On the Critical Transmission Range in One-Dimensional

Sensor Networks Under Non-Uniform Node Placement

Professor Armand M. Makowski
Department of Electrical and Computer Engineering and
the Institute for Systems Research
University of Maryland, College Park, MD 20742
armand@isr.umd.edu

We consider n sensors independently placed at points X1,...,Xn on the unit interval[0,1]according to some probability distribution function F. Two sensor nodes communicate with each other if their distance is less than some given transmission range ρ > 0. We deŽne the critical transmission range Rn as the smallest transmission range such that the nodes X1,...,Xn form a connected graph (under the notion of adjacency implied by the ability of nodes to communicate). Since the distribution of Rn is usually not tractable, we are interested in developing an asymptotic theory for Rn as n becomes large: We seek a deterministic sequence ρ* : N0R+ such that the ratio Rnn converges to some non-trivial limit L in an appropriate sense. When available, such results suggest ρn L as a proxy or approximation for Rn.

We assume that F admits a continuous density f, and two qualitatively dierent cases are identiŽed, namely f > 0 and f = 0 with f = inf(f(x),x [0,1]). In each case, we present results on the form of ρ*n and

L. In the process we make contact with the existence and nature of critical thresholds for the property of graph connectivity in the underlying geometric random graph. Engineering implications for power allocation are discussed.

Wednesday, August 26, 2009

3-4 pm

Room 1200 EECS

 
This is joint work with former Ph.D. student Guang Han (now at Motorola).