A compact ultra low-power pulse delay and extension circuit for neuromorphic processors

Carsten Nielsen, Ning Qiao, G. Indiveri
{"title":"A compact ultra low-power pulse delay and extension circuit for neuromorphic processors","authors":"Carsten Nielsen, Ning Qiao, G. Indiveri","doi":"10.1109/BIOCAS.2017.8325234","DOIUrl":null,"url":null,"abstract":"Although silicon neurons communicate among each other using fast spikes, neuromorphic architectures often require long delays and pulse lengths to process temporal signals. In this paper we present a compact and power efficient pulse extension circuit that can convert short spike events into delayed pulses with configurable delay and pulse lengths that range from fractions of microseconds up to tens of milliseconds. The circuit proposed can be used to realize programmable axonal delays in neuromorphic architectures and to support the generation of synaptic dynamics with biologically plausible pulse lengths in mixed-signal analog/digital circuits. To validate the proposed scheme, we designed the pulse delay and extension circuit using a standard 0.18 μm CMOS process and performed post-layout Monte Carlo simulations. We describe the circuit and demonstrate how it can be configured to obtain biological long configurable delays and extension periods. We assess its operation via circuit simulation results and present an analysis of the Monte Carlo simulations that shows how the proposed circuit is resistant to mismatch with a standard deviation in the produced delay and pulse periods of less than 2%.","PeriodicalId":361477,"journal":{"name":"2017 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 IEEE Biomedical Circuits and Systems Conference (BioCAS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/BIOCAS.2017.8325234","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6

Abstract

Although silicon neurons communicate among each other using fast spikes, neuromorphic architectures often require long delays and pulse lengths to process temporal signals. In this paper we present a compact and power efficient pulse extension circuit that can convert short spike events into delayed pulses with configurable delay and pulse lengths that range from fractions of microseconds up to tens of milliseconds. The circuit proposed can be used to realize programmable axonal delays in neuromorphic architectures and to support the generation of synaptic dynamics with biologically plausible pulse lengths in mixed-signal analog/digital circuits. To validate the proposed scheme, we designed the pulse delay and extension circuit using a standard 0.18 μm CMOS process and performed post-layout Monte Carlo simulations. We describe the circuit and demonstrate how it can be configured to obtain biological long configurable delays and extension periods. We assess its operation via circuit simulation results and present an analysis of the Monte Carlo simulations that shows how the proposed circuit is resistant to mismatch with a standard deviation in the produced delay and pulse periods of less than 2%.
一种用于神经形态处理器的紧凑型超低功耗脉冲延迟和扩展电路
虽然硅神经元之间使用快速尖峰进行通信,但神经形态架构通常需要较长的延迟和脉冲长度来处理时间信号。在本文中,我们提出了一种紧凑和高效的脉冲扩展电路,它可以将短脉冲事件转换为具有可配置延迟和脉冲长度的延迟脉冲,其范围从微秒到几十毫秒不等。所提出的电路可用于实现神经形态结构中的可编程轴突延迟,并支持在混合信号模拟/数字电路中产生具有生物学上合理脉冲长度的突触动力学。为了验证所提出的方案,我们采用标准的0.18 μm CMOS工艺设计了脉冲延迟和扩展电路,并进行了布局后的蒙特卡罗模拟。我们描述了电路并演示了如何配置它以获得生物长的可配置延迟和延长周期。我们通过电路仿真结果评估其操作,并对蒙特卡罗模拟进行了分析,显示了所提出的电路如何在产生的延迟和脉冲周期小于2%的标准偏差下抵抗不匹配。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信