Embedded Algorithmic Noise-Tolerance for Signal Processing and Machine Learning Systems via Data Path Decomposition

Research output: Contribution to journalArticlepeer-review


Low overhead error-resiliency techniques such as algorithmic noise-tolerance (ANT) have been shown to be particularly effective for signal processing and machine learning kernels. However, the overhead of conventional ANT can be as large as 30% due to the use of explicit estimator block. To overcome this overhead, embedded ANT (E-ANT) is proposed [S. Zhang and N. Shanbhag, 'Embedded error compensation for energy efficient DSP systems,' in Proc. GlobalSIP, Dec. 2014, pp. 30-34], where the estimator is embedded in the main computation block via data path decomposition (DPD). E-ANT reduces the logic overhead to be below 8% as compared with the 20%-30% associated with conventional reduced precision replica (RPR) ANT system while maintaining the same error compensation functionality. DPD was first proposed in our original paper [Zhang and Shanbhag, 2014] where its benefits were studied in the case of a simple multiply-accumulator (MAC) kernel. This paper builds upon [Zhang and Shanbhag, 2014] by 1) providing conditions for the existence of DPD, 2) demonstrating DPD for a variety of commonly employed kernels in signal processing and machine learning applications, and 3) evaluating the robustness improvement and energy savings of DPD at the system level for an SVM-based EEG seizure classification application. Simulation results in a commercial 45 nm CMOS process show that E-ANT can compensate for error rates up to 0.38 for errors in FE only, and 0.17 for errors in FE and CE, while maintain a true positive rate ${p}-{\rm tp} > 0.9$ and a false positive rate ${p}-{\rm fp}\leq 0.01$. This represents a greater than 3-orders-of-magnitude improvement in error tolerance over the conventional system. This error tolerance is employed to reduce energy via the use of voltage overscaling (VOS). E-ANT is able to achieve 51% energy savings when errors are in FE only, and up to 43% savings when errors are in both FE and CE.

Original languageEnglish (US)
Article number7439866
Pages (from-to)3338-3350
Number of pages13
JournalIEEE Transactions on Signal Processing
Issue number13
StatePublished - Jul 1 2016

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering


Dive into the research topics of 'Embedded Algorithmic Noise-Tolerance for Signal Processing and Machine Learning Systems via Data Path Decomposition'. Together they form a unique fingerprint.

Cite this