Set appropriate wifi txpower for openwrt

[sarcasticadmin]

Description

For 23x whats the appropriate power for the 2.4ghz and 5ghz?

Without setting country code:

~# iwinfo phy0-ap0 txpowerlist
   0 dbm (   1 mW)
   1 dbm (   1 mW)
   2 dbm (   1 mW)
   3 dbm (   1 mW)
   4 dbm (   2 mW)
   5 dbm (   3 mW)
   6 dbm (   3 mW)
   7 dbm (   5 mW)
   8 dbm (   6 mW)
   9 dbm (   7 mW)
  10 dbm (  10 mW)
  11 dbm (  12 mW)
  12 dbm (  15 mW)
  13 dbm (  19 mW)
  14 dbm (  25 mW)
  15 dbm (  31 mW)
  16 dbm (  39 mW)
  17 dbm (  50 mW)
  18 dbm (  63 mW)
  19 dbm (  79 mW)
* 20 dbm ( 100 mW)

After setting option country 'US' in /etc/config/network:

root@OpenWrt:/etc/config# iwinfo phy1-ap1 info
phy1-ap1 ESSID: "cosmonaut_guest5"
         Access Point: 20:05:b6:ff:03:4a
         Mode: Master  Channel: 36 (5.180 GHz)  HT Mode: HE80
         Center Channel 1: 42 2: unknown
         Tx-Power: 23 dBm  Link Quality: 59/70
         Signal: -51 dBm  Noise: -91 dBm
         Bit Rate: 6.0 MBit/s
         Encryption: WPA-PSK (CCMP)
         Type: nl80211  HW Mode(s): 802.11ax/ac/n
         Hardware: embedded [MediaTek MT7981]
         TX power offset: none
         Channel offset: none
         Supports VAPs: yes  PHY name: phy1

root@OpenWrt:/etc/config# iwinfo phy1-ap1 txpowerlist
   0 dbm (   1 mW)
   1 dbm (   1 mW)
   2 dbm (   1 mW)
   3 dbm (   1 mW)
   4 dbm (   2 mW)
   5 dbm (   3 mW)
   6 dbm (   3 mW)
   7 dbm (   5 mW)
   8 dbm (   6 mW)
   9 dbm (   7 mW)
  10 dbm (  10 mW)
  11 dbm (  12 mW)
  12 dbm (  15 mW)
  13 dbm (  19 mW)
  14 dbm (  25 mW)
  15 dbm (  31 mW)
  16 dbm (  39 mW)
  17 dbm (  50 mW)
  18 dbm (  63 mW)
  19 dbm (  79 mW)
  20 dbm ( 100 mW)
  21 dbm ( 125 mW)
  22 dbm ( 158 mW)
* 23 dbm ( 199 mW)

ref: openwrt/openwrt#17489

AC

• wifi works better

[davidelang]

@sarcasticadmin

we want to run as low a power output as we can get away with to allow us to re-use the same channel at a closer physical distance. This is also why we deploy low and in the middle of the seating, so that the signal gets attenuated by the bodies around it (in cell phones, this is what they are doing when they talk about micro-cells)

It doesn't do any good for the AP to transmit at higher power levels than the client.

doing a quick search, I see that many mobile devices transmit around 15db (~30mw) and will reduce power if they are able to maintain a good connection while doing so.

I found this set of recommendations:

due to the number of APs that we have deployed, we should not need to operate through walls, so the suggestion of 6-9db for 2.4GHz and the open space suggestion of 12-15db for 5GHz sounds reasonable to me.

We currently have them both set for 8db, and scanning that Peter did in the expo floor a year or so ago does indicate that is at the low end of a good signal/noise ratio, so going to 15db for 5GHz seems reasonable, but I would stay where we are for 2.4GHz

I have wanted us to scan during the show to see signal strength with everyone in the rooms for several years, but we have not ended up doing much of that.

[sarcasticadmin]

I found this set of recommendations:

Source?

I have wanted us to scan during the show to see signal strength with everyone in the rooms for several years, but we have not ended up doing much of that.

Please identify what tooling we can use #1038 and well add this to our schedule of things to do as we setup for 23x

[davidelang]

https://airheads.hpe.com/discussion/transmit-power-recommendation

[MrHamel]

The classroom numbers are actually very good defaults to use, coming from the days I had to support co-working office buildings.

[MrHamel]

If possible, can we set 2.4Ghz to 10dbm, and 5Ghz to 15dbm? This is ideal for the APs being on the floor during the show, where attendees will attenuate the signal. When the rooms are unoccupied during the week and overnight, TX will be too much, and surveying at that time will not reflect reality.

In a perfect world, we'd get these APs off the floor by way of tripods or something of the sort. That would allow us to turn down the TX by almost half during the day.

[davidelang]

Ryan Hamel wrote:

MrHamel left a comment (socallinuxexpo/scale-network#1037) If possible, can we set 2.4Ghz to 10dbm, and 5Ghz to 15dbm?

I don't object to this, I don't think we need 5GHz quite that high, but there are enough channels avaialble that it should be ok.

This is ideal for the APs being on the floor during the show, where attendees will attenuate the signal. When the rooms are unoccupied during the week and overnight, TX will be too much, and surveying at that time will not reflect reality. In a perfect world, we'd get these APs off the floor by way of tripods or something of the sort. That would allow us to turn down the TX by almost half during the day.

I disagree with getting the APs up on tripods, part of the purpose of having them on the ground is so that the attendees attenuate the signal. We want the footprint of each AP to be as small as possible so that the frequency can be re-used sooner. The limiting factor is the airtime available, so more APs, each with a smaller footprint, allows for more channel re-use and therefor more airtime available across the show. it does mean that when there are fewer people, there is more contention between APs. but that's also when there's less demand. David Lang

[davidelang]

The user density of a classroom/trade show means that the optimum network design is going to be very different than when you have a much smaller number of users like in a home or even an office. In thos latter cases, the smaller number of users lets you optimize for individual users bandwidth and have APs cover a wider area vs a classroom/trade show where you have to go for best overall airtime utilization, at the cost of individual AP coverage and even the bandwidth available to any one individual. As far as AP coverage goes, this is like the cell phone network setting up microcells, low power, small footprint towers that cover a small area so that the same channels can be used multiple times in the same area that had been served by a single tower years before. In our densest classrooms, this is still not enough when everyone tries to download something big at the same time David Lang

[davidelang]

If I were equipping a site permanently for huge numbers of users (Say a stadium, school, or conference center), I would mount my APs in the ceiling, with very directional antennas pointed down to each cover a small footprint (possibly multiple radios per AP, but each with separate sets of antennas covering different areas) But we can't do that for temporary setups like scale and doing the antenna aiming is very labor intensive, so more APs at low level where they will be attenuated by the users is an 'almost as good and FAR esier' approach. David Lang

[sarcasticadmin]

Output from netgear wndr3800 (as an example):

root@ballh-2:~# iwinfo
phy0-ap0  ESSID: "scale-public-slow"
          Access Point: E0:46:9A:5A:C0:0B
          Mode: Master  Channel: 6 (2.437 GHz)  HT Mode: HT20
          Center Channel 1: 6 2: unknown
          Tx-Power: 12 dBm  Link Quality: unknown/70
          Signal: unknown  Noise: -86 dBm
          Bit Rate: unknown
          Encryption: WPA2 PSK (CCMP)
          Type: nl80211  HW Mode(s): 802.11b/g/n
          Hardware: 168C:0029 168C:A095 [Atheros AR9223]
          TX power offset: none
          Frequency offset: none
          Supports VAPs: yes  PHY name: phy0

phy1-ap0  ESSID: "scale-public-fast"
          Access Point: E0:46:9A:5A:C0:09
          Mode: Master  Channel: 153 (5.765 GHz)  HT Mode: HT20
          Center Channel 1: 153 2: unknown
          Tx-Power: 12 dBm  Link Quality: unknown/70
          Signal: unknown  Noise: -95 dBm
          Bit Rate: unknown
          Encryption: WPA2 PSK (CCMP)
          Type: nl80211  HW Mode(s): 802.11a/n
          Hardware: 168C:0029 168C:A094 [Atheros AR9220]
          TX power offset: none
          Frequency offset: none
          Supports VAPs: yes  PHY name: phy1
root@ballh-2:~# iwinfo phy0-ap0 txpowerlist
   0 dBm (   1 mW)
   1 dBm (   1 mW)
   2 dBm (   1 mW)
   3 dBm (   1 mW)
   4 dBm (   2 mW)
   5 dBm (   3 mW)
   6 dBm (   3 mW)
   7 dBm (   5 mW)
   8 dBm (   6 mW)
   9 dBm (   7 mW)
  10 dBm (  10 mW)
  11 dBm (  12 mW)
* 12 dBm (  15 mW)
  13 dBm (  19 mW)
  14 dBm (  25 mW)
  15 dBm (  31 mW)
  16 dBm (  39 mW)
  17 dBm (  50 mW)
  18 dBm (  63 mW)
  19 dBm (  79 mW)
  20 dBm ( 100 mW)
  21 dBm ( 125 mW)
  22 dBm ( 158 mW)
  23 dBm ( 199 mW)
  24 dBm ( 251 mW)
  25 dBm ( 316 mW)
  26 dBm ( 398 mW)
  27 dBm ( 501 mW)
  28 dBm ( 630 mW)
  29 dBm ( 794 mW)


root@ballh-2:~# iwinfo phy1-ap0 txpowerlist
   0 dBm (   1 mW)
   1 dBm (   1 mW)
   2 dBm (   1 mW)
   3 dBm (   1 mW)
   4 dBm (   2 mW)
   5 dBm (   3 mW)
   6 dBm (   3 mW)
   7 dBm (   5 mW)
   8 dBm (   6 mW)
   9 dBm (   7 mW)
  10 dBm (  10 mW)
  11 dBm (  12 mW)
* 12 dBm (  15 mW)
  13 dBm (  19 mW)
  14 dBm (  25 mW)
  15 dBm (  31 mW)
  16 dBm (  39 mW)
  17 dBm (  50 mW)
  18 dBm (  63 mW)
  19 dBm (  79 mW)
  20 dBm ( 100 mW)
  21 dBm ( 125 mW)
  22 dBm ( 158 mW)
  23 dBm ( 199 mW)
  24 dBm ( 251 mW)