Sixsoid: A new paradigm for $k$-coverage in 3D Wireless Sensor Networks

TL;DR

This paper introduces the Devilsoid, a new 3D shape that improves $k$-coverage guarantees and simplifies deployment in wireless sensor networks compared to previous models like the Reuleaux Tetrahedron.

Contribution

The paper proposes the Devilsoid shape for 3D $k$-coverage, demonstrating its superior coverage volume and easier deployment strategy over existing shapes.

Findings

Devilsoid guarantees more coverage volume than Reuleaux Tetrahedron.

Devilsoid enables simpler and more pragmatic deployment strategies.

The paper provides the construction and volume calculation of Devilsoid.

Abstract

Coverage in 3D wireless sensor network (WSN) is always a very critical issue to deal with. Coming up with good coverage models implies more energy efficient networks. $K$-coverage is one model that ensures that every point in a given 3D Field of Interest (FoI) is guaranteed to be covered by $k$ sensors. When it comes to 3D, coming up with a deployment of sensors that gurantees $k$-coverage becomes much more complicated than in 2D. The basic idea is to come up with a geometrical shape that is guaranteed to be $k$-covered by taking a specific arrangement of sensors, and then fill the FoI will non-overlapping copies of this shape. In this work, we propose a new shape for the 3D scenario which we call a \textbf{Devilsoid}. Prior to this work, the shape which was proposed for coverage in 3D was the so called \textbf{Reuleaux Tetrahedron}. Our construction is motivated from a construction that can be applied to the 2D version of the problem \cite{MS} in which it imples better guarantees over the \textbf{Reuleaux Triangle}. Our contribution in this paper is twofold, firstly we show how Devilsoid gurantees more coverage volume over Reuleaux Tetrahedron, secondly we show how Devilsoid also guarantees simpler and more pragmatic deployment strategy for 3D wireless sensor networks. In this paper, we show the constuction of Devilsoid, calculate its volume and discuss its effect on the $k$-coverage in WSN.