Scheduling problems in interference-aware wireless sensor networks

Abstract

In this paper, we study the Minimum Latency Aggregation Scheduling problem in two interference models, the collision-interference-free graph model and the physical interference model known as Signal-to-Interference-Noise-Ratio (SINR), with power control. The main issue is to compute schedules with the minimum number of timeslots such that data can be aggregated without any collision or interference. While existing works studied the problem under the uniform power model or the unlimited power model, we investigate the problem assuming a more realistic non-uniform power assignment where the maximum power level is bounded. We propose a constant factor approximation algorithm with O(R + X) timeslots, where R is the network radius and X is the link length diversity. Under a reasonable assumption about the link length diversity, the number of timeslots is bounded by O(R + log n) which gives a constant approximation ratio since the lower bound is max{ R, log n}, where n is the number of nodes. Along with the problem of constructing minimum latency data aggregation schedules, we also study two other related optimization problems, namely the Scheduling and Weighted One-Shot Scheduling problems with power control in the SINR model, and provide constant approximation algorithms.