Fine Grain Precision Scaling for Datapath Approximations in Digital Signal Processing Systems - VLSI-SoC: At the Crossroads of Emerging Trends Access content directly
Conference Papers Year : 2015

Fine Grain Precision Scaling for Datapath Approximations in Digital Signal Processing Systems

Seogoo Lee
  • Function : Author
  • PersonId : 990375
Andreas Gerstlauer
  • Function : Author
  • PersonId : 990376

Abstract

Finding optimal word lengths in digital signal processing systems has been one of the primary mechanisms for reducing complexity. Recently, this topic has been explored in a broader approximate computing context, where architectures allowing for fine-grain control of hardware or software accuracy have been proposed. One of the obstacles for adoption of fine-grain scaling techniques is that they require determining the precision of all intermediate values at all possible operation points, making simulation-based optimization infeasible. In this chapter, we study efficient analytical heuristics to find optimal sets of word lengths for all variables and operations in a dataflow graph constrained by mean squared error type of metrics. We apply our method to several industrial-strength examples. Our results show a more than 5,000x improvement in optimization time compared to an efficient simulation-based word length optimization method with less than 10 % estimation error across a range of target quality metrics.
Fichier principal
Vignette du fichier
367527_1_En_6_Chapter.pdf (514.82 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01380301 , version 1 (12-10-2016)

Licence

Attribution

Identifiers

Cite

Seogoo Lee, Andreas Gerstlauer. Fine Grain Precision Scaling for Datapath Approximations in Digital Signal Processing Systems. 21th IFIP/IEEE International Conference on Very Large Scale Integration - System on a Chip (VLSI-SoC), Oct 2013, Istanbul, Turkey. pp.119-143, ⟨10.1007/978-3-319-23799-2_6⟩. ⟨hal-01380301⟩
70 View
115 Download

Altmetric

Share

Gmail Facebook X LinkedIn More