Improving the Reliability of RPL Using Hybrid Deep Learning and Objective Function-Based DODAG Structure for AMI

Abstract

Advanced Metering Infrastructure (AMI) is the prime smart grid application that connects smart meters and electric power stations. Routing Protocol for Low-Power and Lossy Networks (RPL) is the most familiar lightweight routing protocol for AMI networks. The reliability of RPL routing is a potential problem for the efficient deployment of AMI networks. This paper ensures attack and network reliability for AMI-RPL and proposes Hybrid Deep Learning based Intrusion Detection System (HDL-IDS) for attack reliability and a Multi-Objective Function-based Reliable RPL (MOR-RPL) for network reliability. The HDL-IDS assures AMI-RPL reliability against attacks by analyzing and eliminating the attack traffic successfully. The MOR-RPL method improves the RPL construction reliability by procuring the multi-objective function-based reliability metrics in the DODAG building. The simulation results show that the attack and network reliability are significantly enhanced with balanced energy consumption.