1. Introduction

An optical fiber is completely described by a finite set of measurable parameters, and every transmission-system design decision traces back to one of them. Attenuation sets the span budget and amplifier spacing. Chromatic dispersion sets the pulse-broadening rate that the receiver must absorb. Polarization mode dispersion sets a statistical timing uncertainty that grows with the square root of distance. Effective area and the nonlinear refractive index together set the launch power at which the fiber itself begins to distort the signal it carries. Geometry and mechanical parameters set splice loss, bend tolerance, and the twenty-five-year reliability of the installed plant. A standard single-mode fiber meeting ITU-T G.652.D carries a measured attenuation near 0.19 dB/km at 1550 nm in field-installed links, a chromatic dispersion coefficient of approximately 17 ps/(nm·km) at the same wavelength, and a PMD link design value of at most 0.20 ps/√km — three standard-anchored numbers that between them determine most of what a DWDM line system can do over that fiber.

This reference covers every parameter in that set: what it is physically, which mechanism produces it, what value ITU-T specifies or the field typically measures, when it binds a design, and why an engineer selects one value over another. The treatment runs from the guidance parameters that make a fiber single-mode — numerical aperture, V-number, cutoff wavelength, mode field diameter — through the transmission impairment parameters of attenuation, dispersion, and PMD, into the nonlinear parameter set that governs launch-power optimization, and out to the geometrical, mechanical, and bending specifications that govern installation and lifetime. The ITU-T G.652 through G.657 fiber family is then compared parameter by parameter, application-based selection rules are derived, and the ITU-T G.650-series measurement methods that verify each parameter are mapped to field practice.

The parameter values quoted throughout carry their evidence class in the same sentence: standard-specified limits come from ITU-T Recommendations, measured values from published field measurement campaigns and laboratory records, vendor claims from product specifications, and theoretical limits from the underlying physics. Where a value genuinely varies by implementation, it is stated as a typical range rather than false precision. Readers building link budgets on these parameters will find the companion treatments of fiber attenuation and loss and Q-factor improvement techniques useful next steps.

2. Historical Development of Fiber Attenuation and Capacity

Fiber attenuation history is a fifty-year descent of nearly three orders of magnitude followed by a four-decade plateau. In 1966, Kao and Hockham proposed that glass fiber could serve communication if attenuation could be brought below 20 dB/km — at the time, available glasses measured near 1000 dB/km, so the proposal was a materials-purity target, not a demonstrated capability. In 1970, a fused-silica fiber first reached approximately 20 dB/km at visible wavelengths (measured), proving the purity path viable. Through the 1970s, chemical vapor deposition processes removed transition-metal and hydroxyl contamination, and by 1979 laboratory silica fiber reached approximately 0.20 dB/km at 1550 nm (measured), close to the intrinsic limit set by Rayleigh scattering.

From there the curve flattened. The laboratory silica record improved only from 0.154 dB/km in 1985 to 0.1396 dB/km in 2024 (measured records, reported in the hollow-core literature published in Nature Photonics) — a 9% gain across thirty-nine years, because Rayleigh scattering in the glass itself is the residual loss and can only be reduced by lowering the fictive temperature and dopant concentration of the core, both of which approach material limits in pure-silica-core designs. Commercial fiber tracked the records at a small offset: standard germanium-doped G.652 product ships at 0.18 dB/km, ultra-low-loss silica-core G.652 at 0.16 dB/km, and silica-core G.654.B/D product at 0.15 dB/km (vendor product specifications across the commercial low-loss fiber class).

In September 2025, the plateau broke — outside silica. A hollow-core fiber using a double nested antiresonant nodeless (DNANF) geometry, developed by University of Southampton researchers with Microsoft Azure Fiber, measured 0.091 dB/km at 1550 nm, held below 0.1 dB/km across an 18 THz window, and stayed below 0.2 dB/km across 66 THz (measured, published in Nature Photonics). Because the light propagates through air rather than glass, Rayleigh scattering nearly vanishes and the group index falls close to 1, so signals also propagate approximately 45% faster than in silica (measured group-velocity difference between air and glass guidance). Section 13 treats the deployment implications; the historical point here is that the intrinsic-loss floor that defined silica fiber engineering for forty years is no longer the floor for fiber as a category.

Capacity history followed the attenuation history with a lag: once loss stopped limiting reach, dispersion did, and once coherent detection with digital signal processing absorbed dispersion, nonlinearity became the binding constraint. That sequence explains the fiber family tree of Section 10 — G.653 and G.655 were engineered for the dispersion-limited era, while G.654.E and large-effective-area designs answer the nonlinearity-limited era. The dispersion-shifted fiber and NZDSF history covers that middle era in depth.

3. Light Guidance and Modal Parameters

Guidance parameters define whether a fiber confines light, how much light it accepts, and how many modes it supports. Five parameters carry the whole story: the core and cladding refractive indices and their relative difference Δ, the numerical aperture NA, the normalized frequency V, the cutoff wavelength λ_c, and the mode field diameter MFD. Each one is fixed at fiber design time and each one has a direct operational consequence — coupling efficiency, splice loss, bend sensitivity, and the single-mode guarantee itself.