Abstract not found

We consider the problem of route discovery in a mesh network with faulty nodes. The number and the positions of the faulty nodes are unknown. It is known that a flooding strategy like expanding ring search can route a message linear in the minimum number of steps d while it causes a traffic (i.e. the total number of messages) of O(d^2). For optimizing traffic a single-path strategy is optimal producing traffic O(d + p), where p is the number of nodes that are adjacent to faulty nodes. In TR-RSFB-05-077 “Online Multi-Path Routing in a Maze” we presented a deterministic multi-path online routing algorithm that delivers a message within O(d) steps causing traffic O((d + p) log^3 d). Here, we show an improvement of the traffic bound to O((d + p) log^2 d). This algorithm is asymptotically as fast as flooding and nearly traffic-optimal up to a polylogarithmic factor.

Abstract not found

This short report is a journal of my thoughts and ruminations on selected parts of the course material presented in the lectures and in the prescribed reading for the course “Making Decisions in Spite of Ignorance ” given by Peter Damaschke. The discussion here is intended to be superficial in nature, just presenting some general ideas and reflections on the various

Abstract: Wireless Multimedia Sensor Networks (WMSNs) are generally used for surveillance applications, intrusion detection, environmental and building monitoring, etc. Routing has become an important challenge to WMSNs from the standpoint of supporting multimedia applications due to the constraints on energy and computational capabilities of sensor nodes, and acquiring of the global network knowledge for disseminating to nodes packets forwarding purposesµ. In this paper, we propose an online multipath routing protocol for use in WMSNs. The proposed protocol uses sensor nodes ‟ positions to make packet-forwarding decisions at each hop. These decisions are made in real-time, in such a way that there is no need for having the knowledge about the entire network topology. This new routing protocol achieves load-balancing of traffic and minimizes energy consumption among nodes by using: (1) smart greedy forwarding scheme based on adaptive compass for selecting the most appropriate next hop for forwarding the traffic and (2) walking back forwarding scheme to bypass network holes. Performance comparison of the proposed protocol with Two-Phase Geographical Greedy Forwarding (TPGF) and Greedy Perimeter Stateless Routing (GPSR) shows that they: (a) maximize the overall network lifespan by not draining energy from some specific nodes, (b) provide quality of service delivery for video streams by influencing the best node along the route to destination, and (c) scale better in densely deployed wireless sensors network.

Abstract not found

We propose four simple algorithms for routing on planar graphs using virtual coordinates. These algorithms are superior to existing algorithms in that they are oblivious, work also for non-triangular graphs, and their virtual coordinates are easy to construct.