Exploring Formal Strategy Framework for the Security in IoT towards e-Health Context using Computational Intelligence

• Published in: Internet of Things and Big Data Technologies for Next Generation Healthcare pp. 63 - 90
• Main Authors: Ould-Yahia, Youcef, Banerjee, Soumya, Bouzefrane, Samia, Boucheneb, Hanifa
• Format: Book Chapter
• Language: English
• Published: Cham Springer International Publishing 2017
• Series: Studies in Big Data
• Subjects: Host Identity Protocol; Security Solution; Sensor Node; Social Insect; Sybil Attack
• ISBN: 3319497359; 9783319497358
• ISSN: 2197-6503; 2197-6511; 2197-6511
• DOI: 10.1007/978-3-319-49736-5_4

This chapter proposes a novel strategic framework and computationally intelligent model to measure possible vulnerabilities for security context in e-health. In order to keep track of security of e-health paradigm, the chapter conceives a bio-inspired model comprising the collective intelligence of social insects e.g. ant colony. Ant colony optimization is a computationally intelligent meta-heuristics, which takes care-off the different random and uncertain behavior of different sensors deployed towards e-health measures. The essential input is provided from interconnected wireless sensors under Internet of Things (IoT) paradigm and intelligent social insects that could sense the possibility of threats for a patient moving in different physical locations during his medical diagnosis. Social insect ants can sense and communicate through a chemical, known as pheromone, remotely from their nest towards collection of food. The intensity of pheromone measured for different interconnected graphs of e-health could lead to a consolidated algorithm and finally the differences of intensities can infer on the affected or safe path for propagation of medical information. Modelling the pheromone dynamics can be a precise measure to quantify the different e-health security issues like Sinkhole threat or sybil attack under IoT environment. The proposed pheromone alert is presented and compared statistically in terms of precision to identify the classification of possible vulnerabilities.