适于高速光互联和大规模数字集成光路的超高频光学时钟关键技术

At present, the data interchange among different information processing or transmission units or systems becomes a bottle neck for promoting the speed and capacity of information processing and transmission. It becomes more and more urgent to build up ultra high-speed and large-capacity optical mutual connections among processors in multi-nuclei computers, among motherboards in super computing systems, among computers in a cloud network, among optical communication systems, and among optical units in large-scale optical integrated circuits. The realization of these ultra high-speed and large-capacity mutual optical connections depends on the ultra-high-frequency optical clock circuits and techniques which are undeveloped until now. This project is devoted to explore theoretical and technical solutions to the two key problems for ultra high-frequency optical clocks with clock frequencies on the order of THz - the generation of optical ultra-high-frequency amplitude-varying wave form and pulse shaping of optical ultra-high-frequency amplitude-varying wave form. Nonlinear waveguides in photonic crystals and feedback controlling through optical logic gates are to be considered for solving the problems. On the basis of electromagnetic wave theory, the photonic bandgap theory, nonlinear optical theory, optical waveguide theory, and the theory of optical integration, numerical simulations and investigations will be done by combining the finite-difference time-domain method, plane-wave expansion method, transmission-matrix method and finite-element method. A breakthrough is expected for the technique and theory of ultra high-frequency optical clock circuit, to lay the basis for ultra high-speed and large-capacity optical mutual connections.

当前，信息处理单元之间、信息系统之间的信息交换已成为阻碍信息处理和传输的速度和容量提升的瓶颈。在多核计算机内的各处理器之间，在大型计算中心的各主机之间，在云计算网的各计算机之间，在高速光通信网之间，以及在单个光学大规模集成光路内部的各单元光路之间，建立光学互联，以实现高速信息交换，变得越来越迫切。这些高速光学互联的实现依赖于目前尚未解决的超高频光学时钟集成光路和技术。本课题基于二维光子晶体非线性波导和光学逻辑门反馈控制光路，基于光子能带理论、光路集成技术和光子晶体波导理论，以时域有限差分法为主，结合平面波展开法、传输矩阵法和有限元法，对适于高速光互联和大规模集成光路的、时钟频率达太赫兹以上的超高频光学时钟的2项关键技术-超高频光学周期脉动波形产生技术和超高频光学脉动波形整形技术-开展探索研究，将实现超高频光学时钟集成光路和技术的突破，为急需的各种超高速光学互联技术的发展提供理论和技术基础。

本项目主要探索和研究适于高速光互联和大规模数字集成光路的频率在太赫兹量级的超高频光学时钟的2项关键技术—超高频光学周期脉动波形产生技术和超高频光学脉动波形整形技术—的解决思路和方法，并进行相关技术和应用的研究，取得了吸引力的研究进展，提出和建立了一套有关超高频光学周期脉动波形产生技术和超高频光学脉动波形整形技术的理论和技术解决方法，发表了一批高水平学术论文，获发明专利授权45件（含12件美国发明专利），申请发明专利152件，其中含美国专利申请9件， PCT发明专利申请65件。本项目对于发展高速光子芯片技术,尤其是对于发展集成光学芯片技术有重要科学和应用价值。