STAMP: A Real-Time and Low-Power Sampling Error Based Stochastic Number Generator

Mahmut Burak Karadeniz; Ismail Cevik; Mustafa Altun

doi:10.1109/ACCESS.2021.3125672

STAMP: A Real-Time and Low-Power Sampling Error Based Stochastic Number Generator

Mahmut Burak Karadeniz^*, Ismail Cevik, Mustafa Altun

^*Corresponding author for this work

Department of Electronics and Communication Engineering

Istanbul Technical University

Research output: Contribution to journal › Article › peer-review

Abstract

In this paper, we introduce STAMP - a real-time and low-power sampling error based stochastic number generator - for stochastic computing circuits. STAMP exploits the stochastic nature of sampling error; its name is derived from 'stochastic' and 'sampling'. A unique feature of STAMP which distinguishes it from other random generators is the ability to control the output probability of the generated stochastic bit stream in real-time with no area overhead. STAMP is implemented in 180 nm CMOS. Measurements have shown that STAMP passes all tests in the suit of the National Institute of Standards and Technology (NIST) and outperforms the benchmark random number generators in terms of randomness quality. STAMP performs 140 Mb/s throughput with energy consumption of 77 pJ/bit.

Original language	English
Pages (from-to)	151363-151372
Number of pages	10
Journal	IEEE Access
Volume	9
DOIs	https://doi.org/10.1109/ACCESS.2021.3125672
Publication status	Published - 2021

Bibliographical note

Publisher Copyright:
© 2013 IEEE.

Funding

This work was supported in part by the Scientific and Technological Research Council of Turkey (TUBITAK) through the Support Programme for Scientific and Technological Research Projects (1001) under Grant 116E250 and Grant 119E507, and in part by the Istanbul Technical University Scientific Research Project (BAP) under Grant 42435.

Funders	Funder number
Istanbul Technical University Scientific Research Project
TUBITAK	119E507, 116E250
British Association for Psychopharmacology	42435
Türkiye Bilimsel ve Teknolojik Araştirma Kurumu

Keywords

CMOS
LFSR
random number generator
TRNG

Access to Document

10.1109/ACCESS.2021.3125672

Cite this

@article{bfdfe5079b2f421a8bb1d2c2e273a919,

title = "STAMP: A Real-Time and Low-Power Sampling Error Based Stochastic Number Generator",

abstract = "In this paper, we introduce STAMP - a real-time and low-power sampling error based stochastic number generator - for stochastic computing circuits. STAMP exploits the stochastic nature of sampling error; its name is derived from 'stochastic' and 'sampling'. A unique feature of STAMP which distinguishes it from other random generators is the ability to control the output probability of the generated stochastic bit stream in real-time with no area overhead. STAMP is implemented in 180 nm CMOS. Measurements have shown that STAMP passes all tests in the suit of the National Institute of Standards and Technology (NIST) and outperforms the benchmark random number generators in terms of randomness quality. STAMP performs 140 Mb/s throughput with energy consumption of 77 pJ/bit.",

keywords = "CMOS, LFSR, random number generator, TRNG",

author = "Karadeniz, {Mahmut Burak} and Ismail Cevik and Mustafa Altun",

note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",

year = "2021",

doi = "10.1109/ACCESS.2021.3125672",

language = "English",

volume = "9",

pages = "151363--151372",

journal = "IEEE Access",

issn = "2169-3536",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - STAMP

T2 - A Real-Time and Low-Power Sampling Error Based Stochastic Number Generator

AU - Karadeniz, Mahmut Burak

AU - Cevik, Ismail

AU - Altun, Mustafa

PY - 2021

Y1 - 2021

N2 - In this paper, we introduce STAMP - a real-time and low-power sampling error based stochastic number generator - for stochastic computing circuits. STAMP exploits the stochastic nature of sampling error; its name is derived from 'stochastic' and 'sampling'. A unique feature of STAMP which distinguishes it from other random generators is the ability to control the output probability of the generated stochastic bit stream in real-time with no area overhead. STAMP is implemented in 180 nm CMOS. Measurements have shown that STAMP passes all tests in the suit of the National Institute of Standards and Technology (NIST) and outperforms the benchmark random number generators in terms of randomness quality. STAMP performs 140 Mb/s throughput with energy consumption of 77 pJ/bit.

AB - In this paper, we introduce STAMP - a real-time and low-power sampling error based stochastic number generator - for stochastic computing circuits. STAMP exploits the stochastic nature of sampling error; its name is derived from 'stochastic' and 'sampling'. A unique feature of STAMP which distinguishes it from other random generators is the ability to control the output probability of the generated stochastic bit stream in real-time with no area overhead. STAMP is implemented in 180 nm CMOS. Measurements have shown that STAMP passes all tests in the suit of the National Institute of Standards and Technology (NIST) and outperforms the benchmark random number generators in terms of randomness quality. STAMP performs 140 Mb/s throughput with energy consumption of 77 pJ/bit.

KW - CMOS

KW - LFSR

KW - random number generator

KW - TRNG

UR - http://www.scopus.com/inward/record.url?scp=85119718793&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2021.3125672

DO - 10.1109/ACCESS.2021.3125672

M3 - Article

AN - SCOPUS:85119718793

SN - 2169-3536

VL - 9

SP - 151363

EP - 151372

JO - IEEE Access

JF - IEEE Access

ER -

STAMP: A Real-Time and Low-Power Sampling Error Based Stochastic Number Generator

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this