Load Balancing in RPL to Avoid Hotspot Problem for Improving Data Aggregation in IoT

Abstract

Data aggregation plays a vital role in the Internet of Things (IoT), and it aggregates the collected sensor data from devices to suppress redundant data transmissions. Many-to-one traffic pattern in the IoT induces hotspot problem and inefficient data aggregation. The Routing protocol for low-power and lossy networks (RPL) in the network layer impacts the hotspot problem due to the frequent usage of forwarding nodes and load imbalance. The processes of network layer protocol, such as trickle algorithm and Objective Functions (OF) for Destination Oriented Directed Acyclic Graph (DODAG) construction, need more attention to avoid hotspot for efficient data aggregation. This work proposes a Load Balanced RPL (LoB-RPL) protocol to avoid hotspot creation using a composite metric based parent selection, DODAG construction, and local topology adaptive decision on trickle parameters. The LoBRPL improves the Minimum Hop with Hysteresis Objective Function(MRHOF) using the composite metric based parent selection and tunes the parameters of the Trickle algorithm. It ensures efficient maintenance of DODAG structure, hotspot avoidance, and unnecessary DIO transmissions. Beyond the advantages of composite metric based parent selection, consideration of dynamic parameters may induce frequent parent switching in RPL. To avoid frequent changes in the DODAG structure, the LoB-RPL optimally decides the parent switching threshold. Thus, the proposed work ensures a load-balanced and an energy-efficient RPL routing as well as data aggregation in the IoT environment. The LoB-RPL delivers outperforming results compared to the base RPL under various inter-packet interval time over 50 node topologies.