INSPIRE Track 1: An Optical Turing Machine for Native Computing Using High-Capacity Transmission Modulation Format Symbols

该INSPIRE奖的部分资金来自计算机与信息科学与工程理事会、计算机与信息科学与工程理事会、工程理事会计算机与网络系统分部的NeTS项目。该项目将设计和建造“光学图灵机”（OTM），这是一种数字设备，使用与长距离高速传输兼容的光学数据调制格式进行计算，以支持每秒太比特。它通过使用用于传输的相同符号进行计算来集成计算和通信，从而实现网络数据的本地光处理，同时避免昂贵的光-电子-光转换或调制格式转换（例如，串行到并行）。针对NSF INSPIRE计划，OTM探索了网络内高速处理，这需要跨网络、计算和光学设备的并行跨学科发展。它涉及到支持计算的调制的新视角，将算法转换为串行架构，以及基于光能力限制下的网络需求的系统设计，而光能力不能从任何单一的有利位置或焦点发展。OTM激发了串行计算、高阶逻辑算法、同态变换、大数据分析新方法和新型有源光学器件的研究。OTM以更低的功耗和复杂性实现更快的通信和计算，以远远超过当前电子处理能力的速率支持增强的网络内隐私、纠错和分组服务功能，从而使社会受益。如果成功，它也提供了一个机会来重新审视几十年来光学计算尝试的基本假设。跨CS和EE的交叉发表将增强这种效果。

This INSPIRE award is partially funded by the NeTS program in the Division of Computer and Network Systems in the Directorate for Computer and Information Science and Engineering, the Directorate for Computer and Information Science and Engineering, and the Directorate for Engineering.This project will design and build "The Optical Turing Machine" (OTM) a digital device that computes using a optical data modulation format compatible with long-distance, high-speed transmission to support terabit-per-second. It integrates computation and communication by computing with the same symbols used for transmission, which enables native optical processing of network data while avoiding costly optical-electronic-optical conversion or modulation format conversion (e.g., serial to parallel). Specific to the NSF INSPIRE program, OTM explores in-network high-speed processing that necessitates concurrent cross-disciplinary development across networking, computation, and optical devices. It involves new perspectives on modulation to support computation, the transformation of algorithms into serial architectures and system design based on the needs of networking within the constraints of optical capabilities that cannot be developed from any single vantage point or focus.OTM motivates research in serial computation, algorithms in higher-order logic, homomorphic transforms, new methods for big data analysis, and new active optical devices. OTM benefits society by enabling faster communication and computation at potentially lower power and complexity, supporting enhanced in-network capabilities for privacy, error correction, and packet services at rates far in excess of those capable by current electronic processing. If successful, it also provides an opportunity to reexamine the fundamental assumptions of several decades of attempts at optical computing. Cross-publication across both CS and EE will enhance this effect.