Worst-Case Interactive Communication and Enhancing Sensor Network Lifetime

TL;DR

This paper introduces scalable interactive communication protocols to maximize the worst-case lifetime of sensor networks by optimizing data gathering through formal models and clustering strategies.

Contribution

It develops a formalism for worst-case interactive communication and applies it to sensor networks, proposing a novel clustering protocol that outperforms traditional methods.

Findings

Proposed protocols improve data gathering efficiency.

Clustering-based protocol extends sensor network lifetime.

Demonstrated superiority over traditional clustering approaches.

Abstract

We are concerned with the problem of maximizing the worst-case lifetime of a data-gathering wireless sensor network consisting of a set of sensor nodes directly communicating with a base-station.We propose to solve this problem by modeling sensor node and base-station communication as the interactive communication between multiple correlated informants (sensor nodes) and a recipient (base-station). We provide practical and scalable interactive communication protocols for data gathering in sensor networks and demonstrate their efficiency compared to traditional approaches. In this paper, we first develop a formalism to address the problem of worst-case interactive communication between a set of multiple correlated informants and a recipient. We realize that there can be different objectives to achieve in such a communication scenario and compute the optimal number of messages and bits exchanged to realize these objectives. Then, we propose to adapt these results in the context of single-hop data-gathering sensor networks. Finally, based on this proposed formalism, we propose a clustering based communication protocol for large sensor networks and demonstrate its superiority over a traditional clustering protocol.