Fiber Optics — Practice Q&A
Q: Why does fiber use total internal reflection, and what happens if the light hits the core-cladding boundary at too steep an angle?
A: Total internal reflection keeps light confined to the fiber's core by reflecting it back inward whenever it hits the core-cladding boundary at a sufficiently shallow angle relative to that boundary. If the light instead hits the boundary at too steep an angle (beyond the fiber's critical angle), it refracts out through the cladding instead of reflecting back in — meaning it escapes the core entirely and is lost from the signal. Fiber and connector design keep the light paths within the angles that guarantee total internal reflection, which is what makes long-distance guided transmission through the core possible at all.
Q: Why is single-mode fiber preferred for long-haul but multi-mode still used at all, given SMF's better distance performance?
A: SMF's small core eliminates modal dispersion, giving it much better long-distance performance, but it's also more expensive to manufacture and requires more precise (costlier) alignment at every connection point. MMF's larger core is cheaper and more tolerant of imprecise alignment, and over short distances (like within a data center) modal dispersion doesn't accumulate enough to meaningfully degrade the signal — so MMF remains the cost-effective choice specifically where the distance limitation isn't actually a constraint.
Q: What's the practical difference DWDM makes to a telecom operator's capacity planning, versus just laying more fiber?
A: DWDM lets an operator multiply the usable capacity of fiber they've already deployed by adding more wavelength channels to existing strands, rather than needing to physically lay, trench, and splice entirely new fiber runs to add capacity — a process that's dramatically more expensive and slower than upgrading the DWDM equipment at each end of an existing fiber. This is why most backbone capacity growth over the past two decades has come from DWDM channel-count and per-channel-rate improvements rather than from proportional growth in physical fiber-route mileage.
Q: Why is a PON architecture described as "passive," and why does that property matter economically?
A: The optical splitter that divides one feeder fiber into multiple subscriber connections requires no electrical power and no active electronic components — it's a purely passive optical device. This matters economically because it means the operator doesn't need to run power to, or maintain active equipment at, the splitter's physical location out in the field between the central office and subscribers, substantially lowering both the capital cost of the access network and its ongoing operational/maintenance burden compared to an architecture using powered active equipment at that intermediate point.
Q: Why does a dense 5G deployment still depend heavily on fiber, even though 5G is a wireless technology?
A: Because the radio link only covers the "last hop" from a cell site to the end user — every cell site itself still needs a high-capacity, reliable connection back to the core network (backhaul), and radio spectrum alone can't carry the aggregate data volume a dense 5G deployment generates back to the core network economically or reliably at scale. Fiber is the default backhaul medium wherever it's available, meaning fiber infrastructure quality is directly load-bearing for 5G network performance even though the end-user connection itself is wireless.
Q: What's the actual engineering tradeoff in choosing a higher PON split ratio?
A: A higher split ratio lets one feeder fiber serve more subscribers, reducing the per-subscriber cost of the shared upstream infrastructure — but it also means more subscribers are sharing the same total bandwidth on that feeder fiber, so each subscriber's available bandwidth during periods of simultaneous high demand across the split group is lower than it would be with a smaller split ratio. It's a direct cost-versus-guaranteed-capacity tradeoff, not a free efficiency gain.