How to increase 5GHz wireless range for RT-AC68U

[Sir Patriot]

Hi
I am using RT-AC68U in a house that farthest place to the router is about 15 meter away but with 3 walls between it and my cellphone. I am getting weak signal and sometimes it disconnects from 5GHz band but 2.4GHz is working fine with full bars. I know that 5GHz is weaker compare to 2.4GHz to pass the walls but I think 15 meter and 3 walls should not cause that much problem.
Is there any way to make its signal stronger?

 

[L&LD]

Mobile devices like cell phones do not have the best antennae nor the best wireless performance period for battery life concerns. I would not judge the quality of a wireless network with a mobile device such as you're doing. And yes, depending on the construction of the walls and the 15 metres distance, that is enough to cause the problems you are seeing on a sub-par wireless client.

If you can't physically move the router to a more central location or make sure it is at least 3 metres above ground level or more with no obstructions within a metre all around it, then consider using either john9527's firmware or hggomes fork (both based on RMerlin's firmware) instead.

After you have tested each channel and channel width possible to find the best balance of throughput and range.

john9527's firmware relies on older wireless drivers while hggomes offers higher power.

Both will require a reset to factory defaults and a minimal and manual (do not use a cfg backup file to restore your settings) to configure the router to secure it and connect to your ISP.

Create new ssid's for both bands to completely take advantage of the new capabilities the new firmware offers.

 

[Sir Patriot]

There is not any options with RMerlin's firmware to increase antennas power?

 

[bilboSNB]

I am in almost the same situation with my ac68u and Samsung note 10.1, all the different firmwares made hardly any difference, however my laptop n 5G only in the same location absolutely flys.

 

[L&LD]

There is not any options with RMerlin's firmware to increase antennas power?

No, you can decrease it from 80mW. But you cannot increase it.

hggomes firmware offers up to 708mW of power, but it is up to you to use it up to the legal limit in your jurisdiction.

 

[dlandiss]

There is not any options with RMerlin's firmware to increase antennas power?

The challenge with increasing the router's transmitter power is that it does not improve reception. The limit is more likely the power of the mobile device. Better router antennas improve the link in both directions.

 

[kvic]

TL;DR

"Forms of the reciprocity theorems are used in many electromagnetic applications, such as analyzing electrical networks and antenna systems. For example, reciprocity implies that antennas work equally well as transmitters or receivers, and specifically that an antenna's radiation and receiving patterns are identical."

 

[L&LD]

The challenge with increasing the router's transmitter power is that it does not improve reception. The limit is more likely the power of the mobile device. Better router antennas improve the link in both directions.

That is not true of clients that are within a reasonable distance of the router. It may be true when they are very, very far away.

What higher transmitting power offers is the chance for the client to lock onto the signal and for the router to then listen for that client going forward. So in that sense, it does improve reception.

We've already discussed this in these forums (sxf2000?) and more power on just the router equals better reception both ways.

 

[dlandiss]

That is not true of clients that are within a reasonable distance of the router

I guess that depends on what you mean by "a reasonable distance." I live in a 113-year-old house with thick walls and am used to having to implement good connections (ie, good S/N ratio) in both directions.

Physics is physics. Adding power to the router alone cannot improve its reception of the client.

 

[L&LD]

I guess that depends on what you mean by "a reasonable distance." I live in a 113-year-old house with thick walls and am used to having to implement good connections (ie, good S/N ratio) in both directions.

Physics is physics. Adding power to the router alone cannot improve its reception of the client.

No, it can't. But it can make a much more stable connection in the first place which effectively does the same.

A router depends on a client to connect, not the other way around. If the router signal is too low, the client won't. And it looks like the router can't 'hear' it. But if the signal is high enough, the router then pays attention and more times than not, it can hear the client well enough to make a difference in the throughput and range that the client can communicate with the router in, where it otherwise would not.

 

[dlandiss]

But if the signal is high enough, the router then pays attention and more times than not, it can hear the client well enough to make a difference in the throughput and range that the client can communicate with the router in, where it otherwise would not.

I respectfully disagree. This is not like straight voice communications where a radio operator can "listen through the noise" to get the gist of what the other is saying. Both the range and throughput of modern quadrature modulation methods depend heavily on signal-to-noise ratio, and if the S/N is too low in either direction the connection will be unreliable or simply be broken.

 

[L&LD]

I respectfully disagree. This is not like straight voice communications where a radio operator can "listen through the noise" to get the gist of what the other is saying. Both the range and throughput of modern quadrature modulation methods depend heavily on signal-to-noise ratio, and if the S/N is too low in either direction the connection will be unreliable or simply be broken.

You may disagree. But it is exactly like your example. Until you have a listeners attention, they will ignore you. A stronger signal gets attention.

 

[dlandiss]

You may disagree. But it is exactly like your example. Until you have a listeners attention, they will ignore you. A stronger signal gets attention.

Could you refer me to a reference that discusses this effect as it applies to the 802.11n or ac protocols so I can learn more about it? Thank you!

 

[L&LD]

Could you refer me to a reference that discusses this effect as it applies to the 802.11n or ac protocols so I can learn more about it? Thank you!

Sorry, no references except experience. The theory as you state it, I agree with.

How it applies to the world in general? Well, that is another matter and that is what we're discussing here.

 

[sfx2000]

We've already discussed this in these forums (sxf2000?) and more power on just the router equals better reception both ways.

Oh, jeez, someone invoked me - yikes!



Well, I've always been in favor of moving the AP to where the clients are... WiFi is like real estate, location matters..

Throwing more power at the chipset level doesn't help, as that is only in one direction - and many chipsets these days, they have a fixed level, so going into a 3rd Party SW and turning the volume knob to 11 doesn't do anything. In any event, turning up power is one way - transmit - it doesn't cause a reciprocal improvement on the client side when trying to close the link...

Higher Gain Antennas - can help, but it's borrowing from Peter to pay Paul - the higher gain antennas will change the beam patterns - like SteveCH is so fond of saying (correctly I must add), higher gain antennae squish the RF donut - get more peak power, but in a narrower pattern, and anything not in the pattern loses out...

What I do suggest is on an AP level, going to a three stream AP (like AC1750/AC1900 class) will provide improvement due to MIMO and coding gain/Tx-RX diversity gain, even with single stream clients that are 11n/11ac

 

[sfx2000]

I respectfully disagree. This is not like straight voice communications where a radio operator can "listen through the noise" to get the gist of what the other is saying. Both the range and throughput of modern quadrature modulation methods depend heavily on signal-to-noise ratio, and if the S/N is too low in either direction the connection will be unreliable or simply be broken.

I see where you're coming from - but 11n and 11ac do offer additional EC capabilities and gains thru MIMO that cut thru the noise to some extent...

WiFi, like any other radio, ultimately is noise limited - that's why we look at things like CINR vs. RSSI - CINR being the correlated signal over noise (which is different that basic SNR), and RSSI is just overall energy in the band - some look at it on the front end, some look at it on the baseband - depends on implementation...

I always look at RSSI as total energy, and CINR as being the good signal...

 

[kvic]

Higher Gain Antennas - can help, but it's borrowing from Peter to pay Paul - the higher gain antennas will change the beam patterns

If a user roughly know how the beam patterns change, and make right use of it. That's sound engineering..

 

[joegreat]

There is not any options with RMerlin's firmware to increase antennas power?

The simplest solution can be found in my footer...
Just put a 2nd router on a good place and you have (as I do) full coverage of your home with 2.4 and 5 GHz signals!

With kind regards
Joe

 

[dlandiss]

Sorry, no references except experience.

No references usually means that nobody has been able to duplicate your observations. It is risky to anthropomorphize an integrated circuit or communications protocol.

I am not aware of any scenario in 802.11 fitting "What higher transmitting power offers is the chance for the client to lock onto the signal and for the router to then listen for that client going forward."

As far as I understand it router broadcasts its SSID about 10 times a second. If a client wants to log in, it must respond with a password and its protocol capabilities. The router must then respond to the client that it is logged in and a two-way connection exists.

If that initial handshake does not occur, there is no connection. If the S/N ratio of the client side of that handshake, as received by the router, is too low; the handshake never completes.

Listen to sfx2000 -- he understands the server/client relationship and handshaking better than I do.

 

[dlandiss]

I see where you're coming from - but 11n and 11ac do offer additional EC capabilities and gains thru MIMO that cut thru the noise to some extent...

WiFi, like any other radio, ultimately is noise limited - that's why we look at things like CINR vs. RSSI - CINR being the correlated signal over noise (which is different that basic SNR), and RSSI is just overall energy in the band - some look at it on the front end, some look at it on the baseband - depends on implementation...

I always look at RSSI as total energy, and CINR as being the good signal...

No disagreement here. RSSI is what is usually reported by the hardware and displayed by WiFi scanners. Metageek has a Wi-Spy dongle that shows the actual energy as a function of frequency in spectrum analyzer fashion. We can assume that the energy shown by the Wi-Spy is interference + noise, and thus get some idea of S/N ratio by comparing RSSI to the I + N. Here is some non-WiFi interference clobbering channel 4: