L-RPL: RPL powered by laplacian energy for stable path selection during link failures in an Internet of Things network

Abstract

The routing protocol RPL for low power and lossy networks (LLNs) suffers from reduced reliability, increase in control overhead, longer latency and higher energy utilization on experiencing high node density, link or node failures in the network. This paper aims to elaborately study the behaviour of RPL’s expected transmissions count (ETX) based path selection for lengthier routes and the impact of node density and link or node failures on them. Our study finds that the path selection along long routes suffers due to the lack of decisive information on the neighbourhood links. As an alternate, a laplacian energy based path selection is proposed to enrich RPL by bringing in additional neighbourhood connection information into the path selection process to provide stable paths. Laplacian energy drop, a two-walk neighbourhood connectivity measure that is incorporated in L-RPL helps a node to pick the most resilient path. By simulation and testbed experiment results, it is shown that the proposed L-RPL improves path selection and reduces control overhead (29.05% and 39.68% reduction in (Destination Advertisement Object) control data when 9% of nodes fail and during normal conditions respectively), latency and also improves energy utilization.