PhD Dissertation: Duygu Karaoğlan
  • FENS
  • PhD Dissertation: Duygu Karaoğlan

You are here





Computer Science and Engineering, PhD Dissertation, 2016


Thesis Jury

Prof. Dr. Albert Levi (Thesis Advisor), Prof. Dr. Berrin Yanıkoğlu, Assoc. Prof. Ayhan Bozkurt, Prof. Dr. Ali Aydın Selçuk, Prof. Dr. Fatih Alagöz



Date & Time: 23th, November, 2016 –  15:00

Place: FENS 2008

Keywords : Body Area Network Security, Cryptographic Key Generation, Physiological Signals, Bio-cryptography, Key Agreement




Advances in lightweight, small-size and low-power sensors led to the development of wearable biosensors, thus, to the accurate monitoring of human periphery. On top of this, pervasive computing has been improved and technologies have been matured enough to build the plug-and-play Body Area Networks (BANs). In a BAN, the main functionality of a node is to effectively and efficiently collect data from vital body parts, share it with the neighbours and make decisions accordingly. Because of the fact that the captured phenomenon is highly sensitive against privacy breaches in addition to being transmitted using the wireless communication medium, BANs require a security infrastructure. How- ever, due to the extreme energy scarcity, bandwidth and storage constraints of the nodes, conventional solutions are inapplicable for BANs.

In this dissertation, we propose a novel security infrastructure that is designed specifically for the intra-BAN communication. In this regard, we propose a novel key agreement pro- tocol, SKA-PS (Secure Key Agreement using Physiological Signals), which is based on the set reconciliation paradigm. Our protocol generates symmetric shared keys using the physiological parameters derived from the physiological signals of the users, such as electrocardiogram and blood pressure. We also identify 4 different appropriate phys- iological parameters that can be used as cryptographic keys and propose the techniques of generating them. In the security infrastructure that we have developed for the intra- BAN communication, (i) secure node-to-host association is satisfied, (ii) performance enhancing characteristics of bio-cryptography is brought in the foreground, (iii) adopted physiological parameters are random and distinctive enough, along with possessing low error rates, (iv) key agreement protocol works dynamically, possessing remarkably high true match and exceedingly low false match rates, and (v) key agreement protocol re- sists against brute-force, replay and impersonation attacks, together with possessing low communication, computational and storage costs.