Title: Transforming grid to cloud services for multimodal biometrics

Authors: Sitalakshmi Venkatraman

Addresses: School of Science, Information Technology and Engineering, University of Ballarat, P.O. Box 663, Ballarat, VIC 3353, Australia

Abstract: Various barriers due to performance, security and privacy issues associated with shared heterogeneous data centres impede the diffusion of multimodal biometrics that could enhance secured systems in our everyday lives. This paper identifies the issues associated with multimodal fusion and privacy risks and provides means of addressing them. It presents the need for transforming from grid-based services to cloud-based services in order to address these issues, paving the way towards a successful user acceptance of biometric systems. Though grid services could provide heterogeneous resource sharing capabilities and security, multimodal biometrics require scalability and pricing flexibility for gaining user-acceptance and easy adoption. By leveraging 'on-the-demand' services offered in cloud computing, reliable and context-sensitive privacy policies could be applied for multimodal biometrics to cater to the different business and user requirements. Further, the paper combines the concept of cloud services along with momentum back propagation induced artificial neural network (MBP-ANN) model of machine learning for efficient multimodal fusion and flexible authentication schemes. The paper also describes a case scenario of how this model could be adopted by applying the varying security requirements and user privacy preferences for a successful diffusion of multimodal biometrics in the cloud.

Keywords: multimodal biometrics; artificial neural networks; ANNs; grid services; cloud services; cloud computing; grid computing; multimodal fusion; privacy protection; privacy preservation; privacy risks; user acceptance; machine learning; flexible authentication.

DOI: 10.1504/IJCSE.2016.077728

International Journal of Computational Science and Engineering, 2016 Vol.13 No.1, pp.1 - 12

Received: 17 Oct 2011
Accepted: 16 Apr 2012
Published online: 14 Jul 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article