Abstract

In this paper, we present a new multilevel hierarchical (Tree-based) coarse-grained FPGA architecture. This architecture comprises two unidirectional interconnects, a downward interconnect and an upward interconnect. The proposed architecture can support various kinds of coarse-grained blocks. These coarse-grained blocks are defined using an architecture description file. A new software flow has been developed to evaluate the proposed architecture. New tools are developed to support this software flow and they have resulted in the successful placement and routing of different DSP benchmarks on the proposed architecture. A comparison of coarse-grained Treebased and fine-grained Tree-based architectures is performed. This comparison reveals an average area gain of 41% for coarsegrained Tree-based architecture over fine-grained Tree-based architecture. Similarly a comparison of Tree-based and Meshbased coarse-grained architectures shows an average area saving of 60% for Tree-based coarse-grained architectures over Meshbased coarse-grained architectures.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.