Terabit-per-Second Coherent Optical Fiber Communications

光纤通信系统在全球电信网络中广泛使用，为商业、教育、医疗保健和娱乐行业提供几乎所有服务。对这些服务的需求不断增长，其中许多服务直到最近还无法想象，正在推动该网络以惊人的速度扩展，并将最先进的商业系统的光信号的每通道比特率推高到100 Gbit/s。这些系统通过基于相干检测和数字信号处理的根本不同的方法来实现。目前正在开发每信道比特率为400 Gbit/s的下一代系统，并且可以肯定的是，未来需要1 Tbit/s的比特率。 迄今为止报道的对相干光纤通信的研究主要集中在对使用50 GHz的固定信道间隔的常规波分复用系统中的单个信道执行信号处理功能。为了将每信道比特率增加到1 Tbit/s，由于系统设计约束和技术限制，基于简单地增加调制速率的策略不适用。使用超级信道可以实现1 Tbit/s的比特率，超级信道由多个紧密间隔的光副载波组成，这些光副载波共同被视为单个实体。由于超级信道与当前的固定网格波分复用系统不兼容，因此弹性光网络的概念最近已经出现，作为适应具有不同调制格式和符号速率的信道并最大化频谱效率的手段。拟议的研究包括这些最新的创新，专注于弹性光网络中的1 Tbit/s的超级信道。这种信道在光纤上的共存和超级信道的独特特性给系统设计和性能评估带来了新的挑战。三个博士学生将通过这项研究进行培训，该研究涉及评估系统性能的技术，利用系统独特特性的数字信号处理算法，以及支持实验密集调查所需的关键子系统。

Optical fiber communication systems are used extensively throughout the global telecommunications network to provision virtually all services in the business, educational, healthcare and entertainment sectors. The rising demand for these services, many of which were unimaginable until recently, is driving the expansion of this network at dramatic rates and has pushed the per-channel bit rate of optical signals to 100 Gbit/s for state-of-the-art commercial systems. These systems are enabled by a fundamentally different approach based on coherent detection and digital signal processing. Next generation systems with a per-channel bit rate of 400 Gbit/s are currently being developed, and the need for a bit rate of 1 Tbit/s in the future is seen as certain. The research on coherent optical fiber communications reported to date has largely concentrated on performing the signal processing functions for a single channel in a conventional wavelength division multiplexed system that uses a fixed channel spacing of 50 GHz. To increase the per-channel bit rate to 1 Tbit/s, a strategy based on simply increasing the modulation rate is not applicable due to system design constraints and technology limitations. A bit rate of 1 Tbit/s can be achieved using a superchannel, which is comprised of multiple closely spaced optical subcarriers which collectively are considered as a single entity. Since superchannels are not compatible with current fixed-grid, wavelength division multiplexed systems, the concept of elastic optical networks has recently emerged as a means of accommodating channels with different modulation formats and symbol rates, and maximizing spectral efficiency. The proposed research embraces these recent innovations by focusing on 1 Tbit/s superchannels in elastic optical networks. The co-existence of such channels on a fiber and the distinct features of superchannels introduce new challenges in system design and performance evaluation. Three Ph.D. students will be trained through this research which addresses techniques for evaluating the system performance, digital signal processing algorithms that exploit the unique properties of the systems, and key subsystems required to support the experimentally intensive investigation.