Repository logo
 

DUAL SEGMENTED AND RECONFIGURABLE APPROXIMATE MULTIPLIERS FOR ERROR-TOLERANT APPLICATIONS

dc.contributor.authorLi, Ling
dc.contributor.copyright-releaseYesen_US
dc.contributor.degreeMaster of Applied Scienceen_US
dc.contributor.departmentDepartment of Electrical & Computer Engineeringen_US
dc.contributor.ethics-approvalReceiveden_US
dc.contributor.external-examinern/aen_US
dc.contributor.graduate-coordinatorDr. Dmitry Trukhacheven_US
dc.contributor.manuscriptsYesen_US
dc.contributor.thesis-readerDr. William Phillipsen_US
dc.contributor.thesis-readerDr. Jason Guen_US
dc.contributor.thesis-supervisorDr. Kamal El-Sankaryen_US
dc.date.accessioned2021-07-06T17:28:42Z
dc.date.available2021-07-06T17:28:42Z
dc.date.defence2021-06-24
dc.date.issued2021-07-06T17:28:42Z
dc.description.abstractApproximate multiplier circuit designs have shown substantial advantages in improving many operational features, such as power, area and delay, in many error-resilient applications such as image processing and deep learning applications. Existing approximate multiplier circuits in this thesis are first reviewed, evaluated, and compared. The comparison results show that the segment-based multiplier has a good trade-off between accuracy and performance by adjusting segment size. A dual segmentation approximate multiplier is then proposed. Compared to the dynamic segment method (DSM)-based approximate multiplier, the proposed design can reduce the energy by 37.90% for 32-bit multipliers, and by 16.68% for 16-bit multipliers. The DSM and proposed multipliers have almost identical accuracy. A merged approximate multiplier with two configurable precisions is proposed for improving the multiplication performance in fixed point convolutional neural networks (CNN) accelerators. Compared with the single-precision approximate multiplier, the merged approximate multiplier achieves significant performance enhancements with minimal accuracy loss.en_US
dc.identifier.urihttp://hdl.handle.net/10222/80583
dc.subjectApproximate Computingen_US
dc.subjectApproximate Multiplieren_US
dc.subjectAccuracy Configurableen_US
dc.subjectHigh Speeden_US
dc.subjectEnergy Efficienten_US
dc.subjectLow Poweren_US
dc.titleDUAL SEGMENTED AND RECONFIGURABLE APPROXIMATE MULTIPLIERS FOR ERROR-TOLERANT APPLICATIONSen_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
LingLi2021.pdf
Size:
2.5 MB
Format:
Adobe Portable Document Format
Description:
Main article

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description: