Novel Dispersion-Division-Multiplexing Technique for Enabling Double the Spectral Efficiency in Transmission and Routing of Photonic and Microwave Networks

长期以来，光纤有无限的带宽，因此我们有带宽可烧，这一直是一个令人信服的谬论。随着波分复用器(WDM)的使用，多通道通信的爆炸性增长，可用带宽现在占据了中心舞台。与其他类型的通信相比，光通信在频谱效率(即比特/赫兹)方面远远落后，甚至还没有接近香农容量极限。事实上，更高的频谱效率可能是该领域内创造性活动最少的领域，但它拥有未来系统容量革命性增加的前景。光通信中带宽浪费最大且具有许多潜在应用的技术领域之一是在单个波长载波上对许多低速信道进行副载波多路复用(SCM)。在SCM中，一半的数据带宽被浪费，因为：(I)在每个边带承载接收所需的全部信息的情况下，发送双边带信号实质上将复制正在发送的数据，或者(Ii)发送单边带信号要求载波另一侧(即，第二边带将位于的位置)上的镜像频率保持不被任何其他数据信号占用。对于未来的光通信系统，副载波多路复用具有许多优点，包括：(A)在单个大容量波长上传输许多低速数字数据信号，(B)传输许多模拟调制信号，(C)用于容纳基于无线微波的光信号的网络的传输介质，以及(D)在分组交换网络中用于有效路由的控制标签的传输。我们建议实施一种新的多路复用方案，使使用副载波多路复用的光学系统的频谱效率翻倍。该方案被称为色散分割多路复用(DDM)，并且依赖于将两个数据信道同时放置在远离光载波的同一频率空间中。例如，常规地传输一个双边带数据信道，并且在第一个信道的阴影中传输第二个双边带信道。由于光纤的色散而产生阴影。由于色散，单个通道的两个双边带中的每一个都将以略有不同的速度传播。这些边带将周期性地相对于光载波彼此异相，从而相互抵消并导致射频功率衰落(即，在接收器处给定信道的功率几乎完全消失)。一个通道将衰落，而另一个通道将是同相的并具有全功率。通过使用可调谐的色散诱导元件，接收器可以改变每个通道的两个边带之间的相对相位，从而恢复两个频率共存的通道中的任何一个。我们的研究计划将调查使用频谱高效色散分割多路复用器时的独特功能、机会和限制，这些可能在5-10年的时间框架内取得显著成果。我们的系统和网络研究计划的主要基本特征如下。我们将：(A)演示许多WDM信道的高频谱效率DDM，(B)演示DDM在基于无线的光网络中互连节点的效用，(C)演示WDM光网络中的SCM标签交换和路由，以及(D)调查在考虑SNR、消光比和数据带宽时生成、发送和接收DDM信道的基本限制。

It has long been a comfortable fallacy that the optical fiber has infinite bandwidth, and, therefore,we have bandwidth to burn. With the explosion of multiple-channel communications through the usewavelength-division-multiplexing (WDM), available bandwidth now takes center stage. Whencomparing optical communications with other types of communications, we lag far behind in terms ofspectral efficiency (i.e., bits/Hz) and we are not even close to the Shannon capacity limit. In fact, higherspectral efficiency is probably the area of least creative activity within the field, and yet it holds thepromise for revolutionary increases in future systems capacity. One of the technical areas within optical communications that has shown the most bandwidthwaste and has many future potential applications is subcarrier multiplexing (SCM) of many lower-speedchannels on a single wavelength carrier. In SCM, half the data bandwidth is wasted since either:(i) transmitting double-sideband signals will, in essence, duplicate the data being sent given that eachsideband carries the full information needed for reception, or (ii) transmitting single-sideband signalsrequires that the mirror frequencies on the other side of the carrier wave (i.e., where the second sidebandwould have been located) remain unoccupied by any other data signals. Subcarrier multiplexing has many advantages for future optical communications systems,including: (a) the transmission of many lower-speed digital data signals on a single high-capacitywavelength, (b) the transmission of many analog-modulated signals, (c) the transmission medium fornetworks that accommodate wireless microwave-based optical signals, and (d) the transmission of controllabels for efficient routing in packet-switched networks. We propose implementing a new multiplexing scheme that enables double the spectral efficiencyfor optical systems that use subcarrier multiplexing. The scheme is called dispersion-division-multiplexing(DDM) and relies on placing two data channels simultaneously in the same frequency spaceaway from an optical carrrier. For example, one double-sideband data channel is transmittedconventionally, and a second double-sideband channel is transmitted in the shadow of the firstchannel. Shadowing occurs due to the chromatic dispersion of the optical fiber. Due to dispersion, eachof the two double sidebands of a single channel will travel at slightly different speeds. These sidebandswill periodically be out-of-phase will each other relative to the optical carrier wave, thereby cancelingeach other and inducing RF power fading (i.e., a nearly complete disappearing of the given channel'spower at a receiver). One channel would be faded when the other channel would be in-phase and havefull power. By using a tunable dispersion-inducing element, a receiver can change the relative phasebetween each cannel's two sidebands and thus recover either of the two frequency-co-located channels. Our research program will investigate unique functionalities, opportunities, and limitations whenusing spectrally-efficient dispersion-division-multiplexing that could bear significant fruit in the 5-10year time frame. The key fundamental features of our systems and networking research program are asfollows. We will:(a) demonstrate highly-spectrally-efficient DDM for many WDM channels,(b) demonstrate the utility of DDM for interconnecting nodes in a wireless-based photonicnetwork,(c) demonstrate SCM label swapping and routing in a WDM photonic network, and(d) investigate the fundamental limitations of generating, transmitting, and receiving DDMchannels when considering SNR, extinction ratio, and data bandwidth.TABLE OF CONTENTS