Multibiometric Cryptosystem Based on Decision Level Fusion for File Uploading in Cloud

Abstract

Proper identification is important for providing secure facilities and services to user. Cloud computing is involved in our day to day life as it offers several services, storage spaces, applications etc. As the usage increases, vulnerability and security threats arose need to be studied and rectified. As a part of this, proper authentication need to be done before allowing file uploading to cloud. Compared with traditional identification methods, authentication system using biometric are more reliable and universal. Since biometric cannot be lost or forgotten it provides more reliability and security. So biometric provide more security because it requires physical person to be authenticated. Stealing the biometric values by hacker attack will be protected by biometric cryptosystem. This enable users to upload a file safely from their computer and have it saved on the web server's file system. This would provide the basis for easy ways of storing and retrieving data directly within the database which involves a bit more code than working with data through the file system. Additionally, in order for a user to view or download data they must be presented with a URL to that data. If the data resides on the web server's file system, the URL would be a straightforward process. And if the data is stored in the database, a web page needs to be created and retrieve and return the data from the database. Further, if the data is stored on the file system, the demand and network congestion between the database server and web server will be less than when the data is stored directly within the database. Multi-biometric cryptosystem offer high authentication accuracy, wider population coverage and strong security compared to single biometric cryptosystem.