Biometric systems are a vital part of physical access control systems. Biometric systems ensure that access to use a system or service is done by a valid user. These systems are used in various applications like ATMs, voting system, aadhar card, attendance management system etc. The biometric systems are in operation for a quite long time and their use is increasing. There is a requirement for low cost, scalable systems with high accessibility. In this paper a cloud based secure biometric system architecture is proposed, to make the system efficient and economical for remote enrollment on Raspberry Pi, which is a low cost computer running on Linux. The system is capable of capturing multimodal biometric traits such as face and fingerprints and send them to cloud service by end-to–end encryption process. An enhanced encryption method with AES-256 algorithm is used for protecting the biometric traits on insecure communication path within TLS. SHA-256 is used for integrity and RSA for key exchange. Improvement is done in AES by altering the S-box. The principal component of AES block cipher is the S-Box. This S-box is used to provide the confusion factor for AES. The aim of this paper is to design dynamic S-box which depends on the secret key. To increase the difficulty of the system, AES is used in Round structure. The proposed AES algorithm can be enhanced to 1024bit key length for stronger security required in banking system. The proposed architecture is leveraging the power of cloud to make the system scalable and pluggable. Thus the use of Raspberry Pi makes the biometric system low cost and portable and the use of the Encryption algorithm makes the system viable for secure communication and authentication.

1. G. Senthilkumar, K. Gopalakrishnan, V. Sathish Kumar, "Embedded Image Capturing System Using Raspberry Pi System", International Journal of Emerging Trends & Technology in Computer Science, vol. 3, issue 2, April 2014
2. Rushikesh Tade, "Embedded Cloud For Antipiracy", International Journal of Science and Technology Reasearch vol. 2, issue 6, Jun 2013.
3. A. Abugharsa, A. Basari, H. Almangush, "A New Image Encryption Approach using The Integration of A Shifting Technique and The Aes Algorithm", International Journal of Computer Applications, vol. 42-No. 9, Mar 2012
4. R. Hosseinkhani& H. Javadi, "Using Cipher Key to Generate Dynamic S-Box in AES Cipher System", IJCSS, vol. 6, issue 1, 2012
5. E. Mahmoud, A. Hafez, T. Elgarf, A. Zekry, "Dynamic AES-128 with Key-Dependent S-box", IJERA vol. 3, Issue 1, January -February 2013, pp. 1662-1670
6. S. Arrag, A. Hamdoun, A. Tragha, S. Khamlich, "Implementation of Stronger Aes by Using Dynamic S-Box Dependent of Master Key", Journal of Theoretical and Applied Information Technology, vol. 53 No. 2, 20th July 2013.
7. Raspberry Pi, Available: http://en. Wikipedia. org/wiki/Raspberry_Pi
8. Available: http://www. siongboon. com/projects/2013-07-08_raspberry_pi/images/raspberry_pi_circuit. jpg
9. Available: http://en. wikipedia. org/wiki/Advanced_Encryption_Standard
10. Soup to Nuts: From Raw Hardware to Cloud-Enabled Device, Available: http://msdn. microsoft. com/en-us/magazine/dn781356. aspx
11. Camera Module, Available: http://www. raspberrypi. org/ documentation/ usage/camera/README. md.
12. RPi Verified Peripherals, Available: http://elinux. org /RPi_VerifiedPeripherals#Fingerprint_Scanners

Raspberry Pi AES cloud computing Microsoft Azure dynamic S-Box embedded systems.