A. Linares-Barranco, J. Sevillano, M. Obaidat, N. Ferrando, J. Cerdá, Daniel Cascado Caballero, G. Jiménez-Moreno, A. C. Balcells
{"title":"AER filtering using GLIDER: VHDL cellular automata description","authors":"A. Linares-Barranco, J. Sevillano, M. Obaidat, N. Ferrando, J. Cerdá, Daniel Cascado Caballero, G. Jiménez-Moreno, A. C. Balcells","doi":"10.1109/ICECS.2008.4674928","DOIUrl":null,"url":null,"abstract":"Cellular automata (CA) is a bio-inspired processing model for problem solving, initially proposed by Von Neumann. This approach modularizes the processing by dividing the solution into synchronous cells that change their states at the same time in order to get the solution. The communication between them is crucial to achieve the correct solution. On the other hand, the address-event-representation (AER) is a neuromorphic communication protocol for transferring asynchronous events between VLSI chips. These neuro-inspired implementations have been used to design sensor chips (retina, cochleas), processing chips (convolutions, filters) and learning chips, which makes it possible to develop complex, multilayer, multichip neuromorphic systems. This paper presents the fusion of both bio-inspired solutions for implementing a vision filter based on 3times3 convolutions. The GLIDER software tool for developing CA has been used to implement the filter in VHDL and synthesize it into the Spartan II 200 of the USB-AER.","PeriodicalId":404629,"journal":{"name":"2008 15th IEEE International Conference on Electronics, Circuits and Systems","volume":"5 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 15th IEEE International Conference on Electronics, Circuits and Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICECS.2008.4674928","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
Cellular automata (CA) is a bio-inspired processing model for problem solving, initially proposed by Von Neumann. This approach modularizes the processing by dividing the solution into synchronous cells that change their states at the same time in order to get the solution. The communication between them is crucial to achieve the correct solution. On the other hand, the address-event-representation (AER) is a neuromorphic communication protocol for transferring asynchronous events between VLSI chips. These neuro-inspired implementations have been used to design sensor chips (retina, cochleas), processing chips (convolutions, filters) and learning chips, which makes it possible to develop complex, multilayer, multichip neuromorphic systems. This paper presents the fusion of both bio-inspired solutions for implementing a vision filter based on 3times3 convolutions. The GLIDER software tool for developing CA has been used to implement the filter in VHDL and synthesize it into the Spartan II 200 of the USB-AER.
细胞自动机(CA)是一种解决问题的生物启发处理模型,最初由冯·诺伊曼提出。这种方法通过将解决方案划分为同步单元来模块化处理,这些单元同时更改其状态以获得解决方案。他们之间的沟通对于达成正确的解决方案至关重要。另一方面,地址事件表示(AER)是一种神经形态的通信协议,用于在VLSI芯片之间传输异步事件。这些受神经启发的实现已被用于设计传感器芯片(视网膜、耳蜗)、处理芯片(卷积、滤波器)和学习芯片,这使得开发复杂的、多层的、多芯片的神经形态系统成为可能。本文提出了两种仿生解决方案的融合,以实现基于3times3卷积的视觉滤波器。利用GLIDER开发CA的软件工具在VHDL中实现该滤波器,并将其合成到USB-AER的Spartan II 200中。