Resilient greedy routing on GPS-free surface sensor networks

Buri Ban,Xuan Li,Miao Jin

doi:10.1177/1550147720913810

Abstract

We design a greedy routing scheme specifically for GPS-free large-scale wireless sensor networks deployed on surfaces of complex-connected three-dimensional settings. Compared with other greedy embedding–based surface network routing scheme, the proposed one is cut free such that no pair of nodes suffers a long detour to reach each other. The routing scheme is designed to be resilient to node or link failures especially under random node or link failure model where each node in a network has an equal and independent probability of failure during some time interval. The proposed algorithm is fully distributed and scalable to both the size and the topological complexity of a network. Each sensor node requires only limited and constant storage. Simulation results show the proposed routing scheme with a higher successful delivery ratio, a lower average stretch factor, and a lower normalized communication cost compared with other resilient routing methods.

Highlights

IntroductionA wireless sensor network may be deployed on a twodimensional (2D) plane[1] (e.g. for crop sensing in fields or wildlife tracking on plains), or in a three-dimensional (3D) volume[2,3,4] (for underwater or space reconnaissance), or on a 3D surface[5,6] (e.g. with sensors mounted on ocean floor, mountain surface, or the surface of various man-made structures)
A wireless sensor network may be deployed on a twodimensional (2D) plane[1], or in a three-dimensional (3D) volume[2,3,4], or on a 3D surface[5,6]
This article focuses on designing a routing algorithm resilient to random node or link failures in a large-scale wireless sensor network deployed on the surface of a complexconnected 3D setting

Summary

Introduction

A wireless sensor network may be deployed on a twodimensional (2D) plane[1] (e.g. for crop sensing in fields or wildlife tracking on plains), or in a three-dimensional (3D) volume[2,3,4] (for underwater or space reconnaissance), or on a 3D surface[5,6] (e.g. with sensors mounted on ocean floor, mountain surface, or the surface of various man-made structures). Multipath routing schemes are effective for geographically correlated failures Alternative paths, especially those disjoint with the failed one, has a high probability to avoid a network failure region. It is obvious that multipath routing schemes, especially those choosing alternative paths based on disjoint or different homotopy types, have intrinsically a very low resilience to independent node failures. Each sensor node stores a set of planar coordinates and uses it to enable greedy routing These methods suffer a large stretch factor in average. Our contributions can be summarized as follows: We design a greedy routing scheme for GPS-free wireless sensor networks deployed on surfaces of complex-connected 3D settings. Compared with other greedy embedding–based surface network routing scheme, the proposed approach is cut free such that no pair of nodes suffers a long detour to reach each other. As long as, for each face fijk, the edge lengths satisfy the triangle inequality: lij + ljk . lki

X fijk 2F u u jk i

Findings

Conclusion