Coarse-grained fabrics are comprised of multi-bit configurable logic blocks and configurable interconnect. This work is focused on area and energy optimization techniques for coarse-grained reconfigurable fabric architectures. In this work, a variety of design techniques have been explored to improve the utilization of computational resources and increase energy savings. This includes splitting, folding, multi-level vertical interconnect. In addition to this, I have also studied fully connected homogeneous and heterogeneous architectures, and 3D architecture. I have also examined some of the hybrid strategies of computation unit’s arrangements. In order to perform energy and area analysis, I selected a set of signal and image processing benchmarks from MediaBench suite. I implemented various fabric architectures on 90nm ASIC process from Synopsys. Results show area improvement with energy savings as compared to baseline architecture.
| Identifer | oai:union.ndltd.org:unt.edu/info:ark/67531/metadc103413 |
| Date | 12 1900 |
| Creators | Yadav, Anil |
| Contributors | Mehta, Gayatri, Namuduri, Kamesh, Gomathisankaran, Mahadevan |
| Publisher | University of North Texas |
| Source Sets | University of North Texas |
| Language | English |
| Detected Language | English |
| Type | Thesis or Dissertation |
| Format | Text |
| Rights | Public, Yadav, Anil, Copyright, Copyright is held by the author, unless otherwise noted. All rights Reserved. |
Page generated in 0.347 seconds