Challenges in Estimating the Information Capacity of the Fiber-Optic Channel

Abstract

Since its early commercial deployment in the late 1980s, optical fiber has evolved to become the predominant carrier of the globe's communications. Yet, after accommodating the world's exponentially growing appetite for transmitted data for more than three decades, its ability to continue doing so is being challenged by fundamental factors. In this article, we review these factors and examine their consequences in terms of information capacity. In particular, we review the difficulties that are imposed by the nonlinear nature of fiber-optic transmission on the assessment of the capacity and on the definition of fundamental concepts, such as bandwidth and spectral efficiency. We discuss relevant approximations and regimes of operation in which bounds for the capacity can be effectively assessed while covering a broad range of applications ranging from interdatacenter communications to links spanning transoceanic distances. We relate to a broad variety of transmission schemes and discuss the potential benefits of spatial multiplexing with multimode and multicore fibers.State-of-the-art transmission experiments are also reviewed and compared with theoretical capacity bounds.

The Role of Parallelism in the Evolution of Optical Fiber Communication Systems

“…Estimating capacity limits of various fiber-optic transmission systems has been a topic of increasing interest within the fiber-optics community over the past 15 years [1], [7], [61], [62], [63], [64], [65], [66]. An invited paper on this topic written by leading experts in the field is included in this Special Issue [67]. This section summarizes key considerations relevant to the capacity limit of various types of optical fiber by taking into account frequently ignored but practically highly relevant limitations due to a transponder's practically unavoidable noise floor.…”

“…Note that, in line with common industry terminology, we loosely use the term "capacity" throughout this article to denote the net information throughput of a fiber-optic transmission system. In this context, the term "capacity" as used here may denote a commercially specified system capacity, an experimentally measured system capacity, or may more rigorously refer to a system's achievable information rate (AIR), whose upper bound is given by the information-theoretically stricter definition of "capacity" (see [67] and [68] for an overview of various capacity metrics used in the analysis of fiber-optic transmission systems). The factor of 2 in (1) represents polarization multiplexing, 2 M is the number of parallel spatial paths (spatial parallelism), N is the number of WDM channels (spectral parallelism), and Rmn is the net information bit rate of a WDM channel.…”

“…where RS represents the symbol rate and SNR is the electrical signal-to-noise-ratio (i.e., the total signal power divided by the total noise power 4 within a bandwidth corresponding to the symbol rate RS, assuming the use of Nyquist pulses). For a detailed discussion of the information theory of fiber-optic transmission systems, the reader may refer to, e.g., [7] and [67]. In practical systems, the SNR is furthermore affected by an implementation penalty η (η ≤ 1) accounting for effects, such as nonideal hardware and nonideal FEC decoding in the receiver, which are difficult to capture in the SNR (see, e.g., [76]), resulting in…”

“…RRC pulses have a finite spectral roll-off that broadens their spectral extent beyond RS. Pulse shaping involves a balance between digital shaping complexity, WDM channel spacing, WDM crosstalk, and bandwidth requirements of the transponder 3 If the channel is not AWGN, higher capacities are in principle achievable [67]. The fact that the exact statistics of some optical noise sources are unknown hints to the difficulties in speaking of a true information-theoretic "Shannon limit" of the fiber-optic channel and let expressions such as (6) only be Shannon limit estimates [73].…”

“…2(a). Typical optical signal launch powers in terrestrial long-haul systems are chosen to maximize the SNR, at a point where P NLIN is not negligible but also not dominant, and the SNR peaks at an optimum launch power [7], [67]. Using this peak SNR in the Shannon formula [cf.…”