FT8 Signal Report Explained: What the dB Numbers Really Mean
Sooner or later every FT8 operator looks at a log and asks a simple question: what does a report of -14 mean? On SSB a report of 59 is a statement of good copy, and on CW the number 599 says the same thing in a ritual way. In FT8 the report is a number in decibels with a minus sign, and the same station can send you -03 in the morning and -17 an hour later without moving a wire. This is the FT8 signal report: an automatic estimate of the signal to noise ratio, made by the decoder at the receiving end. It is useful, but it is easy to misread. This guide explains how the number is obtained, what it is measured against, how it relates to the S units that you know from voice modes and what to conclude from the figures in your decoded list.
What the Number Actually Is
The report in an FT8 exchange is the signal to noise ratio, abbreviated SNR, that the decoding program calculated for the received signal. WSJT-X shows it in the column “dB” next to each decoded line. The measurement is relative to the noise that would be present in a bandwidth of 2500 Hz, which is a typical width of an SSB filter. The Wikipedia article on the signal-to-noise ratio explains the general concept: it is the ratio of the power of the wanted signal to the power of the noise, usually expressed in decibels.
The choice of a 2500 Hz reference is a convention, and it is important to understand that it is only a convention. The FT8 signal itself occupies about 50 Hz. The noise in 50 Hz is much smaller than the noise in 2500 Hz, by a factor of 50, or 17 dB. A report of -21 dB therefore means that the signal is 21 dB below the noise measured in 2500 Hz, but only about 4 dB below the noise in the bandwidth that the decoder really uses. That is the reason why FT8 can produce a correct decode when the signal is invisible to your ears: the real ratio in its narrow bandwidth is close to zero, and the coding and the averaging over 15 seconds complete the job.
Range, Limits and Typical Values
The values that you see in the decoded list cover a wide range. A report around +10 to +20 dB is a very strong signal, from a station that is close or has a great antenna. Reports from -5 to -15 are typical for a normal contact with a good signal. Values below -20 are weak signals, and the decoder starts to miss messages below about -21 dB under ordinary conditions, while the use of a priori information can push the limit lower, which is what a priori (AP) decoding in WSJT-X does. The exact threshold depends on the noise, on the frequency stability and on how many signals overlap.
| Reported SNR | What it means in practice | Chance of a decode |
|---|---|---|
| +10 dB and above | Very strong; check your own level for distortion if you are close | Certain |
| 0 to -10 dB | Good signal; most contacts complete easily | Very high |
| -10 to -18 dB | Normal for a busy DX contact; copy is reliable | High |
| -18 to -22 dB | Weak; some messages are missed, repeats are likely | Medium |
| Below -22 dB | At the edge; decoding depends on a clean band and AP help | Low |
These bands are a guide rather than a rule, because the point at which a message is lost depends on the local conditions. Fading, interference and a drifting transmitter all raise the level of signal needed.
Why It Is Not the Same as an S Meter
An S meter measures the strength of the signal at the input of the receiver, in units of about 6 dB each, with S9 as a reference level. The FT8 report measures the ratio of the signal to the noise, and a strong signal in a very noisy band can give the same number as a weak signal on a quiet band. This difference is the reason why you cannot translate a figure in dB directly into an S unit. Our article on RST reports and the S meter explains the units that are used for voice and CW; the FT8 number is a ratio in the decoder, not a reading of strength.
It follows that two stations can receive the same signal and give different reports. One has a quiet location with a low noise floor and reports -10, the other lives in a city with strong interference from electronics and reports -18 for the same transmitted power. Neither is wrong. The report tells you about the combination of the signal and the receiving environment at the other end. It does not tell you the power or the distance, and it cannot be used as a calibrated measurement of your field strength.
Why Reports Are Not Reciprocal
On a path in the HF bands you might send -15 and receive -05. The difference of 10 dB does not mean that your signal is weaker by that much. There are many reasons for an asymmetry:
- Different noise levels. The noise at each end is different, and the report is relative to the local noise.
- Different antennas and power. A station with 100 watts and a beam will be reported better than a station with 10 watts and a wire, which does not say anything about the receiving ability of the first one.
- Local interference. The other end may hear a strong neighbouring signal that raises the noise in the decoding window.
- AGC and filter settings. The way in which the receiver treats a strong signal near the passband changes the noise level in the audio, and the SNR estimate with it.
- Timing and fading. A signal that fades in the middle of a 15 second period gets a lower estimate than the same signal with a stable level.
The practical lesson is that you should not compare your report with the report of another station as a measure of the quality of your station. If you want an objective comparison, use the reports on the same path at the same time from several stations, as shown by a spotting map, where the difference between your signal and that of a reference station is more meaningful.
What Goes Into the Message
The report is not only an indication on the screen. It is also part of the information that the stations exchange. In a standard FT8 contact one station sends the report to the other, who replies with a report preceded by the letter R, which means “received”, and then the exchange is closed with RR73 or 73. The numbers are those that the program has calculated, and in WSJT-X they are filled in automatically by the macros. You can change them by hand, but the common practice is to leave the value that the software has measured. If you want to understand the steps of a full exchange, our guide to setting up WSJT-X describes the screen and the buttons.
The report that you log will become part of your confirmation on the logbook sites, and a station that checks the entries sees the numbers. A logged number that makes no sense, for example +30 on a path of 10,000 km with a power of 5 watts, may signal a mistake. Another check is whether your sent and received reports are plausible in relation to each other. A difference of 20 dB between the two directions is possible but unusual, and a big gap is a reason to look at your own station.
Using the Report to Improve Your Station
The number can be a powerful tool if you know how to use it. Compare your reports from the same station on the same band at different times, and watch how they change. If your reports are consistently lower on one band than another, it points to an antenna problem, a noisy connection or a poor match on that band. If the reports vary a lot in a few minutes, the cause may be a rig that is overheating and lowering its power. If the reports are good but you do not get answers, the problem lies at the other end, in the pileup or in the choice of the frequency.
It also helps to use a reporting network. The decoded signals from your station are shown on a map, with the reports from many receivers at once. Our article on PSK Reporter shows how to read it. If you change an antenna, you can compare the average report from the same area before and after the change, and an improvement of 3 dB is clearly visible in the numbers.
Reading the Decoded List in Practice
A busy band shows dozens of lines in each period, and the report column helps you to choose whom to call. A station with a strong report in your list is audible in your receiver, which means that the path is open at that moment, and it is a better target than a station that appears once at -22. Watch the same call sign over several periods. If its report rises from period to period, the path is improving, and if it falls steadily, the opening is closing. A DX station that shows +05 on your screen but never answers your call may be working a large pileup, and your own transmit level is not the cause. Keep a short note of the best reports that you see on each band at each hour, and after a few weeks you will have a personal propagation chart that is more useful than a general forecast.
Common Misunderstandings
The first misunderstanding is that a minus sign means a bad signal. Because the reference is a wide bandwidth, almost all the signals in a normal FT8 contact are negative, and a positive value is the exception. The second is that the report depends on the power of your transmitter alone. It depends on the whole chain: power, antenna, path, noise and the receiving station. The third is that you should send “better” reports to be polite. Do not edit the number to make the other station happy; an honest value is more useful and it is what the software expects. The fourth is that a report of -24 is the limit. The limit is a statistical matter, and good conditions allow decodes of weaker signals, while a crowded band causes failure of stronger ones.
The Bottom Line
An FT8 signal report is the decoder’s estimate of the signal to noise ratio, referred to a 2500 Hz bandwidth although the signal itself occupies only about 50 Hz. A figure of -21 dB means roughly 4 dB below the noise in the real bandwidth, which is why weak signals can still be decoded. The report is not an S meter reading, it is not calibrated, and it is rarely the same in both directions because the noise, antennas and interference differ at each end. Use it to compare your own results over time and across bands, trust the trend more than a single value and leave the numbers in your messages as the software has measured them.
Frequently Asked Questions
What does a report of -15 mean in FT8?
It means the decoder estimated the signal at 15 dB below the noise power in a 2500 Hz bandwidth. It is a normal value for a good contact and is easily decoded.
Why is the FT8 report negative?
The reference is the noise in a wide bandwidth, much larger than the 50 Hz that the signal needs. The signal is therefore below the reference noise in most cases, and the number is negative.
How low can an FT8 signal be and still decode?
Under ordinary conditions the limit is around -21 dB. Better conditions and the use of a priori decoding may allow weaker signals, and a crowded band raises the limit.
Can I convert a dB report into an S unit?
No. An S unit measures signal strength, and the FT8 report measures the ratio of signal to noise. The same report can correspond to very different signal levels.
Why does my report differ from the one I receive?
The noise level, the antennas, the power and the local interference at the two ends are different, so the ratio is different in each direction. This is normal and does not mean that one side is wrong.
Should I change the report in the message?
No. Leave the value that the software calculated. It is part of the standard exchange and gives the other operator an honest indication of how well you were heard.