Thesis (Ph. D.)--Carnegie Mellon University, 2003. / Title from PDF title page (viewed Dec. 18, 2009). Includes bibliographical references (p. 49-52).

Thesis (M.S. in Computer Science)--Naval Postgraduate School, September 2003. / Thesis advisor(s): Geoffrey Xie, John Gibson. Includes bibliographical references (p. 75-76). Also available online.

Thesis (M.B.A.)--University of Hong Kong, 1987.

Current home networks typically involve two or more machines sharing network resources. The vision for the home network has grown from a simple computer network, to every day appliances embedded with network capabilities. In this environment devices and services within the home can interoperate, regardless of protocol or platform. Network clients can discover required resources by performing network discovery over component descriptions. Common approaches to this discovery process involve simple matching of keywords or attribute/value pairings. Interest emerging from the Semantic Web community has led to ontology languages being applied to network domains, providing a logical and semantically rich approach to both describing and discovering network components. In much of the existing work within this domain, developers have focused on defining new description frameworks in isolation from existing protocol frameworks and vocabularies. This work proposes an ontology-based description framework which takes the ontology approach to the next step, where existing description frameworks are in- corporated into the ontology-based framework, allowing discovery mechanisms to cover multiple existing domains. In this manner, existing protocols and networking approaches can participate in semantically-rich discovery processes. This framework also includes a system architecture developed for the purpose of reconciling existing home network solutions with the ontology-based discovery process. This work also describes an implementation of the approach and is deployed within a home-network environment. This implementation involves existing home networking frameworks, protocols and components, allowing the claims of this work to be examined and evaluated from a ‘real-world’ perspective.

No description available.

The dissertation addresses the communication issues of interconnecting the different government sectors LANs, and access to the global Internet. Network capacities are being purposely overengineered in today's commercial Internet. Any network provider, be it a commercial Internet Service Provider (ISP) or Information Technology Service department at government sector, company or university site, will design network bandwidth resources in such a way that there will be virtually no data loss, even during the worst possible network utilization scenario. Thus, the service delivered by today's end-to-end wide area Internet would be perfect if it wasn't for the inter-domain connections, such as Internet access link to the ISP, or peering points between ISPs. The thesis studies the performance of the network in Eritrea, displaying the problems of Local Area Networks (LANs) and Wide Area Networks (WAN) and suggesting initial solutions and investigating the performance of (WAN) through the measured traffic analysis between Asmara LAN and Massawa LAN, using queuing theory system (M/M/1 and M/M/2) solution. The dissertation also uses OPNET IT Guru simulation software program ·to study the performance of LAN and WLAN in Eritrea. The items studied include traffic, collision, packet loss, and queue delay. Finally in order to follow the current trends, we study the performance ofVOIP links in Eritrean WANs environment, with a focus on five different link capacities: 28 kbps, 33 kbps, 64 kbps, and 128 kbps for voice and 256/512 kbps for voice and data. Using the R value as a measure of mean opinion score (MOS), we determine that the 33 kbps link would be adequate for Eritrean WANs. / Thesis (M.Sc.Eng.)-University of KwaZulu-Natal, Durban, 2006.

No description available.

Traffic anomaly detection is a standard task for network administrators, who with experience can generally differentiate anomalous traffic from normal traffic. Many approaches have been proposed to automate this task. Most of them attempt to develop a sufficiently sophisticated model to represent the full range of normal traffic behaviour. There are significant disadvantages to this approach. Firstly, a large amount of training data for all acceptable traffic patterns is required to train the model. For example, it can be perfectly obvious to an administrator how traffic changes on public holidays, but very difficult, if not impossible, for a general model to learn to cover such irregular or ad-hoc situations. In contrast, in the proposed method, a number of models are gradually created to cover a variety of seen patterns, while in use. Each model covers a specific region in the problem space. Any novel or ad-hoc patterns can be covered easily. The underlying technique is a knowledge acquisition approach named Ripple Down Rules. In essence we use Ripple Down Rules to partition a domain, and add new partitions as new situations are identified. Within each supposedly homogeneous partition we use fairly simple statistical techniques to identify anomalous data. The special feature of these statistics is that they are reasonably robust with small amounts of data. This critical situation occurs whenever a new partition is added. We have developed a two knowledge base approach. One knowledge base partitions the domain. Within each domain statistics are accumulated on a number of different parameters. The resultant data are passed to a knowledge base which decides whether enough parameters are anomalous to raise an alarm. We evaluated the approach on real network data. The results compare favourably with other techniques, but with the advantage that the RDR approach allows new patterns of use to be rapidly added to the model. We also used the approach to extend previous work on prudent expert systems - expert systems that warn when a case is outside its range of experience. Of particular significance we were able to reduce the false positive to about 5%.

With the extremely rapid development of the Internet in recent years, emerging peer-to-peer network overlays are meeting the requirements of a more sophisticated communications environment, providing a useful substrate for applications such as scalable file sharing, data storage, large-scale multicast, web-cache, and publish-subscribe services. Due to its design flexibility, peer-to-peer networks can offer features including self-organization, fault-tolerance, scalability, load-balancing, locality and anonymity. As the Internet grows, there is an urgent requirement to understand real-time network performance degradation. Measurement tools currently used are ping, traceroute and variations of these. SNMP (Simple Network Management Protocol) is also used by network administrators to monitor local networks. However, ping and traceroute can only be used temporarily, SNMP can only be deployed at certain points in networks and these tools are incapable of sharing network measurements among end-users. Due to the distributed nature of networking performance data, peer-to-peer overlay networks present an attractive platform to distribute this information among Internet users. This thesis aims at investigating the desirable locality property of peer-to-peer overlays to create an application to share Internet measurement performance. When measurement data are distributed amongst users, it needs to be localized in the network allowing users to retrieve it when external Internet links fail. Thus, network locality and robustness are the most desirable properties. Although some unstructured overlays also integrate locality in design, they fail to reach rarely located data items. Consequently, structured overlays are chosen because they can locate a rare data item deterministically and they can perform well during network failures. In structured peer-to-peer overlays, Tapestry, Pastry and Chord with proximity neighbour selection, were studied due to their explicit notion of locality. To differentiate the level of locality and resiliency in these protocols, P2Psim simulations were performed. The results show that Tapestry is the more suitable peer-to-peer substrate to build such an application due to its superior localizing data performance. Furthermore, due to the routing similarity between Tapestry and Pastry, an implementation that shares network measurement information was developed on freepastry, verifying the application feasibility. This project also contributes to the extension of P2Psim to integrate with GT-ITM and link failures.

In the past few years Wireless Mesh Networks (WMNs) have developed as a promising technology to provide flexible and low-cost broadband network services. The Expected Transmission Count (ETX) routing metric has been put forward recently as an advanced routing metric to provide high QoS for static WMNs. Most previous research in this area suggests that ETX outperforms other routing metrics in throughput and efficiency. However, it has been determined that ETX is not immune to load sensitivity and route oscillations in a single radio environment. Route oscillations refer to the situation where packet transmission switches between two or more routes due to congestion. This has the effect of degrading performance of the network, as the routing protocol may select a non optimal path. In this thesis we avoided the route oscillation problem by forcing data transmission on fixed routes. This can be implemented in the AODV (Ad hoc On-demand Distance Vector) protocol by disabling both error messages and periodic updating messages (the HELLO scheme). However, a critical factor for our approach is that ETX must determine a high quality initial route in AODV. This thesis investigates whether the ETX metric improves initial route selection in AODV compared to the HOPS metric in two representative client-server applications: the Traffic Control Network (TCN) and the Video Stream (VS) network. We evaluate the ETX and HOPS metrics in a range of scenarios which possess different link qualities and different traffic loads. We find the ETX metric greatly improves initial route selection in AODV compared to the HOPS in the network in which only single flow exists. For networks in which there are multiple simultaneous flows, ETX behaves similar to HOPS in initial route selection. Based on these results, we find the solution of route stabilization to route oscillations in the context of ETX is only useful in the single flow case. To address this problem, we propose a modified solution of repeatedly broadcasting RREQ (Route Request) packets. Simulation results show that our modified solution allows ETX to be useful in the initial route selection in both single flow and multiple simultaneous flows cases.