Design of an energy-efficient binarized convolutional neural network accelerator using a nonvolatile field-programmable gate array with only-once-write shifting
- 25 March 2021
- journal article
- research article
- Published by IOP Publishing in Japanese Journal of Applied Physics
- Vol. 60 (SB), SBBB07
- https://doi.org/10.35848/1347-4065/abe682
Abstract
This paper presents an energy-efficient hardware accelerator for binarized convolutional neural networks (BCNNs). In such BCNN accelerator, a data-shift operation becomes dominant to effectively control input/weight-data streams under limited memory bandwidth. A magnetic-tunnel-junction (MTJ)-based nonvolatile field-programmable gate array (FPGA), where the number of stored-data updating is minimized in configurable logic block, is well suited hardware platform for implementing such BCNN accelerator. Owing to the nonvolatile storage capability of the NV-FPGA, not only power consumption in data-shift operation, but also standby power consumption in idle function block is reduced without losing internal data. It is demonstrated under 45nm CMOS/MTJ process technologies that the energy consumption of the proposed BCNN accelerator is 50.7% lower than that of a BCNN accelerator using a conventional SRAM-based FPGA.Keywords
This publication has 28 references indexed in Scilit:
- Standby-Power-Free Integrated Circuits Using MTJ-Based VLSI ComputingProceedings of the IEEE, 2016
- Edge Computing: Vision and ChallengesIEEE Internet of Things Journal, 2016
- Subthreshold Operation of CAAC-IGZO FPGA by Overdriving of Programmable Routing Switch and Programmable Power SwitchIEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2016
- A spin–orbit torque switching scheme with collinear magnetic easy axis and current configurationNature Nanotechnology, 2016
- Magnetization switching by spin–orbit torque in an antiferromagnet–ferromagnet bilayer systemNature Materials, 2016
- Can we connect trillions of IoT sensors in a sustainable way? A technology/circuit perspective (Invited)Published by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- Fabrication of a 3000-6-input-LUTs embedded and block-level power-gated nonvolatile FPGA chip using p-MTJ-based logic-in-memory structurePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- High-speed simulator including accurate MTJ models for spintronics integrated circuit designPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2012
- A perpendicular-anisotropy CoFeB–MgO magnetic tunnel junctionNature Materials, 2010
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998