A Novel Delay-Aware Packing Algorithm for FPGA Architecture Using RFET
Authors:
- 11 August 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 362-366
Abstract
Reconfigurable devices are gaining increasing attention as a viable alternative and supplementary solution to traditional CMOS technology. In this paper, we develop a more efficient 2-input look-up table (LUT) based on the reconfigurable field-effect transistors (RFETs), leading to a smaller transistor usage and a smaller critical path delay. The cells are organized into regular matrices, known as MClusters, with a fixed interconnection pattern to replace LUTs in field-programmable gate arrays (FPGAs). To improve the efficiency of utilizing this structure, we design a SAT-based delay-aware packing algorithm to better utilize logical gates for the MCluster structure. Finally, we combine this algorithm with FPGA simulation tools to form a comprehensive benchmarking flow. A series of benchmark tests show that under the optimal design, up to 35% and 30% reduction can be achieved in delay and energy-delay product (EDP), respectively, compared to the traditional CMOS FPGAs.Keywords
This publication has 17 references indexed in Scilit:
- Towards Area Efficient Logic Circuit: Exploring Potential of Reconfigurable Gate by Generic Exact SynthesisIEEE Open Journal of the Computer Society, 2023
- Reconfigurable field effect transistors: A technology enablers perspectiveSolid-State Electronics, 2022
- A wired-AND transistor: Polarity controllable FET with multiple inputs2018
- Hybrid Topologies for Reconfigurable Matrices Based on Nano-Grain Cells2017
- ASAP7: A 7-nm finFET predictive process design kitMicroelectronics Journal, 2016
- A novel design of a memristor-based look-up table (LUT) for FPGA2014
- A Proposal for a Novel Hybrid Interconnect Technology for the End of RoadmapIEEE Electron Device Letters, 2013
- ABC: An Academic Industrial-Strength Verification Tool2010
- Activity Estimation for Field-Programmable Gate Arrays2006
- Architecture and CAD for Deep-Submicron FPGAS1999