• Complex
  • Title
  • Keyword
  • Abstract
  • Scholars
  • Journal
  • ISSN
  • Conference
成果搜索

author:

Wang, Qin (Wang, Qin.) [1]

Indexed by:

EI Scopus

Abstract:

With the advancement of technology, the energy consumption of computing devices continues to rise. As a new paradigm to address energy challenges in error-tolerant environments, approximate computing holds significant importance. This paper proposes an approximate multiplier-the Low-Carry Unsigned Approximate Multiplier (UAM8-FP), an 8×8 approximate multiplier design. The key innovation of UAM8-FP lies in its exclusive use of carry-free compressors, which drastically reduces carry propagation during partial product reduction and significantly accelerates partial product accumulation. The second core feature of UAM8-FP's compressor strategy is its phase-dependent adaptive compression technique, where compressors of varying approximation densities are deployed based on the sparsity of '1's at different reduction stages. This approach not only enhances computational speed but also minimizes precision deviation probability. To further optimize performance, a third strategy is introduced: customized compressor allocation, where exact and approximate compressors are dynamically selected based on partial product distribution. By integrating carry reduction, staged approximation, and column-wise optimization, UAM8-FP achieves a balanced trade-off among speed, hardware cost, and accuracy. Compared to existing approximate multipliers, it demonstrates strong competitiveness, offering both academic and practical value. © 2025 IEEE.

Keyword:

Compressors Economic and social effects Energy utilization Frequency multiplying circuits Green computing Multiplying circuits

Community:

  • [ 1 ] [Wang, Qin]College of Physics and Information Engineering, Fuzhou University, Fuzhou, China

Reprint 's Address:

  • 王钦

Email:

Show more details

Version:

Related Keywords:

Source :

Year: 2025

Page: 278-282

Language: English

Cited Count:

WoS CC Cited Count:

SCOPUS Cited Count:

ESI Highly Cited Papers on the List: 0 Unfold All

WanFang Cited Count:

Chinese Cited Count:

30 Days PV: 3

Online/Total:758/13855253
Address:FZU Library(No.2 Xuyuan Road, Fuzhou, Fujian, PRC Post Code:350116) Contact Us:0591-22865326
Copyright:FZU Library Technical Support:Beijing Aegean Software Co., Ltd. 闽ICP备05005463号-1