Band steering - "priority" to 5Ghz

Reading comprehension…

My main problem with MIKROTIK WIRELESS is the “giving up in crowded environments”.

I gave in to the idea… “Well if you have S–t internet service… It’s not gonna be as apparent.” That’s how I tried to justify the lousy throughput once you have more than like 5 devices.

THE PROBLEM IS CONNECTIVITY!!! I had the same problem again and again. The 2.4 radio would show all sorts of disconnect messages. And of course things didn’t work. But EVEN WORSE would be clients that “appeared fine”. Caps-man would show the devices as connected with good numbers. But things don’t actually pass.

The cure… You have to restart the radio.

How do you detect that device are no longer working??? You have to find something else connected to that radio to net watch. (Like a printer) When it drops… Cycle that radio. Then magically… That OTHER device, that “looked fine” starts passing traffic again.

With more and more devices relying on wifi… Basing the network on Mikrotik wireless… Doesn’t make any sense.

We have many devices not capable of using DFS channels. Try disabling them in your router and see if your devices come up in the 5ghgz band.

Good afternoon,

can please anybody advise why my 5Ghz interfaces always shows up 17dBm though I’m changing the Tx power value in Channels sections and I’m setting it to 20dBm or other
https://imgur.com/a/CjovlwZ

You might want to read this:
https://help.mikrotik.com/docs/display/ROS/Wireless+Interface#WirelessInterface-802.11nwirelesschipsetsrepresentpowerperchainandthe802.11ac

If you have 2 chains, 20dbm - 3dbm = 17dbm

Well the power settings on a MT are not simple with the later devices and higher RouterOS versions.

What I assume as what it does …

  • TX power, certainly in Europe, is limited by the regulatory domain max EIRP (20dBm for 2.4GHz, 23dBm and 20 dBm for 5GHz indoor only freq, 27 dBm for 5GHz also outdoor freq)
    See CLI: " interface wireless info country-info etsi" or replace etsi with your country
  • TX power is limited at the higher MCS encoding because of the needed signal quality (see data sheets)
  • TX power used to be given as information in “Current TX power”, and power could be set per MCS. Now only “all rate fixed” can be used. And no feedback is given.

The only line of information visible in RouterOS GUI is in the “status”. All possible values of this can be seen with: “/interface wireless info allowed-channels”
It is the calculation of the EIRP limit, reduced with the antenna gain filled in or set by default. In text below wAP has “antenna gain=2” set
As far as I verified there is no reduction for the multiple antenna in the status info, not even for 2.4GHz “n” where it is set “per antenna”, versus “ac” where it is total (see wiki)

The EIRP table values get sometimes updated with ROS releases

[admin@Omnitik] > interface wireless info country-info etsi
  ranges: 2402-2482/b,g,gn20,gn40(20dBm)
          5170-5250/a,an20,an40,ac20,ac40,ac80,ac160,ac80+80(23dBm)/passive,indoor
          5170-5330/a,an20,an40,ac20,ac40,ac80,ac160,ac80+80(20dBm)/dfs,passive,indoor
          5250-5330/a,an20,an40,ac20,ac40,ac80,ac160,ac80+80(20dBm)/dfs,passive,indoor
          5490-5710/a,an20,an40,ac20,ac40,ac80,ac160,ac80+80(27dBm)/dfs,passive
          5190-5310/a-turbo(20dBm)/dfs
          5180-5300/a-turbo(20dBm)/dfs
          5520-5680/a-turbo(27dBm)/dfs,passive
          5510-5670/a-turbo(27dBm)/dfs,passive
          902-927/b,g,g-turbo,gn20,gn40(30dBm)


[admin@MktwAPac] /interface wireless info> allowed-channels 
interface: wlan2
  channels: 5180/20-Ce/ac/P(21dBm),5185/20-Ce/ac/P(21dBm),5190/20-Ce/ac/P(21dBm),5195/20-Ce/ac/P(21dBm),
            5200/20-Ce/ac/P(21dBm),5205/20-Ce/ac/P(21dBm),5210/20-Ce/ac/P(21dBm),5215/20-Ce/ac/P(21dBm),
            5220/20-Ce/ac/P(21dBm),5180/20-Ce/ac/DP(18dBm),5185/20-Ce/ac/DP(18dBm),5190/20-Ce/ac/DP(18dBm),
            5195/20-Ce/ac/DP(18dBm),5200/20-Ce/ac/DP(18dBm),5205/20-Ce/ac/DP(18dBm),5210/20-Ce/ac/DP(18dBm),
            5215/20-Ce/ac/DP(18dBm),5220/20-Ce/ac/DP(18dBm),5225/20-Ce/ac/DP(18dBm),5230/20-Ce/ac/DP(18dBm),
            5235/20-Ce/ac/DP(18dBm),5240/20-Ce/ac/DP(18dBm),5245/20-Ce/ac/DP(18dBm),5250/20-Ce/ac/DP(18dBm),
            5255/20-Ce/ac/DP(18dBm),5260/20-Ce/ac/DP(18dBm),5265/20-Ce/ac/DP(18dBm),5270/20-Ce/ac/DP(18dBm),
            5275/20-Ce/ac/DP(18dBm),5280/20-Ce/ac/DP(18dBm),5285/20-Ce/ac/DP(18dBm),5290/20-Ce/ac/DP(18dBm),
            5295/20-Ce/ac/DP(18dBm),5300/20-Ce/ac/DP(18dBm),5260/20-Ce/ac/DP(18dBm),5265/20-Ce/ac/DP(18dBm),
            5270/20-Ce/ac/DP(18dBm),5275/20-Ce/ac/DP(18dBm),5280/20-Ce/ac/DP(18dBm),5285/20-Ce/ac/DP(18dBm),
            5290/20-Ce/ac/DP(18dBm),5295/20-Ce/ac/DP(18dBm),5300/20-Ce/ac/DP(18dBm),5500/20-Ce/ac/DP(25dBm),
            5505/20-Ce/ac/DP(25dBm),5510/20-Ce/ac/DP(25dBm),5515/20-Ce/ac/DP(25dBm),5520/20-Ce/ac/DP(25dBm),
            5525/20-Ce/ac/DP(25dBm),5530/20-Ce/ac/DP(25dBm),5535/20-Ce/ac/DP(25dBm),5540/20-Ce/ac/DP(25dBm),
            5545/20-Ce/ac/DP(25dBm),5550/20-Ce/ac/DP(25dBm),5555/20-Ce/ac/DP(25dBm),5560/20-Ce/ac/DP(25dBm),
            5565/20-Ce/ac/DP(25dBm),5570/20-Ce/ac/DP(25dBm),5575/20-Ce/ac/DP(25dBm),5580/20-Ce/ac/DP(25dBm),
            5585/20-Ce/ac/DP(25dBm),5590/20-Ce/ac/DP(25dBm),5595/20-Ce/ac/DP(25dBm),5600/20-Ce/ac/DP(25dBm),
            5605/20-Ce/ac/DP(25dBm),5610/20-Ce/ac/DP(25dBm),5615/20-Ce/ac/DP(25dBm),5620/20-Ce/ac/DP(25dBm),
            5625/20-Ce/ac/DP(25dBm),5630/20-Ce/ac/DP(25dBm),5635/20-Ce/ac/DP(25dBm),5640/20-Ce/ac/DP(25dBm),
            5645/20-Ce/ac/DP(25dBm),5650/20-Ce/ac/DP(25dBm),5655/20-Ce/ac/DP(25dBm),5660/20-Ce/ac/DP(25dBm),
            5665/20-Ce/ac/DP(25dBm),5670/20-Ce/ac/DP(25dBm),5675/20-Ce/ac/DP(25dBm),5680/20-Ce/ac/DP(25dBm)

[admin@MktwAPac] /interface wireless info> allowed-channels 
interface: wlan1
  channels: 2412/20/gn(18dBm),2417/20/gn(18dBm),2422/20/gn(18dBm),2427/20/gn(18dBm),2432/20/gn(18dBm),
            2437/20/gn(18dBm),2442/20/gn(18dBm),2447/20/gn(18dBm),2452/20/gn(18dBm),2457/20/gn(18dBm),
            2462/20/gn(18dBm),2467/20/gn(18dBm),2472/20/gn(18dBm)

The manual TX-power “all rates fixed” filled in value, does not change this status line. So there you can not see what the TX-power used actually is. The value in the status line will be the maximum used ! But as the wiki says you should set the TX-power per antenna for all protocols except “ac” chipsets where you should give the total. “802.11n wireless chipsets represent power per chain and the 802.11ac wireless chipsets represent the total power”. This does not fit with the above data, which is always the total power, or not? The multiple antenna must be compensated for, so with 2 it is 3dBm lower per antenna. And with no feedback, what are we doing when we set manual TX-power to 17 , on “n” chipsets and what on “ac” chipsets? The max is ‘status_dBm - 3dBm’ ??

Well the power settings on a MT are not simple with the later devices and higher RouterOS versions.

What I assume as what it does …

  • TX power, certainly in Europe, is limited by the regulatory domain max EIRP (20dBm for 2.4GHz, 23dBm and 20 dBm for 5GHz indoor only freq, 27 dBm for 5GHz also outdoor freq)
    See CLI: " interface wireless info country-info etsi" or replace etsi with your country
  • TX power is limited at the higher MCS encoding because of the needed signal quality (see data sheets)
  • TX power used to be given as information in “Current TX power”, and power could be set per MCS. Now only “all rate fixed” can be used. And no feedback is given.

The only line of information visible in RouterOS GUI is in the “status”. All possible values of this can be seen with: “/interface wireless info allowed-channels”
It is the calculation of the EIRP limit, reduced with the antenna gain filled in or set by default. In text below wAP has “antenna gain=2” set
As far as I verified there is no reduction for the multiple antenna in the status info, not even for 2.4GHz “n” where it is set “per antenna”, versus “ac” where it is total (see wiki)

The EIRP table values get sometimes updated with ROS releases
Code: Select all

[admin@Omnitik] > interface wireless info country-info etsi
ranges: 2402-2482/b,g,gn20,gn40(20dBm)
5170-5250/a,an20,an40,ac20,ac40,ac80,ac160,ac80+80(23dBm)/passive,indoor
5170-5330/a,an20,an40,ac20,ac40,ac80,ac160,ac80+80(20dBm)/dfs,passive,indoor
5250-5330/a,an20,an40,ac20,ac40,ac80,ac160,ac80+80(20dBm)/dfs,passive,indoor
5490-5710/a,an20,an40,ac20,ac40,ac80,ac160,ac80+80(27dBm)/dfs,passive
5190-5310/a-turbo(20dBm)/dfs
5180-5300/a-turbo(20dBm)/dfs
5520-5680/a-turbo(27dBm)/dfs,passive
5510-5670/a-turbo(27dBm)/dfs,passive
902-927/b,g,g-turbo,gn20,gn40(30dBm)


[admin@MktwAPac] /interface wireless info> allowed-channels
interface: wlan2
channels: 5180/20-Ce/ac/P(21dBm),5185/20-Ce/ac/P(21dBm),5190/20-Ce/ac/P(21dBm),5195/20-Ce/ac/P(21dBm),
5200/20-Ce/ac/P(21dBm),5205/20-Ce/ac/P(21dBm),5210/20-Ce/ac/P(21dBm),5215/20-Ce/ac/P(21dBm),
5220/20-Ce/ac/P(21dBm),5180/20-Ce/ac/DP(18dBm),5185/20-Ce/ac/DP(18dBm),5190/20-Ce/ac/DP(18dBm),
5195/20-Ce/ac/DP(18dBm),5200/20-Ce/ac/DP(18dBm),5205/20-Ce/ac/DP(18dBm),5210/20-Ce/ac/DP(18dBm),
5215/20-Ce/ac/DP(18dBm),5220/20-Ce/ac/DP(18dBm),5225/20-Ce/ac/DP(18dBm),5230/20-Ce/ac/DP(18dBm),
5235/20-Ce/ac/DP(18dBm),5240/20-Ce/ac/DP(18dBm),5245/20-Ce/ac/DP(18dBm),5250/20-Ce/ac/DP(18dBm),
5255/20-Ce/ac/DP(18dBm),5260/20-Ce/ac/DP(18dBm),5265/20-Ce/ac/DP(18dBm),5270/20-Ce/ac/DP(18dBm),
5275/20-Ce/ac/DP(18dBm),5280/20-Ce/ac/DP(18dBm),5285/20-Ce/ac/DP(18dBm),5290/20-Ce/ac/DP(18dBm),
5295/20-Ce/ac/DP(18dBm),5300/20-Ce/ac/DP(18dBm),5260/20-Ce/ac/DP(18dBm),5265/20-Ce/ac/DP(18dBm),
5270/20-Ce/ac/DP(18dBm),5275/20-Ce/ac/DP(18dBm),5280/20-Ce/ac/DP(18dBm),5285/20-Ce/ac/DP(18dBm),
5290/20-Ce/ac/DP(18dBm),5295/20-Ce/ac/DP(18dBm),5300/20-Ce/ac/DP(18dBm),5500/20-Ce/ac/DP(25dBm),
5505/20-Ce/ac/DP(25dBm),5510/20-Ce/ac/DP(25dBm),5515/20-Ce/ac/DP(25dBm),5520/20-Ce/ac/DP(25dBm),
5525/20-Ce/ac/DP(25dBm),5530/20-Ce/ac/DP(25dBm),5535/20-Ce/ac/DP(25dBm),5540/20-Ce/ac/DP(25dBm),
5545/20-Ce/ac/DP(25dBm),5550/20-Ce/ac/DP(25dBm),5555/20-Ce/ac/DP(25dBm),5560/20-Ce/ac/DP(25dBm),
5565/20-Ce/ac/DP(25dBm),5570/20-Ce/ac/DP(25dBm),5575/20-Ce/ac/DP(25dBm),5580/20-Ce/ac/DP(25dBm),
5585/20-Ce/ac/DP(25dBm),5590/20-Ce/ac/DP(25dBm),5595/20-Ce/ac/DP(25dBm),5600/20-Ce/ac/DP(25dBm),
5605/20-Ce/ac/DP(25dBm),5610/20-Ce/ac/DP(25dBm),5615/20-Ce/ac/DP(25dBm),5620/20-Ce/ac/DP(25dBm),
5625/20-Ce/ac/DP(25dBm),5630/20-Ce/ac/DP(25dBm),5635/20-Ce/ac/DP(25dBm),5640/20-Ce/ac/DP(25dBm),
5645/20-Ce/ac/DP(25dBm),5650/20-Ce/ac/DP(25dBm),5655/20-Ce/ac/DP(25dBm),5660/20-Ce/ac/DP(25dBm),
5665/20-Ce/ac/DP(25dBm),5670/20-Ce/ac/DP(25dBm),5675/20-Ce/ac/DP(25dBm),5680/20-Ce/ac/DP(25dBm)

[admin@MktwAPac] /interface wireless info> allowed-channels
interface: wlan1
channels: 2412/20/gn(18dBm),2417/20/gn(18dBm),2422/20/gn(18dBm),2427/20/gn(18dBm),2432/20/gn(18dBm),
2437/20/gn(18dBm),2442/20/gn(18dBm),2447/20/gn(18dBm),2452/20/gn(18dBm),2457/20/gn(18dBm),
2462/20/gn(18dBm),2467/20/gn(18dBm),2472/20/gn(18dBm)


The manual TX-power “all rates fixed” filled in value, does not change this status line. So there you can not see what the TX-power used actually is. The value in the status line will be the maximum used ! But as the wiki says you should set the TX-power per antenna for all protocols except “ac” chipsets where you should give the total. “802.11n wireless chipsets represent power per chain and the 802.11ac wireless chipsets represent the total power”. This does not fit with the above data, which is always the total power, or not? The multiple antenna must be compensated for, so with 2 it is 3dBm lower per antenna. And with no feedback, what are we doing when we set manual TX-power to 17 , on “n” chipsets and what on “ac” chipsets? The max is ‘status_dBm - 3dBm’ ??

Thank you for your detailed answer. Based on that I can conclude that I won’t be able to increase the power. Then what are other ways to improve the situation that devices switch to 2.4GHz though 5GHz RSSI is quite acceptable?
http://prntscr.com/26thu4f

Thank you for your detailed answer. Based on that I can conclude that I won’t be able to increase the power. Then what are other ways to improve the situation that devices switch to 2.4GHz though 5GHz RSSI is quite acceptable?
http://prntscr.com/26thu4f

It seems like this is somehow implemented, although the details are obscure,

https://help.mikrotik.com/docs/display/ROS/WifiWave2#WifiWave2-Steeringproperties
the effect seems dubious to me as i am still having issues with 2.4G being the selected band most of the time.

My main problem with MIKROTIK WIRELESS is the “giving up in crowded environments”.

When Mikrotik (WLAN driver) is in competition with other brands ( or with MT on wifiwave2), then MT has a handicap of 3dB. That is if the EIRP level is the power limiting factor (EIRP is very low in Europe and is the limiting factor).

MT WLAN driver (has to? why?) follow the non-TPC permitted EIRP levels, which are 3 dB lower than the TPC permitted EIRP levels.
Other brand AP will be the preferred AP, because of the stronger signal.

ROS 7.12 beta

*) wifiwave2 - list APs with a higher maximum data rate as more preferable roaming candidates;

But how ? Used in 802.11k ?
Or will MT finally add QBSS fields in the beacons?
“The QBSS (QOS enhanced basic service set) information element is an 802.11e construct that enables an access point to communicate its channel usage”

Missing QBSS with MT. Yet another potential reason why the MT AP is/was not the preferred AP for wireless devices.

AsI understand it, the ‘preference’ mostly works by delaying response frames for the 2.4GHz channel, so that the client receives the 5G channel frames first.
I simple workaround that has worked well for me has been to set a longer beacon interval for 2G than 5G.
Additionally an access policy blocking 5G for low RSSI would help insure 2G is selected for low signal strength.

Band steering MT, yuck yuck… Is this the comedy forum ???

Just dont hand out passwords to 2.4 no connection.

Re:
RSSI of 2.4Ghz: about -50
RSSI of 5Ghz: about -70

Move your wireless client device much closer to your dual-band AP. Then see what you default connect to.

A -50 connection on 2.4 GHz might be a faster connection than a -70 connection on 5 GHz.

Normally , connection rates will shift up to faster rates. Each time they shift to a faster rate , the TX power will normally decrease and the RX receive signal strength will also decrease ( for both the client and the AP ). These connection rates will try to achieve the fastest rate possible based on the client and AP signal quality. However , if/when a link has problems ( including problems at faster connection rates after it has already connected ), it may disconnect and auto-reconnect to a different AP.

MikroTik gave us vendor drivers (wifi-qcom-ac) starting with ROS 7.13. No issues with roaming anymore. Clients choose 5ghz when relatively near AP, smoothly roam over to 2ghz when moving away from AP, smoothly roam back to 5ghz when moving near AP again. Works like a charm.

:heart: Thank you Mikrotik for making it happen. :heart:

Yes.. but… if only is on Same SSID. I tried to roam to another SSID using fixed steering group, but with no success.

I need to have 2 separates SSIDs

Can’t do that.
1 steering group per SSID.

Wifi roaming is only possible within the same SSID.

Steering in MK is junk, I have 15 access points in AX mode and FT enabled. Steering makes devices to roam every second, most of them were losing internet access and falling back to cellular, some were stuck at 5 Ghz without roaming back to 2.4 and eventually losing connection due to low signal.
“Sa Query timeout” where flooding the log.
It’s a chaos. I had to disable it.

In my case - Steering option fixed my problem, that devices would prefer sometimes 2.4ghz instead of 5ghz.
I have observed that Mikrotik is pushing “too much” for 5ghz and devices sometimes simply just disconnect instead of connecting to 2.4ghz. However this might due to super crowded 2.4ghz band around me.

In short - if 5ghz drops for some reason shortly or device preferred 2.4 instead of 5ghz, now it’s working as supposed to and clients prefer 5ghz as soon as it’s there.
Mikrotik is overly complicated for non professionals. I think ubiquity, for instance is more plug and play and that’s why if we are not professional enough assume that other manufacturers are better. They are not. IMHO.