Why Does the Bathroom Have the Highest WiFi Signal Attenuation?

[EleSuki]

There is a Japanese IT news website that I've been researching about the strength of wireless LAN signals. I would like you to take a look at an image where the attenuation rate of the signal is low only in the area marked as "6," which is the bathroom. Why is the wireless signal strength attenuating to such an extent specifically in the bathroom?

The attenuation rate of the 5GHz signal is most significantly evident in the bathroom.

 

[Baluncore]

Maybe it is because the walls of the bathroom are more substantial than other walls in the house. That may be because it is a wet area and so has heavy tiles, or a composite cladding on the walls.

 

[berkeman]

Why is the wireless signal strength attenuating to such an extent specifically in the bathroom?

What is the difference between the blue and red bars in the graphic? Different frequency bands maybe?

And did they mention how they made these measurements? For the wavelengths involved in WiFi, you need to take multipath into account to be sure you are not conflating attenuation and multipath nulls...

 

[EleSuki]

Maybe it is because the walls of the bathroom are more substantial than other walls in the house. That may be because it is a wet area and so has heavy tiles, or a composite cladding on the walls.

Thank you. It may be worth considering the material of the walls. In Japanese bathrooms, there are tiles, but mortar is also used. Silicon carbide has the property of absorbing electromagnetic waves. Does mortar have such a property as well?

 

[berkeman]

In Japanese bathrooms, there are tiles

What other material is in bathroom walls, and what does that have to do with my multipath comments above?

 

[EleSuki]

What is the difference between the blue and red bars in the graphic? Different frequency bands maybe?

And did they mention how they made these measurements? For the wavelengths involved in WiFi, you need to take multipath into account to be sure you are not conflating attenuation and multipath nulls...

I apologize for not mentioning the frequency of the bar graph. The red color represents a frequency of 5GHz, while the blue color represents a frequency of 2.4GHz. Although it is unclear what kind of wireless LAN measurement device was used, it seems that a professional measurement device capable of accurately measuring the radio wave strength was used, as even smartphones can measure signal strength.

In the experiment, it is likely that separate measurements were taken for 2.4GHz and 5GHz frequencies. However, it is certain that even considering multipath nulling, the signal strength is noticeably lower in the bathroom.

 

[EleSuki]

What other material is in bathroom walls, and what does that have to do with my multipath comments above?

I was researching multipath nulling since I'm not very familiar with it. It seems that when multiple frequency bands coexist, the signal strength of a specific frequency, even if it is, for example, 100, can be received as 60 due to the multipath nulling function.

This is related to the sensitivity adjustment of the wireless LAN device on the receiving end. Am I correct? From my perspective, in order to avoid the issue of multipath nulling, it is possible to choose only one of the 2.4GHz and 5GHz wireless LAN frequencies and disable the other through the router settings.

I speculate that the experiment was conducted by enabling one frequency to mitigate the multipath nulling issue.

 

[berkeman]

It seems that when multiple frequency bands coexist,

Multipath is a single frequency issue. When there are reflections of the main signal from objects, that causes multiple receive signals of different phases and amplitudes, so you can get reinforcement or nulling of the combined receive signal. When making attenuation measurements where multipath signals exist, it is necessary to move the receive antenna spatially a wavelength or two to find the extent of the high and low receive strengths. Perhaps the numbers being shown in the diagram are the average values of those spatial extremes (hopefully, if they did a careful job measuring the actual attenuation).

To overcome multipath issues, you can employ spatial diversity and/or frequency diversity. Spatial diversity is illustrated by devices that have multiple antennas spaced a bit apart...

https://www.coolbox.pe/router-tplink-tl-mr6400apac-wifi-300-mbps-tl-mr6400apac/p

 

[EleSuki]

Multipath is a single frequency issue. When there are reflections of the main signal from objects, that causes multiple receive signals of different phases and amplitudes, so you can get reinforcement or nulling of the combined receive signal. When making attenuation measurements where multipath signals exist, it is necessary to move the receive antenna spatially a wavelength or two to find the extent of the high and low receive strengths. Perhaps the numbers being shown in the diagram are the average values of those spatial extremes (hopefully, if they did a careful job measuring the actual attenuation).

To overcome multipath issues, you can employ spatial diversity and/or frequency diversity. Spatial diversity is illustrated by devices that have multiple antennas spaced a bit apart...

View attachment 327745
https://www.coolbox.pe/router-tplink-tl-mr6400apac-wifi-300-mbps-tl-mr6400apac/p

Thank you once again. I see! So even if only a single frequency is emitted, the multipath nulling problem can still occur. I've learned something new. The website where this measurement was conducted is PC Watch, which covers news on IT gadgets and other related topics. They have been conducting speed tests for wireless LAN devices based on standards like IEEE 802.11b and 802.11a, so they are likely aware of the multipath nulling issue (although they may have taken a stance of measuring only the rough signal strength without precision).

However, in that case, like you, I can only speculate whether the bathroom experienced multipath nulling, making the "apparent signal strength appear lower," or if the bathroom walls possess characteristics that absorb radio waves and attenuated the wireless LAN signals. It seems I have grasped the intentions behind your question. I hadn't even considered the possibility of the existence of the multipath nulling problem, so this is quite enlightening. Even with a signal strength of 100, it can nullify itself to around 30.

 

[berkeman]

I think most likely the tile material of the bathroom may be causing the higher attenuation, but there are more metal pipes in the walls of the bathroom compared to other rooms, so that may also be part of the problem. Do you know if the vent pipes are metal or plastic?

 

[EleSuki]

I think most likely the tile material of the bathroom may be causing the higher attenuation, but there are more metal pipes in the walls of the bathroom compared to other rooms, so that may also be part of the problem. Do you know if the vent pipes are metal or plastic?

If i try to research the details of tile materials, may find out what factors are involved in the attenuation of radio waves by tiles.

In foreign movies, I remember seeing exposed pipes in bathrooms, but in Japan, bathrooms typically only have pipes for faucets and showers. The faucets also protrude slightly from the wall.Irregularly shaped metallic substances are also inconvenient for radio signals.

 

[tech99]

The bathroom seems to be nearly above the LAN router, so the low low signal strength could be because the vertical antenna of the router will have an overhead minimum.
Microwave propagation in and around buildings and near to the ground is always subject to multipath. It is usual to consider a statistical distribution of received field strength with position. One reading means nothing.

 

[Baluncore]

The bathroom seems to be nearly above the LAN router, so the low low signal strength could be because the vertical antenna of the router will have an overhead minimum.

I believe that is a trick of the projection. Both the router and the bathroom are on the first floor = 1F diagram.

 

[berkeman]

The bathroom seems to be nearly above the LAN router,

I believe that is a trick of the projection. Both the router and the bathroom are on the first floor = 1F diagram.

I dunno, maybe tech99 can read Japanese, and what does "F" mean in Japanese anyway?...

 

[TonyStewart]

TL:DR The bathroom is high in humidity which absorbs and reflects microwave energy. Ceramic does not. Just try to heat up your empty coffee/tea mug with another cup of water in the microwave oven.But avoid painted cups with gold rims. you'll get arcs where it burns out. The bathroom also has more reflective surfaces from fixtures, tubs water etc.There are several well-defines principles that cause signal fading in the home from low power routers to higher carrier (5GHz vs 2.5GHz) to reflections in weaker areas called Ricean Fading. This is compensated by Router choice, router setup in WiFi using browser, location near reflected surfaces including walls, floor, mobile WiFi auto equalization which attempts to correct errors, retry or re-train channel, lower baud rate, switch channel in order to avoid some major reflections. But most often just moving a couple mm or more is all you need to get out of a deadspot. Deadspots are the reception of carrier signals delayed by half-wave length from any and every surface ( humidity and metals more than others) near the same amplitude. So in the fringe zones deadpots will drop 5 to 15 dB just by moving 1 or 2 mm away from the deadspot.

https://www.speedtest.net.

In addition, you can try adjusting the retry or re-training channel, lowering the baud rate, or switching channels to avoid major reflections or rely on the automatic setup. However, sometimes simply moving a few millimeters can be enough to escape a deadspot, as these are many tiny zones where carrier signals are delayed by half-wavelength from surfaces and experience significant signal degradation. The one wavelength that cancels out is a relative term and can be a broad range within the 10M, 20M, 40MHz etc band at 2400 MHz or 5GHz. Where the signals is strong reflections are not a problem, it is where the signal is weaker ,then reflections are every half wave from the damp wall in the bathroom.
Anecdotal
In a previous home when we moved in, we could not get a modem hookup for 5 days. So I found a home several doors down that I could connect with by resting a laptop on a table near the rear window and aiming a large tree of moisture-filled leaves to reflect the signal and get internet when the signals were weaker 20 yrs ago less likely to all be locked up. If I moved the laptop moe than 1 degree or 1mm , I would lose the signal. So it can be used additively but usually is just a nuyciance with droputs.

In terms of Wi-Fi bandwidth, there are different combinations of 20 MHz bands that can be grouped to achieve higher bitrates. However, to take advantage of these higher bitrates (>54 Mbps), you generally need to be in close proximity to the router. If you prefer a more reliable connection, you can choose the lowest common rate, such as 11 Mbps on the 2.4 GHz frequency. Factors such as Friis Loss, which is the loss in signal strength over distance, and Shannon's Law, which defines the relationship between bandwidth and signal speed, play a significant role in Wi-Fi performance. Furthermore, Ricean Fading loss becomes more dominant with greater distance and more reflections, causing signal degradation.

If your Wi-Fi speed drops below 54 Mbps or 11 Mbps, it indicates that there may be issues with power levels, echoes, reflections, location, or antenna orientation. Achieving speeds of 150 Mbps or using 5GHz frequency can be challenging unless you have a clear line of sight to the router, and it may not perform as well as the 2.4 GHz frequency in certain environments.

In the past, a Windows application could be used to plot signal levels in dBm while slowly moving objects or walking in the signal path. This method allowed for visualizing signal strengths and potential deadspots.

Factors such as Ricean Fading (reflections), co-channel collisions, and the higher threshold of the 5GHz frequency contribute to Rx noise sources. When the 2.4GHz carrier signal drops below -75 dBm, it becomes easier for the signal to drop below the demodulator noise threshold. The SNR (Signal-to-Noise Ratio) threshold depends on various factors, including router power, mode, channel, bandwidth, selection criteria in setup, antenna orientation, location near walls, mobile device orientation, motion near deadspots, etc. To determine if packet errors are occurring, you can conduct tests such as transferring videos on the LAN or using websites like speedtest.net.

When excessive retries, dropouts, or lower efficiency are observed, the mobile device automatically adapts its algorithm to correct the issues by changing the channel bandwidth (20, 40, 60, 80 MHz) through manual or automated methods selected in the router setup. Experienced users can also improve the router's power, central location, and Wi-Fi driver setup to enhance performance.

Router setup is typically done through a web page accessed via the local gateway. Mobile devices can be configured to change from the default automated setup, which can be useful in weak signal areas.

The theory for line-of-sight transmission without ground reflections can be computed using Friis Loss calculators, which show that the loss is proportional to frequency and inversely squared with distance. However, higher data rates with compressed bandwidth require stronger signals due to the Shannon-Hartley law. Indoors, the effects of multipath fading, known as Ricean Loss, dominate Wi-Fi communication since signals reflect to some extent for all frequencies and rates, albeit at different positions. The ratio between the Wi-Fi signal (S) and Ricean Noise (N), known as S/N or SNR, is crucial. Technically, it is the Carrier-to-Noise ratio (CNR) that gets demodulated to the baseband signal SNR. Outdoors, the range between transmitter (Tx) and receiver (Rx) is limited by the receiver's white noise level threshold. The signal is calibrated when not moving to equalize the losses and phase shifts caused by echoes or reflections.

When high error rates occur, the modem in the mobile device should step down the data rates to tolerate a slightly lower SNR. For example, good performance can be achieved with a 2.4GHz 54 Mbps signal when the signal strength is above -75 dBm. If it falls below -80 dBm, you can try adjusting antenna locations or stabilizing a laptop, as even a 1 mm change can make a 6 dB difference in signal strength (e.g., from -82 to -76 dBm), potentially improving the data connection from little or no data to almost error-free.

If the aforementioned methods do not improve the situation, you can consider adding an old router to connect to the new router/modem. This setup can act as an extender or provide dedicated coverage to another area, such as basement corners or the third floor, by utilizing both 2.4GHz and 5GHz options and selecting the best configuration.

Reflections are present everywhere and can be particularly problematic when using a low-power modem/router or when the router is not centrally located due to other computer connections.

 

[renormalize]

TL:DR The bathroom is high in humidity which absorbs and reflects microwave energy.

I just want to correct this one misstatement. The power-loss due to atmospheric humidity is tiny at WiFi frequencies, as can be seen from the following graph:

Even at $100\text{% RH}$, the attenuation at $2.45$ and $5\text{ GHz}$ is less than $10^{-2}\text{ dB/km}$, and so the loss is utterly negligible across the dimensions of a bathroom.
Source: https://www.mdpi.com/2079-9292/8/12/1521

 

[tech99]

It looks to me that for coverage of the bathroom the signal must pass through more walls.

 

[EleSuki]

The bathroom seems to be nearly above the LAN router, so the low low signal strength could be because the vertical antenna of the router will have an overhead minimum.
Microwave propagation in and around buildings and near to the ground is always subject to multipath. It is usual to consider a statistical distribution of received field strength with position. One reading means nothing.

Thank you. I understand that there may be some spatial issues, but the bathroom is not directly above the LAN router; it is located on the same floor. The bathroom is indicated as "Red 65, Blue 66." Japanese can be ambiguous, making it easy for non-native speakers to misinterpret.

 

[EleSuki]

TL:DR The bathroom is high in humidity which absorbs and reflects microwave energy. Ceramic does not. Just try to heat up your empty coffee/tea mug with another cup of water in the microwave oven.But avoid painted cups with gold rims. you'll get arcs where it burns out. The bathroom also has more reflective surfaces from fixtures, tubs water etc.There are several well-defines principles that cause signal fading in the home from low power routers to higher carrier (5GHz vs 2.5GHz) to reflections in weaker areas called Ricean Fading. This is compensated by Router choice, router setup in WiFi using browser, location near reflected surfaces including walls, floor, mobile WiFi auto equalization which attempts to correct errors, retry or re-train channel, lower baud rate, switch channel in order to avoid some major reflections. But most often just moving a couple mm or more is all you need to get out of a deadspot. Deadspots are the reception of carrier signals delayed by half-wave length from any and every surface ( humidity and metals more than others) near the same amplitude. So in the fringe zones deadpots will drop 5 to 15 dB just by moving 1 or 2 mm away from the deadspot.

https://www.speedtest.net.View attachment 327778

In addition, you can try adjusting the retry or re-training channel, lowering the baud rate, or switching channels to avoid major reflections or rely on the automatic setup. However, sometimes simply moving a few millimeters can be enough to escape a deadspot, as these are many tiny zones where carrier signals are delayed by half-wavelength from surfaces and experience significant signal degradation. The one wavelength that cancels out is a relative term and can be a broad range within the 10M, 20M, 40MHz etc band at 2400 MHz or 5GHz. Where the signals is strong reflections are not a problem, it is where the signal is weaker ,then reflections are every half wave from the damp wall in the bathroom.
Anecdotal
In a previous home when we moved in, we could not get a modem hookup for 5 days. So I found a home several doors down that I could connect with by resting a laptop on a table near the rear window and aiming a large tree of moisture-filled leaves to reflect the signal and get internet when the signals were weaker 20 yrs ago less likely to all be locked up. If I moved the laptop moe than 1 degree or 1mm , I would lose the signal. So it can be used additively but usually is just a nuyciance with droputs.

In terms of Wi-Fi bandwidth, there are different combinations of 20 MHz bands that can be grouped to achieve higher bitrates. However, to take advantage of these higher bitrates (>54 Mbps), you generally need to be in close proximity to the router. If you prefer a more reliable connection, you can choose the lowest common rate, such as 11 Mbps on the 2.4 GHz frequency. Factors such as Friis Loss, which is the loss in signal strength over distance, and Shannon's Law, which defines the relationship between bandwidth and signal speed, play a significant role in Wi-Fi performance. Furthermore, Ricean Fading loss becomes more dominant with greater distance and more reflections, causing signal degradation.

If your Wi-Fi speed drops below 54 Mbps or 11 Mbps, it indicates that there may be issues with power levels, echoes, reflections, location, or antenna orientation. Achieving speeds of 150 Mbps or using 5GHz frequency can be challenging unless you have a clear line of sight to the router, and it may not perform as well as the 2.4 GHz frequency in certain environments.

In the past, a Windows application could be used to plot signal levels in dBm while slowly moving objects or walking in the signal path. This method allowed for visualizing signal strengths and potential deadspots.

Factors such as Ricean Fading (reflections), co-channel collisions, and the higher threshold of the 5GHz frequency contribute to Rx noise sources. When the 2.4GHz carrier signal drops below -75 dBm, it becomes easier for the signal to drop below the demodulator noise threshold. The SNR (Signal-to-Noise Ratio) threshold depends on various factors, including router power, mode, channel, bandwidth, selection criteria in setup, antenna orientation, location near walls, mobile device orientation, motion near deadspots, etc. To determine if packet errors are occurring, you can conduct tests such as transferring videos on the LAN or using websites like speedtest.net.

When excessive retries, dropouts, or lower efficiency are observed, the mobile device automatically adapts its algorithm to correct the issues by changing the channel bandwidth (20, 40, 60, 80 MHz) through manual or automated methods selected in the router setup. Experienced users can also improve the router's power, central location, and Wi-Fi driver setup to enhance performance.

Router setup is typically done through a web page accessed via the local gateway. Mobile devices can be configured to change from the default automated setup, which can be useful in weak signal areas.

The theory for line-of-sight transmission without ground reflections can be computed using Friis Loss calculators, which show that the loss is proportional to frequency and inversely squared with distance. However, higher data rates with compressed bandwidth require stronger signals due to the Shannon-Hartley law. Indoors, the effects of multipath fading, known as Ricean Loss, dominate Wi-Fi communication since signals reflect to some extent for all frequencies and rates, albeit at different positions. The ratio between the Wi-Fi signal (S) and Ricean Noise (N), known as S/N or SNR, is crucial. Technically, it is the Carrier-to-Noise ratio (CNR) that gets demodulated to the baseband signal SNR. Outdoors, the range between transmitter (Tx) and receiver (Rx) is limited by the receiver's white noise level threshold. The signal is calibrated when not moving to equalize the losses and phase shifts caused by echoes or reflections.

When high error rates occur, the modem in the mobile device should step down the data rates to tolerate a slightly lower SNR. For example, good performance can be achieved with a 2.4GHz 54 Mbps signal when the signal strength is above -75 dBm. If it falls below -80 dBm, you can try adjusting antenna locations or stabilizing a laptop, as even a 1 mm change can make a 6 dB difference in signal strength (e.g., from -82 to -76 dBm), potentially improving the data connection from little or no data to almost error-free.

If the aforementioned methods do not improve the situation, you can consider adding an old router to connect to the new router/modem. This setup can act as an extender or provide dedicated coverage to another area, such as basement corners or the third floor, by utilizing both 2.4GHz and 5GHz options and selecting the best configuration.

Reflections are present everywhere and can be particularly problematic when using a low-power modem/router or when the router is not centrally located due to other computer connections.

Thank you. The bathroom does indeed have high humidity, but based on the comments from the person who conducted the measurements below the image, it mentions that it is natural for the bathroom, surrounded by mortar walls, to have poor signal reception. While the humidity increases when using the bathroom, many households open windows when it is not in use, and within about an hour, the humidity levels become similar to the outside air.

I have also experienced similar issues with signal strength and directionality. In order to extend the WiFi signal to a shed located a short distance away, I used a repeater. To ensure that the repeater emitted a directional signal towards it, I created a cover using aluminum foil-lined boards, directing the signal towards the repeater.

Regarding the method of measuring the signal, I believe they likely used a professional measuring instrument, so the possibility of misjudging dead spots is low. What concerns me at the moment is why only rooms with mortar walls or tiles are experiencing multipath null problems or potential signal attenuation caused by the walls absorbing the signal (although this possibility seems unlikely at present).

 

[TonyStewart]

I have seen no evidence to suggest mortar has any more reflective or absorptive properties than kiln-dried wood. But moisture absorption has a significant effect.

 

[Nik_2213]

Although the air in wash-room will hold too little moisture to absorb WiFi, the material of the walls may be a different matter.
Tangential, this house's century-old local bricks swallow higher frequencies better than low. A cordless or DECT phone is okay, TXT is a single bar at best. '2' and, worse, '5' WiFi are 'line of sight' only.

I resorted to a 'tamed & tethered' LTE/router hung in window to get TXTs for 2FA at this 'Desk with No Bars'...

A thought: Just as 'WINDOW', tinsel cut to 1/4 wavelength, was used to thwart radar, could a conductive fibrous agent in the wash-room's tiles and/or cement be involved ??

 

[Baluncore]

A thought: Just as 'WINDOW', tinsel cut to 1/4 wavelength, was used to thwart radar, could a conductive fibrous agent in the wash-room's tiles and/or cement be involved ??

I see no reason why not. Anything is possible.
The length of the nails used to attach the wall cladding, the size of the door handle, or the copper pipes and house wiring, will all have some effect.

 

[TonyStewart]

Iloss is utterly negligible across the dimensions of a bathroom.
Source: https://www.mdpi.com/2079-9292/8/12/1521

The measurement is not the transmission loss but reflection coefficient from a change in impedance from condensation or absorption. If you haven’t tried to measure the loss or gain of reflection of moist surfaces I suggest you try some . The interfaces of moist surfaces creates significantly more Ricean Fading. I have measured with these reflective and Fading Losses many times both my advantage and disadvantage. Of course metal fixtures a more obvious sources for Ricean Fading Loss and others have shared their expertise too. One case was where ice melted and a trans-Canadian microwave tower went L.O.S. from this so that ground aperture had to be blocked.
To demonstrate with your mobile , record instanenous speedtest.net deviations while moving slowly in different orientations in your bathroom or any remote room . These variations are due to Ricean Fading at marginal point dropouts that may be mm wide.
The same problems exist with GPS when amidst skyscrapers with Ricean Fading off glass exterior buildings. This can create false position and/or LOS.

 

[renormalize]

The measurement is not just transmission loss but reflection coefficient. If you have’s tried to measure the gain of reflection of moist surfaces and the RiceanLosses this creates, try it. I have with wet leaves dropping >10dB and others as well where ice melted and a trans-Canadian microwave tower went LOS from this so that ground aperture had to be blocked.

The single statement by you that I criticized was "The bathroom is high in humidity which absorbs and reflects microwave energy." I take "humidity" to mean moisture in the atmosphere. The graph I posted accounts for all loss (reflection/scattering and absorption) suffered by WiFi microwave signals propagating through a humid atmosphere, and shows this loss to be totally insignificant in the context of the air in a bathroom. That was my only point, nothing about how liquid water on the surface of, or absorbed into, a solid material may affect WiFi strength.

 

[Baluncore]

Without an inspection, this discussion will be fruitless. For all we know, the bathroom walls may have been rendered with a waterproof mortar, supported on galvanised chicken wire.

At RF, liquid water has a dielectric constant of about Er=80. Solid water is ice, closer to Er=4. Snow is a foam dielectric of; ice 10% + air 90%. A room full of water-saturated air is transparent to RF. Water bound in biological tissues has a dielectric constant closer to Er=15, as are some clays that bind water. A damp mortar wall ... ?

 

[TonyStewart]

Significant reflection coefficients occur from exceeding the critical angle in a dielectric mismatch or metallic will reduce SNR but long haul less significant. Also ceramic tiles are generally transparent unless condensation exists and low Dk unlike wood in high humidity environments. But we can only speculate this result.

 

[EleSuki]

I have seen no evidence to suggest mortar has any more reflective or absorptive properties than kiln-dried wood. But moisture absorption has a significant effect.

Thank you. The hygroscopicity of mortar, which means that moisture adheres to the surface of the walls, may indeed be related to the attenuation of radio waves. Even if I have the bathroom window open, it seems that the moisture clinging to the walls cannot evaporate completely.

 

[Baluncore]

A thought: Just as 'WINDOW', tinsel cut to 1/4 wavelength, was used to thwart radar, ...

The tinsel scatterers, used as a radar countermeasure during WWII, were code-named: Window, Chaff, or Düppel. It was cut to a length of λ/2 to make a resonant dipole, not λ/4.

 

[Nik_2213]

Sorry, confused with anti-RADAR stuff on U-Boots' Schnorkels.
Blame the heat, as even our cats are sulking...

 

[TonyStewart]

Bottom line:

Do you want the fastest WiFi link? All the time? or some of the time? Switching bands manually.

If you need it some of the time > 54 Mbps burst rate with an average 50% thruput is best on 2.4GHz and then switch to 5 GHz when in rooms -60 to -40 dBm and expect wide variations when < -60 and complete dropouts < -70 on the highest data rate, otherwise it will choke down to 11 Mbps or slower

Whereas 54 MBps on 2.4GHz is reliable to -75 dBm then switch to 11 MBps manual if you are moving around and want overall consistent speed without video dropouts. -80 dBm will have high-error rate with occasions good bursts at 11 Mbps. At these weak signal levels, stable orientation is critical, then orient for max best speedtest.net.3.4Gbps advertised , you might as well have a 5 m ethernet cable because that's how close you might have to be unless your router has you locked with beamforming and you don't move.

Your Wifi adapter won't choose the fastest channel unless you dropout or manual switch 2.4 to 5 when close by.