To overcome these operational limitations, the research team constructed and evaluated two contrasting architectural frameworks: a two-layer hierarchical model and a baseline flat model.
In the hierarchical design, incoming network traffic first encounters a binary classification stage tasked solely with distinguishing benign communications from malicious activity.
Beyond top-line detection rates, the hierarchical model provided significant advantages in resource conservation, memory footprint, and latency, which represent vital deployment metrics for battery-powered medical hardware.
Subject of Research: Machine learning architectures for network intrusion detection in healthcare Internet of Things systems Article Title: Hierarchical versus flat machine learning model for intrusion detection in secure IoT healthcare environment Article References: ElRushaidat, D., Sammak, T., Ghannam, Y., & Alkhalil, B.
Hierarchical versus flat machine learning model for intrusion detection in secure IoT healthcare environment.