Securing Internet of Bio-Nano Things: ML-Enabled Parameter Profiling of Bio-Cyber Interfaces

Abstract

The Internet of Bio-Nano Things (IoBNT) is an emerging paradigm at the cross-roads of Internet of Things (IoT), E-healthcare and synthetic biology. IoB Nt relies on innovating molecular communication between biologically synthesized nanodevices to facilitate in vivo drug delivery, remote monitoring and healthcare management via. the Internet. IoBNT therefore, requires robust security primitives to adequately address patient privacy concerns, increase physician confidence as well as limit (any) malicious operation resulting in bio-terrorism. The present work focused on securing the bio-cyber interfacing of IoB Nt technologies and considered three prevalent bio-electrical transduction techniques including bio-luminescence, redox modality and biological field effect transistors (BioFETs). Using the state-of-the-art machine learning (ML) algorithms, the proposed security framework employed parameter profiling the distinct operating features of bio-cyber interfaces and aimed to identify anomalous operation. During validation, an optimal profiling accuracy ranging between 88–91 % was recorded. The corresponding time complexity, scalability and qualitative analysis of the proposed parameter profiling-based framework leads to us to recommend it for further adoption in improving IoBNT security.