Pub Date : 1994-07-06DOI: 10.1109/LEOSST.1994.700419
B. Macdonald
introduction European collaborative programs being carried out under the RACE (Research and Dcvelopnient of Advanced Communications in Europe) banner are aimed at the realisation of an Integrated Broadband Communications (IBC) network. These programs are the drivers for new optical system applications such as the use of optical backplane interconnect, wave division multiplexing techniques, optical broadband switching and crucial to the commercial succcss. low cost networks. System projects within RACE are addressing the introduction of optical technology into telecommunication networks with developments broadly grouped into four theme areas the core network, the access network, switching fabric and radio/mobile. Othcr projects are developing a comprehensive range of advanced optoelectronic components that iire needed to satisfy the system requirements and only major advance in the difficult packaging technology has enabled the necessary performance, size and cost targets to be real i sed. Component packaging/interconnect developments can be catergorised into a number of key areas that are linked to the system themes mentioned above: fibre alignment and fixing to device arrays (core, switching, access) .high speed packaging design (core, switching) backplane interconnect and data interconnect techniques (switching) low cost technologies (access, radio) manufacture and downstream issues ( an important area linked to all components) The three examples below give a sample representation of component development in these critical areas.
{"title":"Packaging Developments For Optoelectronic Components In Broadband Communication Networks","authors":"B. Macdonald","doi":"10.1109/LEOSST.1994.700419","DOIUrl":"https://doi.org/10.1109/LEOSST.1994.700419","url":null,"abstract":"introduction European collaborative programs being carried out under the RACE (Research and Dcvelopnient of Advanced Communications in Europe) banner are aimed at the realisation of an Integrated Broadband Communications (IBC) network. These programs are the drivers for new optical system applications such as the use of optical backplane interconnect, wave division multiplexing techniques, optical broadband switching and crucial to the commercial succcss. low cost networks. System projects within RACE are addressing the introduction of optical technology into telecommunication networks with developments broadly grouped into four theme areas the core network, the access network, switching fabric and radio/mobile. Othcr projects are developing a comprehensive range of advanced optoelectronic components that iire needed to satisfy the system requirements and only major advance in the difficult packaging technology has enabled the necessary performance, size and cost targets to be real i sed. Component packaging/interconnect developments can be catergorised into a number of key areas that are linked to the system themes mentioned above: fibre alignment and fixing to device arrays (core, switching, access) .high speed packaging design (core, switching) backplane interconnect and data interconnect techniques (switching) low cost technologies (access, radio) manufacture and downstream issues ( an important area linked to all components) The three examples below give a sample representation of component development in these critical areas.","PeriodicalId":379594,"journal":{"name":"Proceedings of IEE/LEOS Summer Topical Meetings: Integrated Optoelectronics","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131800024","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 : 1994-07-06DOI: 10.1109/LEOSST.1994.700426
N. Miyauchi, T. Nagahama, N. Ohtsuka, N. Okabayashi, K. Nakao, Z. Tani
{"title":"Hologram laser for CD-ROM drive","authors":"N. Miyauchi, T. Nagahama, N. Ohtsuka, N. Okabayashi, K. Nakao, Z. Tani","doi":"10.1109/LEOSST.1994.700426","DOIUrl":"https://doi.org/10.1109/LEOSST.1994.700426","url":null,"abstract":"","PeriodicalId":379594,"journal":{"name":"Proceedings of IEE/LEOS Summer Topical Meetings: Integrated Optoelectronics","volume":"185 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131443684","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 : 1994-07-06DOI: 10.1109/LEOSST.1994.700536
K. Ozasa, Eun Kyu Kim, Tianchun Ye, Y. Aoyagi
{"title":"Selective Epitaxy With In Situ Mask Processing","authors":"K. Ozasa, Eun Kyu Kim, Tianchun Ye, Y. Aoyagi","doi":"10.1109/LEOSST.1994.700536","DOIUrl":"https://doi.org/10.1109/LEOSST.1994.700536","url":null,"abstract":"","PeriodicalId":379594,"journal":{"name":"Proceedings of IEE/LEOS Summer Topical Meetings: Integrated Optoelectronics","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115683554","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 : 1994-07-06DOI: 10.1109/LEOSST.1994.700503
K. Hall, K. Rauschenbach, J. Livas, G. Raybon
High-speed transmission systems f i r local-area and wide-area network are currently being investigated. All-optical implementations of these high-speed systems require nonlinear optical loop mirror (NOLM) [1) or semiconductor waveguide [2] demultiplexers. Future systems, exhibiting flexible conversion of signals from high-rate soliton systems to long-haul WDM systems will require alhptical pulse width and wavelength conversion in addition to all-optical demultiplexing. In this paper, we investigate the performance of NOLM's as pulse width and wavelength converters. A NOLM can be configured as a pulse width and wavelength converter by utilizing pulse walk-through of the control and signal pulses in the device. Besides jitter tolerance j31, pulse walk-through allows the narrow control pulse to walk through the wider signal pulse, imposing a nonlinear phase shift across the entire signal pulse. Thus a narrow pulse at one wavelength may be used to switch-out a wider pulse at another wavelength. We investigate experimentally and theoretically, the range of pulse widths and wavelengths that can be used in the NOLM converter. Also, we present 10 Gb/s bitetror-rate @ER) measurements of the performance of the all-optical pulsewidth and wavelength converter.
{"title":"All-optical Pulse Width And Wavelength Conversion","authors":"K. Hall, K. Rauschenbach, J. Livas, G. Raybon","doi":"10.1109/LEOSST.1994.700503","DOIUrl":"https://doi.org/10.1109/LEOSST.1994.700503","url":null,"abstract":"High-speed transmission systems f i r local-area and wide-area network are currently being investigated. All-optical implementations of these high-speed systems require nonlinear optical loop mirror (NOLM) [1) or semiconductor waveguide [2] demultiplexers. Future systems, exhibiting flexible conversion of signals from high-rate soliton systems to long-haul WDM systems will require alhptical pulse width and wavelength conversion in addition to all-optical demultiplexing. In this paper, we investigate the performance of NOLM's as pulse width and wavelength converters. A NOLM can be configured as a pulse width and wavelength converter by utilizing pulse walk-through of the control and signal pulses in the device. Besides jitter tolerance j31, pulse walk-through allows the narrow control pulse to walk through the wider signal pulse, imposing a nonlinear phase shift across the entire signal pulse. Thus a narrow pulse at one wavelength may be used to switch-out a wider pulse at another wavelength. We investigate experimentally and theoretically, the range of pulse widths and wavelengths that can be used in the NOLM converter. Also, we present 10 Gb/s bitetror-rate @ER) measurements of the performance of the all-optical pulsewidth and wavelength converter.","PeriodicalId":379594,"journal":{"name":"Proceedings of IEE/LEOS Summer Topical Meetings: Integrated Optoelectronics","volume":"64 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120868772","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 : 1994-07-06DOI: 10.1109/LEOSST.1994.700431
Y. Akahori, Y. Murantoto, K. Kato, M. Ikeda, A. Kozen, Y. Itaya
{"title":"Monolithically Integrated Long-wavelength High-speed Waveguide P-i-n Hemt Receiver","authors":"Y. Akahori, Y. Murantoto, K. Kato, M. Ikeda, A. Kozen, Y. Itaya","doi":"10.1109/LEOSST.1994.700431","DOIUrl":"https://doi.org/10.1109/LEOSST.1994.700431","url":null,"abstract":"","PeriodicalId":379594,"journal":{"name":"Proceedings of IEE/LEOS Summer Topical Meetings: Integrated Optoelectronics","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127031019","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 : 1994-07-06DOI: 10.1109/LEOSST.1994.700454
R. A. Novotny, A. Lentine, L. Chirovsky, T. K. Woodward
{"title":"Comparison Of Logic Circuits Simulated In The FET-SEED Technology","authors":"R. A. Novotny, A. Lentine, L. Chirovsky, T. K. Woodward","doi":"10.1109/LEOSST.1994.700454","DOIUrl":"https://doi.org/10.1109/LEOSST.1994.700454","url":null,"abstract":"","PeriodicalId":379594,"journal":{"name":"Proceedings of IEE/LEOS Summer Topical Meetings: Integrated Optoelectronics","volume":"125 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123722328","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 : 1994-07-06DOI: 10.1109/LEOSST.1994.700478
T. Szymanski, H. S. Hinton
Smart pixel arrays for a dynamic optical backplane called the HyperPlane are described. Arrays of "programmable" smart pixels are used to iinject or extract optical signals into the parallel oDtical channels of a free-space optical b'ackplane. By sett ing pixel states appropriately, any network (e.g. crossbar, mesh, hypercube, etc.) can be dynamically embedded into the backplane. Summary: A free-space optical backplane consists of a large number of parallel optical channels (10,000 to 100,000) spaced a few hundred microns apart [ 1][2]. To access these optical channels each printed circuit board will contain one or more smart pixel arrays [ l ] . This paper describes the smart pixel arrays for a free-space optical backplane architecture which we call the "HyperPlane". Fiber . . . . ?itching Node SmaA bixei Arravs 'v Parallel Optical Channels In this architecture the smart pixel arrays are used manage access to the optical channels available in the free-space optical backplane. A smart pixel consists of Figwe 1: A fre-space optical backplane. an incoming window, an out-going window, a latch, two multiplexers and an address bit comparator, as shown in o;t;;f bitto bltto fig. 2. Pixels can be programmed to be in one of three basic states, the "transparent", "transmitting" and "receiving" states, as shown in fig. 3.. The state of a pixel can be changed by down-loading a bit-stream from an associated message-processor. The pixels can also be programmed to receive messages for any destination by down-loading the appropriate address bits. Each smart pixel requires 12 logic gates and they are organized into a 2 dimensional array called a "communication slice" as shown in fig. 4. The data for configuring the slice is loaded in bit-serially from the sides; parallel data to be transmitted enters from the top, and parallel data being received exits from the bottom. The communication slices can be generalized to allow multiple transmissions and/or receptions Of paralld data simultaneously as shown in fig. 5b. Each white box represents a smart pixel (i.e., an optical 1-bit data-path), and each black box represents an electrical 1-bit data-path (i.e., bonding pad). A single die capable of containing 1,024 pixels can be organized in various formats, i.e., one 32-by-32 slice, sixteen 8b y 4 slices, or thirty-two 4-by-8 slices, etc. These organizations allow the architect to vary the ratio of electrical-to-optical IO bandwidth of the die and the architectural aspects of the Hyperplane. Multiple smart pixel arrays form the basis of the HyperPZane architecture. The Hyperplane can embed any conventional interconnection network by programming the pixels accordingly. Optimal embeddings for arrays, meshes, hypercubes and various other networks have been identified and 85 receive Vansmit 4 4 4
{"title":"A Smart Pixel Design For A Dynamic Free-space Optical Backplane","authors":"T. Szymanski, H. S. Hinton","doi":"10.1109/LEOSST.1994.700478","DOIUrl":"https://doi.org/10.1109/LEOSST.1994.700478","url":null,"abstract":"Smart pixel arrays for a dynamic optical backplane called the HyperPlane are described. Arrays of \"programmable\" smart pixels are used to iinject or extract optical signals into the parallel oDtical channels of a free-space optical b'ackplane. By sett ing pixel states appropriately, any network (e.g. crossbar, mesh, hypercube, etc.) can be dynamically embedded into the backplane. Summary: A free-space optical backplane consists of a large number of parallel optical channels (10,000 to 100,000) spaced a few hundred microns apart [ 1][2]. To access these optical channels each printed circuit board will contain one or more smart pixel arrays [ l ] . This paper describes the smart pixel arrays for a free-space optical backplane architecture which we call the \"HyperPlane\". Fiber . . . . ?itching Node SmaA bixei Arravs 'v Parallel Optical Channels In this architecture the smart pixel arrays are used manage access to the optical channels available in the free-space optical backplane. A smart pixel consists of Figwe 1: A fre-space optical backplane. an incoming window, an out-going window, a latch, two multiplexers and an address bit comparator, as shown in o;t;;f bitto bltto fig. 2. Pixels can be programmed to be in one of three basic states, the \"transparent\", \"transmitting\" and \"receiving\" states, as shown in fig. 3.. The state of a pixel can be changed by down-loading a bit-stream from an associated message-processor. The pixels can also be programmed to receive messages for any destination by down-loading the appropriate address bits. Each smart pixel requires 12 logic gates and they are organized into a 2 dimensional array called a \"communication slice\" as shown in fig. 4. The data for configuring the slice is loaded in bit-serially from the sides; parallel data to be transmitted enters from the top, and parallel data being received exits from the bottom. The communication slices can be generalized to allow multiple transmissions and/or receptions Of paralld data simultaneously as shown in fig. 5b. Each white box represents a smart pixel (i.e., an optical 1-bit data-path), and each black box represents an electrical 1-bit data-path (i.e., bonding pad). A single die capable of containing 1,024 pixels can be organized in various formats, i.e., one 32-by-32 slice, sixteen 8b y 4 slices, or thirty-two 4-by-8 slices, etc. These organizations allow the architect to vary the ratio of electrical-to-optical IO bandwidth of the die and the architectural aspects of the Hyperplane. Multiple smart pixel arrays form the basis of the HyperPZane architecture. The Hyperplane can embed any conventional interconnection network by programming the pixels accordingly. Optimal embeddings for arrays, meshes, hypercubes and various other networks have been identified and 85 receive Vansmit 4 4 4","PeriodicalId":379594,"journal":{"name":"Proceedings of IEE/LEOS Summer Topical Meetings: Integrated Optoelectronics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130533110","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 : 1994-07-06DOI: 10.1109/LEOSST.1994.700524
A. Mendez, J. Lambert
suw(ARp Optical CDMA has recognized advantages for bursty, concurrent, asynchronous, non-blocking digital data communications as might be encountered in a high performance computer environment (1,2,3,4). Also, it is believed to be protocol transparent (4). However, optical CDMA tends to require laser pulsewidths much narrower than the bit time, whether coherent (4,5) or incoherent (6,7) linear codes are used (this is the origin of the "time penalty"). In addition, the concurrent communication requires broadcasting which, together with the fiber optic implementation of the codes, gives link losses in excess of the link margin for most optical communications transceiver candidates (which generaly are designed for point-to-point applications). Since 1990 we have been concentrating on matrix CDMA codes which have significantly reduced time penalties (8,9). The link loss problem can be solved by means of fiber optic amplifiers. We have integrated a 4x4 matrix CDMA system breadboard (9) with a 1550nm communications grade laser, an Ar/Ti:Sapphire 980 nm pump, and an erbium doped fiber; see Figure 1. The fiber doped amplifier (FDA) was a linear configuration. The laser diode was gain switched with a step recovery diode (SRD) and signal generator at various frequencies between 100and 500MHz. Without optical amplification, the correlation signal was below the electronic noise of the detector. With optical amplification, it has good signal to noise (SNR) and signal to clutter characteristics as shown in Figure 2. This paper discusses the analysis and experiments of optical CDMA enabled by optical amplifiers of the FDA type.
{"title":"Optically Amplified Optical Code Division Mxmtiple Access(CDMA) Experiments","authors":"A. Mendez, J. Lambert","doi":"10.1109/LEOSST.1994.700524","DOIUrl":"https://doi.org/10.1109/LEOSST.1994.700524","url":null,"abstract":"suw(ARp Optical CDMA has recognized advantages for bursty, concurrent, asynchronous, non-blocking digital data communications as might be encountered in a high performance computer environment (1,2,3,4). Also, it is believed to be protocol transparent (4). However, optical CDMA tends to require laser pulsewidths much narrower than the bit time, whether coherent (4,5) or incoherent (6,7) linear codes are used (this is the origin of the \"time penalty\"). In addition, the concurrent communication requires broadcasting which, together with the fiber optic implementation of the codes, gives link losses in excess of the link margin for most optical communications transceiver candidates (which generaly are designed for point-to-point applications). Since 1990 we have been concentrating on matrix CDMA codes which have significantly reduced time penalties (8,9). The link loss problem can be solved by means of fiber optic amplifiers. We have integrated a 4x4 matrix CDMA system breadboard (9) with a 1550nm communications grade laser, an Ar/Ti:Sapphire 980 nm pump, and an erbium doped fiber; see Figure 1. The fiber doped amplifier (FDA) was a linear configuration. The laser diode was gain switched with a step recovery diode (SRD) and signal generator at various frequencies between 100and 500MHz. Without optical amplification, the correlation signal was below the electronic noise of the detector. With optical amplification, it has good signal to noise (SNR) and signal to clutter characteristics as shown in Figure 2. This paper discusses the analysis and experiments of optical CDMA enabled by optical amplifiers of the FDA type.","PeriodicalId":379594,"journal":{"name":"Proceedings of IEE/LEOS Summer Topical Meetings: Integrated Optoelectronics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131330282","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：