匿名技术的研究进展

对匿名技术的研究情况进行了综述。统一给出了关联性、匿名性、假名等概念,探讨了Crowds、洋葱路由、Mix nets等匿名通信的实现机制,介绍了有关的研究进展,指出了匿名性度量、环签名和群签名等匿名签名算法、匿名应用中信赖的建立等技术中需要进一步研究的问题。     

降低射频电路串扰噪声的布线方法

研究并解决了无障碍多层无网格射频电路的布线问题,主要包括三个部分:设计规则、串扰噪声限制和布线方法.首先通过设计规则计算连线参数,运用RLc模型估算串扰噪声,然后,按照需要连接的实节点个数进行归类并提出相应的布线算法.实验结果表明,该布线方法可有效地降低射频电路的串扰噪声.     

Bandwidth‐guaranteed QoS routing of multiple parallel paths in CDMA/TDMA ad hoc wireless networks

This paper investigates the issues of QoS routing in CDMA/TDMA ad hoc networks. Since the available bandwidth is very limited in ad hoc networks, a QoS request between two nodes will be blocked if there does not exist a path that can meet the QoS requirements, even though there is enough free bandwidth in the whole system. In this paper, we propose a new scheme of using multiple paths between two nodes as the route for a QoS call. The aggregate bandwidth of the multiple paths can meet the bandwidth requirement of the call and the delays of these paths are within the required bound of the call. We also propose three strategies by which to choose a set of paths as the route, namely, shortest path first (SPF), largest bandwidth first (LBF), and largest hop‐bandwidth first (LHBF). Extensive simulations have been conducted to evaluate the performance of the three strategies in comparison with a traditional single path routing algorithm. The simulation results show that the proposed multiple paths routing scheme significantly reduces the system blocking rates in various network environments, especially when the network load is heavy. Copyright © 2005 John Wiley & Sons, Ltd.     

一种分布式的洋葱路由新机制

洋葱路由技术在公网上提供双向、实时的匿名通信，其面向连接和严格的源路由方式导致效率低、扩展性差。利用分布式的概念，采用分段连接技术思想提出了一种安全、高效的分布式洋葱路由机制(DORS)，给出了DORS路由实现算法，并从方案实现效率、安全性方面对DORS进行了分析。分析结果表明DORS具有良好的扩展性、鲁棒性、简单性和有效性。     

基于关键节点和多播节点的多播路由算法

在视频会议等多播应用中，降低多播树网络费用非常重要。本文提出了基于关键节点和目的节点的多播路由KDDMC算法。由于在算法中优先考虑采用关键节点，实现更多链路的共享，从而降低网络费用。在随机网络上的仿真结果表明，KDDMC算法的多播树网络费用优于SPH算法。同时证明了KDDMC算法的复杂度为O(n^3)，且利用所提出的路由表算法易于分布式实现。     

多粒度全光网中的多层联合路由

多粒度交换技术利用波带级路由以及光纤级路由,可有效降低光交叉连接器的复杂度.但多粒度交换增加了光网络的逻辑层次结构,使得网络中的路由与资源分配问题更为复杂,因此,在多粒度全光网中实现多层联合路由是提高网络性能的关键.本文对多粒度交叉连接结构进行了分析,并对多粒度全光网中的多层联合路由问题进行了研究.     

传感器网络中基于二元簇的节能路由算法

传感器网络中的分簇算法已经被证明能够有效地节约能量。据此提出了一种基于二元簇的路由算法,其基本思想是网络中的每个节点同时属于2个不同簇,利用二元性来减少在路由发现中的资源浪费,并能够更加确定地发现路由以节约资源。理论分析和模拟实验都表明,该算法在能量消耗和网络生命周期延长方面较优。     

The Role of Network Topologies in the Optical Core of IP-over-WDM Networks with Static Wavelength Routing

In this paper, we present a performance analysis of network topologies for the optical core of IP-over-WDM networks with static wavelength routing. The performance analysis is focused on regular degree four topologies, and, for comparison purposes, degree three topologies are also considered. It is shown that the increase of the nodal degree from three (degree three topology with smallest diameter) to four (degree four topology with smallest diameter) improves the network performance if a larger number of wavelengths per link is available. However, the influence of wavelength interchange on the nodal degree gain is small. The performance of regular degree four topologies with smallest diameter is also compared with the performance of mesh–torus topologies (which are also degree four topologies), and it is shown that the blocking probability of degree four topologies with smallest diameter is about two orders of magnitude lower than the blocking probability of mesh–torus topologies. It is also presented a performance comparison of WDM-based networks with nodal degrees ranging from two to five and it is shown that the increase of the nodal degree from two to three leads to high nodal degree gains, while de increase of the nodal degree from four to five leads to low nodal degree gains. These results show that degree three and degree four topologies are very attractive for use in the optical core of IP-over-WDM networks.     

Experience with Applying Formal Methods to Protocol Specification and System Architecture

In the last three years or so we at Enterprise Platforms Group at Intel Corporation have been applying formal methods to various problems that arose during the process of defining platform architectures for Intel's processor families. In this paper we give an overview of some of the problems we have worked on, the results we have obtained, and the lessons we have learned. The last topic is addressed mainly from the perspective of platform architects.