模糊双向联想记忆网络的有效学习算法

基于模糊取大运算和爱因斯坦S-模提出新的模糊双向联想记忆网络模型（Max—SesFBAM），并为该网络提出了一种新的学习算法。在理论上严格证明了，任意给定的模式对集，只要存在有连接权矩阵对使其为Max—SesFBAM的平衡态集，则依该学习算法所确定的连接权矩阵对（W^-，U^-）是所有这样的连接权矩阵对中的最大者；且该最大连接权矩阵对能使Max—SesFBAM对任意输入在一步内就进入平衡态。     

多模式对连接权矩阵的一种神经网络学习算法

该文提出多模式对连接权矩阵的一种神经网络学习算法,并给出了严格的理论证明。该算法能够将多个模糊模式对可靠地编码存储到尽可能少的连接权矩阵中,从而大大地减少存储空间,而且容易实现,并举例验证了它的有效性。     

一个新的二进前向多层网学习算法及布尔函数优化实现

本文首先给出二进前向多层网几何学习算法的一个改进策略,提高了原算法的学习效率,然后同个新的神经网络启发式遗传几何学习算法。ＨＧＧＬ算法采用面向知识的交叉算子和变异算子对几何超平面进行优化的划分,同时确定隐层神经元的个数及连接权系数和阈值,对任意布尔函数,ＨＧＧＬ算法可获得迄今为止隐节点数量少的神经网络结构。     

基于实虚型连续多值复数Hopfield神经网络的QAM盲检测

针对统计量算法盲检测QAM信号的缺陷,该文提出了一个实虚型连续多值复数Hopfield神经网络算法,该网络的实部、虚部各含一个连续多值实激活函数.该文构造了适用于该网络的能量函数,并分别在异步和同步更新模式下证明了该神经网的稳定性.当该神经网的权矩阵借助接收数据补投影算子构成时,该实虚型连续多值复数Hopfield神经网络可有效地实现QAM信号盲检测.仿真试验表明:该算法采用较短接收数据即可到达全局真解点,并且适用于含公零点信道.     

一种遗传-梯度混合算法

本文提出一种函数全局优化的遗传-梯度混合算法。该算法由选择、交叉和诱导变异算子组成。通过对自适应滤波和前向神经网络权值训练两个典型问题的数字仿真，表明该算法是一种快速有效的全局优化算法。     

神经网络在ESB自适应波束形成中的应用

基于特征空间(ESB)自适应波束形成算法性能优良,但需要进行矩阵特征分解,运算量大。提出了一种基于神经网络的ESB自适应波束形成算法。该算法仅将Toeplitz化后的采样协方差矩阵的第一列元素作为网络输入,从而降低了输入矢量的维数。利用广义回归神经网络逼近权矢量,神经网络的并行计算可提高运算速度。计算机仿真结果表明此方法是有效的。     

最小Rayleigh熵恒模信号分离算法

提出了一种最小Rayleigh熵恒模信号分离算法(MRE-AOCMSS)。该算法通过对代价函数作适当的变换, 得到新的协方差矩阵,然后构造出协方差矩阵的Rayleigh熵,再通过最小化Rayleigh熵得到用以恢复恒模信号的最佳权矢量。该算法能够准确地分离出期望信号并抑制掉干扰,从而有效地克服了常规恒模算法(CMA)对同信道干扰(CCI)功率大于信号功率的情况敏感这一缺点。计算机仿真结果说明了该算法的有效性。     

模糊对向传播神经网络的学习算法

模糊对向传播神经网络的学习算法由输入层至竞争层的连接权向量和竞争层到输出层的连接权向量两部分的学习组成,对于前者,分别选用聚类法和工下降法,本文研究了模糊对向传播神经网络的两种学习算法从理论上分析了这两种算法的性质,把算法应用于著名Ｍａｃｋｅｙ－Ｇｌａｓｓ混沌时间序列预测问题中,实验结果表明后一种算法的学习精度及泛化能力较前一种算法要好,但前者的学习速度要快。     

基于广义能量函数的扩频码序列盲估计算法

本文提出一种基于广义能量函数(GEF)的直接序列扩频(DS/SS)信号扩频码序列的盲估计方法.广义能量函数通过引入一个加权矩阵来优化线性神经网络的连接权矢量,可以推导出一种新的递归最小二乘(RLS)学习算法.该算法能高效提取一个输入信号相关矩阵的多个主分量,可对同步和非同步模型下的PN码序列实现盲估计.该算法具有收敛快、稳健性好等优点,其运算量和储存量远远小于特征值分解算法,收敛速度、相关性能和运算复杂度等恢复性能优于压缩投影逼近子空间跟踪(PASTd)算法和改进神经网络(MHR)算法.计算机仿真证明,该算法能在较低的信噪比条件下,实时高效地恢复PN码序列,具有优异的性能.     

基于伪逆的协同神经网络学习算法

本文改进了Ｈａｋｅｎ协同神经网络的算法。该学习算法在Ｈａｋｅｎ算法的基础上引进反馈机制,对权值矩阵反复训练,使权值矩阵能更有效地进行图像识别,并增大了网络容量。     

基于极大权的最小连通支配集启发式算法

Ad hoc无线网络中基于最小连通支配集(MCDS)的路由是一个引人瞩目的方法,文中提出了一种基于极大权的MCDS的启发式算法,确保了性能强的主机担任网关节点的角色,能更好的协调管理网络中其他的节点,从而保持MCDS的相对稳固性并为全网中的广播和路由操作提供一个高效的通信基础.仿真结果表明,该算法能在保证生成权和极大的连通支配集的同时也确保它的极小性,因此能有效地用于基于MCDS的路由设计中.     

模糊汉明神经网络及其实现的研究

由于传统汉明神经网络未解决模式重叠和识别算法是否一定收敛的问题,也未充分利用输入模式与其他神经元之间的靠近程度信息,本文提出一种模糊汉明神经网络.模糊汉明神经网络可接受二值和非二值输入;使用模糊类隶属度子网解决模式重叠问题和充分利用靠近程度信息;采用比较子网保证算法的收敛和减少互连.其模块式的电路设计也便于网络的VLSI实现和扩展.     

Nonlinear blind signal separation with intelligent controlled learning

The paper proposes a new nonlinear blind source separation algorithm with hybridisation of fuzzy logic based learning rate control and simulated annealing to improve the global solution search. Benefits of fuzzy systems and simulated annealing are incorporated into a multilayer perceptron network. Fuzzy logic control allows adjustments of learning rate to enhance the rate of convergence of the algorithm. Simulated annealing is implemented to avoid the algorithm becoming trapped in local minima. A simple and computationally efficient method for controlling learning rate and ensuring a global solution is proposed. The performance of the proposed algorithm in terms of convergence of entropy, is studied alongside other techniques of learning rate adaptation. Simulations show that the proposed nonlinear algorithm outperforms other existing nonlinear algorithms based on fixed learning rates.     

基于YOLOv5网络模型的人员口罩佩戴实时检测

近年来,随着硬件算力的提升和人工智能算法的创新发展,使得深度学习算法在目标检测方面有着广泛的应用。针对现有人工方式查看人员口罩佩戴情况的不足,提出了一种基于深度学习YOLOv5算法实现对口罩佩戴情况的实时检测。算法首先将数据集进行归一化处理,再将数据接入YOLOv5网络进行迭代训练,并将最优权重数据保存用作测试集测试,算法通过tensorboard可视化显示训练和测试结果。实验结果表明,所提算法检测的准确性高,实时性强,满足实际使用需求。     

一种通过结构边界进行贝叶斯网络学习的算法

贝叶斯网络是智能算法领域重要的理论工具,其结构学习问题被认为是NP-hard问题。该文通过混合学习算法的方式,从分析低阶条件独立性测试提供的信息入手,给出了构造目标网络结构空间边界的方法,并给出了完整的证明。在此基础上执行打分搜索算法获得最终的网络结构。仿真结果表明该算法与同类算法相比具有更高的精度和更好的执行效率。     

A Dominating-Set-Based Routing Scheme in Ad Hoc Wireless Networks

Efficient routing among a set of mobile hosts (also called nodes) is one of the most important functions in ad hoc wireless networks. Routing based on a connected dominating set is a promising approach, where the searching space for a route is reduced to nodes in the set. A set is dominating if all the nodes in the system are either in the set or neighbors of nodes in the set. In this paper, we propose a simple and efficient distributed algorithm for calculating connected dominating set in ad hoc wireless networks, where connections of nodes are determined by their geographical distances. We also propose an update/recalculation algorithm for the connected dominating set when the topology of the ad hoc wireless network changes dynamically. Our simulation results show that the proposed approach outperforms a classical algorithm in terms of finding a small connected dominating set and doing so quickly. Our approach can be potentially used in designing efficient routing algorithms based on a connected dominating set.     

Weighted Steiner Connected Dominating Set and its Application to Multicast Routing in Wireless MANETs

In this paper, we first propose three centralized learning automata-based heuristic algorithms for approximating a near optimal solution to the minimum weight Steiner connected dominating set (WSCDS) problem. Finding the Steiner connected dominating set of the network graph is a promising approach for multicast routing in wireless ad-hoc networks. Therefore, we present a distributed implementation of the last approximation algorithm proposed in this paper (Algorithm III) for multicast routing in wireless mobile ad-hoc networks. The proposed WSCDS algorithms are compared with the well-known existing algorithms and the obtained results show that Algorithm III outperforms the others both in terms of the dominating set size and running time. Our simulation experiments also show the superiority of the proposed multicast routing algorithm over the best previous methods in terms of the packet delivery ratio, multicast route lifetime, and end-to-end delay.     

Betweenness centrality based connectivity aware routing algorithm for prolonging network lifetime in wireless sensor networks

Network lifetime is the key design parameter for wireless sensor network protocols. In recent years, based on energy efficient routing techniques numerous methods have been proposed for enhancing network lifetime. These methods have mainly considered residual energy, number of hops and communication cost as route selection metrics. This paper introduces a method for further improvement in the network lifetime by considering network connectivity along with energy efficiency for the selection of data transmission routes. The network lifetime is enhanced by preserving highly connected nodes at initial rounds of data communication to ensure network connectivity during later rounds. Bassed on the above mentioned concept, a connectivity aware routing algorithm: CARA has been proposed. In the proposed algorithm, connectivity factor of a node is calculated on the basis of Betweenness centrality of a node and energy efficient routes are found by using fuzzy logic and ant colony optimization. The simulation results show that the proposed algorithm CARA performs better than other related state-of-the-art energy efficient routing algorithms viz. FML, EEABR and FACOR in terms of network lifetime, connectivity, energy dissipation, load balancing and packet delivery ratio.     

Connectivity constrained wireless sensor deployment using multiobjective evolutionary algorithms and fuzzy decision making

Deployment of sensor nodes is an important issue in designing sensor networks. The sensor nodes communicate with each other to transmit their data to a high energy communication node which acts as an interface between data processing unit and sensor nodes. Optimization of sensor node locations is essential to provide communication for a longer duration. An energy efficient sensor deployment based on multiobjective particle swarm optimization algorithm is proposed here and compared with that of non-dominated sorting genetic algorithm. During the process of optimization, sensor nodes move to form a fully connected network. The two objectives i.e. coverage and lifetime are taken into consideration. The optimization process results in a set of network layouts. A comparative study of the performance of the two algorithms is carried out using three performance metrics. The sensitivity analysis of different parameters is also carried out which shows that the multiobjective particle swarm optimization algorithm is a better candidate for solving the multiobjective problem of deploying the sensors. A fuzzy logic based strategy is also used to select the best compromised solution on the Pareto front.