习的ｄｅｌｔａ规则的快速学习算法．用改进后的结构和学习方法初始化前馈神经网络的权值；使该算法性能显著提高．最后本文通过仿真实例验证了该方法．     

一种基于改进BP神经网络的物体识别方法

提出基于自适应学习速率动量梯度下降的BP算法进行物体识别,并以修正的Hu不变矩特征作为BP神经网络的输入,通过训练对网络的权值和阈值进行调整.该算法使BP神经网络在学习速率和稳定性上有了进一步的提高.仿真结果表明该方法对物体的平移、旋转、缩放都具有不变性,从而验证了该方法的有效性.     

基于e-修正神经网络的直接自适应控制设计

针对一般非线性不确定系统设计了一种e-修正神经网络直接自适应控制方法。首先采用虚拟控制量的方法,并将其分解成参考模型输出、线性动态补偿输出与神经网络自适应输出三项;然后针对传统σ-修正神经网络在权值更新时的不足,设计了一种基于e-修正方法的权值自适应更新律,并设计了输出反馈误差观测器用以对神经网络进行训练;最后对基于σ-修正与e-修正两种权值自适应更新律进行仿真对比。仿真结果表明基于e-修正神经网络方法在跟踪误差、不确定性逼近等效果上均优于基于σ-修正神经网络方法。     

基于LVQ网络优化的雷达目标识别算法

目标一维距离像在雷达目标识别领域中具有很高的研究价值,神经网络有很强的自适应能力,被广泛应用于目标识别领域中。通过研究分析,将学习向量量化（Learning Vector Quantization, LVQ）神经网络应用于雷达目标一维距离像识别。针对LVQ神经网络对初始连接权值敏感的问题和如何增强网络的分类识别性能,提出利用粒子群优化（Particle Swarm Optimization, PSO）算法对其进行优化。在此基础上提出了基于PSO-LVQ神经网络的雷达目标一维距离像识别新方法。通过3类飞机实测数据实验,验证了PSO算法优化LVQ神经网络初始连接权值的可行性和PSO-LVQ识别算法的有效性。     

基于小波神经网络的自适应均衡器的研究

提出了基于小波神经网络的通信信道自适应均衡器 ,给出了这种均衡器的结构和训练算法。理论分析和计算机仿真均表明 ,与线性均衡器相比 ,基于小波神经网络的均衡器具有更快的收敛速度 ,是一种前景广阔的均衡器。     

A multilayer complex neural network training algorithm and its application in adaptive equalization

In this paper, the layer-by-layer optimizing algorithm for training multilayer neural network is extended for the case of a multilayer neural network whose inputs, weights, and activation functions are all complex. The updating of the weights of each layer in the network is based on the recursive least squares method. The performance of the proposed algorithm is demonstrated with application in adaptive complex communication channel equalization.     

Complex-Valued Recurrent Neural Network with IIR Neuron Model: Training and Applications

In this paper, based on the digital filter theory and approach, a new algorithm for training a complex-valued recurrent neural network, is proposed. Each recurrent neuron is modeled as an infinite impulse response (IIR) filter. The network weights are updated by optimizing the IIR filter coefficients, and the optimization is based on the layer-by-layer optimizing procedure as well as the recursive least-squares method. The performance of the proposed algorithm is demonstrated with application to a complex communication channel equalization. Our approach provides a new way to perform fast training of complex-valued recurrent neural networks.     

一种适用于稀疏多径信道的自适应均衡算法

针对传统的自适应均衡算法在稀疏多径信道下性能表现不佳的问题,提出了一种基于基追踪降噪的自适应均衡算法。该算法利用稀疏多径信道下均衡器权值的稀疏性,将自适应均衡器的训练过程看作压缩感知理论中稀疏信号对字典的加权求和,并利用重构算法直接对稀疏权值进行求解,解决了迭代参数设置和收敛慢的问题。采用基追踪降噪作为重构算法并选用变量分离近似稀疏重构对该最优化问题进行求解,既提高了权值的重构精度又降低了计算的复杂度。仿真结果表明,所提算法能够以较低的计算量和较少的训练序列达到更优性能,这对提升系统的通信性能具有参考价值。     

一种新的神经网络均衡器:结构、算法与性能

本文根据克服数字通信中码间干扰（ISI）的最佳均衡解一般表达式，提出了一种新的自适应神经网络均衡器结构，然后导出了基于该结构的一种自适应算法和相应的学习规则，最后对提出的自适应神经网络均衡器性能进行了计算机模拟，模拟结果与分析表明：本文提出的神经网络均衡器用于实现最佳信道均衡非常有效，比传统线性均衡器和Gibson等人[1]提出的多层感知均衡器（MLPE）性能更优越，更具实用性．     

A new training strategy for neural network using shuffled frog-leaping algorithm and application to channel equalization

This paper makes use of shuffled frog-leaping algorithm (SFLA) as a training algorithm to train multi-layer artificial neural network (ANN). Next, The SFLA ANNs are used for channel equalization. We, in this paper, also introduce SFLA for channel equalization that is formulated as an optimization problem. In short, this paper introduces a novel strategy for training of ANN and also proposes two novel approaches for channel equalization problem using shuffled frog-leaping algorithm (SFLA). The proposed strategies are tested both in time-invariant and time varying channels and interestingly yield better performance than contemporary approaches as evidenced by simulation results.     

A Supervised Constant Modulus Algorithm for Blind Equalization

Blind equalization is a technique for adaptive equalization of a communication channel without the aid of the usual training sequence. Although the Constant Modulus Algorithm (CMA) is one of the most popular adaptive blind equalization algorithms, it suffers from slow convergence rate. A novel enhanced blind equalization technique based on a supervised CMA (S-CMA) is proposed in this paper. The technique is employed to initialize the coefficients of a linear transversal equalizer (LTE) filter in order to provide a fast startup for blind training. It also presents a computational study and simulation results of this newly proposed algorithm compared to other CMA techniques such as conventional CMA, Normalized CMA (N-CMA) and Modified CMA (M-CMA). The simulation results have demonstrated that the proposed algorithm has considerably better performance than others.     

Functional link neural network cascaded with Chebyshev orthogonal polynomial for nonlinear channel equalization

Nonlinear intersymbol interference (ISI) leads to significant error rate in nonlinear communication and digital storage channel. In this paper, therefore, a novel computationally efficient functional link neural network cascaded with Chebyshev orthogonal polynomial is proposed to combat nonlinear ISI. The equalizer has a simple structure in which the nonlinearity is introduced by functional expansion of the input pattern by trigonometric polynomial and Chebyshev orthogonal polynomial. Due to the input pattern and nonlinear approximation enhancement, the proposed structure can approximate arbitrarily nonlinear decision boundaries. It has been utilized for nonlinear channel equalization. The performance of the proposed adaptive nonlinear equalizer is compared with functional link neural network (FLNN) equalizer, multilayer perceptron (MLP) network and radial basis function (RBF) along with conventional normalized least-mean-square algorithms (NLMS) for different linear and nonlinear channel models. The comparison of convergence rate, bit error rate (BER) and steady state error performance, and computational complexity involved for neural network equalizers is provided.