Mathematical Decomposition Techniques for Distributed Cross-Layer Optimization of Data Networks

Network performance can be increased if the traditionally separated network layers are jointly optimized. Recently, network utility maximization has emerged as a powerful framework for studying such cross-layer issues. In this paper, we review and explain three distinct techniques that can be used to engineer utility-maximizing protocols: primal, dual, and cross decomposition. The techniques suggest layered, but loosely coupled, network architectures and protocols where different resource allocation updates should be run at different time-scales. The decomposition methods are applied to the design of fully distributed protocols for two wireless network technologies: networks with orthogonal channels and network-wide resource constraints, as well as wireless networks where the physical layer uses spatial-reuse time-division multiple access. Numerical examples are included to demonstrate the power of the approach.     

CAN总线共享时钟调度算法的应用与改进

控制器局域网（CAN）协议被广泛运用在分布式嵌入式系统中,研究表明共享时钟调度算法能有效的将基于CAN的微控制器与时间触发网络结构结合起来。文章指出了共享时钟调度算法存在的一些不足,并且提出一种改进的共享时钟调度算法,通过构建一个实际系统验证了两种算法在CAN通信中的性能,结果显示改进的共享时钟调度算法是对原算法的一种有效补充。     

Ad Hoc网络协议栈通用要求研究

Ad Hoc网络是由一组带有无线收发装置的移动节点组成的一个多跳的临时性的自治系统。Ad Hoc网络与有线网络相比有很多特殊点，会导致传统的网络协议栈在Ad Hoc下不能有效使用。解决的办法是使Ad Hoc网络的网络协议栈满足比传统协议栈更高的要求：物理层要能实现分布式多点发起的同步，媒体访问控制（MAC）层要能管理多个分布站点随机的收发行为，网络层要能高效地管理分布式的路由协议，传输层要针对各种不同的业务进行优化。     

Balancing transport and physical Layers in wireless multihop networks: jointly optimal congestion control and power control

In a wireless network with multihop transmissions and interference-limited link rates, can we balance power control in the physical layer and congestion control in the transport layer to enhance the overall network performance while maintaining the architectural modularity between the layers? We answer this question by presenting a distributed power control algorithm that couples with existing transmission control protocols (TCPs) to increase end-to-end throughput and energy efficiency of the network. Under the rigorous framework of nonlinearly constrained utility maximization, we prove the convergence of this coupled algorithm to the global optimum of joint power control and congestion control, for both synchronized and asynchronous implementations. The rate of convergence is geometric and a desirable modularity between the transport and physical layers is maintained. In particular, when congestion control uses TCP Vegas, a simple utilization in the physical layer of the queueing delay information suffices to achieve the joint optimum. Analytic results and simulations illustrate other desirable properties of the proposed algorithm, including robustness to channel outage and to path loss estimation errors, and flexibility in trading off performance optimality for implementation simplicity. This work presents a step toward a systematic understanding of "layering" as "optimization decomposition," where the overall communication network is modeled by a generalized network utility maximization problem, each layer corresponds to a decomposed subproblem, and the interfaces among layers are quantified as the optimization variables coordinating the subproblems. In the case of the transport and physical layers, link congestion prices turn out to be the optimal "layering prices.".     

无线传感器网络中基于效用模型的分布式功率控制机制

本文将效用模型引入无线传感器网络的功率控制设计中,提出了一种基于效用模型的分布式功率控制机制(简称UMDPC).该机制建立了网络中所有传感器节点的功率与效用模型的对应关系,将链路可靠性、网络能耗归纳到统一的网络效用优化框架中,并证明该效用优化问题是凸优化问题,构造基于对偶分解的分布式的优化算法,获得网络效用最大化条件下各节点的优化发射功率.最后,通过模拟实验对所提机制及其实现算法的性能进行比较和评价.实验结果表明,本文所提机制最大化了网络的效用,提高了网络的能量利用效率.     

灵活透明光网络的自适应传输控制策略

灵活透明光网络是传送网的发展演进方向之一。目前,智能传送网的协议标准主要用于选路与交换过程的控制,若在自动交换光网络(ASON)架构基础上引入对传送平面物理特性的智能控制功能,基于损伤感知对控制平面进行功能扩展,可以实现端到端传输质量最优化。传输自适应的控制过程可以用可调节补偿单元信息搜集、链路传输性能监测、自适应调节计算、可调节补偿单元控制等4个步骤来实现。支持光通路自适应传输过程的协议扩展可采用分布式路由方案和分布式信令方案两种解决方案。     

基于SDN的农垦EDRP路由协议研究

随着SDN技术的发展,新型的网络架构提出一种将网络控制平面与数据平面分离的设计思路,从而更灵活地实现网络流量控制、网络安全,并可以结合自身的组网环境和组网结构,对网络系统进行改进,结合农垦的网络现状,提出一种新的路由协议EDRP（Enhance Deter-mined Routing Protocol）,以简化网络的管理和运维成本,并提高网络的安全性和健壮性.     

Service applications for SONET DCS distributed restoration

This paper determines the scope of network applications and services that could be offered using a SONET DCS-based self-recovering mesh architecture with distributed control. The study includes an outage impact analysis on network services and a determination of how network restoration time objectives will affect the applicability for the distributed controlled DCS network architecture. It is concluded that using SONET DCS distributed control architectures to provide more complete survivability of a network would support numerous applications. Future services will demand a fault-tolerant network with complete survivability; this may only be reached through integration of SONET DCS distributed control architectures with other survivable architectures such as cell relay networks (e.g., supporting SMDS) and self-healing rings     

基于信令控制的ASON分布式连接管理技术

自动交换光网络(ASON)已经成为下一代光网络的发展方向,而其控制平面基于信令控制的分布式连接管理功能则是实现光网络智能性的关键.文章介绍了ASON和网络结构与控制接口、分布式的光路连接管理策略,讨论了ASON的功能结构,并详细分析了基于信令的呼叫和连接处理的实现过程.     

A review and prospect on the network congestion control

Network congestion has become more critical due to the rapid increase in the total number of Internet users and applications. This paper gives a survey for the research development of network congestion control in the past twenty years. The evaluation index of the ideal congestion control protocol is introduced, such as stability, fairness, efficiency, etc. Several network models and analysis methods are also given for the development of the congestion control, such as the nonlinear model, the control theory, the optimization method. On the basis of massive research of the congestion control issues, there is the generalization of the end-to-end protocol and intermediate nodes. Some open problems have been discussed for the further research of network congestion control.     

Optimized multipath network coding in lossy wireless networks

Network coding has been a prominent approach to a series of problems that used to be considered intractable with traditional transmission paradigms. Recent work on network coding includes a substantial number of optimization based protocols, but mostly for wireline multicast networks. In this paper, we consider maximizing the benefits of network coding for unicast sessions in lossy wireless environments. We propose Optimized Multipath Network Coding (OMNC), a rate control protocol that dramatically improves the throughput of lossy wireless networks. OMNC employs multiple paths to push coded packets to the destination, and uses the broadcast MAC to deliver packets between neighboring nodes. The coding and broadcast rate is allocated to transmitters by a distributed optimization algorithm that maximizes the advantage of network coding while avoiding congestion. With extensive experiments on an emulation testbed, we find that OMNC achieves more than two-fold throughput increase on average compared to traditional best path routing, and significant improvement over existing multipath routing protocols with network coding. The performance improvement is notable not only for one unicast session, but also when multiple concurrent unicast sessions coexist in the network.     

A Session Layer Concept for Overlay Networks

Current day mobility solutions are characterized by a number of essential restrictions that prevent them from being successful in an overlay network environment. This paper proposes 4 properties that mobility solutions must have to address these restrictions: application cooperation, higher layer mobility awareness, general applicability and protocol heterogeneity. Subsequently, a mobility solution that realizes all these properties is described. The solution introduces a session layer in the protocol stack that consists of two subsystems: the connection abstraction system and the address management system. Other mobility solutions found in the literature, session layer and others, typically only realize a subset of the four properties. Tom Mahieu holds a Bachelor's and Master's degree in computer sciences that he obtained at the Catholic University of Leuven. After obtaining his Master's degree in 1997, he was admitted into the Distrinet research group at the Department of Computer Science of the Catholic University of Leuven in Belgium as a research and teaching assistant, where he is current working on flexible component architectures for protocol stacks under the supervision of Prof. Pierre Verbaeten and Prof. Wouter Joosen. His Ph.D. dissertation deals with protocol stack support for applications that operate in dynamically changing network environments, such as mobile and wireless networks. Pierre Verbaeten is the head of the DistriNet research group. His main research interest is on system support and software engineering support for distributed systems. Under his supervision, DistriNet has grown to a group of 40 researchers (both Ph.D. students and postdocs) and 6 professors. He is responsible for teaching courses on computer architecture, computer networks and distributed systems. Wouter Joosen received a Ph.D. degree at the K.U. Leuven in 1996. His main research interests are in the areas of software architecture for distributed systems, aspect-oriented and component-based development, middleware, software security and embedded systems. Over the last 10 years, Wouter Joosen has been researching software development technologies (processes, architectures, notations and prototypes) that enable the construction and management of distributed applications that need support for a number of non-functional requirements, including security, real-time aspects, reliability and load balancing.     

Self-Configuring Applications for Heterogeneous Systems: Program Composition and Optimization Using Cognitive Techniques

This paper describes several challenges facing programmers of future edge computing systems, the diverse many-core devices that will soon exemplify commodity mainstream systems. To call attention to programming challenges ahead, this paper focuses on the most complex of such architectures: integrated, power-conserving systems, inherently parallel and heterogeneous, with distributed address spaces. When programming such complex systems, new concerns arise: computation partitioning across functional units, data movement and synchronization, managing a diversity of programming models for different devices, and reusing existing legacy and library software. We observe that many of these challenges are also faced in programming applications for large-scale heterogeneous distributed computing environments, and current solutions as well as future research directions in distributed computing can be adapted to commodity computing environments. Optimization decisions are inherently complex due to large search spaces of possible solutions and the difficulty of predicting performance on increasingly complex architectures. Cognitive techniques are well suited for managing systems of such complexity, citing recent trends of using cognitive techniques for code mapping and optimization support. Combining these, we describe a fundamentally new programming paradigm for complex heterogeneous systems, where programmers design self-configuring applications and the system automates optimization decisions and manages the allocation of heterogeneous resources.     

移动Ad Hoc网络安全性研究

移动Ad Hoc网络安全性问题已经成为无线通信领域研究的热点之一.文中介绍了Ad Hoc网络的体系结构和特点,分析了它存在的安全问题,并介绍了分布式CA方案,通过分析攻击流模型和动态安全性评估模型,对分布式CA方案的安全性进行评估.     

Design and analysis of RT-Ring: a protocol for supporting real-time communications

Distributed applications with quality of service (QoS) requirements are more and more used in several areas (e.g., automated factory networks, embedded systems, conferencing systems). These applications produce a type of traffic with hard timing requirements, i.e., transmissions must be completed within specified deadlines. To handle these transmissions, the communication system must use real-time protocols to provide a communication service that is able to satisfy the QoS requirements of the distributed applications. In this paper, the authors propose a new real-time protocol, called RT-Ring, able to support transmissions of both real-time and generic traffic over a ring network. RT-Ring provides both network guarantees and high network resource utilization, while ensuring the compatibility with the emerging differentiated service architectures. Network guarantees are fully proved and high network utilization is highlighted by a comparative study with the FDDI protocol. This comparison shows that RT-Ring network capacities are greater than the corresponding FDDI capacities. In fact, by assuming the FDDI frames with a length equal to the RT-Ring slot size and by using the same traffic load the authors show that the capacities of FDDI are equal to the lower bound capacities of RT-Ring.     

PPmcast: a novel P2P-based application-level multicast using Kamulia

The rapid growth and popularity of networked meeting environments such as video conference in recent years have spawned a series of research interests in constructing large-scale environments. For increasing scalability and decreasing the cost of management and deployment, more and more studies propose using peer-to-peer (P2P) architectures to construct large-scale application-level multicast algorithm for games, multimedia and other applications. In order to improve the applications' performance, high efficiency multicast algorithm is required. This paper focuses on developing a novel P2P application-level multicast algorithm based on Kamulia protocol. In the proposed algorithm, all nodes of P2P network are organized into structured overlay network by distributed hash table (DHT). The distance metric between two nodes is obtained by computing the exclusive or (XOR) value. The theoretical analysis indicates the proposed algorithm has better time complexity degree compared with other similar algorithms. The simulation experiments also show that the proposed algorithm has better performance on the following aspects, control expenses, loss ratio after failure, and waiting time to the first packet.     

ADSL宽带接入系统

ADSL技术是基于铜双绞线的一种重要的宽带接入方式,在实现宽带到家庭、因特网接入中有着广泛的应用前景。文章首先给出了基于ATM over ADSL接入网的网络参考模型,然后分析了几种不同组网方式的特征及协议栈模型。考虑到SOHO方式的普遍性,文章还提出了支持SOHO的两种网络结构,并比较了不同结构的差异。     

Network of the future

Telecommunications networks of the future will exploit two new network architecture concepts that are currently being implemented, or soon will be. These are the Intelligent Network and ISDN, the Integrated Services Digital Network, which together will support a full range of voice, data, and image services that Information Age telecommunications users will demand. These new network architectures, operating synergistically with intelligence in terminal systems, will constitute a framework in which users and service providers will link together standardized functional components to create customized services. These components, along with interfaces and signaling protocols at the interfaces and within the network will result from continuing national and international standardization efforts. In the planning of these new architectures, a few major goals are of paramount importance: • the achievement of a flexible network structure in which functionality is distributed among the network components in a way which supports the timely and economic introduction of new services in response to user needs; • the establishment of industry standards at the interfaces between network elements such that service suppliers can choose among a set of available systems products in building their networks and avoid dependence on a small set of suppliers, • the development of standard user interfaces supporting signaling procedures which can provide the user with increased control of, and access to, services to satisfy his needs; Achievement of these goals will result in the realization of an Open Network Architecture. The ISDN and Intelligent Network architecture concepts are described in this paper.     

Cross-layer feedback architecture for mobile device protocol stacks

Applications using traditional protocol stacks (e.g., TCP/IP) from wired networks do not function efficiently in mobile wireless environments. This is primarily due to the layered architecture and implementation of protocol stacks. One mechanism to improve the efficiency of the stack is cross-layer feedback, that is, making information from within one layer available to another layer of the stack. For example, TCP retransmissions can be reduced by making it aware of network disconnections or handoff events. We highlight the need for a cross-layer feedback architecture and identify key design goals for an architecture. We present our ECLAIR architecture, which satisfies these design goals. We describe a prototype implementation that validates ECLAIR. We also discuss other cross-layer architectures and provide a cross-layer design guide.     

Transport layer protocols and architectures for satellite networks

Designing efficient transmission mechanisms for advanced satellite networks is a demanding task, requiring the definition and the implementation of protocols and architectures well suited to this challenging environment. In particular, transport protocols performance over satellite networks is impaired by the characteristics of the satellite radio link, specifically by the long propagation delay and the possible presence of segment losses due to physical channel errors. The level of impact on performance depends upon the link design (type of constellation, link margin, coding and modulation) and operational conditions (link obstructions, terminal mobility, weather conditions, etc.). To address these critical aspects a number of possible solutions have been presented in the literature, ranging from limited modifications of standard protocols (e.g. TCP, transmission control protocol) to completely alternative protocol and network architectures. However, despite the great number of different proposals (or perhaps also because of it), the general framework appears quite fragmented and there is a compelling need of an integration of the research competences and efforts. This is actually the intent of the transport protocols research line within the European SatNEx (Satellite Network of Excellence) project. Stemming from the authors' work on this project, this paper aims to provide the reader with an updated overview of all the possible approaches that can be pursued to overcome the limitations of current transport protocols and architectures, when applied to satellite communications. In the paper the possible solutions are classified in the following categories: optimization of TCP interactions with lower layers, TCP enhancements, performance enhancement proxies (PEP) and delay tolerant networks (DTN). Advantages and disadvantages of the different approaches, as well as their interactions, are investigated and discussed, taking into account performance improvement, complexity, and compliance to the standard semantics. From this analysis, it emerges that DTN architectures could integrate some of the most efficient solutions from the other categories, by inserting them in a new rigorous framework. These innovative architectures therefore may represent a promising solution for solving some of the important problems posed at the transport layer by satellite networks, at least in a medium‐to‐long‐term perspective. Copyright © 2006 John Wiley & Sons, Ltd.