Biometric Security Algorithms Based On Real-time Electrocardiogram Features In Wireless Body Sensor Networks Framework

Recently,biometric security becomes an essential building block for securing the transmission of medical information in wireless body sensor networks(WBSNs)for healthcare applications.In the context of modern health monitoring systems,one of the mandatory requirements is to provide a balance between secure communication and energy-efficiency due to several constraints and heterogeneous nature of WBSNs.On the one hand,in order to offer stability in WBSNs,it is significant to develop efficient biometric random binary sequences(RBSs)which are beneficial for several security aspects such as entity identifiers(EIs),secret key generation,and key agreement.The main problem of existing techniques for generating biometric RBSs is that they not only utilize complex feature extraction procedures,but they also extract less number of entropic bits from each heartbeat,so more processing time and energy are required.On the other hand,it is also a challenging task in developing biometric security algorithms for WBSNs in which secret keys must fulfill properties of randomness and distinctiveness,and also medical data encryption procedures ought to require as less as possible resource consumption.This research aims to address above-mentioned problems,so we developed energy-efficient security solutions for WBSNs based on biometric RSBs by using time-domain features of Heart rate Variability(HRV).Overall,this study has three main contributions to provide secure communication of health information by consuming less energy and processing time for WBSNs.The research contributions of this study are divided into sever-dependent biometric security algorithms(SRR-based and IPI-based)and server-independent communication for EIs(Fuzzy Vault-based).Initially,we developed a simple key generation procedure based on the logical operation between Standard Deviation of NN intervals(SDNN)and Root-Mean Squared of the Successive Differences(RMSSD)termed as SRR.The NN interval is the duration between two consecutive R-R peaks in ECG waveform.Furthermore,SRR-based keys are applied to develop biometric security algorithm for data encryption to secure transmission of medical information in WBSNs.After that,we developed RBSs generation technique by incorporating finite monotonic increasing sequences generation mechanism using inter-pulse intervals(IPIs)of heartbeats and cyclic block encoding procedure for decreasing the measurement errors to ensure extraction of entropic bits up to more significant extent from each IPI.These RBSs are utilized to develop privacy preserving and security algorithm to assure medical data confidentiality and authenticity in WBSNs.Finally,we proposed an efficient fuzzy vault based security algorithm(EFVSA)based on real-time ECG for entities identification to offer inter-sensor communication or sever-independent communication in WBSNs.EFVSA utilizes time-domain analysis of ECG for features extraction,and EI generation thus consumes fewer resources than frequency-domain analysis.In conclusion,this study presents solutions to critical issues in implementing the security in WBSNs that is how to safeguard the security aspects like confidentiality,and authentication of a patient's data over networking environment in an energy-efficient manner.In this research,to minimize processing time and energy utilization for implementing security in WBSNs,using medical information becomes an attractive approach since it is already available in the system mainly common to different wireless sensor devices of the network.From the experimental results,it is analyzed that performance of proposed RBSs approach can be up to four times faster than the existing heartbeat-based schemes.Additionally,in order to validate those RBSs can be exercised for security aspects,nine NIST tests and hamming distance are experimented to measure randomness and distinctiveness of biometric RBSs,respectively.Furthermore,by comparative analysis,it is observed that IPI-based proposed security algorithm consume less processing time(0.196ms)and power consumption(8.901 m W)than SRR-based developed security algorithm that is 0.207 ms and 9.64 m W,respectively.However,in order to provide inter-sensor communication with in-network proposed EFVSA is the suitable choice since it consumes fewer resources such as transmission time and power utilization than traditional node recognition approaches i-e 0.168 ms and 8.12 m W,respectively.