Compact and accurate stochastic circuits with shared random number sources

Ichihara, Hideyuki; Ishii, Shota; Daiki, Sunamori; Iwagaki, Tsuyoshi; Inoue, Tomoo

doi:10.1109/iccd.2014.6974706

Cited by 68 publications

(41 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also, using one RNG will avoid one of the drawbacks mentioned by Hayes [12], which is conversion circuit area cost. In the case that requires an uncorrelated random number sequence such as generating MUX selector SN, the circular shift RNG sharing scheme is used as proposed in [13].…”

Section: Correlating Stochastic Numbers Methodsmentioning

confidence: 99%

“…However, sometimes uncorrelated bit-streams are required, such as the MUX selector SN. In this case the circular shifting RNG sharing scheme should be used as proposed in [13]. Therefore, a single RNG is enough for the complete CSC system.…”

Section: Correlated Stochastic Computing System Design Methodologymentioning

confidence: 99%

“…Although SC's benefit is the low area cost, the conversion circuits can take up to 80% of the overall area, particularly when independent RNGs are needed for SN generations [12]. Some works proposed sharing schemes to share one RNG with different SNGs [4,[13][14][15]. For example, if k different SNGs share only one RNG as in the case in the circular shifting in [4], the area cost of the RNGs is decreased by 1/k .…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Accurate and compact stochastic computations by exploiting correlation

Abdellatef¹,

Khalil-Hani²,

Shaikh-Husin³

2019

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Recent studies have shown, contrary to what was previously believed, that by exploiting correlation in stochastic computing (SC) designs, more accurate SC circuits with low area cost can be realized. However, if these basic SC circuits or blocks are cascaded in series to form a large complex system, correlation between stochastic numbers (SNs) from one block to the next would be lost; thus, inaccuracies are introduced. In this study, we propose correlating circuits to be used in building complex correlated SC systems. One of the circuits is the correlator that restores lost correlations between two SNs due to previous processing. In addition, a correlated SN generator is introduced to generate SN correlated to a specific SN. Experimental results show that our methods have improved the accuracy of stochastic computation and preserved the stochastic computing correlation without the need for conversion from SC to the conventional binary-encoded computing, and vice versa. Consequently, lower latency and lower area cost are achieved.

show abstract

Section: Correlating Stochastic Numbers Methodsmentioning

confidence: 99%

Section: Correlated Stochastic Computing System Design Methodologymentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Accurate and compact stochastic computations by exploiting correlation

Abdellatef¹,

Khalil-Hani²,

Shaikh-Husin³

2019

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…Note that, a new "LFSR" is not necessarily required for generation of random bit streams in 2 n RRR duplicators. Instead, an LFSR can be reused and can be considered to generate another random bit stream [7]. Reference [7] shows that one SN generator can share LFSR with other SN generators.…”

Section: Instances Of the 2 N Rrr And Their Hardware Costsmentioning

confidence: 99%

Scalable Stochastic Number Duplicators for Accuracy-flexible Arithmetic Circuit Design

Ishikawa

Tawada

Yanagisawa

et al. 2020

IPSJ Transactions on System LSI Design Methodology

View full text Add to dashboard Cite

Stochastic computing is a computation method which can implement arithmetic operations by simple logic circuits. Stochastic numbers are used in this method, whose values are defined by their bit streams' appearance rates of 1's. As a nature of stochastic computing, changing the length of the input stochastic numbers will change the whole circuit's accuracy. However, in some implementations with re-convergence paths, the circuit itself will cause errors, i.e., the length of the input stochastic numbers does not change that circuit's accuracy. This paper proposes a stochastic number duplicator whose outputs differ every time and are all independent. This stochastic number duplicator has a scalable structure by changing the numbers of flip-flops for bit re-arrangement. From the experimental evaluations and discussions, we clarify that the proposed stochastic number duplicator enables accuracy-flexible circuits. interests are combinatorics and graph theory, computational geometry, VLSI design and verification, and network analysis and design. He is a member of IEEE, ACM, IPSJ, and IEICE.

show abstract

“…However, the extensive use of stochastic computation is still limited, because of its long run-time and inaccuracy. Recent improvements have mainly focused on improving the accuracy and performance of the stochastic circuits by sharing consecutive bit streams, sharing the stochastic number generators, using true random generators, exploiting the correlation and using the spectral approach for stochastic circuit synthesis [11][12][13][14][15][16][17]. This paper also explores new methods to improve the accuracy and performance of stochastic circuits.…”

Section: Introductionmentioning

confidence: 99%

Implementation Of Reconfigurability Number Generators For Digital Bit Stream Signals

2018

IJRTER

View full text Add to dashboard Cite

The high performance of FPGA (Field Programmable Gate Array) in image processing applications is justified by its flexible reconfigurability, its inherent parallel nature and the availability of a large amount of internal memories. Lately, the Stochastic Computing (SC) paradigm has been found to be significantly advantageous in certain application domains including image processing because of its lower hardware complexity and power consumption. However, its viability is deemed to be limited due to its serial bit stream processing and excessive runtime requirement for convergence. To address these issues, a novel approach is proposed in this work where an energy-efficient implementation of SC is accomplished by introducing fast-converging Quasi-Stochastic Number Generators (QSNGs) and parallel stochastic bit stream processing, which are well suited to leverage FPGA's reconfigurability and abundant internal memory resources. the proposed approach has been tested on the Virtex-4 FPGA, and results have been compared with the serial and parallel implementations of conventional stochastic computation using the well-known SC edge detection and multiplication circuits. Results prove that by using this approach, execution time, as well as the power consumption are decreased by a factor of 3.5 and 4.5 for the edge detection circuit and multiplication circuit, respectively.

show abstract

Compact and accurate stochastic circuits with shared random number sources

Cited by 68 publications

References 10 publications

Accurate and compact stochastic computations by exploiting correlation

Accurate and compact stochastic computations by exploiting correlation

Scalable Stochastic Number Duplicators for Accuracy-flexible Arithmetic Circuit Design

Implementation Of Reconfigurability Number Generators For Digital Bit Stream Signals

Contact Info

Product

Resources

About