基于复杂网络的联合作战指挥体系研究

以作战指挥理论需求为牵引,从复杂网络的角度对联合作战指挥体系问题进行网络抽象,将指挥体系中的指挥机构作为网络的节点,将指挥关系作为网络的边,按照指挥体系历史发展的规律,从简单到复杂,分别构建了传统层级纵长树状型的指挥体系网络模型,在树状模型基础上,增加了横向联系但依然具有层级结构的信息共享改进型的指挥体系网络模型,以及...     

雷达组网系统责任区抗干扰优化部署

针对有源压制干扰对雷达组网系统的威胁,深入分析了雷达组网系统的抗干扰优化部署问题,建立了雷达组网系统责任区抗干扰优化布站决策模型,并给出了具体的算法和求解分析.仿真结果验证了该方法的有效性,可为雷达组网系统的实际优化部署提供参考.     

基于遗传算法的无线传感器网络路由研究

无线传感器网络路由算法面临着节省能量、延长网络寿命、提高可靠性等方面的挑战。提出了一种基于遗传的无线传感器网络路由算法。对算法中的算子编码、适应度函数设计、编码方式及参数选取进行了细致研究。仿真结果表明,本算法减少了能耗、延长了网络生存时间并提高了网络的可靠性。     

扩展信念网络模型的军事情报检索系统设计

着眼信息化战争条件下军事情报分类、处理困难的现实,在深入分析军事情报检索系统特点基础上,提出一个基于分类的自动存储和利用查询术语同义词关系扩展信念模型的检索军事情报检索原型系统,详细描述了其存储和检索设计思路.     

首都周边维稳预警机制研究

在构建首都大周边维稳预警机制中,情报工作具有监测、预测和辅助决策功能。社情监测离不开情报的收集,维稳社情预测依赖于情报的分析、研判、传递与通报,危害社会安全稳定案件预控离不开情报支撑。做好首都周边地区维稳预警工作,需要健全和完善情报收集制度、情报分析和研判制度、情报送报制度、相关情报工作制度以及社会舆情信息汇集与分析制度。     

新疆边境地区恐怖活动特点及打击对策

近年来以“东突”为主的恐怖势力在新疆边境地区活动十分猖獗,呈现出联合,渗透对象多元、低龄等特点。公安边防部队地处边境一线,任务艰巨,责任重大,要通过加强反恐情报建设、健全装备保障体制、加强战法研究与战术训练、加强边境管理以及加强国际警务合作等多种对策,不断提高部队打击和防范恐怖活动的能力与水平,为维护边疆稳定做出贡献。     

无线Mesh网络技术研究

无线Mesh网络是新近出现的一种无线多跳网状拓扑网络,在多领域被广泛的应用。简要介绍了无线Mesh网络的结构、特点及应用,并重点分析了影响无线Mesh网性能的主要技术因素。     

IED casualties mask the real problem: it's us

Casualty figures suggest that the US/Allied Counter Improvised Explosive Device (C-IED) policy and the present allocation of national assets, resources, and intellectual capital have not been very successful. A number of explanations for why this has been the case are discussed and critiqued here.     

Crime facilitation purposes of social networking sites: A review and analysis of the ‘cyberbanging’ phenomenon

There have been growing claims in media circles and law-enforcement settings that street gangs and criminal groups are turning to Internet-based social networking sites for various reasons ranging from the showcasing of their images and exploits to the suspected recruitment of members. The present study investigates whether such a trend is, in fact, in place. The presence of street gangs on these Internet sites is referred to as cyberbanging. While there is some anecdotal evidence suggesting that gangs are turning to social networking sites, there is little available research on exactly how street gangs and criminal groups use the Internet. Our main argument is that gang culture is in many ways an individualized phenomenon and this feature ties in directly with recent assessments of the Internet as a setting that is governed by a process of networked individualism. This theoretical link between the individualized gang setting and the presence of gang members on social networking sites helps us understand why recruitment is improbable even in a context where people are openly diffusing their image and exploits to a growing number of Internet users. The empirical segment of this research adds to this general outlook. Based on a keyword search of over 50 street gang names, the three main social networking sites (Twitter, Facebook, and MySpace) were monitored for street gang presence. Results illustrate that gang presence on social networking sites is linked primarily to promoting a general gang or street culture through individual displays. In regard to the visitors to such sites, there is no evidence that they are being tricked or manipulated in any way. They are, however, showing their curiosity in regard to such groups and, for those who share their comments and opinions, signs of support are evident. Thus, whereas criminal gangs are not proactively using the Internet to convert anyone into being gang members, social networking sites are creating a new venue for people who share or are sensitive to the values underlying street gang lifestyle to come together. These sites essentially create a new convergence setting for gang members to interact with a wider number of people who would probably never have been exposed to their lifestyles and exploits through physical interactions. The study's conclusion extends these findings toward further research in this area, as well as outlining the more relevant implications for law-enforcement monitoring of this growing phenomenon.     

信息推荐系统中的朋友关系预测算法设计

伴随着互联网规模的不断扩展,信息过载问题越来越突出。信息推荐系统被视为解决信息过载问题的最有效方法。然而目前的方法大多数仅考虑用户独立的反馈,而忽略用户的社会属性对推荐的重要作用,这对信息推荐系统的性能会造成巨大的影响。为此,本文提出了基于朋友关系预测的信息推荐算法,将用户的社会关系预测引入信息推荐过程中,分别基于用户的拓扑信息及历史交互信息建立用户社会关系的存在性判定及关系类型判定,并利用线性回归分析方法和逻辑回归分析方法实现了基本特征的融合。最后,通过在Epinions和Slashdot真实数据集上的实验证明,本方法能够有效提高用户社会关系预测的准确性。