{"title":"用于未来信号处理系统的光模块","authors":"J.N. Lee","doi":"10.1109/HICSS.1989.47189","DOIUrl":null,"url":null,"abstract":"Analog optical modules must be part of a larger signal-processing system, and it is difficult to pass information from an optical module operating near maximum throughput to the other parts of a signal-processing system. The basis for this difficulty is often thought to be the high bandwidth and large number of channels of the optical module, but more quantitatively, optical modules do not yet have the capability to perform nonlinear operations on partially processed data that would greatly reduce the module output rates. Moreover, they require additional processing for equalization of channel response rather than for representation of useful information, and this processing is often left to the digital postprocessor. Recent work carried out to address the interface difficulty is reported. One avenue is the use of additional optical processing operations at the output of existing optical-processing modules, such as data compression adaptive data thresholding, and various multiplexed readout schemes. Another avenue is to reduce precision requirements by using various adaptive learning techniques inherent in some parallel models of computation, such as neural-net models. A third effort involves development of 2-D, three-terminal, spatial light modulator devices that can provide the needed nonlinear transfer functions, the capability to cascade optical-processing operations, and throughputs greater than those attainable with all-electronic approaches.<<ETX>>","PeriodicalId":300182,"journal":{"name":"[1989] Proceedings of the Twenty-Second Annual Hawaii International Conference on System Sciences. Volume 1: Architecture Track","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1989-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Optical modules for future signal processing systems\",\"authors\":\"J.N. Lee\",\"doi\":\"10.1109/HICSS.1989.47189\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Analog optical modules must be part of a larger signal-processing system, and it is difficult to pass information from an optical module operating near maximum throughput to the other parts of a signal-processing system. The basis for this difficulty is often thought to be the high bandwidth and large number of channels of the optical module, but more quantitatively, optical modules do not yet have the capability to perform nonlinear operations on partially processed data that would greatly reduce the module output rates. Moreover, they require additional processing for equalization of channel response rather than for representation of useful information, and this processing is often left to the digital postprocessor. Recent work carried out to address the interface difficulty is reported. One avenue is the use of additional optical processing operations at the output of existing optical-processing modules, such as data compression adaptive data thresholding, and various multiplexed readout schemes. Another avenue is to reduce precision requirements by using various adaptive learning techniques inherent in some parallel models of computation, such as neural-net models. A third effort involves development of 2-D, three-terminal, spatial light modulator devices that can provide the needed nonlinear transfer functions, the capability to cascade optical-processing operations, and throughputs greater than those attainable with all-electronic approaches.<<ETX>>\",\"PeriodicalId\":300182,\"journal\":{\"name\":\"[1989] Proceedings of the Twenty-Second Annual Hawaii International Conference on System Sciences. Volume 1: Architecture Track\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1989-01-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"[1989] Proceedings of the Twenty-Second Annual Hawaii International Conference on System Sciences. Volume 1: Architecture Track\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/HICSS.1989.47189\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"[1989] Proceedings of the Twenty-Second Annual Hawaii International Conference on System Sciences. Volume 1: Architecture Track","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/HICSS.1989.47189","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Optical modules for future signal processing systems
Analog optical modules must be part of a larger signal-processing system, and it is difficult to pass information from an optical module operating near maximum throughput to the other parts of a signal-processing system. The basis for this difficulty is often thought to be the high bandwidth and large number of channels of the optical module, but more quantitatively, optical modules do not yet have the capability to perform nonlinear operations on partially processed data that would greatly reduce the module output rates. Moreover, they require additional processing for equalization of channel response rather than for representation of useful information, and this processing is often left to the digital postprocessor. Recent work carried out to address the interface difficulty is reported. One avenue is the use of additional optical processing operations at the output of existing optical-processing modules, such as data compression adaptive data thresholding, and various multiplexed readout schemes. Another avenue is to reduce precision requirements by using various adaptive learning techniques inherent in some parallel models of computation, such as neural-net models. A third effort involves development of 2-D, three-terminal, spatial light modulator devices that can provide the needed nonlinear transfer functions, the capability to cascade optical-processing operations, and throughputs greater than those attainable with all-electronic approaches.<>