Computer Science – Networking and Internet Architecture
Scientific paper
2009-08-22
Physica A: Statistical Mechanics and its Applications, Volume 389, Issue 14, 15 July 2010, Pages 2841-2848
Computer Science
Networking and Internet Architecture
13 pages, 7 figures, research paper
Scientific paper
10.1016/j.physa.2010.03.005
This work shows potentials for rapid self-organisation of sensor networks where nodes collaborate to relay messages to a common data collecting unit (sink node). The study problem is, in the sense of graph theory, to find a shortest path tree spanning a weighted graph. This is a well-studied problem where for example Dijkstra's algorithm provides a solution for non-negative edge weights. The present contribution shows by simulation examples that simple modifications of known distributed approaches here can provide significant improvements in performance. Phase transition phenomena, which are known to take place in networks close to percolation thresholds, may explain these observations. An initial method, which here serves as reference, assumes the sink node starts organisation of the network (tree) by transmitting a control message advertising its availability for its neighbours. These neighbours then advertise their current cost estimate for routing a message to the sink. A node which in this way receives a message implying an improved route to the sink, advertises its new finding and remembers which neighbouring node the message came from. This activity proceeds until there are no more improvements to advertise to neighbours. The result is a tree network for cost effective transmission of messages to the sink (root). This distributed approach has potential for simple improvements which are of interest when minimisation of storage and communication of network information are a concern. Fast organisation of the network takes place when the number $k$ of connections for each node ({\em degree}) is close above its critical value for global network percolation and at the same time there is a threshold for the nodes to decide to advertise network route updates.
No associations
LandOfFree
Rapid self-organised initiation of ad hoc sensor networks close above the percolation threshold does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Rapid self-organised initiation of ad hoc sensor networks close above the percolation threshold, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Rapid self-organised initiation of ad hoc sensor networks close above the percolation threshold will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-190444