模型检测中可变攻击者模型的构造

提出了一个可变攻击者模型构造方案. 该方案通过定义抽象项的概念及其运算规则,大大降低了攻击者进行代数运算的复杂度. 定义了攻击者行为库和攻击规则选择算法,使检测者能根据不同的协议构造不同的攻击者模型. 由于攻击者行为可任意组合,故实现了攻击者模型的可变性. 可变攻击者模型保证了模型检测工具对协议分析的效率和准确性.     

UNIX操作系统安全再分析

当今UNIX操作系统已得到广泛的普及和应用,因而UNIX的安全性是很重要的课题,其关键在于要不断增强用户的安全意识,并从系统本身来提高安全性,本文从逻辑安全角度出发,针对调整用户标识程序,超级用户两方面,仍主要就BSD4 UNIX及其兼容版本,探讨各种可能的安全违例。     

正交检测技术在周界入侵探测系统中的应用

在分析了正交信号检测原理的基础上,将正交检测技术应用于周界入侵探测装置中,搭建了正交信号检测电路,应用Multisim8软件进行了仿真和分析。     

基于鱼眼镜头的无盲区周界入侵识别系统研究

周界入侵识别已经慢慢成为智能视频监控系统的一项所必需的功能,使用鱼眼镜头作为视频采集节点的光学系统,较大型的监控场合只需布置少量的视频采集节点便可得到无盲区的周界环境信息。采用图像拼接技术和识别技术将服务器端收集到的信息拼接成全景视频图像并进行入侵行为识别,从而有效避免了针对多节点独立进行入侵识别的并行运算,可大大减轻系统对多CPU运算平台的依赖性,提高系统的可靠性。现场测试结果表明,使用该方案设计的周界入侵识别系统,可有效保证不出现识别盲区,在使用同一识别算法和同等级识别延时的前提下,大大降低对监控系统服务器的配置要求,特别是对处理器个数的要求。     

基于角色冒充对数字签名协议的攻击

实现对具有数字签名的身份认证协议进行验证.从攻击者的角度,基于角色冒充的方法,使用逆向归结的推理方式,采用JAVA来实现验证算法.对于一些具有数字签名的协议,可以进行有效的验证.     

Metareasoning about Security Protocols using Distributed Temporal Logic

We introduce a version of distributed temporal logic for rigorously formalizing and proving metalevel properties of different protocol models, and establishing relationships between models. The resulting logic is quite expressive and provides a natural, intuitive language for formalizing both local (agent specific) and global properties of distributed communicating processes. Through a sequence of examples, we show how this logic may be applied to formalize and establish the correctness of different modeling and simplification techniques, which play a role in building effective protocol tools.     

Towards a Quantitative Analysis of Security Protocols

This paper contributes to further closing the gap between formal analysis and concrete implementations of security protocols by introducing a quantitative extension of the usual Dolev-Yao intruder model. This extended model provides a basis for considering protocol attacks that are possible when the intruder has a reasonable amount of computational power, in particular when he is able, with a certain probability, to guess encryption keys or other particular kind of data such as the body of a hashed message. We also show that these extensions do not augment the computational complexity of the protocol insecurity problem in the case of a finite number of interleaved protocol sessions.     

Research on human intruder detection and localization based on LCX sensor

Many aspects of the intruder detection system (IDS) based on the LCX sensor are still unknown due to its complex propagation characteristics. In order to study the field disturbance mechanism of human intruder and to reduce the phase error caused by the initial state of the detection signal, a novel method of improving the localization accuracy is proposed. At the frequencies of 40, 100, and 200 MHz, the IDS based on the LCX sensor is proposed by analyzing the scattering characteristics of the radar cross section (RCS) of human intruder and using IQ demodulation method. According to the characteristics of the IDS, the RCS is obtained by irradiating the human intruder from three typical radiation directions, which are front, side, and low‐side direction in the mentioned frequency range. Combined with the method of pulse accumulation and synchronous subtraction, the intruder localization can be easily realized by using pulse delay positioning method. The results demonstrate the improvement in localization accuracy and the decrement in false positive rate, and the positioning error is less than 1 m.     

云平台下入侵人员位置实时监测方法研究

云平台数据遭受攻击时,入侵人员可伪造网络地址或通过间接攻击方式隐藏真实地址。此时,管理员不但要修复漏洞,还要获得证据线索和溯源。为此,提出云平台下入侵人员位置实时监测方法。分析云平台入侵监测系统设计需求,建立入侵监测模型;从数据源、安全警报等方面设置入侵人员位置监测系统结构,综合考虑各类干扰因素,设计系统电路与软件程序...