Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.tuc4
A. Dickinson
The high end of microprocessor performance is currently dominated by Reduced Instruction Set Computer (RISC) architectures. These machines execute one or more instructions per clock cycle. A processor such as the i8601 [1] runs with a 40MHz clock - requiring that on average an instruction must be delivered to the CPU every 25nS. With DRAM access times currently at around 100nS, timely instruction delivery has become a critical constraint on processor speed.
{"title":"An Optical Respite from the Von Neumann Bottleneck","authors":"A. Dickinson","doi":"10.1364/optcomp.1991.tuc4","DOIUrl":"https://doi.org/10.1364/optcomp.1991.tuc4","url":null,"abstract":"The high end of microprocessor performance is currently dominated by Reduced Instruction Set Computer (RISC) architectures. These machines execute one or more instructions per clock cycle. A processor such as the i8601 [1] runs with a 40MHz clock - requiring that on average an instruction must be delivered to the CPU every 25nS. With DRAM access times currently at around 100nS, timely instruction delivery has become a critical constraint on processor speed.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115606648","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.me1
P. Mathey, G. Pauliat, J. Launay, G. Roosen
A large number of proposed or demonstrated architectures for optical computing takes advantage of the unique properties of photorefractive crystals.
大量的提议或展示了体系结构的光学计算利用光折变晶体的独特性能。
{"title":"Huge optical amplification by applying pulsed electric fields to photorefractive crystals","authors":"P. Mathey, G. Pauliat, J. Launay, G. Roosen","doi":"10.1364/optcomp.1991.me1","DOIUrl":"https://doi.org/10.1364/optcomp.1991.me1","url":null,"abstract":"A large number of proposed or demonstrated architectures for optical computing takes advantage of the unique properties of photorefractive crystals.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115501784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.tua3
C. Waterson, B. K. Jenkins
Several abstract models of parallel computation have been developed and studied by the computer science and parallel processing communities [1, 2]. The shared memory models are among the most computationally powerful of these models. They benefit from substantial theoretical foundations, and many algorithms have been mapped onto these models in order to characterize theoretically optimum parallel performance. A number of attempts have been made to develop electronic parallel architectures based on the shared memory model. Most of them have been unsuccessful, primarily due to the complexity of the interconnection network hardware and its associated control.
{"title":"Shared Memory Optical/Electronic Computer: Architecture Design","authors":"C. Waterson, B. K. Jenkins","doi":"10.1364/optcomp.1991.tua3","DOIUrl":"https://doi.org/10.1364/optcomp.1991.tua3","url":null,"abstract":"Several abstract models of parallel computation have been developed and studied by the computer science and parallel processing communities [1, 2]. The shared memory models are among the most computationally powerful of these models. They benefit from substantial theoretical foundations, and many algorithms have been mapped onto these models in order to characterize theoretically optimum parallel performance. A number of attempts have been made to develop electronic parallel architectures based on the shared memory model. Most of them have been unsuccessful, primarily due to the complexity of the interconnection network hardware and its associated control.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129176782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.md2
D. Jared, R. Turner, K. Johnson
This paper discusses several design and fabrication issues surrounding VLSI, ferroelectric liquid crystal (FLC) spatial light modulators (SLMs). These SLMs consist of a VLSI CMOS backplane and FLC modulators as shown in Fig. (1). The FLC is sandwiched between the CMOS backplane and a sheet of glass coated with a transparent conductor. The design and fabrication issues that are described include: FLC material selection, alignment of the FLC, installation of the glass cover, and the design of photodetectors, amplifiers, and pad drivers. An electrically addressed dynamic RAM SLM with 64 × 64 pixels and three optically addressed SLMs with 32 × 32 pixels are described to discuss these issues.
{"title":"Design and fabrication of VLSI ferroelectric liquid crystal spatial light modulators","authors":"D. Jared, R. Turner, K. Johnson","doi":"10.1364/optcomp.1991.md2","DOIUrl":"https://doi.org/10.1364/optcomp.1991.md2","url":null,"abstract":"This paper discusses several design and fabrication issues surrounding VLSI, ferroelectric liquid crystal (FLC) spatial light modulators (SLMs). These SLMs consist of a VLSI CMOS backplane and FLC modulators as shown in Fig. (1). The FLC is sandwiched between the CMOS backplane and a sheet of glass coated with a transparent conductor. The design and fabrication issues that are described include: FLC material selection, alignment of the FLC, installation of the glass cover, and the design of photodetectors, amplifiers, and pad drivers. An electrically addressed dynamic RAM SLM with 64 × 64 pixels and three optically addressed SLMs with 32 × 32 pixels are described to discuss these issues.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125918269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.tud5
H. Rajbenbach, S. Bann, J. Huignard
Technology advances in solid state lasers, spatial light modulators and nonlinear optical materials are centrally important for the construction of optoelectronics processors that combine the massive interconnectivity and parallelism of optics with the accuracy and flexibility of digital electronics. In pattern recognition applications, hybrid optical-digital approaches in which optics performs correlation operations and electronics processes the output correlation plane for classification have already been demonstrated(1−2). Today, the performances of semiconductor lasers, diode-pumped YAG lasers, two dimensional liquid crystal light modulators and photorefractive materials allow the introduction of compact and more flexible optical hardware in optoelectronic processors. In this paper, we present a compact and reconfigurable multichannel joint transform optical correlator designed and constructed for industrial recognition applications. The principle of operation is shown in Fig.1. The object to be identified S(x,y) is display on one half of the input scene. The other half of the input, allocated to the reference R(x,y) is split in N subarrays, or channels, each containing a reference object or a calculated version of reference object. The sum R(x,y) + S(x,y) is Fourier transformed and the spectrum is recorded in a dynamic holographic medium. The complex light field produced by reading out the joint-transform power spectrum contains the cross-correlation component R(x,y) ⊗ S (x-2a, y), where 2a is the separation between signal and reference and ⊗ denotes the correlation operation(3). The identification is performed by detecting the position and relative intensities of the correlation peaks in the corresponding subarrays of the output plane.
{"title":"A compact photorefractive joint transform correlator for industrial recognition tasks","authors":"H. Rajbenbach, S. Bann, J. Huignard","doi":"10.1364/optcomp.1991.tud5","DOIUrl":"https://doi.org/10.1364/optcomp.1991.tud5","url":null,"abstract":"Technology advances in solid state lasers, spatial light modulators and nonlinear optical materials are centrally important for the construction of optoelectronics processors that combine the massive interconnectivity and parallelism of optics with the accuracy and flexibility of digital electronics. In pattern recognition applications, hybrid optical-digital approaches in which optics performs correlation operations and electronics processes the output correlation plane for classification have already been demonstrated(1−2). Today, the performances of semiconductor lasers, diode-pumped YAG lasers, two dimensional liquid crystal light modulators and photorefractive materials allow the introduction of compact and more flexible optical hardware in optoelectronic processors. In this paper, we present a compact and reconfigurable multichannel joint transform optical correlator designed and constructed for industrial recognition applications. The principle of operation is shown in Fig.1. The object to be identified S(x,y) is display on one half of the input scene. The other half of the input, allocated to the reference R(x,y) is split in N subarrays, or channels, each containing a reference object or a calculated version of reference object. The sum R(x,y) + S(x,y) is Fourier transformed and the spectrum is recorded in a dynamic holographic medium. The complex light field produced by reading out the joint-transform power spectrum contains the cross-correlation component R(x,y) ⊗ S (x-2a, y), where 2a is the separation between signal and reference and ⊗ denotes the correlation operation(3). The identification is performed by detecting the position and relative intensities of the correlation peaks in the corresponding subarrays of the output plane.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"402 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131852747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.me25
J. Pezzaniti, R. Chipman
Several free space optical interconnects for digital optical computing which use polarization beam combining are currently being implemented 1-5 These architectures interconnect 2-D arrays of optical logic devices by imaging arrays of spots, generated by binary phase gratings, from one logic device to the next. Polarization beam combining addresses the need to combine input beams and separate output beams by using space-variant mirrors in conjunction with polarizing beam splitters and waveplates. The throughput of the interconnect is limited primarily by the polarizing beam splitters and the waveplates.
{"title":"A polarization metrology for optical interconnects which use polarization beam combining","authors":"J. Pezzaniti, R. Chipman","doi":"10.1364/optcomp.1991.me25","DOIUrl":"https://doi.org/10.1364/optcomp.1991.me25","url":null,"abstract":"Several free space optical interconnects for digital optical computing which use polarization beam combining are currently being implemented 1-5 These architectures interconnect 2-D arrays of optical logic devices by imaging arrays of spots, generated by binary phase gratings, from one logic device to the next. Polarization beam combining addresses the need to combine input beams and separate output beams by using space-variant mirrors in conjunction with polarizing beam splitters and waveplates. The throughput of the interconnect is limited primarily by the polarizing beam splitters and the waveplates.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126791850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.me8
S. A. Brodsky, C. Guest
Optical implementations of neural networks can combine advantages of neural network adaptive parallel processing and optical free-space connectivity. Binary valued Backpropagation1, a supervised learning algorithm related to standard Backpropagation2, significantly reduces interconnection storage and computation requirements. This implementation of binary valued Backpropagation used optical matrix-vector multiplication3 to represent the forward information flow between network layers. Previous analog optical network memory systems have been described4.
{"title":"Optical Matrix-Vector Implementation of Binary Valued Backpropagation","authors":"S. A. Brodsky, C. Guest","doi":"10.1364/optcomp.1991.me8","DOIUrl":"https://doi.org/10.1364/optcomp.1991.me8","url":null,"abstract":"Optical implementations of neural networks can combine advantages of neural network adaptive parallel processing and optical free-space connectivity. Binary valued Backpropagation1, a supervised learning algorithm related to standard Backpropagation2, significantly reduces interconnection storage and computation requirements. This implementation of binary valued Backpropagation used optical matrix-vector multiplication3 to represent the forward information flow between network layers. Previous analog optical network memory systems have been described4.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127619521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.ma3
Julian Cheng, G. Olbright, R. Bryan
In this paper, we describe the design and operation of optical logic gates based on heterojunction phototransistor (HPT) and vertical-cavity surface-emitting laser (VCSEL) structures. We call the HPT/VCSEL structure a surface-emitting laser logic device. These structures will find use in optical communication systems as well as in parallel optical computing architectures. We illustrate complete sets of optical logic functions, upon which arithmetical logic units (ALU) are based, and provide specific examples of binary arithmetic operations based on optical symbolic substitution.
{"title":"Binary Arithmetic Using Optical Symbolic Substitution and Cascadable Surface-Emitting Laser Logic Devices","authors":"Julian Cheng, G. Olbright, R. Bryan","doi":"10.1364/optcomp.1991.ma3","DOIUrl":"https://doi.org/10.1364/optcomp.1991.ma3","url":null,"abstract":"In this paper, we describe the design and operation of optical logic gates based on heterojunction phototransistor (HPT) and vertical-cavity surface-emitting laser (VCSEL) structures. We call the HPT/VCSEL structure a surface-emitting laser logic device. These structures will find use in optical communication systems as well as in parallel optical computing architectures. We illustrate complete sets of optical logic functions, upon which arithmetical logic units (ALU) are based, and provide specific examples of binary arithmetic operations based on optical symbolic substitution.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131588566","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.tud2
Yao Li, B. Ha
In this paper, a study of a ring array processor distribution topology for optical digital processing and interconnect is presented. The work was motivated by the facts that (1) conventional optical imaging elements such as lenses are circularly symmetric about optic axes, and (2) the existing linear/rectangular array distribution topology is sometimes inefficient in terms of optical implementation and synchronization. The proposed new free-space optical ring array topology based processing and interconnect schemes can solve various existing problems in optical processing and interconnects.
{"title":"Ring Array Processor Distribution Topology for Optical Processing and Interconnect","authors":"Yao Li, B. Ha","doi":"10.1364/optcomp.1991.tud2","DOIUrl":"https://doi.org/10.1364/optcomp.1991.tud2","url":null,"abstract":"In this paper, a study of a ring array processor distribution topology for optical digital processing and interconnect is presented. The work was motivated by the facts that (1) conventional optical imaging elements such as lenses are circularly symmetric about optic axes, and (2) the existing linear/rectangular array distribution topology is sometimes inefficient in terms of optical implementation and synchronization. The proposed new free-space optical ring array topology based processing and interconnect schemes can solve various existing problems in optical processing and interconnects.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122859187","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1992-05-22DOI: 10.1364/optcomp.1991.me3
C. Gu, P. Yeh
Thresholding and Max operations are essential elements in the implementation of neural networks. Although there have been several optical implementations of neural networks, the thresholding functions are performed electronically [1-3]. Optical thresholding and Max operations have the advantages of parallelism and cascadability without resorting to optoelectronic conversion. Unfortunately, there has been very limited work in this area. In this paper, we propose and study the properties of self-oscillation in nonlinear optical (NLO) four-wave mixing (FWM) and NLO resonators for parallel optical thresholding and Max operations.
{"title":"Optical Thresholding and Max Operation","authors":"C. Gu, P. Yeh","doi":"10.1364/optcomp.1991.me3","DOIUrl":"https://doi.org/10.1364/optcomp.1991.me3","url":null,"abstract":"Thresholding and Max operations are essential elements in the implementation of neural networks. Although there have been several optical implementations of neural networks, the thresholding functions are performed electronically [1-3]. Optical thresholding and Max operations have the advantages of parallelism and cascadability without resorting to optoelectronic conversion. Unfortunately, there has been very limited work in this area. In this paper, we propose and study the properties of self-oscillation in nonlinear optical (NLO) four-wave mixing (FWM) and NLO resonators for parallel optical thresholding and Max operations.","PeriodicalId":302010,"journal":{"name":"Optical Computing","volume":"57 4 Suppl 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116515790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：