Towards Efficient Load Balancing Strategy for RPL Routing Protocol in IoT Networks

Abstract

The IPv6 Routing Protocol for Low-Power and Lossy Networks (RPL) has been considered as the new standard routing protocol designed to meet the requirements of wide range of Low Power and Lossy Networks (LLNs) applications including industrial and environmental monitoring, smart grid, and wireless sensor networks. However, due to the uneven deployment of sensor nodes in large-scale networks and the heterogeneous traffic patterns, some sensor nodes have much heavier workload than others. The lack of load balancing mechanism results in these sensor nodes quickly exhausting their energy, therefore shorten the network lifetime of battery-powered wireless sensor networks. To overcome this problem, we propose a skewness and load balancing routing protocol based on the RPL protocol, named SB-RPL that exploits various routing metrics including link quality and skewness among subtrees of the network in support topology construction. In this work, we first investigate the load balancing and related issues of RPL both via numerical simulations and via actual large-scale testbed. Performance analysis results show that RPL trees suffer from severe skewness regardless of routing metrics in randomly generated networks. Through extensive computer simulations and actual experiments, we demonstrate that SB-RPL significantly improves end-to-end packet delivery performance and tree balance compared to the standard RPL.