APPLICATIONS OF MULTIMODAL BIOMETRICS AUTHENTICATION FOR ENHANCING THE IOT SECURITY USING DEEP LEARNING

Abstract

The  Internet of Things (IoT) integrates billions of electronic devices into computer networks toprovide advanced and intelligent services that enable devices to communicate with each other byexchanging information with minimal human interaction. The security issue is at higher risk in IoTsystems  than  in  other  computing  systems.  Maintaining  the  security  requirement  when  attackingthe  physical  surface  of  the  IoT  system  device  is  a  challenging  task.  Implementing  securitymechanisms like authentication and access control for the IoT ecosystem is necessarily needed toensure the securityof IoTdevices. The key used for security may be stolen, forgotten, or forged.Also,  the  key  may  be  generated  by  intruders  or  men  in  the  middle  of  traditional  securitymechanisms. Biometric security is becoming more advanced and sophisticated with technologicaladvancements  and  is  mostly  used  in  authentication  systems.  In  unimodal  biometrics,  only  onebiometrics character can be applied which does not apply to ensure the security of IoT systems. Inthis paper, Multimodal biometrics authentication  was used for securing edge devices in the  IoTecosystems.  Face  image  and  fingerprint  image  were  used  as  multimodal  biometrics  systems  forauthenticating users to secured IoT devices. A pi-camera module and fingerprint sensor were usedto  capture  biometric  data.  Image  processing  techniques  were  then  applied  to  the  images.  ThenCNN algorithms were used for feature extraction and model creation. During model creation, theRELU function was used as an activation function, soft-max for image classification, and Max-pooling forimage dimensional reduction which helped the model speed up the training process.Experimental results show that the accuracy of the face image and fingerprint image is 92% and89% respectively, which is a promising result that achieves the objective of the study.

Keywords : Internet of Things, Multimodal Biometrics, Authentication, CNN, Deep Learning