The National ICT R&D Fund (http://www.ictrdf.org.pk/) has been working jointly with the industry and the academia since its establishment. Apart from the projects being transferred from the PTCL R&D Fund, many projects are initiated by the National ICT R&D fund itself. The National ICT R&D Fund has facilitated a large number of projects by providing funds for the successful execution of these projects. One of these research projects initiated by Center for Advanced Studies in Engineering (CASE) is to develop a Massively Parallel Fingerprint Recognition System Architecture for Performing Multi Million Marches per Second. For more details, visit www.ictrdf.org.pk/fp-mpfrs.htm.

Accurate automatic personal identification is critical in a variety of applications in an electronically interconnected society. The need and interest in personal identification systems has increased manifolds due to security concerns at national and international level.

Among all the biometric techniques, fingerprint-based identification systems have received the most attention because of the long history of fingerprints and their extensive use in forensics. An automatic recognition of people based on fingerprints requires that the input fingerprint be matched with a large number of fingerprints in a database and high speed recognition is required.

There are not many companies which are providing High Density AFIS System. The offered systems are very expensive in the tune of few US Million dollars and thus present a good business opportunity for positioning. Most of these systems are SW based solution and employ a cluster of computers for high density matching. This research project will develop a high speed FPGA-based scalable Fingerprint Recognition System Architecture. The architecture will be scalable for higher throughput applications. The project aims to develop matching algorithms, which can be efficiently mapped in Hardware for subsequent mapping on FPGA-based system. The project proposes to develop this massively parallel fingerprint recognition system architecture for performing multimillion matches per second for high-density applications.