L
Admin
[Technical article / FAQ, rfforum Editorial Team]
RF-over-fiber link NF is often written as if the RF LNA were the whole story. On many links the laser relative intensity noise (RIN), shot noise at the photodiode, and thermal terms set the floor once photocurrent is high — and adding RF gain after a noisy optical hop does not buy the Friis win people expect.
1. What RIN actually does
- RIN is a relative optical-power fluctuation; after detection it becomes an RF noise density that scales with (photocurrent)² in the usual small-signal picture
- At high received optical power, RIN (or shot) can dominate over the RF LNA noise figure you carefully bought
- EDFA / booster stages add their own noise figure into the optical hop; treating "optics" as zero-NF gain is a common budget error
2. Freeze list for an RFoF NF claim
- Optical power / photocurrent at the detector, laser RIN class, and whether an EDFA sits in the path — on the same sheet as RF NF
- Whether the RF LNA is before the laser or after the photodiode, and which placement Friis actually favors for *this* photocurrent
- Modulation depth / OMI and whether the NF number was taken at the same optical operating point as the linearity number
- Fiber length and reflections (connector APC vs UPC) if multi-path optical noise is in play
3. Habit
Never publish "RFoF NF" without photocurrent and optical-noise terms next to the RF cascade. An LNA that looked decisive on a short dark fiber can be irrelevant on a bright, amplified hop.
Discussion: on your last RFoF noise hunt, which term won (laser RIN, shot, EDFA, RF LNA) — and do you require a photocurrent line on every NF traveler?
— rfforum 编辑部 / Editorial Team
RF-over-fiber link NF is often written as if the RF LNA were the whole story. On many links the laser relative intensity noise (RIN), shot noise at the photodiode, and thermal terms set the floor once photocurrent is high — and adding RF gain after a noisy optical hop does not buy the Friis win people expect.
1. What RIN actually does
- RIN is a relative optical-power fluctuation; after detection it becomes an RF noise density that scales with (photocurrent)² in the usual small-signal picture
- At high received optical power, RIN (or shot) can dominate over the RF LNA noise figure you carefully bought
- EDFA / booster stages add their own noise figure into the optical hop; treating "optics" as zero-NF gain is a common budget error
2. Freeze list for an RFoF NF claim
- Optical power / photocurrent at the detector, laser RIN class, and whether an EDFA sits in the path — on the same sheet as RF NF
- Whether the RF LNA is before the laser or after the photodiode, and which placement Friis actually favors for *this* photocurrent
- Modulation depth / OMI and whether the NF number was taken at the same optical operating point as the linearity number
- Fiber length and reflections (connector APC vs UPC) if multi-path optical noise is in play
3. Habit
Never publish "RFoF NF" without photocurrent and optical-noise terms next to the RF cascade. An LNA that looked decisive on a short dark fiber can be irrelevant on a bright, amplified hop.
Discussion: on your last RFoF noise hunt, which term won (laser RIN, shot, EDFA, RF LNA) — and do you require a photocurrent line on every NF traveler?
— rfforum 编辑部 / Editorial Team
