No description available.

ITC/USA 2007 Conference Proceedings / The Forty-Third Annual International Telemetering Conference and Technical Exhibition / October 22-25, 2007 / Riviera Hotel & Convention Center, Las Vegas, Nevada / The Eastern and Western Ranges along with DISA share a similar vision around Net-Centricity such that "Anyone, anywhere can get to any data source and exploit the information they are authorized to access." Their legacy infrastructure is built around TDM and ATM transport networks, which are link based and connection oriented. To achieve the vision the infrastructure must evolve towards a packet switched network (PSN) that is meshed based. Consequently, a means to interwork non-IP enabled services is required. Pseudo Wire protocol encapsulation provides the means for extending telemetry, data, voice, and video services in native formats over Ethernet, IP, and MPLS networks in a reliable way that delivers greater operational efficiency and a smooth migration to a single converged network.

Transport Control Protocol (TCP) is the dominant protocol in modern communication networks, in which the issues of reliability, flow, and congestion control must be handled efficiently. This thesis studies the impact of the next-generation bufferless optical burst-switched (OBS) networks on the performance of TCP congestion-control implementations (i.e., dropping-based, explicit-notification-based, and delay-based). The burst contention phenomenon caused by the buffer-less nature of OBS occurs randomly and has a negative impact on dropping-based TCP since it causes a false indication of network congestion that leads to improper reaction on a burst drop event. In this thesis we study the impact of these random burst losses on dropping-based TCP throughput. We introduce a novel congestion control scheme for TCP over OBS networks, called Statistical Additive Increase Multiplicative Decrease (SAIMD). SAIMD maintains and analyzes a number of previous round trip times (RTTs) at the TCP senders in order to identify the confidence with which a packet-loss event is due to network congestion. The confidence is derived by positioning short-term RTT in the spectrum of long-term historical RTTs. The derived confidence corresponding to the packet loss is then taken in to account by the policy developed for TCP congestion-window adjustment. For explicit-notification TCP, we propose a new TCP implementation over OBS networks, called TCP with Explicit Burst Loss Contention Notification (TCP-BCL). We examine the throughput performance of a number of representative TCP implementations over OBS networks, and analyze the TCP performance degradation due to the misinterpretation of timeout and packet-loss events. We also demonstrate that the proposed TCP-BCL scheme can counter the negative effect of OBS burst losses and is superior to conventional TCP architectures in OBS networks. For delay-based TCP, we observe that this type of TCP implementation cannot detect network congestion when deployed over typical OBS networks since RTT fluctuations are minor. Also, delay-based TCP can suffer from falsely detecting network congestion when the underlying OBS network provides burst retransmission and/or deflection. Due to the fact that burst retransmission and deflection schemes introduce additional delays for bursts that are retransmitted or deflected, TCP cannot determine whether this sudden delay is due to network congestion or simply to burst recovery at the OBS layer. In this thesis we study the behaviour of delay-based TCP Vegas over OBS networks, and propose a version of threshold-based TCP Vegas that is suitable for the characteristics of OBS networks. The threshold-based TCP Vegas is able to distinguish increases in packet delay due to network congestion from burst contention at low traffic loads. The evolution of OBS technology is highly coupled with its ability to support upper-layer applications. Without fully understanding the burst transmission behaviour and the associated impact on the TCP congestion-control mechanism, it will be difficult to exploit the advantages of OBS networks fully.

Transport Control Protocol (TCP) is the dominant protocol in modern communication networks, in which the issues of reliability, flow, and congestion control must be handled efficiently. This thesis studies the impact of the next-generation bufferless optical burst-switched (OBS) networks on the performance of TCP congestion-control implementations (i.e., dropping-based, explicit-notification-based, and delay-based). The burst contention phenomenon caused by the buffer-less nature of OBS occurs randomly and has a negative impact on dropping-based TCP since it causes a false indication of network congestion that leads to improper reaction on a burst drop event. In this thesis we study the impact of these random burst losses on dropping-based TCP throughput. We introduce a novel congestion control scheme for TCP over OBS networks, called Statistical Additive Increase Multiplicative Decrease (SAIMD). SAIMD maintains and analyzes a number of previous round trip times (RTTs) at the TCP senders in order to identify the confidence with which a packet-loss event is due to network congestion. The confidence is derived by positioning short-term RTT in the spectrum of long-term historical RTTs. The derived confidence corresponding to the packet loss is then taken in to account by the policy developed for TCP congestion-window adjustment. For explicit-notification TCP, we propose a new TCP implementation over OBS networks, called TCP with Explicit Burst Loss Contention Notification (TCP-BCL). We examine the throughput performance of a number of representative TCP implementations over OBS networks, and analyze the TCP performance degradation due to the misinterpretation of timeout and packet-loss events. We also demonstrate that the proposed TCP-BCL scheme can counter the negative effect of OBS burst losses and is superior to conventional TCP architectures in OBS networks. For delay-based TCP, we observe that this type of TCP implementation cannot detect network congestion when deployed over typical OBS networks since RTT fluctuations are minor. Also, delay-based TCP can suffer from falsely detecting network congestion when the underlying OBS network provides burst retransmission and/or deflection. Due to the fact that burst retransmission and deflection schemes introduce additional delays for bursts that are retransmitted or deflected, TCP cannot determine whether this sudden delay is due to network congestion or simply to burst recovery at the OBS layer. In this thesis we study the behaviour of delay-based TCP Vegas over OBS networks, and propose a version of threshold-based TCP Vegas that is suitable for the characteristics of OBS networks. The threshold-based TCP Vegas is able to distinguish increases in packet delay due to network congestion from burst contention at low traffic loads. The evolution of OBS technology is highly coupled with its ability to support upper-layer applications. Without fully understanding the burst transmission behaviour and the associated impact on the TCP congestion-control mechanism, it will be difficult to exploit the advantages of OBS networks fully.

Pocket Switched Networks (PSNs) offer a means of infrastructureless inter-human communication by utilising Delay and Disruption Tolerant Networking (DTN) technology. However, creating PSNs involves solving challenges which were not encountered in the Deep Space Internet for which DTN technology was originally intended.End-to-end communication over multiple hops in PSNs is a product of short range opportunistic wireless communication between personal mobile wireless devices carried by humans. Opportunistic data delivery in PSNs is far less predictable than in the Deep Space Internet because human movement patterns are harder to predict than the orbital motion of satellites. Furthermore, PSNs require some scheme for efficient neighbour discovery in order to save energy and because mobile devices in PSNs may be unaware of when their next encounter will take place.This thesis offers novel solutions for neighbour discovery and opportunistic data delivery in PSNs that make practical use of dynamic inter-human encounter patterns.The first contribution is a novel neighbour discovery algorithm for PSNs called PISTONS which relies on a new inter-probe time calculation (IPC) and the bursty encounter patterns of humans to set the time between neighbour discovery scans. The IPC equations and PISTONS also give participants the ability to easily specify their required level of connectivity and energy saving with a single variable.This thesis also contains novel distributed spatio-temporal clustering and opportunistic data delivery algorithms for PSNs which can be used to deliver data over multiple hops. The spatio-temporal clustering algorimths are also used to analyse the social networks and transient groups which are formed when humans interact.

Wavelength division multiplexed (WDM) optical networks are rapidly becoming the technology of choice in next-generation Internet architectures. This dissertation addresses the important issues of improving four aspects of optical networks, namely, routing efficiency, fairness, differentiated quality of service (QoS) and throughput. A new approach for implementing efficient routing and wavelength assignment in WDM networks is proposed and evaluated. In this approach, the state of a multiple-fiber link is represented by a compact bitmap computed as the logical union of the bitmaps of the free wavelengths in the fibers of this link. A modified Dijkstra's shortest path algorithm and a wavelength assignment algorithm are developed using fast logical operations on the bitmap representation. In optical burst switched (OBS) networks, the burst dropping probability increases as the number of hops in the lightpath of the burst increases. Two schemes are proposed and evaluated to alleviate this unfairness. The two schemes have simple logic, and alleviate the beat-down unfairness problem without negatively impacting the overall throughput of the system. Two similar schemes to provide differentiated services in OBS networks are introduced. A new scheme to improve the fairness of OBS networks based on burst preemption is presented. The scheme uses carefully designed constraints to avoid excessive wasted channel reservations, reduce cascaded useless preemptions, and maintain healthy throughput levels. A new scheme to improve the throughput of OBS networks based on burst preemption is presented. An analytical model is developed to compute the throughput of the network for the special case when the network has a ring topology and the preemption weight is based solely on burst size. The analytical model is quite accurate and gives results close to those obtained by simulation. Finally, a preemption-based scheme for the concurrent improvement of throughput and burst fairness in OBS networks is proposed and evaluated. The scheme uses a preemption weight consisting of two terms: the first term is a function of the size of the burst and the second term is the product of the hop count times the length of the lightpath of the burst.

Sensor devices and the emergent networks that they enable are capable of transmitting information between data sources and a permanent data sink. Since these devices have low-power and intermittent connectivity, latency of the data may be tolerated in an effort to save energy for certain classes of data. The BUBBLE routing algorithm developed by Hui et al. in 2008 provides consistent routing by employing a model which computes individual nodes popularity from sets of nodes and then uses these popularity values for forwarding decisions. This thesis considers enhancements to BUBBLE based on the hypothesis that nodes do form groups and certain centrality values of nodes within these groups can be used to improve routing decisions further. Built on this insight, there are two algorithms proposed in this thesis. First is the Community-Based- Forwarding (CBF), which uses pairwise group interactions and pairwise node-to-group interactions as a measure of popularity for routing messages. By having a different measure of popularity than BUBBLE, as an additional factor in determining message forwarding, CBF is a more conservative routing scheme than BUBBLE. Thus, it provides consistently superior message transmission and delivery performance at an acceptable delay cost in resource constrained environments. To overcome this drawback, the concept of unique interaction pattern within groups of nodes is introduced in CBF and it is further renewed into an enhanced algorithm known as Hybrid-Community-Based- Forwarding (HCBF). Utilizing this factor will channel messages along the entire path with consideration for higher probability of contact with the destination group and the destination node. Overall, the major contribution of this thesis is to design and evaluate an enhanced social based routing algorithm for resource-constrained Pocket Switched Networks (PSNs), which will optimize energy consumption related to data transfer. It will do so by explicitly considering features of communities in order to reduce packet loss while maintaining high delivery ratio and reduced delay.

A unique all-optical 1x2 self-routed switch is introduced. This switch routes an optical packet from one input to one of two possible outputs. The header and payload are transmitted separately in the system, and the header bits are processed in parallel thus increasing the switching speed as well as reducing the amount of buffering required for the payload. A 1x2 switching operation is analysed and a switching ratio of up to 14dB is obtained. The objective of the research was to investigate a unique all-optical switch. The switch works by processing the optical bits in a header packet which contains the destination address for a payload packet. After the destination address is processed the optical payload packet gets switched to one of two outputs depending on the result of the optical header processing. All-optical packet switching in the optical time domain was accomplished by making use of all-optical parallel processing of an optical packet header. This was demonstrated in experiments in which a three bit parallel processing all-optical switching node was designed, simulated and used to successfully demonstrate the concept. The measure of success that was used in the simulated experiments was the output switching ratio, which is the ratio between the peak optical power of a high bit at the first output and the peak optical power of a high bit at the second output. In all experimental results the worst case scenario was looked at, which means that if there was any discrepancy in the peak value of the output power then the measurement’s minimum/maximum value was used that resulted in a minimum value for the switching ratio. The research resulted in an optical processing technique which took an optical bit sequence and delivered a single output result which was then used to switch an optical payload packet. The packet switch node had two optical fibre inputs and two optical fibre outputs. The one input fibre carried the header packet and the other input fibre carried the payload packet. The aim was to switch the payload packet to one of the two output fibres depending on the bit sequence within the header packet. Also only one unique address (header bit sequence) caused the payload packet to exit via one of the outputs and all the other possible addresses caused the payload packet to exit via the other output. The physical configuration of the all-optical switches in the parallel processing structure of the switching node determined for which unique address the payload packet would exit via a different output than when the address was any of the other possible combinations of sequences. Only three Gaussian shaped bits were used in the header packet at a data rate of 10 Gbps and three Gaussian shaped bits in the payload packet at a data rate of 40 Gbps, but in theory more bits can be used in the payload packet at a decreased bit length to increase throughput. More bits can also be used in the header packet to increase the number of addresses that can be reached. In the simulated experiments it was found that the payload packet would under most circumstances exit both outputs, and at one output it would be much larger than at the other output (where it was normally found to be suppressed when compared to the other output’s optical power). The biggest advantage of this method of packet-switching is that it occurs all-optically, meaning that there is no optical to electronic back to optical conversions taking place in order to do header processing. All of the header processing is done optically. One of the disadvantages is that the current proposed structure of the all-optical switching node uses a Cross-Gain Modulator (XGM) switch which is rather expensive because of the Semiconductor Optical Amplifier (SOA). In this method of packet-switching the length of the payload packet cannot exceed the length of one bit of the header packet. This is because the header processing output is only one header bit length long and this output is used to switch the payload packet. Thus any section of the payload packet that is outside this header processing output window will not be switched correctly / Dissertation (MEng (Electronic Engineering))--University of Pretoria, 2007. / Electrical, Electronic and Computer Engineering / unrestricted

Network infrastructures often need to stage content so that it is accessible to consumers. The standard solution, deploying the content on a centralised server, can be inadequate in several situations. Our thesis is that information encoded in social networks can be used to tailor content staging decisions to the user base and thereby build better data delivery infrastructures. This claim is supported by two case studies, which apply social information in challenging situations where traditional content staging is infeasible. Our approach works by examining empirical traces to identify relevant social properties, and then exploits them. The first study looks at cost-effectively serving the ``Long Tail'' of rich-media user-generated content, which need to be staged close to viewers to control latency and jitter. Our traces show that a preference for the unpopular tail items often spreads virally and is localised to some part of the social network. Exploiting this, we propose Buzztraq, which decreases replication costs by selectively copying items to locations favoured by viral spread. We also design SpinThrift, which separates popular and unpopular content based on the relative proportion of viral accesses, and opportunistically spins down disks containing unpopular content, thereby saving energy. The second study examines whether human face-to-face contacts can efficiently create paths over time between arbitrary users. Here, content is staged by spreading it through intermediate users until the destination is reached. Flooding every node minimises delivery times but is not scalable. We show that the human contact network is resilient to individual path failures, and for unicast paths, can efficiently approximate flooding in delivery time distribution simply by randomly sampling a handful of paths found by it. Multicast by contained flooding within a community is also efficient. However, connectivity relies on rare contacts and frequent contacts are often not useful for data delivery. Also, periods of similar duration could achieve different levels of connectivity; we devise a test to identify good periods. We finish by discussing how these properties influence routing algorithms.

Orientador: Helio Waldman / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-06T02:56:20Z (GMT). No. of bitstreams: 1 Campelo_DivanilsonRodrigodeSousa_D.pdf: 945421 bytes, checksum: 52541c616bcfcfd8ae2d0a50c597fcc3 (MD5) Previous issue date: 2006 / Resumo: Investigamos o problema de estimar valores de probabilidades de bloqueio em redes de caminhos ópticos com topologias lineares. Apresentamos um melhor substituto para a suposição de independência de enlaces em redes de topologia linear: a suposição de independência de objetos. Apresentamos a prova assintótica desta suposição para redes lineares infinitas com um único canal, e mostramos que a expressão assintótica é uma aproximação muito boa para anéis finitos de qualquer tamanho. Para o caso de múltiplos comprimentos de onda, apresentamos novas aproximações de carga reduzida para anéis WDM com restrição de continuidade de comprimento de onda. Para anéis com conversão plena de comprimentos de onda, propomos um método matricial inovador que permite cálculos exatos de probabilidades de bloqueio e taxa de ocupação nestas redes. Um método "escalável" para a obtenção da constante de normalização do modelo clássico de Erlang também é apresentado. Por fim, analisamos o desempenho de meios lineares bloqueantes. Apresentamos expressões exatas para o throughput em meios compartimentalizados e não-compartimentalizados, e quantificamos os ganhos de compartimentalização em meios lineares / Abstract: We address the problem of estimating blocking probabilities in optical path networks with linear topologies. We present a better substitute for the link independence assumption in networks with linear topology: the object independence assumption. We present an asymptotic proof of this assumption for in?nite single-channel networks, and we show that the asymptotic expression is a very good approximation for ?nite rings with any size. In the case of multiple wavelengths, we present new reduced load approximations for WDM rings with wavelength continuity constraint. For rings with full wavelength conversion, we propose an innovative matrix-based method for calculating exact values of blocking probabilities and occupancy rates in such networks. A scalable method for deriving the normalization constant of the Erlang¿s classical model is also presented. Finally, we analyze the performance of linear blocking media. We present exact expressions for the throughput in slotted and unslotted media, and we quantify the slotting gains in linear media / Doutorado / Telecomunicações e Telemática / Doutor em Engenharia Elétrica