Browsing by Author "Dr. David Thuente, Committee Member"

Now showing 1 - 11 of 11

Comparison of End-to-End QoS Reservation Schemes in Next Generation Networks
(2009-05-27) Ramaswamy, Pavithra; Dr. David Thuente, Committee Member; Dr. Khaled Harfoush, Committee Member; Dr. Harry. G. Perros, Committee Chair
Increasing demand for network services has led the network providers and equipment vendors to consider ways to provide better Quality of Service (QoS) for these services. In the Internet today, there is a little or no interaction between the network access providers and the network service providers. Several organizations such as, the Telecommunication standardization sector of International Telecommunication Union (ITU-T), the European Telecommunications Standards Institute (ETSI), and the 3rd Generation Partnership Project (3GPP) are involved in the standardization of a general QoS control architecture to bring the network access and application services to one common framework known as the Next Generation Network (NGN). In this thesis, we have studied the establishment of a connection with end-to-end QoS assurances that spans over a number of heterogeneous wireless and wireline networks, namely, a Wireless Local Area Network (WLAN), a Worldwide Interoperability for Microwave Access (WiMAX), a Metro Ethernet, a Multiprotocol Label Switching (MPLS)-based and a Differentiated Services (Diffserv)-based Wide Area Networks (WANs). The end-to-end QoS architecture that we have studied follows the ITU-T NGN QoS control architecture where the connection setup signaling uses the 3GPP IP Multimedia Subsystem (IMS) signaling. Within this context, we have studied the QoS interactions between the component network architectures, developed a mapping of QoS parameters across the various networking technologies, and identified the four different schemes for QoS reservation. These schemes are: local segmentation push scheme, local segmentation pull scheme, end-to-end push scheme and end-to-end pull scheme. Using simulation techniques, we compared the connection setup time performance of these four schemes. Our results indicate that the pull mode schemes that involve the user terminals to initiate the QoS reservation perform better than the push mode schemes that involve the IMS core network elements to initiate QoS reservation. This combined with the advantage of the localized QoS reservation makes the local segmentation pull mode QoS reservation scheme has a lower connection setup time compared to other schemes.
DGoogle: A Full-Text Search Engine in Large-Scale Peer-to-Peer Systems.
(2006-08-17) Lal, Akshay; Dr. David Thuente, Committee Member; Dr. Khaled Harfoush, Committee Chair; Dr. Mihail L. Sichitiu, Committee Member
Full-text search engines like Google serve an important role in accessing Internet resources. In such engines, a search for web pages, matching a user's query, are typically carried on a set of co-administered, physically co-located clusters of servers. Full-text search capabilities are also needed in distributed, peer-to-peer systems. Such systems are inherently not co-administered due to the limited capacity and the instability of their members, and to provide censorship resilience and member anonymity. Full-text search in large-scale peer-topeer systems is challenging since the search can potentially span millions of peers in different administrative domains. In this paper, we introduce a full-text search engine, DGoogle, designed for largescale peer-to-peer systems. DGoogle is simple to implement, does not require an organization of maintained resources based on their semantics, and can deal with text queries of arbitrary format. Simulation results show that DGoogle offers faster response to user queries, and consumes lesser network bandwidth when compared to the existing techniques such as Inverted Indices.
Differential Capacity p-Cycles
(2009-03-05) Jaikumar, Prashant; Dr. David Thuente, Committee Member; Dr. Injong Rhee, Committee Member; Dr. Rudra Dutt, Committee Chair
Survivability has become a central part of modern optical network design as the hundreds of wavelengths get multiplexed on fibers carrying data at Tbps speeds in DWDM networks. Provisioning for 100% restoration on failure using minimum amount of resources has become an important design problem. p-Cycles have emerged as a useful fault tolerance mechanism that operate at the speed of SONET rings, but also have low mesh-like spare capacity requirement. In this thesis, a modified version of p-cycle, called differential capacity p-cycle, is proposed that improve spare capacity efficiency beyond what is provided by a set of traditional p-cycles. Different variants of differential capacity p-cycles are proposed, analogous to some of the traditional p-cycle variants. The designs of the various types of differential capacity p-cycles are formulated using integer linear programs, and the spare capacity usage of these new structures are compared with that of traditional p-cycles and their variants.
Exploration and Analysis of a Method for Estimating the Rank of a Matrix
(2007-04-16) Orlowski, Nicholas Richard; Dr. Donald Bitzer, Committee Member; Dr. Robert Funderlic, Committee Chair; Dr. David Thuente, Committee Member
The singular value decomposition (SVD) provides important information about a matrix and its rank, including its singular values and singular vectors. Because of noise in a matrix and the limitations of binary representation, the calculated SVD of a matrix is necessarily an estimate of the true SVD of that matrix. Our method approximates the true singular values of a matrix by gathering samples of the calculated singular values of the matrix. With these approximations, we can use hypothesis testing to make quantitative statements about the magnitude of each singular value and the rank of the matrix.
A framework for Quality of Service analysis of IP based video networks
(2003-10-23) Chandrasekhar, Vinay; Dr. Khaled Harfoush, Committee Member; Dr. David Thuente, Committee Member; Dr. Mladen Vouk, Committee Chair
Applications that use real time video are becoming increasingly necessary for effective communication over the Internet. Their popularity is increasing in areas such as distance learning, distributed research and video conferencing. Since real time video has special Quality of Service (QoS) requirements for it to be acceptable to end users, administrators are faced with challenges that involve the end-to-end ability of participating systems to support real time video communication. Unfortunately, the tools that exist in the market today are either expensive and closed source, or are generic and do not explicitly consider the characteristics of real time video. This work proposes a framework for assessment of end-to-end Quality of Service capabilities for support of real-time Variable Bit Rate (VBR) compressed video communication. This framework evaluates the endpoints and their inter-connectivity and determines the extent of their ability to support compressed video over User Datagram Protocol (UDP). The framework generates VBR traffic that mimics compressed low bit rate video, maintains and reports detailed and summary statistics of each test. User configurable options include specifying activity levels and bandwidth upper bounds. The framework itself is modular and extensible, allowing for functionalities to be added or replaced as the testing requirements change. The framework defines a methodology to conduct video QoS tests, and describes how the test cycles are to be conducted for determining the Quality of Service support. Various metrics that are indicative of the Quality of Service for real time video are listed and described. Traffic generation and transmission requirements for mimicking video traffic are explained. At each participating end point, the values of the various QoS metrics are maintained in real time using specific Simple Network Management Protocol (SNMP) Management Information Bases (MIBs) designed for use with this framework. A tool is implemented based on this framework, and its performance as a video QoS measurement tool is studied. The tool uses an Iperf based traffic generation module, that reads from trace files in order to generate test traffic. The trace files contain datagram size and timing information required to generate video-like VBR traffic. An SNMP MIB implementation is provided to record the QoS metrics determined during the tests. A separate monitoring program queries the endpoint MIBs, tabulates and displays the QoS information. This information can then be used to determine end-to-end support.
Network Design and Optimization - with Applications in Optical and Wireless Networks
(2006-09-07) Huang, Shu; Dr. Carla Savage, Committee Member; Dr. David Thuente, Committee Member; Dr. Harry Perros, Committee Member; Dr. Rudra Dutta, Committee Chair
Typically, enterprise networks may incorporate Wide Area Networks (WANs),Metro Area Networks (MANs) and Local Area Networks (LANs). These networks have very different characteristics in terms of the physical media, data link layer and network layer protocols they use. In all these networks, well-designed network architectures are the key to achieving high performance at reasonable costs. We study the optimization problems that have arisen in the design of different networks, specifically, optical networks and wireless networks. In optical networks, we study the traffic grooming problem in Wavelength Division Multiplexing (WDM) networks with dynamic traffic demands. We present of detailed study of current research in this area and propose a new design problem for both single-link and multi-link networks. In wireless networks, we present new formulations for the design problem in Wireless Mesh Networks (WMN) that take different interference models into consideration and propose algorithmic methods to solve them.
Performance Evaluation of Handoff between UMTS/802.11b based on Mobile IP and Stream Control Transmission Protocol
(2005-08-18) Song, Jung Kee; Dr. Wenye Wang, Committee Chair; Dr. David Thuente, Committee Member; Dr. Arne A. Nilsson, Committee Member
Various wireless networks, contemporarily, have evolved as prime communication methods, encountering convergence paradigm among heterogeneous technologies including applications based on IP, which has given enormous impacts in our way of life because of its proved robustness and scalability as well as ample services. With the synergy of the two technologies, ubiquitous access to all-IP information sources has become reality. For wireless IP services, IP mobility is one of the major issues that should be resolved. Especially, the performance of handoff mechanism is a pithy issue that determines the performance of application level services. Although Mobile IP (MIP) and its extensions, as network layer solutions, have been proposed and standardized, their handoff mechanisms bring unavoidable transmission throughput degradation due to packet loss, registration delay, and transport layer blocking. Moreover, to accommodate MIP, significant quantity of modifications should be brought into each heterogeneous network architecture. In this thesis, we evaluate the performance of a transport layer handoff approach, mobile SCTP (mSCTP), and compare it with that of a network layer solution, MIP. mSCTP is based on Stream Control Transmission Protocol (SCTP), the third general purpose transport layer protocol, standardized by IETF. SCTP conceptually enables seamless handoff in transport layer without any change in IP protocol stack by its multi-homing feature and dynamic address reconfiguration (DAR) extension. We analyze the performance of mSCTP and MIP by introducing handoff delay, end-to-end transmission throughput, and packet loss, and conduct a simulation study of the two protocols in 802.11b WLAN-only and UMTS/802.11b integrated networks using NS-2 network simulator.
Power Conservation in wireless sensor network using receiver switch-off
(2004-03-28) Goel, Apurva Ajay; Dr. David Thuente, Committee Member; Dr. Rudra Dutta, Committee Chair; Dr. Mihail Sichitiu, Committee Member
A wireless sensor network is a network of nodes equipped with sensors and capable of relaying their data to the monitoring station using multi-hop communication. Essentially, they operate using the ad hoc paradigm due to the unpredictable and dynamic topology. The lifetime of the nodes is limited to that of its battery. Thus in this work we try to conserve the battery by putting the nodes to sleep when they are not involved in any communication process. Sensor nodes produce readings every fixed time interval and we try to leverage the predictability of the packet inter-arrival time to obtain the sleep periods. We model the inter-arrival times that a node observes in the form of a probability distribution function. A node may be forwarding packets from multiple sources and it associate this knowledge to every stream it is carrying. After a node captures a packet, using its knowledge it can predict the approximate arrival time of the next packet. It can try to sleep during this period. It thus conserves power. In our study we show that simply forwarding the packet when they are received causes the predictability needed for this approach to be lost. We then propose an alternative approach, which is to maintain some local periodicity. This can be accomplished by delaying the packets. The multi-hop nature of the network causes these delays to accumulate and thus we have to consider the tolerances of the data packets to these delays. We then discuss the relation between the sleep durations possible and the delay tolerances of the packets. Finally, we explore the inter-dependence between packet loss, sleep duration, delay and the ultimate effect on the power savings.
Resource Optimization and Security in Distributed Computing
(2009-12-11) Xiong, Kaiqi; Mihail Devetsikiotis, Committee Member; Matthias Stallmann, Committee Member; Dr. David Thuente, Committee Member; Harry Perros, Committee Chair
With the number of e-Business applications dramatically increasing, service level agreements (SLA) will play an important part in distributed service computing. An SLA is a combination of several quality of service (QoS) metrics, such as security, performance, and availability, agreed between a customer and a service provider. Due to the complexity of these metrics, most existing research typically addresses only one of these QoS metrics. In the case of the response time as a performance metric, the average time to process and complete a job is typically used in the literature. However, this may not be of real interest to a customer. A statistically bounded metric, that is, a percentile response time, is more realistic than the average response time. Moreover, in enterprise service computing, customer requests are typically distinguished by different request characteristics and service requirements. This dissertation includes a study of trustworthiness, percentile response time, service availability, and authentication among service stations or sites that may be owned by different service providers. The first part of this dissertation contains an analysis of percentile response time, which is one of the most important SLA metrics. Effective and accurate numerical solutions for the calculation of the percentile response time in single-class and multi-class queueing networks are obtained. Then, the numerical solution is incorporated in a resource allocation problem. Specifically, we present an approach for the resource optimization that minimizes the total cost of computer resources required while preserving a given percentile of the response time. In the second part of this dissertation, we extend the approach to consider trustworthiness, service availability, and the percentile of response time in Web services. We clearly define these QoS metrics and provide their quantitative analysis. Then, we take into account these QoS metrics in a trust-based resource allocation problem in which a set of computer resources is used by a service provider to host a typical Web services application for single-class customer services and multiple-class customer services respectively. We formulate the trust-based resource allocation problem as an optimization problem under SLA constraints in which we calculate the number of servers in each service site that minimize a cost function that reflects operational costs for single-class customer services and multiple-class customer services respectively. We solve this problem using an efficient numerical procedure. Experimental results show the applicability of the procedure and validate its accuracy. Finally, in the third part of this dissertation we first present a thorough performance evaluation of two notable public key cryptography-based authentication techniques, Public-Key Cross Realm Authentication in Kerberos (PKCROSS) and Public Key Utilizing Tickets for Application Servers (PKTAPP, a.k.a. KX.509/KCA), in terms of computational and communication times. We then demonstrate their performance difference using queueing networks. PKTAPP was proposed to address the scalability issue of PKCROSS. However, our in-depth analysis of these two techniques shows that PKTAPP does not perform better than PKCROSS in a large scale system. Thus, we propose a new public key cryptography-based group authentication technique. Our performance analysis demonstrates that the new technique can scale better than PKCORSS and PKTAPP.
Techniques For Finding Nash Equilibria In Combinatorial Auctions
(2005-07-07) Sureka, Ashish; Dr. Peter Wurman, Committee Chair; Dr. Munindar Singh, Committee Member; Dr. Michael Young, Committee Member; Dr. David Thuente, Committee Member
Auctions that allow participants to bid on a combination of items rather than just the individual items are called combinatorial auctions. For items that exhibit complementarity and substitutability, combinatorial auctions can be used to reach economically efficient allocations of goods and services. There has been a surge of recent research on combinatorial auctions because of the wide variety of practical situations to which they can be applied. There are several instances in which combinatorial auctions have already been applied to allocate scares resources, but there are still some challenging issues that need to be addressed before combinatorial auctions can be much more widely used in practice. Many different combinatorial auctions designs have been proposed by researchers and recently there has been a lot of work on studying the computational and strategic aspects of these auction designs. In this thesis, I analyze combinatorial auctions from a game theoretic perspective and propose techniques for determining pure strategy Nash equilibrium of combinatorial auctions. For a variety of reasons, combinatorial auctions pose serious computational challenges to compute Nash equilibria using current techniques. One problem is that the size of the strategy space in combinatorial auctions is very large and grows exponentially with the number of bidders and items. Another computational issue is that for combinatorial auctions it is computationally expensive to compute the payoffs of the players as a result of the joint actions. This makes it computationally expensive to determine the complete payoff matrix upfront and then determine Nash equilibrium. In this dissertation, we present techniques to overcome these problems. We present algorithms based on meta-heuristic search techniques, best response dynamics and linear programming to tackle these problems. We present empirical and theoretical results to support our claim that the algorithms perform well.
Wireless MAC Protocol Design and Analysis
(2009-06-01) Jang , Beakcheol; Dr. Mihail L. Sichitiu, Committee Chair; Dr. David Thuente, Committee Member; Dr. Khaled Harfoush, Committee Member; Dr. Rudra Dutta., Committee Member
Wireless networks are becoming very common due to their advantages such as rapid deployment and support for mobility. In this dissertation, we design and analyze the Medium Access Control (MAC) protocol for two popular wireless networks: Wireless Sensor Networks (WSNs) and Wireless Local Area Networks (WLANs). For WSNs, we design and analyze an energy efficient MAC protocols. Energy efficiency is a key design factor of a MAC protocol for WSNs. Existing preamble-sampling based MAC protocols have large overheads due to their preambles and are inefficient at large wakeup intervals. Synchronous scheduling MAC protocols minimize the preamble by combining preamble sampling and scheduling techniques; however, they do not prevent energy loss due to overhearing. In this dissertation, we present an energy efficient MAC protocol for WSNs, called AS-MAC, that avoids overhearing and reduces contention and delay by asynchronously scheduling the wakeup time of neighboring nodes. We also provide a multi-hop energy consumption model for AS-MAC. To validate our design and analysis, we implement the proposed scheme on the MICAz and TELOSB platforms. Experimental results show that AS-MAC considerably reduces energy consumption, packet loss and delay when compared with other energy efficient MAC protocols. For WLANs, we present a saturation throughput model for IEEE 802.11, the standard of WLAN, for a simple infrastructure scenario with hidden stations. Despite the importance of the hidden terminal problem, there have been a relatively small number of studies that consider the effect of hidden terminals on IEEE 802.11 throughput. Moreover, existing models are not accurate for scenarios with the short-term unfairness. In this dissertation, we present a new analytical saturation throughput model for IEEE 802.11 for a simple but typical infrastructure scenario with small number of hidden stations. Simulation results are used to validate the model and show that our model is extremely accurate. Lastly, we provide a saturation throughput model for IEEE 802.11 for the general infrastructure scenario with hidden stations. Simulation results show that this generalized model is reasonably accurate.