Title

Approximation algorithms for power-aware scheduling of wireless sensor networks with rate and duty-cycle constraints

Document Type

Conference Proceeding

Publication Date

1-1-2006

Abstract

We develop algorithms for finding the minimum energy transmission schedule for duty-cycle and rate constrained wireless sensor nodes transmitting over an interference channel. Since traditional optimization methods using Lagrange multipliers do not work well and are computationally expensive given the non-convex constraints, we develop fully polynomial approximation schemes (FPAS) for finding optimal schedules by considering restricted versions of the problem using multiple discrete power levels. We first show a simple dynamic programming solution that optimally solves the restricted problem. For two fixed transmit power levels (0 and P), we then develop a 2-factor approximation for finding the optimal fixed transmission power level per time slot, P opt, that generates the optimal (minimum) energy schedule. This can then be used to develop a (2, 1 + ε)-FPAS that approximates the optimal power consumption and rate constraints to within factors of 2 and arbitrarily small ε > 0, respectively. Finally, we develop an algorithm for computing the optimal number of discrete power levels per time slot (O(1/ε)), and use this to design a (1, 1 + ε)-FPAS that consumes less energy than the optimal while violating each rate constraint by at most a 1 + ε factor. © Springer-Verlag Berlin Heidelberg 2006.

Publication Source (Journal or Book title)

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

First Page

463

Last Page

479

This document is currently not available here.

COinS