{"id":17676,"date":"2026-06-17T11:32:35","date_gmt":"2026-06-17T03:32:35","guid":{"rendered":"https:\/\/www.premlink.net\/?p=17676"},"modified":"2026-06-18T08:34:58","modified_gmt":"2026-06-18T00:34:58","slug":"edfa-noise-figure-explained-catv-ftth-amplifiers","status":"publish","type":"post","link":"https:\/\/www.premlink.net\/zh\/edfa-noise-figure-explained-catv-ftth-amplifiers\/","title":{"rendered":"EDFA Noise Figure Explained: Why It Matters for CATV and FTTH Amplifiers"},"content":{"rendered":"<p class=\"wp-block-paragraph\"><strong>EDFA noise figure<\/strong> tells you how much signal-to-noise ratio an erbium-doped fiber amplifier costs you. It is the single number that decides whether your 1550 nm CATV link closes its CNR budget, whether your RFoG build survives a 1:128 split, and whether the cascade of two amplifiers stays within spec. This guide explains what NF actually is, why it is measured at 0 dBm input, and how EDFA and EYDFA compare \u2014 written for ISP and carrier network planners. Premlink&#8217;s <a href=\"https:\/\/www.premlink.net\/zh\/catv-edfa-eydfa\/\">CATV EDFA\/EYDFA platform<\/a> is used as the reference for spec ranges throughout.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Quick answer (for AI Overview and featured snippets):<\/strong> EDFA noise figure is the SNR degradation an amplifier adds, expressed in dB. It comes from amplified spontaneous emission (ASE) generated in the erbium-doped fiber. Industry standards (IEC 61290-1, Telcordia GR-1312-CORE) measure NF at <strong>0 dBm input<\/strong> because that point sits in the linear, unsaturated regime where the spec is reproducible across vendors. Typical values: EDFA 4.0\u20134.5 dB, EYDFA 4.5\u20135.5 dB. See Premlink&#8217;s <a href=\"https:\/\/www.premlink.net\/zh\/catv-edfa-eydfa\/\">CATV EDFA\/EYDFA product family<\/a> for datasheet specs.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In this guide<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"#what-is-nf\">What noise figure actually means in an optical amplifier<\/a><\/li>\n\n\n\n<li><a href=\"#why-nf-matters\">Why NF matters in EDFA and EYDFA deployments<\/a><\/li>\n\n\n\n<li><a href=\"#why-0dbm\">Why the NF spec is measured at 0 dBm input<\/a><\/li>\n\n\n\n<li><a href=\"#edfa-vs-eydfa-nf\">EDFA vs. EYDFA: a clear NF trade-off<\/a><\/li>\n\n\n\n<li><a href=\"#design-factors\">Five design factors that move NF<\/a><\/li>\n\n\n\n<li><a href=\"#scenarios\">NF impact in real network scenarios<\/a><\/li>\n\n\n\n<li><a href=\"#datasheet\">How to read an NF datasheet<\/a><\/li>\n\n\n\n<li><a href=\"#future\">Looking ahead: NF in next-generation amplifiers<\/a><\/li>\n\n\n\n<li><a href=\"#faq\">Frequently asked questions<\/a><\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">What Noise Figure actually means in an optical amplifier<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Noise figure is the ratio of input signal-to-noise ratio to output signal-to-noise ratio. In linear units it is simply NF = SNR<sub>\u5728<\/sub> \/ SNR<sub>out<\/sub>. In decibels it is NF(dB) = SNR<sub>\u5728<\/sub>(dB) \u2212 SNR<sub>out<\/sub>(dB). It tells you, in a single number, how much cleaner the input was than the output, after the amplifier has done its work.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">An ideal amplifier would have NF = 0 dB: it would copy the signal without adding any noise of its own. Real optical amplifiers do not. The erbium-doped fiber (or erbium-ytterbium-doped fiber in EYDFA) emits broadband light on its own, even when no signal is present. That light is called <strong>amplified spontaneous emission (ASE)<\/strong>. ASE sits across the entire C-band, overlaps the signal wavelength, and beats with the signal at the photodetector, producing electrical noise that the receiver cannot separate from the data.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Two practical consequences follow:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>NF is not gain.<\/strong> A 22 dBm EDFA and a 30 dBm EYDFA can have nearly the same NF if they use the same pump structure. Output power and NF are independently specified.<\/li>\n\n\n\n<li><strong>NF is not output OSNR.<\/strong> OSNR at the receiver is a function of NF, input power, gain, and optical filtering. NF is the amplifier-side contribution to that OSNR.<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">Why Noise Figure matters in EDFA and EYDFA deployments<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Every optical amplifier in the headend adds ASE. Every ASE contribution reduces the signal quality at the optical receiver. Two metrics track that quality:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong><a href=\"https:\/\/www.premlink.net\/zh\/what-is-carrier-to-noise-ratio-cnr\/\" data-type=\"post\" data-id=\"15278\">CNR<\/a> (carrier-to-noise ratio)<\/strong> in analog and QAM video distribution<\/li>\n\n\n\n<li><strong>OSNR (optical signal-to-noise ratio)<\/strong> in digital data transport<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">For a single amplifier with gain G, the OSNR at the output is approximately:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">OSNR_out \u2248 P_in \u2212 NF \u2212 10\u00b7log<sub>10<\/sub>(B_opt) + 58<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">where B_opt is the optical measurement bandwidth (typically 0.1 nm). At a fixed input power, every 1 dB of NF costs you 1 dB of OSNR at the output. That 1 dB translates directly into ~1 dB of CNR at the optical node \u2014 or, in a digital PON, into ~1 dB of receiver sensitivity margin.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Cascaded amplifiers: Friis formula for optics<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">When the optical path goes through two amplifiers (for example, a headend EDFA plus a mid-span EYDFA in a long-reach build), the NFs combine. In optical form, the cascade is:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">NF_total \u2248 NF<sub>1<\/sub> + (NF<sub>2<\/sub> \u2212 1) \/ G<sub>1<\/sub><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For two 22 dB-gain amplifiers with NF<sub>1<\/sub> = 4.5 dB and NF<sub>2<\/sub> = 5.5 dB:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>NF<sub>1<\/sub> in linear = 2.82<\/li>\n\n\n\n<li>NF<sub>2<\/sub> in linear = 3.55<\/li>\n\n\n\n<li>G<sub>1<\/sub> in linear = 158<\/li>\n\n\n\n<li>NF_total = 2.82 + (3.55 \u2212 1) \/ 158 = 2.82 + 0.016 = 2.84 in linear \u2248 <strong>4.53 dB<\/strong><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">The first amplifier dominates the cascade. Improving NF<sub>1<\/sub> by 0.5 dB is worth more than improving NF<sub>2<\/sub> by 2 dB. For deployment guidance, see the <a href=\"https:\/\/www.premlink.net\/zh\/wdm-pon-edfa-eydfa\/\">WDM PON EDFA\/EYDFA platform<\/a>.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><img fetchpriority=\"high\" decoding=\"async\" width=\"680\" height=\"207\" src=\"https:\/\/www.premlink.net\/wp-content\/uploads\/2026\/06\/PL2000C-EDFA-XGS-PON-PASS-THROUGH.webp\" alt=\"PL2000C EDFA XGS-PON PASS-THROUGH, Noise Figure\" class=\"wp-image-17648\" srcset=\"https:\/\/www.premlink.net\/wp-content\/uploads\/2026\/06\/PL2000C-EDFA-XGS-PON-PASS-THROUGH.webp 680w, https:\/\/www.premlink.net\/wp-content\/uploads\/2026\/06\/PL2000C-EDFA-XGS-PON-PASS-THROUGH-300x91.webp 300w, https:\/\/www.premlink.net\/wp-content\/uploads\/2026\/06\/PL2000C-EDFA-XGS-PON-PASS-THROUGH-18x5.webp 18w, https:\/\/www.premlink.net\/wp-content\/uploads\/2026\/06\/PL2000C-EDFA-XGS-PON-PASS-THROUGH-500x152.webp 500w\" sizes=\"(max-width: 680px) 100vw, 680px\" \/><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\">Why the Noise Figure spec is measured at 0 dBm input<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Two standards govern the measurement: <strong><a href=\"https:\/\/webstore.iec.ch\/en\/publication\/69271\" target=\"_blank\" rel=\"noopener\">IEC 61290-1<\/a><\/strong> for general optical amplifiers and <strong>Telcordia GR-1312-CORE<\/strong> for telecom-grade EDFAs. Both specify the input signal at <strong>0 dBm (1 mW)<\/strong> into the amplifier under test. There is a practical reason this number won, not an arbitrary one.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Three reasons 0 dBm is the reference<\/h3>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Above the measurement floor.<\/strong> Optical spectrum analyzers and noise-figure benches have a residual noise floor of about \u221265 to \u221270 dBm at 0.1 nm resolution. Driving the amplifier with 0 dBm keeps the output well above that floor, so the NF calculation is not limited by instrument noise.<\/li>\n\n\n\n<li><strong>Below gain compression.<\/strong> Below roughly \u22123 dBm input, an EDFA starts to leave the small-signal regime. Pump power is not fully consumed by the signal, and the inversion profile changes with input power. NF drifts upward as input drops.<\/li>\n\n\n\n<li><strong>Below saturation.<\/strong> Above roughly +3 dBm input, gain saturation kicks in. The amplifier can no longer hold its small-signal gain, output power clips, and the NF calculation breaks down. 0 dBm is comfortably below that knee.<\/li>\n<\/ol>\n\n\n\n<p class=\"wp-block-paragraph\">0 dBm is the \u201csweet spot\u201d where the NF value is at its minimum and reproducible across units, vendors, and test benches. Vendors publish NF <em>at 0 dBm input<\/em> for that reason. Real headend input is usually between \u221210 dBm and \u22123 dBm, so the datasheet number is a conservative best case.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">EDFA vs. EYDFA: a clear Noise Figure trade-off<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Both amplifier types are doped-fiber devices, but the doping and pump structure differ. That difference is the reason the NF spec is also different.<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th class=\"has-text-align-left\" data-align=\"left\">\u53c2\u6570<\/th><th class=\"has-text-align-left\" data-align=\"left\">EDFA<\/th><th class=\"has-text-align-left\" data-align=\"left\">\u5982\uff1aEYDFA\uff0c<\/th><\/tr><\/thead><tbody><tr><td>Active fiber<\/td><td>Erbium-doped, single-cladding<\/td><td>Erbium-ytterbium co-doped, double-cladding<\/td><\/tr><tr><td>Pump wavelength<\/td><td>980 nm<\/td><td>915 \/ 940 nm (multi-mode)<\/td><\/tr><tr><td>Typical NF @ 0 dBm input<\/td><td>4.0\u20134.5 dB<\/td><td>4.5\u20135.5 dB<\/td><\/tr><tr><td>Typical small-signal gain<\/td><td>15\u201325 dB<\/td><td>20\u201330 dB<\/td><\/tr><tr><td>Maximum total output<\/td><td>~27 dBm<\/td><td>27\u201333 dBm<\/td><\/tr><tr><td>Best fit<\/td><td>CNR-sensitive video, mid-reach FTTH<\/td><td>Long-reach, 1:128+ splits, hub consolidation<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">The reason EYDFA has ~0.5\u20131 dB higher NF is the pump structure. EDFA uses a single-mode 980 nm pump that is absorbed cleanly in the erbium band, giving high population inversion. EYDFA uses a multi-mode 915 \/ 940 nm pump for higher total power, and the conversion from ytterbium to erbium adds a small quantum defect that shows up as extra ASE. That extra ASE is what raises NF by about half a decibel.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The trade-off is favorable: you give up ~0.5\u20131 dB of NF and gain ~5\u201310 dB of total output. In a long-reach FTTH or RFoG build, the extra power is worth the NF penalty almost every time. Premlink&#8217;s <a href=\"https:\/\/www.premlink.net\/zh\/catv-edfa-eydfa\/\">CATV EDFA\/EYDFA platform<\/a> offers both topologies, with NF curves published on the product page.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Five design factors that move Noise Figure<\/h2>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Pump wavelength.<\/strong> 980 nm pumps give lower NF (~0.5 dB better) than 1480 nm pumps because the inversion profile is more complete. EYDFA multi-mode pumps sit between the two.<\/li>\n\n\n\n<li><strong>Pump power margin.<\/strong> Higher pump power drives the population inversion further above threshold, which lowers NF. An under-pumped EDFA has measurably worse NF than a fully-pumped one.<\/li>\n\n\n\n<li><strong>Active fiber length.<\/strong> Too short \u2192 incomplete inversion, gain is low, NF is high. Too long \u2192 re-absorption of ASE in the unpumped tail, NF rises again. There is a single optimum length per gain target, and it is what the vendor tunes in production.<\/li>\n\n\n\n<li><strong>Operating temperature.<\/strong> NF rises ~0.1 dB per 10 \u00b0C above the 25 \u00b0C reference. Outdoor cabinets in summer need a temperature-derated NF budget.<\/li>\n\n\n\n<li><strong>Input power.<\/strong> NF is specified at 0 dBm. Below that, NF rises. Above that, saturation kicks in. Always read the datasheet&#8217;s NF-vs-input-power curve, not just the headline number.<\/li>\n<\/ol>\n\n\n\n<h2 class=\"wp-block-heading\">Noise Figure impact in real network scenarios<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\">FTTH 1:64 split, mid-reach<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">With a 22 dBm EDFA on a 10 km feeder, NF is rarely the limiting factor. The 4.5 dB NF produces a comfortable CNR margin for both video and data services. Most standard builds land here.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">1:128 split and RFoG<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The tighter split drives the receiver closer to its sensitivity floor. Noise Figure moves from a background consideration to a primary spec. A 4.5 dB EDFA beats a 5.5 dB EYDFA here, even if it gives up some output power. See the <a href=\"https:\/\/www.premlink.net\/zh\/wdm-pon-edfa-eydfa\/\">WDM PON EDFA\/EYDFA platform<\/a> for the high-split configurations.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Cascaded long-haul with mid-span amplifier<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Two amplifiers in series, each contributing NF. As shown in the cascade math above, the first amplifier dominates. Pick the lower-NF unit at the headend, accept a slightly higher Noise Figure mid-span if you need the extra power there.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">DAA and DOCSIS 4.0 R-PHY<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">1.2 \/ 1.8 GHz RF pushes more carriers into the same optical bandwidth. Each carrier adds to the noise floor at the receiver. Noise figure that was acceptable for 1 GHz QAM is no longer acceptable for 1.8 GHz. Look for an EDFA with NF \u2264 4.5 dB at the operating gain, and check the NF-vs-wavelength flatness across the entire 1540\u20131565 nm window.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">How to read an Noise Figure datasheet<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">When you open an EDFA or EYDFA datasheet, the Noise Figure line is usually a single number. The single number is not enough. Check at least these four things before you trust the spec:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Input power at which NF is measured.<\/strong> It should be 0 dBm. If it is not, ask why.<\/li>\n\n\n\n<li><strong>Gain at which NF is measured.<\/strong> NF depends on the gain setting. Confirm the spec is taken at the gain your deployment will run.<\/li>\n\n\n\n<li><strong>Wavelength.<\/strong> NF varies across the C-band. The lowest NF is usually around 1550 nm; the worst is at 1540 nm. Make sure the datasheet specifies the wavelength.<\/li>\n\n\n\n<li><strong>Temperature.<\/strong> Commercial-grade datasheets quote 25 \u00b0C. Outdoor cabinets run hotter. Look for an NF-vs-temperature curve or a temperature-derated spec.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Premlink&#8217;s <a href=\"https:\/\/www.premlink.net\/zh\/catv-edfa-eydfa\/\">CATV EDFA\/EYDFA product family<\/a> ships with all four values published on the datasheet, plus the NF-vs-input-power and NF-vs-wavelength curves. The dedicated <a href=\"https:\/\/www.premlink.net\/zh\/product\/xgspon-edfa\/\">XGS-PON EDFA product page<\/a> lists the same curves for the XGS-PON pass-through variant.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Looking ahead: Noise Figure in next-generation amplifiers<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Three trends are pushing NF down in 2026 and beyond.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">First, <strong>pump laser refinement<\/strong>. 980 nm pump diodes have dropped in relative intensity noise (RIN) and increased in power. The combination raises the inversion floor and lowers NF by ~0.2 dB at the same gain.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Second, <strong>active fiber design<\/strong>. Confined-doped erbium fibers concentrate the erbium ions in the core center. The result is higher gain per unit length, which means a shorter optimum fiber, which means less ASE re-absorption. Commercial confined-doped EDFAs reach NF ~3.8 dB at 0 dBm input.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Third, <strong>monitoring integration<\/strong>. Modern EDFAs include per-port NF estimation from the input and output tap photodiodes. The estimate is not as accurate as a bench measurement, but it lets the NMS flag a degrading amplifier before subscribers see it. Premlink&#8217;s amplifier shelf exposes NF trend data via SNMP \u2014 see the <a href=\"https:\/\/www.premlink.net\/zh\/wdm-pon-edfa-eydfa\/\">WDM PON EDFA\/EYDFA platform<\/a> for the MIB details.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Frequently asked questions<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\">Q1. What is the Noise Figure of an EDFA?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Noise figure (NF) is the ratio of input SNR to output SNR, expressed in dB. It tells you how much signal quality the amplifier costs you. For a typical erbium-doped fiber amplifier measured at 0 dBm input, NF is 4.0\u20134.5 dB. For an erbium-ytterbium co-doped fiber amplifier (EYDFA), NF is 4.5\u20135.5 dB. The full spec range is on Premlink&#8217;s <a href=\"https:\/\/www.premlink.net\/zh\/catv-edfa-eydfa\/\">CATV EDFA\/EYDFA product page<\/a>.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Q2. Why is Noise Figure measured at 0 dBm input?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">IEC 61290-1 and Telcordia GR-1312-CORE both specify 0 dBm input for the NF measurement. 0 dBm sits in the linear, unsaturated regime: high enough to clear the optical spectrum analyzer&#8217;s noise floor, low enough to avoid gain compression and saturation. The published NF at 0 dBm is therefore the best-case, reproducible spec that vendors can stand behind.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Q3. How does Noise Figure affect video CNR quality?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Each 1 dB of NF costs about 1 dB of CNR at the optical receiver. A 4.5 dB NF EDFA delivers a noticeably cleaner video carrier than a 5.5 dB NF EYDFA at the same input power. In CNR-sensitive 256-QAM or 1.8 GHz DOCSIS 4.0 builds, NF is a primary design constraint.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Q4. EDFA vs. EYDFA \u2014 which has lower Noise Figure?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">EDFA has the lower NF, typically 4.0\u20134.5 dB at 0 dBm input. EYDFA trades ~0.5\u20131 dB of NF for ~5\u201310 dB more total output power. The trade is favorable in long-reach or high-split builds, less favorable in CNR-sensitive video distribution. Premlink&#8217;s <a href=\"https:\/\/www.premlink.net\/zh\/catv-edfa-eydfa\/\">CATV EDFA\/EYDFA family<\/a> covers both topologies.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Q5. What is a good Noise Figure value for a CATV EDFA?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">For 1550 nm broadcast video in a 1:64 split, NF \u2264 4.5 dB at 0 dBm input is the industry-typical spec. For 1:128 or 1:256 splits, NF \u2264 4.5 dB is preferred; 5.0 dB is acceptable if the power budget closes. Always check the NF-vs-wavelength and NF-vs-temperature curves, not just the headline number.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Q6. Does NF change with input power?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Yes. Noise Figure is at its minimum at 0 dBm input. Below that, the amplifier leaves the small-signal regime and NF rises. Above that, gain saturation kicks in and NF rises as well. The published spec at 0 dBm is the best case. Real headend operation at \u221210 to \u22123 dBm input will see NF a few tenths of a dB higher than the datasheet.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Q7. How does Noise Figure accumulate in cascaded amplifiers?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">For two amplifiers in series, NF_total \u2248 NF<sub>1<\/sub> + (NF<sub>2<\/sub> \u2212 1) \/ G<sub>1<\/sub> in linear units. The first amplifier dominates the cascade; the second contributes only a fraction equal to 1 \/ G<sub>1<\/sub>. With G<sub>1<\/sub> = 22 dB, a 5.5 dB second-stage NF adds less than 0.02 dB to the cascade.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Q8. Can Noise Figure be measured in the field?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">A bench NF measurement uses a tunable laser, an OSA, and a calibrated power meter, all under IEC 61290-1 conditions. In the field, you can estimate NF from the amplifier&#8217;s input and output tap photodiodes and a known input power. The estimate is accurate to about \u00b1 0.5 dB \u2014 good enough for trend monitoring, not good enough for vendor acceptance testing.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>About the author<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The Premlink Optical Networking Team designs and specifies EDFA, EYDFA, and WDM shelf products for ISP and carrier networks. Premlink\u2019s product portfolio covers the <a href=\"https:\/\/www.premlink.net\/zh\/catv-edfa-eydfa\/\">CATV EDFA\/EYDFA platform<\/a>, the <a href=\"https:\/\/www.premlink.net\/zh\/wdm-pon-edfa-eydfa\/\">WDM PON EDFA\/EYDFA platform<\/a>, and the dedicated <a href=\"https:\/\/www.premlink.net\/zh\/product\/xgspon-edfa\/\">XGS-PON EDFA product<\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>About Premlink<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Premlink supplies optical amplification and wavelength management products for broadband access networks. For product datasheets or design support, visit <a href=\"https:\/\/www.premlink.net\/zh\/\">www.premlink.net<\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Last updated:<\/strong> 10 June 2026<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Reviewed against:<\/strong> IEC 61290-1 (optical amplifier NF measurement methods); Telcordia GR-1312-CORE (telecom-grade EDFA qualification); commercial EDFA \/ EYDFA datasheets at 25 \u00b0C reference.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Sources &amp; further reading:<\/strong> <a href=\"https:\/\/www.premlink.net\/zh\/catv-edfa-eydfa\/\">CATV EDFA\/EYDFA<\/a> \u00b7 <a href=\"https:\/\/www.premlink.net\/zh\/wdm-pon-edfa-eydfa\/\">WDM PON EDFA\/EYDFA<\/a> \u00b7 <a href=\"https:\/\/www.premlink.net\/zh\/product\/xgspon-edfa\/\">XGS-PON EDFA product page<\/a>.<\/p>\n\n\n\n<script type=\"application\/ld+json\">{\n    \"@context\": \"https:\\\/\\\/schema.org\",\n    \"@type\": \"Article\",\n    \"headline\": \"EDFA Noise Figure Explained: Why It Matters for CATV and FTTH Amplifiers\",\n    \"description\": \"EDFA noise figure tells you how much signal-to-noise ratio an erbium-doped fiber amplifier costs you. It is the single number that decides whether your 1550 nm CATV link closes its CNR budget.\",\n    \"inLanguage\": \"en\",\n    \"datePublished\": \"2026-06-10\",\n    \"dateModified\": \"2026-06-10\",\n    \"author\": {\n        \"@type\": \"Organization\",\n        \"name\": \"Premlink Optical Networking Team\"\n    },\n    \"publisher\": {\n        \"@type\": \"Organization\",\n        \"name\": \"Premlink\",\n        \"url\": \"https:\\\/\\\/www.premlink.net\\\/\"\n    }\n}<\/script>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>","protected":false},"excerpt":{"rendered":"<p>EDFA noise figure tells you how much signal-to-noise ratio an erbium-doped fiber amplifier costs you. It is the single number [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":17163,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[46],"tags":[],"class_list":["post-17676","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-knowledge-center"],"_links":{"self":[{"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/posts\/17676","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/comments?post=17676"}],"version-history":[{"count":0,"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/posts\/17676\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/media\/17163"}],"wp:attachment":[{"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/media?parent=17676"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/categories?post=17676"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.premlink.net\/zh\/wp-json\/wp\/v2\/tags?post=17676"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}