Modeling and simulation of large scale, high-density photonic integrated circuits.

光子学作为一种使能技术，已经被广泛应用于宽带通信、高速互连、遥感和并行信号处理等领域。尽管分立光子器件仍然是构建复杂光子和光电系统的主流技术，集成光子电路开始作为用于高端应用（例如高数据速率相干光纤通信）的更先进的技术平台而出现。光子学的下一个前沿领域肯定是计算机和消费电子产品的高速光学互连。对于这样的应用，将需要基于硅或其他材料的大规模和高密度光子集成。随着近年来制造技术的进步，高性能和特定应用的光子电路已经被开发和展示。为了进一步降低开发成本，需要强大的计算机辅助设计方法和软件工具来进行此类系统的性能仿真和设计优化。到目前为止，大多数的分析和数值建模和仿真技术，如时域有限差分（FDTD）方法是不可扩展的物理尺寸和/或拓扑结构的复杂性的大规模，高密度的光子集成电路。在这项研究中，我建议研究和开发一个层次的方法，根据最近发现的理论模型，即，复杂的模式。该研究将侧重于硅光子学作为技术平台，并开发一个通用的建模框架，可以扩展并应用于任意物理尺寸和拓扑复杂性的光子集成电路。此外，我们计划利用新兴的云计算技术来进一步加速和增强模拟引擎，使整体模拟工具的效率至少比传统的数值方法高出100倍。这项研究如果成功，将填补未来不同应用的光子集成电路开发和工程的重要空白。

Photonics as an enabling technologies has been deployed in a wide range of applications from broadband communications, high-speed interconnects, remote sensing and massive-parallel signal processing etc. While discrete photonic components are still the main-stream technologies for building complex photonic and optoelectronic systems, integrated photonic circuits started to emerge as more advanced technology platform for high-end applications such as high-data-rate coherent fiber optic communications. The next frontier in photonics will be definitely high-speed optical interconnects for computers and consumer electronics. For such applications, large-scale and high-density photonic integration based on silicon or other materials will be required. With the recent progress in fabrication technologies, high-performance and application-specific photonic circuits have been developed and demonstrated. To further reduce the cost of development, powerful computer aided design methodology and software tools are required for performance simulation and design optimization of such systems. So far, most of the analytical and numerical modeling and simulation techniques such as the finite-difference time-domain (FDTD) method is not scalable to the physical size and/or topology complexity of the large-scale, high-density photonic integrated circuits. In this research, I propose to investigate and develop a hierarchical approach based on recently discovered theoretical model, i.e., the complex modes. The research will focus on silicon photonics as the technology platform and develop a universal modeling framework that can be scaled and applied to photonic integrated circuits of arbitrary physical size and topology complexity. Further, we plan to utilize the emerging cloud-computing technologies to further accelerate and empower the simulation engine so that the overall simulation tools will be at least 100 times more efficient than the conventional numerical approach. The research, if successful, will fill in an important gap in development and engineering of future photonic integrated circuits for different applications.