Steiner Tree Based Distributed Multicast Routing in Networks

Novak, Roman; Rugelj, Jože; Kandus, Gorazd

doi:10.1007/978-1-4613-0255-1_10

Cited by 16 publications

(14 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, a minimum Steiner tree in terms of minimum weight can be a path, which has a long data distribution time by (1). On the other hand, a minimum Steiner tree with bounded end-to-end bound delay can have a higher outgoing degree and a longer data distribution time as well by (1).…”

Section: Heuristic Algorithm For Fast Data Distribution Multicast Treementioning

confidence: 96%

See 1 more Smart Citation

Steiner Tree for Fast Data Distribution

Fan¹,

Chen²,

Kotsireas³

et al. 2011

AIP Conference Proceedings

View full text Add to dashboard Cite

Section: Heuristic Algorithm For Fast Data Distribution Multicast Treementioning

confidence: 96%

“…The problem of finding a minimum multicast tree, also called a multicast routing problem, has been studied in the field of network communications [1]. Kompella [2] defined the constrained Steiner tree problem for multicast as a minimum cost Steiner tree problem in graphs with additional constraints such as end-to-end delay bounds.…”

Section: Introductionmentioning

confidence: 99%

Steiner Tree for Fast Data Distribution

Fan¹,

Chen²,

Kotsireas³

et al. 2011

AIP Conference Proceedings

View full text Add to dashboard Cite

“…As mentioned above, the creation of an optimal aggregation topology in a communication network is an instance of the complete Steiner Minimal Tree problem, whose manifestations in a network have the undesirable property that unalike sets of connected nodes can include different network paths, making reuse of routes by multiple querying sinks unlikely. Instead of using Steiner Trees for data aggregation we propose to use a network wide Minimal Spanning Tree (MST), a known heuristic approach used for the construction of Steiner Minimal Trees [27,39], having a worst case cost ratio of less than or equal to 1/2 to an optimal Minimal Spanning Tree [26]. Sink nodes establish individual aggregation tree views of this MST, incurring minimal global cost for query dissemination and bounded cost for aggregation, while all routing decisions are taken using only local node knowledge.…”

Section: Ongoing Research Sat: Efficient Shared Aggregation Topology mentioning

confidence: 99%

Distributed Middleware for Global Knowledge in Sensor Networks

Hänsel¹,

Mauch²,

Radusch³

et al. 2004

Frequenz

View full text Add to dashboard Cite

In traditional computer networks object-based middleware provides portability between different hardware and software architectures as well as a common application environment for each distributed participant. Thereby, the networking middleware abstracts from different network requirements, tasks, and protocols by offering higher level tools and program language specific objects. These requirements and tasks can include for example marshalling, providing additional object consistency over unreliable network protocols as well as implementing common algorithms. However, the hardware and software found in each sensor node is almost the same due to the nature of sensor networks. But nonetheless the application domains for a specific sensor network can vary heavily. Therefore, all differentiation between sensor nodes happens at run-time. For example, nodes must be able to reduce their work load to sensing, forwarding, or sleeping depending on the node density or energy level. This implies a certain level of reflection within the proposed sensor middleware: Nodes must be able to determine and influence its and other nodes' status and abilities. That demands an appropriate specification which is outlined here. Since the sensor nodes must operate unattended, the middleware must offer new levels of support for automatic configuration and error handling. This includes distributing traditional centralised services like data collection and processing with a rule-based approach. Apart from general sensor network constrains such as low energy, low memory, low bandwidth, the massive number of nodes and the implied self-configuration, and the anticipated high failure rate, two major parameters are important for the selection of networking algorithms: the mobility of individual sensor nodes and the query/ event rate. These parameters determine mainly whether the middleware can choose to spend time gathering information for proactive routing or whether the algorithms must relapse to reactive data dissemination. The sensor networking middleware can help the application developer to abstract from all these parameters while providing the appropriate algorithms for efficient communication. The most common algorithms are presented shortly and classified according to the above parameters. ÜbersichtDie Hauptaufgabe traditioneller objekt-basierter Middlewaresysteme ist zum einen die Bereitstellung von Portabilität zwischen verschiedenen Hard-und Softwarearchitekturen sowie von einer einheitlicher Umgebung für die verteilten Applikationen. Dabei abstrahiert die Middleware von den verschiedenen Netzwerkanforderungen, -aufgaben und -Protokollen und bietet programmiersprachenspezifische Werkzeuge für den einheitlichen Zugriff auf die Netzwerkressourcen. Derartige Anforderungen und Aufgaben sind z. B. Marshalling, Objektkonsistenz über unzuverlässige Netzwerkprotokolle sowie die Implementierung allgemein verwendbarer Algorithmen. Während in Sensometzwerken die Hardware der einzelnen Knoten im allgemeinen gleich ist, können sich die Anwendungen...

show abstract

“…Many multicast routing protocols have been proposed for MANET [1,4,5,7,8,9,11,12,13]. A review paper was given by Cordeiro et.…”

Section: Introductionmentioning

confidence: 99%

K-MCore for Multicasting on Mobile Ad Hoc Networks

Peng

Chu

2006

2006 Seventh International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT'06)

View full text Add to dashboard Cite

A k-cluster of a tree includes a single path and k −1 subpaths growing from that path. A k-MCore is a k-cluster that minimizes the sum of the distances of all vertices to the cluster plus the size of the cluster. This structure is motivated by the applications on overlay multicasting. The overlay multicast protocol constructs a virtual mesh spanning all member nodes of a multicast group. It employs standard unicast routing and forwarding to fulfill multicast functionality. In this paper, we propose effective distributed algorithms for constructing k-MCore on a tree network. The k-MCore is more stable and easier to maintain than the spanning tree in virtual mesh. The simulation results show that our approach handles the flexibility and mobility issues in an overlay multicast protocol effectively, especially when the group size is large.

show abstract

Steiner Tree Based Distributed Multicast Routing in Networks

Cited by 16 publications

References 30 publications

Steiner Tree for Fast Data Distribution

Steiner Tree for Fast Data Distribution

Distributed Middleware for Global Knowledge in Sensor Networks

K-MCore for Multicasting on Mobile Ad Hoc Networks

Contact Info

Product

Resources

About