一种增强移动自组网Internet连接的改进型方案

针对移动自组网因单向链路以及盲目转播网关发现分组而造成的与Internet互联性能下降,提出了一种增强型互联方案.同时结合了先应和后应网关发现策略优点的混合网关发现策略被用来获取良好的Internet连接并保持低开销.而通过交换携带有额外增加的对称邻居节点列表和非对称邻居节点列表字段的AODV hello分组,单向链路得以从路由计算中予以清除而广播风暴也同时得到抑制.基于ns-2的仿真结果表明此改进型Internet连接方案在单向链路数量及节点移动速度均变化的应用环境下能取得较其他方案为优的互联性能.     

神经网络在系统优化设计中的应用

本文首先简介了系统优化问题，然后以Hopfield神经网络在对铺设公路至若干油田的路线进行优化设计中的应用为典型事例，初步探讨了在复杂系统的优化方案设计中运用人工神经网络的可行性，并从中得出拥有并行分布计算优势的神经网络在解决系统组合优化问题时具有优越性这一结论。     

人工神经网络在火灾探测中的应用

火灾的早期探测是较为复杂且具有重要意义的研究课题。针对传统火灾探测方法存在的不足,提出了一种基于人工神经网络技术的火灾自动探测系统,即采用黄金分割优选法对网络隐含层节点数进行优选,采用优于BP算法的LM算法进行网络学习训练。实验结果表明不仅这种系统具有拟合精度高和运算速度快等优点,且探测效率良好。     

基于BP神经网络的心电图三分类

研究了一种基于神经网络的心电图分类方法,该方法利用前向多层神经网络的反向传播算法(Back Propagation),即BP算法,采用MATLAB软件建立用于三分类的BP神经网络,并利用大量的心电图数据训练神经网络,使神经网络对未训练过的心电图有较好的分类能力。实验结果表明,用这种方法用于心电图的三分类取得较好的效果。     

Mathematical Analysis of Energy Efficiency in IEEE 802.11 DCF

The standardized IEEE ,802. II distributed coordination function （ DCF） provides a contention-based distributed channel access mechanism for mobile stations to share the wireless medium. However, when stations are mobile or portable units, power consumption becomes a primary issue since terminals are usually battery driven. This paper proposes an analytical model that calculates the energy efficiency of both the basic and the RTS/CTS access mechanisms of the IEEE 802. II protocol. The model is validated with simulation results using NS-2 simulation package. The effects of the network size, the average packet length, the initial contention window and maximum backoff stages on the energy efficiency of both access mechanisms are also investigated. Results show that the basic scheme has low energy efficiency at large packet length and large network size, and depends strongly on the number of stations and the backoff procedure parameters. Conversely, the RTS/CTS mechanism provides higher energy efficiency when the network size is large, and is more robust to variations in the backoff procedure parameters.     

三角债清理的图论模型分析

三角债是我国当前经济领域中存在的一个突出问题。文章建立了三角债的图论模型,并证明了任一债务网络皆可转换为最小债务网络,从而简化了三角债清理工作。     

基于遗传算法和BP神经网络的优化设计方法

随着生产规模的复杂化、多维化和非线性等复杂特性的增加，对生产中高效的优化技术要求也越来越迫切，利用并行遗传算法和BP网络的优点，提出采用遗传算法的并行搜索和解空间搜索的优点进行网络参数的选取，利用BP网络简单和可塑性强的优点来优化样本空间，以取得整体的优化效率。     

加强信息网络环境下的图书馆建设

信息网络的迅速发展，对图书馆提出了严峻挑战的同时也带来机遇。图书馆要面对时代变化，不断进行自身建设，才能适应社会发展需要。     

EBF网络特征映射和恢复及其在鲁棒话者识别中的应用

The performance of speaker verification systems is often compromised under real-world environments. For example, variations in handset characteristics could cause severe performance degradation. This paper presents a novel method to overcome this problem by using a non-linear handset mapper. Under this method, a mapper is constructed by training an elliptical basis function network using distorted speech features as inputs and the corresponding clean features as the desired outputs. During feature recuperation, clean features are recovered by feeding the distorted features to the feature mapper. The recovered features are then presented to a speaker model as if they were derived from clean speech. Experimental evaluations based on 258 speakers of the TIMIT and NTIMIT corpuses suggest that the feature mappers improve the verification performance remarkably.     

Modeling water and carbon fluxes above summer maize field in North China Plain with back-propagation neural networks

INTRODUCTION Dynamics of carbon and water vapor fluxesexchange between the atmosphere and the ecosystembiosphere, depend on complex and non-liner interplayamong physiological, ecological, biochemical andedaphic factors and meteorological conditions (Jarvis,1995; Leuning et al., 1995). There are many studiesquantifying the fluxes across different time and spacescales as well as assessing the environment con-straints on them by some kinds of biophysical orempirical models whose resul…