Receiver noise figure and sensitivity are often placed beside one another in specifications, but they do not name the same property. Noise figure describes how a receiver degrades signal-to-noise ratio under reference conditions. Sensitivity is an input level that reaches a defined performance threshold in a specified configuration.
The central fictional budget uses 1 MHz bandwidth, 5 dB receiver noise figure and a required 10 dB SNR. Starting from about -174 dBm/Hz at the 290 K reference, the integrated thermal noise is about -114 dBm, input-referred receiver noise about -109 dBm, and simple sensitivity about -99 dBm. Changing bandwidth or required SNR changes sensitivity without changing the hardware noise figure.
These are educational link-budget calculations. They do not certify equipment, replace accredited RF measurement, set legal limits or guarantee field coverage and service.
This specialist guide belongs to the Wireless Communication root. It also connects to The Gold Standard of Education | Learning That Transfers into Life and the How X Works hub. The worked cases are original teaching examples; their numbers are stipulated so that the reasoning can be checked.
eduKateSG · Wireless Communication
Find your next learning step
Use noise figure for receiver-added noise degradation and sensitivity for a weakest-signal test with bandwidth and performance criterion stated.
ROUTE 1 · CHAPTERS 1–5
Define the quantities
Fix the object, unit, denominator and reference condition.
ROUTE 2 · CHAPTERS 6–10
Work the central example
Carry the calculation through a complete, checkable case.
ROUTE 3 · CHAPTERS 11–15
Audit the evidence
Test sensitivity, measurement choices and rival explanations.
ROUTE 5 · CHAPTERS 21–24
Report and continue
Repair shortcuts, state limits and follow verified routes.
Full chapter contents · Wireless Communication subject guide
Chapter contents
Define the quantities · Chapters 1–5
1. Define noise factor and noise figure
Work the central example · Chapters 6–10
6. Show why the numbers are not synonyms
Audit the evidence · Chapters 11–15
11. Audit temperature and frequency
12. Audit gain and AGC settings
Transfer the reasoning · Chapters 16–20
17. Transfer to coding and data rate
18. Transfer to weak-signal measurements
Report and continue · Chapters 21–24
21. Repair lower dBm always means lower noise figure
22. Repair good noise figure guarantees service
Start by naming the quantity that the decision actually needs. Noise factor compares input and output signal-to-noise ratios for a receiver under stated reference conditions; noise figure is ten times the base-ten logarithm of that factor.
Work the teaching case in a visible order. A fictional receiver has noise factor 3.162, giving noise figure 5.0 dB.
Make the reasoning independently checkable. Convert with NF=10 log10(F) and recover F from 10^(NF/10).
Test transfer by changing one condition while holding the others still. Change F to 2.0 and predict the dB value falls to about 3.01.
Keep the claim boundary visible. Noise figure varies with frequency, settings, temperature and reference plane.
Before accepting the result, predict what should move and what should stay fixed. Apply this check: Convert with NF=10 log10(F) and recover F from 10^(NF/10). Then run the changed case: Change F to 2.0 and predict the dB value falls to about 3.01.
The evidence receipt for “Define noise factor and noise figure” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Keep the observation, transformation and conclusion on separate evidence lines. Receiver sensitivity is the input signal level needed to meet a specified performance criterion, bandwidth, modulation and configuration.
Work the teaching case in a visible order. A fictional digital receiver requires SNR 10 dB at its decision point and is evaluated over 1 MHz bandwidth.
Make the reasoning independently checkable. Attach the criterion and bandwidth to the dBm value.
Test transfer by changing one condition while holding the others still. Change required SNR to 6 dB without changing hardware noise figure.
Keep the claim boundary visible. A bare sensitivity number is not comparable across unequal tests.
Build a compact audit receipt with input, unit, method, calculation, observation and limit. Use this check: Attach the criterion and bandwidth to the dBm value. For transfer, Change required SNR to 6 dB without changing hardware noise figure.
The evidence receipt for “Define sensitivity operationally” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Use the short answer only after the reference condition and units are fixed. Available thermal noise for the ideal reference is kTB, commonly represented near 290 K as about -174 dBm/Hz before bandwidth integration.
Work the teaching case in a visible order. For 1 MHz, -174+10log10(1,000,000)≈-114 dBm.
Make the reasoning independently checkable. Recompute from six decades of bandwidth and check the 60 dB addition.
Test transfer by changing one condition while holding the others still. Narrow the bandwidth to 100 kHz.
Keep the claim boundary visible. The -174 dBm/Hz approximation depends on reference temperature and conventions.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Recompute from six decades of bandwidth and check the 60 dB addition. Next, Narrow the bandwidth to 100 kHz.
The evidence receipt for “Calculate thermal noise” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Trace the mechanism in the order another reader can reproduce. Noise figure raises the effective receiver input-referred noise above the ideal thermal reference in the teaching budget.
Work the teaching case in a visible order. With 5 dB NF and 1 MHz, the input-referred noise floor is about -109 dBm.
Make the reasoning independently checkable. Keep dB additions separate from linear power sums.
Test transfer by changing one condition while holding the others still. Improve NF to 3 dB.
Keep the claim boundary visible. External antenna noise, cable loss and interference require additional terms.
Turn the claim into a test that could fail. First, Keep dB additions separate from linear power sums. Then change one consequential condition: Improve NF to 3 dB.
The evidence receipt for “Add receiver noise figure” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Treat a plausible number as a prompt for a check rather than as proof. A simple sensitivity estimate adds required SNR to the input-referred noise floor.
Work the teaching case in a visible order. At -109 dBm noise and 10 dB required SNR, sensitivity is about -99 dBm.
Make the reasoning independently checkable. Check that a more negative threshold represents ability to detect a weaker signal.
Test transfer by changing one condition while holding the others still. Keep NF at 5 dB but require 6 dB SNR, giving about -103 dBm.
Keep the claim boundary visible. Implementation loss, coding, demodulation and measurement criteria must be declared.
Before accepting the result, predict what should move and what should stay fixed. Apply this check: Check that a more negative threshold represents ability to detect a weaker signal. Then run the changed case: Keep NF at 5 dB but require 6 dB SNR, giving about -103 dBm.
The evidence receipt for “Complete the sensitivity budget” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
CHAPTER 6 OF 24 · Work the central example
6. Show why the numbers are not synonyms
Start by naming the quantity that the decision actually needs. Noise figure is a receiver noise degradation metric; sensitivity combines noise, bandwidth and a performance threshold.
Work the teaching case in a visible order. Two receivers can share 5 dB NF but have -99 and -93 dBm sensitivities because their bandwidths or required SNR differ.
Make the reasoning independently checkable. Hold NF fixed and vary one test condition at a time.
Test transfer by changing one condition while holding the others still. Give two receivers equal sensitivity by trading NF against bandwidth.
Keep the claim boundary visible. Equal end numbers do not imply equal internal performance.
Build a compact audit receipt with input, unit, method, calculation, observation and limit. Use this check: Hold NF fixed and vary one test condition at a time. For transfer, Give two receivers equal sensitivity by trading NF against bandwidth.
The evidence receipt for “Show why the numbers are not synonyms” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Keep the observation, transformation and conclusion on separate evidence lines. Loss ahead of the first amplifier degrades system noise performance and shifts the signal available at the receiver input.
Work the teaching case in a visible order. Insert a 2 dB cable loss before the 5 dB receiver in the stipulated chain.
Make the reasoning independently checkable. Define whether sensitivity is measured at antenna connector, cable input or demodulator input.
Test transfer by changing one condition while holding the others still. Move the same loss after a high-gain low-noise amplifier.
Keep the claim boundary visible. A cascade needs gain and noise-factor analysis, not simple relocation by prose.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Define whether sensitivity is measured at antenna connector, cable input or demodulator input. Next, Move the same loss after a high-gain low-noise amplifier.
The evidence receipt for “Track the reference plane” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Use the short answer only after the reference condition and units are fixed. The first active stages strongly influence total noise factor when they provide gain ahead of later stages.
Work the teaching case in a visible order. Use a stipulated low-noise amplifier with F1=1.5 and gain G1=10 followed by F2=4; total F≈1.5+(4-1)/10=1.8.
Make the reasoning independently checkable. Calculate in linear factors before converting to dB.
Test transfer by changing one condition while holding the others still. Reduce first-stage gain to 2.
Keep the claim boundary visible. Mismatch, loss and nonlinearity may require a more complete model.
Turn the claim into a test that could fail. First, Calculate in linear factors before converting to dB. Then change one consequential condition: Reduce first-stage gain to 2.
The evidence receipt for “Use Friis cascade reasoning” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Trace the mechanism in the order another reader can reproduce. Noise power grows with effective noise bandwidth, while filters and detection criteria determine which bandwidth belongs in the test.
Work the teaching case in a visible order. Increasing bandwidth from 100 kHz to 1 MHz adds 10 dB of thermal noise.
Make the reasoning independently checkable. Use effective noise bandwidth rather than a marketing channel label when available.
Test transfer by changing one condition while holding the others still. Keep occupied signal bandwidth fixed but widen the receiver filter.
Keep the claim boundary visible. Filter shape and digital processing can make nominal and effective bandwidth differ.
Before accepting the result, predict what should move and what should stay fixed. Apply this check: Use effective noise bandwidth rather than a marketing channel label when available. Then run the changed case: Keep occupied signal bandwidth fixed but widen the receiver filter.
The evidence receipt for “Keep bandwidth honest” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Treat a plausible number as a prompt for a check rather than as proof. Ask whether the reader is diagnosing receiver-generated noise or the weakest signal that passes a declared test.
Work the teaching case in a visible order. Use noise figure for the former and sensitivity for the latter.
Make the reasoning independently checkable. State performance criterion, bandwidth and reference plane before quoting sensitivity.
Test transfer by changing one condition while holding the others still. Change modulation while leaving receiver hardware unchanged.
Keep the claim boundary visible. Some systems specify energy-per-bit, SINAD, packet error rate or other criteria.
Build a compact audit receipt with input, unit, method, calculation, observation and limit. Use this check: State performance criterion, bandwidth and reference plane before quoting sensitivity. For transfer, Change modulation while leaving receiver hardware unchanged.
The evidence receipt for “Use the short-answer gateway” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Start by naming the quantity that the decision actually needs. Noise figure can change across frequency and temperature, so one room-temperature midpoint value does not own the whole receiver.
Work the teaching case in a visible order. The test plan measures three frequencies and two temperatures.
Make the reasoning independently checkable. Compare like settings and include uncertainty.
Test transfer by changing one condition while holding the others still. Turn a switchable preamplifier off.
Keep the claim boundary visible. Manufacturer maximum, typical and measured values are different claims.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Compare like settings and include uncertainty. Next, Turn a switchable preamplifier off.
The evidence receipt for “Audit temperature and frequency” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Keep the observation, transformation and conclusion on separate evidence lines. Automatic gain control, attenuators and preamplifiers alter the measurement configuration.
Work the teaching case in a visible order. The ITU-style teaching record fixes attenuator minimum and notes preamplifier state.
Make the reasoning independently checkable. Save every receiver setting with the result.
Test transfer by changing one condition while holding the others still. Repeat in low-noise and high-dynamic-range modes.
Keep the claim boundary visible. A mode name is not a substitute for measured conditions.
Turn the claim into a test that could fail. First, Save every receiver setting with the result. Then change one consequential condition: Repeat in low-noise and high-dynamic-range modes.
The evidence receipt for “Audit gain and AGC settings” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Use the short answer only after the reference condition and units are fixed. Analogue sensitivity may use SINAD while digital sensitivity may use bit, frame or packet error rate.
Work the teaching case in a visible order. A -100 dBm value at 12 dB SINAD cannot be compared directly with -100 dBm at 10% packet error.
Make the reasoning independently checkable. Name the output metric, threshold and confidence.
Test transfer by changing one condition while holding the others still. Tighten the packet-error criterion.
Keep the claim boundary visible. A single threshold hides the full response curve.
Before accepting the result, predict what should move and what should stay fixed. Apply this check: Name the output metric, threshold and confidence. Then run the changed case: Tighten the packet-error criterion.
The evidence receipt for “Audit the test criterion” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Trace the mechanism in the order another reader can reproduce. Field performance includes antenna noise, man-made noise and interference that a laboratory noise figure does not capture.
Work the teaching case in a visible order. Add a stipulated external noise contribution that dominates the 5 dB receiver term.
Make the reasoning independently checkable. Separate internal receiver noise from site environment.
Test transfer by changing one condition while holding the others still. Move the receiver from a quiet site to an urban site.
Keep the claim boundary visible. A good receiver noise figure cannot guarantee a clear channel.
Build a compact audit receipt with input, unit, method, calculation, observation and limit. Use this check: Separate internal receiver noise from site environment. For transfer, Move the receiver from a quiet site to an urban site.
The evidence receipt for “Audit external noise and interference” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Treat a plausible number as a prompt for a check rather than as proof. Source calibration, impedance match, gain drift and finite averaging affect measured noise figure and sensitivity.
Work the teaching case in a visible order. Repeat five measurements and report spread instead of a lone rounded value.
Make the reasoning independently checkable. Check calibration dates and traceability.
Test transfer by changing one condition while holding the others still. Change cable connection and repeat.
Keep the claim boundary visible. This article does not define a compliance uncertainty budget.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Check calibration dates and traceability. Next, Change cable connection and repeat.
The evidence receipt for “Audit uncertainty and repeatability” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Start by naming the quantity that the decision actually needs. Sensitivity can be compared with predicted received power to form margin, while noise figure belongs inside the receiving-system noise calculation.
Work the teaching case in a visible order. A predicted -94 dBm signal against -99 dBm sensitivity gives 5 dB teaching margin.
Make the reasoning independently checkable. Keep fading, implementation and required availability assumptions visible.
Test transfer by changing one condition while holding the others still. Add 8 dB fading loss and observe negative margin.
Keep the claim boundary visible. A static link budget does not determine real service reliability alone.
Turn the claim into a test that could fail. First, Keep fading, implementation and required availability assumptions visible. Then change one consequential condition: Add 8 dB fading loss and observe negative margin.
The evidence receipt for “Transfer to link budgets” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Keep the observation, transformation and conclusion on separate evidence lines. Coding and modulation change the required performance threshold and often signal bandwidth without magically changing the hardware noise figure.
Work the teaching case in a visible order. Lower the required SNR from 10 to 6 dB under a stipulated coding change.
Make the reasoning independently checkable. Account for rate, bandwidth and packet criterion.
Test transfer by changing one condition while holding the others still. Use the same code but double occupied bandwidth.
Keep the claim boundary visible. Real coding gains depend on block length, decoder and channel conditions.
Before accepting the result, predict what should move and what should stay fixed. Apply this check: Account for rate, bandwidth and packet criterion. Then run the changed case: Use the same code but double occupied bandwidth.
The evidence receipt for “Transfer to coding and data rate” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Use the short answer only after the reference condition and units are fixed. A narrow measurement bandwidth can reveal weak carriers even when a wideband service receiver has a different sensitivity.
Work the teaching case in a visible order. Compare 1 kHz and 1 MHz noise bandwidths under the same 5 dB NF.
Make the reasoning independently checkable. Calculate the 30 dB bandwidth difference.
Test transfer by changing one condition while holding the others still. Integrate the narrowband result over a wider service signal.
Keep the claim boundary visible. Detector time, phase noise and signal structure can limit narrowband measurements.
Build a compact audit receipt with input, unit, method, calculation, observation and limit. Use this check: Calculate the 30 dB bandwidth difference. For transfer, Integrate the narrowband result over a wider service signal.
The evidence receipt for “Transfer to weak-signal measurements” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Trace the mechanism in the order another reader can reproduce. Antenna loss, preamplifier gain and receiver noise figure combine into a system noise temperature or equivalent factor.
Work the teaching case in a visible order. Keep every stage in order and refer values to one input.
Make the reasoning independently checkable. Recalculate after moving a lossy cable ahead of the amplifier.
Test transfer by changing one condition while holding the others still. Raise the first-stage gain while respecting its linear range.
Keep the claim boundary visible. Gain does not repair saturation, intermodulation or poor selectivity.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Recalculate after moving a lossy cable ahead of the amplifier. Next, Raise the first-stage gain while respecting its linear range.
The evidence receipt for “Transfer to receiver cascades” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
CHAPTER 20 OF 24 · Transfer the reasoning
20. Build a reproducible receiver receipt
Treat a plausible number as a prompt for a check rather than as proof. Record frequency, reference plane, temperature, impedance, bandwidth, configuration, modulation, criterion, source calibration, repetitions, noise figure and sensitivity.
Work the teaching case in a visible order. Another reader reconstructs -114, -109 and -99 dBm in the central budget.
Make the reasoning independently checkable. Remove bandwidth and identify why sensitivity becomes uninterpretable.
Test transfer by changing one condition while holding the others still. Version the receipt after firmware changes filter behavior.
Keep the claim boundary visible. Reproducibility checks the stated test, not every operating environment.
Turn the claim into a test that could fail. First, Remove bandwidth and identify why sensitivity becomes uninterpretable. Then change one consequential condition: Version the receipt after firmware changes filter behavior.
The evidence receipt for “Build a reproducible receiver receipt” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
CHAPTER 21 OF 24 · Report and continue
21. Repair lower dBm always means lower noise figure
Start by naming the quantity that the decision actually needs. A more negative sensitivity can result from narrower bandwidth or an easier criterion even when noise figure is worse.
Work the teaching case in a visible order. Trade 10 dB narrower bandwidth against several dB poorer NF in a changed example.
Make the reasoning independently checkable. Compare components only after aligning conditions.
Test transfer by changing one condition while holding the others still. Hold bandwidth and criterion fixed, then vary NF.
Keep the claim boundary visible. Sensitivity remains a system test outcome, not a direct NF label.
Before accepting the result, predict what should move and what should stay fixed. Apply this check: Compare components only after aligning conditions. Then run the changed case: Hold bandwidth and criterion fixed, then vary NF.
The evidence receipt for “Repair lower dBm always means lower noise figure” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
CHAPTER 22 OF 24 · Report and continue
22. Repair good noise figure guarantees service
Keep the observation, transformation and conclusion on separate evidence lines. Service also depends on received power, interference, selectivity, dynamic range, fading, protocol and implementation.
Work the teaching case in a visible order. The central receiver can have 5 dB NF yet fail when an interferer overloads it.
Make the reasoning independently checkable. Add field counters and spectrum observations.
Test transfer by changing one condition while holding the others still. Remove the interferer while keeping noise figure unchanged.
Keep the claim boundary visible. Noise figure describes one bounded aspect of receiver performance.
Build a compact audit receipt with input, unit, method, calculation, observation and limit. Use this check: Add field counters and spectrum observations. For transfer, Remove the interferer while keeping noise figure unchanged.
The evidence receipt for “Repair good noise figure guarantees service” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Use the short answer only after the reference condition and units are fixed. The method supports separating internal noise degradation from weakest-signal performance and reconstructing a declared sensitivity budget.
Work the teaching case in a visible order. It can reject a comparison between sensitivities measured with different bandwidths and criteria.
Make the reasoning independently checkable. Write the decision and test conditions before ranking receivers.
Test transfer by changing one condition while holding the others still. Move from the fictional budget to an accredited measurement.
Keep the claim boundary visible. Use current standards, manufacturer procedures and qualified RF laboratories for compliance.
Ask a second reader to reproduce the decision from the stated evidence. They should be able to apply: Write the decision and test conditions before ranking receivers. Next, Move from the fictional budget to an accredited measurement.
The evidence receipt for “State the supported decision” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
CHAPTER 24 OF 24 · Report and continue
24. Continue through verified wireless routes
Trace the mechanism in the order another reader can reproduce. This distinction sits beside SNR, Eb/N0, RSRP/RSSI and coherence-bandwidth branches.
Work the teaching case in a visible order. Choose the sibling that owns the next signal, noise or channel question.
Make the reasoning independently checkable. Keep receiver noise, external noise, performance criterion and delivered service separate.
Test transfer by changing one condition while holding the others still. Move to SNR versus Eb/N0 without relabeling sensitivity.
Keep the claim boundary visible. This article owns receiver noise figure versus sensitivity under declared test conditions.
Turn the claim into a test that could fail. First, Keep receiver noise, external noise, performance criterion and delivered service separate. Then change one consequential condition: Move to SNR versus Eb/N0 without relabeling sensitivity.
The evidence receipt for “Continue through verified wireless routes” records the source, input, transformation, changed-context prediction and limit on separate lines. Keep frequency, bandwidth, reference plane, temperature, settings, modulation, criterion, calibration and uncertainty with both measurements.
Contents · Previous chapter · Return to Wireless Communication
Continue through the verified reading route
Signal strength and SNR
Signal strength and SNR. Use this when received power is being mistaken for link quality.
SNR and Eb/N0
SNR and Eb/N0. Use this when bandwidth and bit rate change the normalized noise measure.
RSRP and RSSI
RSRP and RSSI. Use this when LTE reference-signal power is confused with total received power.
Frequently asked questions
Can two receivers have the same noise figure but different sensitivity?
Yes. Bandwidth, required SNR, modulation, coding and test criterion can differ.
Does a lower numerical dBm sensitivity always mean a better noise figure?
No. The sensitivity may use a narrower bandwidth or easier performance threshold.
Does good noise figure guarantee a reliable link?
No. Received power, external noise, interference, selectivity, dynamic range, fading and protocol behavior also matter.
Authoritative sources and further reading
ITU-R SM.1838-1. The in-force recommendation defines test procedures for receiver noise figure and explains its relation to sensitivity.
ITU-R P.372-17. The recommendation defines receiving-system noise factors, temperatures and bandwidth relationships.
NASA Radio JOVE thermal-noise note. NASA explains the kTB thermal-noise relationship used in the teaching budget.
Return to the Wireless Communication subject guide, The Gold Standard of Education, or the How X Works hub.