Improving the robustness of today's intra-domain link-state networks is of increasing importance. This is because it is becoming commonplace for such networks to carry telephony, on-demand video and other kinds of real-time traffic. Since user requirements on real-time traffic are exacting, it is essential that the network is able to respond rapidly to network errors. A way of reaching fast reaction times in response to network errors is to precompute fail-over paths that can be utilized immediately when an error is detected. The fail-over path is thereafter in use until the network has converged, handling traffic that would otherwise have been lost due to the error. During convergence (when new paths are calculated) there is a caveat that needs to be avoided -the formation of temporary loops. Temporary loops can form naturally during convergence in link-state networks and can unfortunately prevent traffic from reaching a fail-over path. Traffic can therefore be lost even though a fail-over path is in place. We present an algorithm that can be used in conjunction with link-state routing, to ensure that temporary loops do not form during convergence. The algorithm improves on previous loop-free algorithms for link-state routing by reducing the number of state transitions necessary before a router can update its routing table to a new, guaranteed to exist, path for a wide variety of topological configurations.

Power-aware routing algorithms aim to extend the lifetime of a wireless network by avoiding to route over nodes that are close to being depleted of energy. Previous work in such algorithms often require each node in the network to be aware of the topology of the entire network. In this paper we investigate power-aware routing in Bluetooth networks. Routing in a Bluetooth network involves creating a scatternet structure in order to facilitate the communication between nodes in different piconets. We propose a new routing algorithm that does not require each node to have any knowledge of the network except for its neighbors. This is done by collecting information regarding potential routing paths in the routing messages themselves. We also present a simplified version of the algorithm for general wireless networks, which does not involve scatternet formation and therefore requires less communication overhead.

We present a modification to the internal border gateway protocol that prevents oscillation. We deal with oscillations caused by multi-exit-discriminators within a single autonomous system. The algorithm keeps track of all valid routes it has received. Then the best route is chosen from the set of valid routes. There is no need for extra messages as compared to the standard internal border gateway protocol. As long as there is just one route to a destination, the algorithm stays dormant and the properties of the standard internal border gateway protocol are preserved.

Godkänd; 2004; 20061127 (ysko)

Layering techniques for video transmission are being actively considered as a way of satisfying receivers in a heterogeneous environment. A layering scheme based on a combination of network constrains and the perceptual quality of video is an attractive proposition. In this paper a feed-forward post-codec controller has been described for sending a video sequence in an optimally layered way. For optimization the controller considers the perceptual quality of video, load on sender and load on network. The correlations used for the controller have been developed experimentally and deductively. The controller so developed has been shown to enhance the perceptual quality of video at the receiver end in a heterogeneous environment.

In reliable multicast protocols, each data packet being sent must be acknowledged. Collecting the acknowledgments centrally at the sources can cause ACK-implosion and can result in poor scalability. To overcome this, clustering algorithms which use virtual structures to gather acknowledgments were proposed. In this work, we analyze the complexities of three such clustering algorithms: Lorax, k-degree, and Self-adjust. We compare the quality of the virtual structures produced by these! algorithms, focusing on the number of clusters, cluster size, cluster radius, and the optimal positioning of cluster leaders. Our simulation showed that the virtual structure produced by Self-adjust is better in terms of cluster radius and the location of cluster leaders. However, due to the self-adjusting nature of the algorithm, it might take longer time to compute than the other two algorithms.

In reliable multicast, the multicast packets must be acknowledged. We propose a clustering structure which can be used by most of the existing reliable multicast protocols for collecting acknowledgements and for making local retransmissions. Given a network N and a multicast routing tree (or a set of trees) T, we consider a subgraph G of N induced by the members of a multicast group. We then form disjoint clusters (local groups) of multicast receivers such that the receivers within a cluster are densely connected in G. The goal is to obtain a balanced clustering structure (dependent on the topology of G) such that the number of clusters is constant and the cluster size is kept low. This structure enables different clusters to process acknowledgments concurrently. It is also used to localize retransmissions. That is, when a packet is missed at a node, we will obtain the lost packet from another node which resides in the same cluster or at a nearby cluster whenever possible.

No abstract available.