Faster Route Update Using Downward Traffic in RPL-based Multi-hop IoT Networks

Abstract

Generally, studies on the low power and lossy networks (LLNs) are conducted under the assumption of sparse upward traffic patterns. Furthermore, in LLNs, the multi-hop transmission is required and link connections are unstable due to the employ of low power devices. To meet the requirements, RPL (IPv6 Routing Protocol for Low-power and Lossy Networks) is standardized in 2012. RPL makes and updates upward routes based on the expected upward packet transmission count (ETX), and makes downward routes by the reverse of upward routes. However, many recent applications demand various kinds of traffic pattern to the LLNs, for example, downward-centric traffic for the electronic shelf label system. This research shows the performance degradation and loop problem that occurs from recursive global repair operations of RPL in downward traffic centric applications. To solve these problems effectively, we suggest a specification compliant packet transmission count delivery scheme that exploits downward packets, and way to update routes through the calculation of ETX using the delivered information. Our results in the multi-hop testbed over IEEE 802.15.4 links indicate that the performance of our scheme is superior to conventional RPL in scenarios with various traffic patterns, in terms of packet delivery ratio, parent change count, control overhead and radio energy consumption.