Human Body Odor Based Authentication Using Machine Learning

Date

2017

Authors

Yang, Bin

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

The wearable devices become very popular in the people's daily lives. Because the wearable device is close to human body and the human biometric is unique for individuals, more and more wearable devices use the biometric parameters such as fingerprint, face, voice, iris, and hand geometry to perform authentication. But the above authentication methods require human intervention to enter authentication information scanning human body. This thesis proposes a new biometric authentication method using human body odor, which makes the authentication more convenient and effective. I show that the human odor based authentication is accurate since every human body odor is determined by the major histocompatibility complex genes through some measures using gas sensors and machine learning techniques. In my analysis, I use three machine learning methods; K-means, Principal Component Analysis, and Neural Network, and show that the authentication accuracy using human body odor is very high as a result of experiments.

Description

This item is available only to currently enrolled UTSA students, faculty or staff. To download, navigate to Log In in the top right-hand corner of this screen, then select Log in with my UTSA ID.

Keywords

authentication, k-means, machine learning, NN, odor, PCA

Citation

Department

Electrical and Computer Engineering