Body Area Network (Ban) Security by Utilyzing ECG Signal Processing and RBS Generation

An rising category of wireless sensor networks is the Wireless Body Sensor Networks (WBSNs). With increased sophistication and advancements in wireless sensor networks, several categories of wireless sensor networks have evolved one among which is the wireless body sensor network (WBSN) or the body area network (BAN) which are used interchangeably in this paper. Wireless body sensor networks can be defined as the connectivity of sensors in the periphery of the human body which can communicate with each other as well as communicate with a central base station or control station. There are however several factors governing the reliability of the network. One of the key challenges faced with WBSNs is securing data transmission since complex encryption algorithms cannot be employed due to the limited resources of processing power and memory. The proposed technique incorporates the deep Markov model for random bit sequence (RBS) generation from the ECG based data which has been used as the counterpart for the actual heart beats. The features which have been extracted for the inter pulse interval (IPI) are RR interval, SS Interval and QRS complex. The database used for the study is the MIT-BIH library wherein the Electro Cardiogram data is available in the form of .mat files and can be processed for analysis. The security is based on the authentication provided by a random binary stream (RBS) which is 128 bits in length. The RBS is generated from the inter-pulse interval (IPI) extracted from the ECG waveform. The computation parameters considered are the entropy and the hamming distance. The performance evaluation parameters for the proposed technique are the entropy and the hamming distance. It has been shown that the proposed technique achieves better results in terms of hamming distance and entropy compared to previous work.