Is Your Router's Transmit Power Juiced?

[sfx2000]

Interesting article...

Couple of things to consider however...

1) Conducted Path Measurements - you can measure each RF chain, but you need to consider that the impendance match for conducted path is 50 ohm, typical free space matching for EIRP is 37 ohm, so the numbers will change.

2) EIRP is a magical number - what is really important is TRP, and this can only be done in a RF chamber - Rohde&Schwarz has a mini chamber that works well (the DST-200), or one can go for the big-dog, the TS8991

Wireless OEM's spend big money on these items and similar, for example the "StarLab" SATIMO chamber -- which is a gold standard in the RF world.

The FCC's rules here, along with the EU, Anatel, and other regulatory agencies - many follow the guidance of the FCC - but at the same time, there is interpretation of these rules... there are weasel words perhaps, and also loopholes to be exploited -- bending, rather than breaking a given rule.

Most OEM's do have a team that is specifically focused on the Regulatory aspects of the products they develop.

I'm not saying that Asus is in the right or wrong, but this article is only one aspect. This is a rat-hole that perhaps SNB doesn't want to get in to - considering the Netgear-Asus litigation - now that the article is published, can't really un-ring the bell there...

sfx

 

[RMHC]

If I'm understanding this correctly, the FCC just increased the allowable power in the U-NII-1 band: http://www.fcc.gov/document/5-ghz-u-nii-ro

 

[Razor512]

Why don't the router companies just give user access to the full range of transmit power that the device supports on all channels. It should be perfectly fine as long as they put a warning. (just like when they allow you to select different regions (some allowing channels 1-14 to be used)

I really don't see the reason behind the 17dBm limit on certain 5GHz channels. the range is already so short (signal is also easily blocked) that even if they allowed 100 watts, it likely would not bother anything else.

From a logical standpoint, the 1 watt limit also makes no sense. (go over 1 watt and the FCC goes crazy, but 2 AP's pumping out 1 watt each is totally fine (not to mention that you can place the AP in just about any location, so what is the limit trying to protect?).
I say they should increase the limit to around 1100 watts. (they have no problem when microwaves use it) .

 

[stevech]

Everyone knows that the FCC regulations (and peers internationally) are necessary to allow unlicensed transmitters to coexist. Share the spectrum.
It's not a human tissue safety issue.

If too many people use excessive radiated power (i.e., inclusive of antenna gain), we all suffer.

Those here old enough to remember the 27MHz Citizens Band in its height of popularity know of what I speak. It was originally strictly licensed - not unlicensed. So that the FCC had recourse if people violated its max power which was 5W. But it became polluted with people buying/using illegal power amps. And when the propagation conditions were just so, one fool could jam a channel for hundreds or thousands of miles. On a lesser scale, we have that risk with 40MHz channels in 2.4GHz and "AMP'd" WiFi and all that rot.

 

[RMerlin]

Why don't the router companies just give user access to the full range of transmit power that the device supports on all channels. It should be perfectly fine as long as they put a warning. (just like when they allow you to select different regions (some allowing channels 1-14 to be used)

For the same reason cars cannot go in whichever direction they want, in whichever lane they want, while on a highway. If you don't organize things and establish some rules, then people will start crashing into one another.

Those restrictions aren't there just to piss you off. If your country needs part of the spectrum to operate their radars, you sure as hell don't want planes to start crashing into one another because somebody decided he needed to use the same channel at a crazy output power just so he could stream Youtube across the street.

 

[RMerlin]

I'm not saying that Asus is in the right or wrong, but this article is only one aspect. This is a rat-hole that perhaps SNB doesn't want to get in to - considering the Netgear-Asus litigation - now that the article is published, can't really un-ring the bell there...

What puzzles me there is that Netgear themselves said that Asus has since resolved (or tried to cover up) the issue through a firmware update, yet Tim's tests were done with the latest firmware release. Something doesn't add up here.

 

[thiggins]

sfx: Are you saying that the measurement technique and comparison was invalid?

 

[thiggins]

What puzzles me there is that Netgear themselves said that Asus has since resolved (or tried to cover up) the issue through a firmware update, yet Tim's tests were done with the latest firmware release. Something doesn't add up here.

Where did NETGEAR say that?

 

[thiggins]

If I'm understanding this correctly, the FCC just increased the allowable power in the U-NII-1 band: http://www.fcc.gov/document/5-ghz-u-nii-ro

I think this is the section?

Power

96.
Section 15.247 allows 1 Watt of total peak conducted power whereas Section 15.407
limits maximum conducted output power to the lesser of 1 Watt or 17 dBm + 10 log B (where B is
bandwidth in MHz).144 In addition to the 1 Watt power limit, there are different PSD limits in Sections
15.247 and 15.407 such that 1 Watt of total power is available only when the 6-dB bandwidth is 500
kilohertz or more under Section 15.247 and when the 26-dB bandwidth is 20 megahertz or more under
Section 15.407.145 Because we are trying to accommodate digitally modulated devices that are currently
permitted under both rules, we proposed in the NPRM to remove the bandwidth dependent term (i.e.,
remove 17 + 10 log B) from Section 15.407, so that the power limit would be 1 Watt.146 In the NPRM,
the Commission did not expect that removing the variable power limit in Section 15.407 would increase
the potential for harmful interference, because under current rules manufacturers are able to certify
equipment that uses up to 1 Watt of power under Section 15.247.
97.
We modify our rules to remove the bandwidth-dependent term from Section 15.407(a)(3)
of our rules as proposed. As we initially suggested and the majority of commenters agree, utilizing the 1-
Watt power limit will not increase the potential for harmful interference because unlicensed devices are
already allowed to operate without the bandwidth-dependent term under Section 15.247.147

 

[georgedone]

Don't understand

I don't understand one thing: how can be factored the physical antenna here. If I buy a router with a tiny antenna but the transmit power right at the limit as dictated by FCC and I fit it instead with a huge antenna with a 15dBm gain, surely this will have an influence in the radiated power. I seem to remember my router (Buffalo WZR-HP-G450) which have 3 antenna with 6dBm gain was using less electrical transmit power to compensate for the antennas with higher than average gain (or at least that's what I understood from FCC certification documentation). Bottom line, a vendor which has more electrical power on the connector might not be violating FCC rules after all if they fit on their routers small, inefficient antennas... Or do they ?

 

[L&LD]

georgedone,

I think that the key is that 'radiated power' is the measurement, like HP at the wheel. Not what the amps (or engine) produces.

 

[RMerlin]

Where did NETGEAR say that?

Your own article

NETGEAR notified ASUS of its intention to sue in early July. Since then, ASUS has posted multiple new versions of firmware for the products named in the suit that NETGEAR says lower transmit power. NETGEAR has also filed a complaint with the FCC.

 

[thiggins]

I don't understand one thing: how can be factored the physical antenna here. If I buy a router with a tiny antenna but the transmit power right at the limit as dictated by FCC and I fit it instead with a huge antenna with a 15dBm gain, surely this will have an influence in the radiated power. I seem to remember my router (Buffalo WZR-HP-G450) which have 3 antenna with 6dBm gain was using less electrical transmit power to compensate for the antennas with higher than average gain (or at least that's what I understood from FCC certification documentation). Bottom line, a vendor which has more electrical power on the connector might not be violating FCC rules after all if they fit on their routers small, inefficient antennas... Or do they ?

Radiated power has its own, different spec limit.

 

[azazel1024]

Radiated power has its own, different spec limit.

Yup and in general, yes that would push it beyond the power limits.

If the transmitted power is at the max, then you can generally use something in the range of about a 6dBi antenna and you'll be at the max emitted power (EIRP). At least if you look at the FCC's tables and guidance, that is about what it works out to.

Beyond 6dBi, the larger the antenna, the lower the transmitted power must be, but the emmitted power (EIRP) can actually be higher, because the emissions are "tighter" and thus will interfere with fewer people.

In general for every 6dBi in antenna gain the transmit power must be reduced by 3dBm. Obviously a router manufacturer isn't going to do anything if you replace their antennas with bigger ones. However, rolling out of the factory, they should be doing that and also in theory as the end user, YOU should be doing that as well.

This DOES mean that in terms of EIRP, a router with 6dBi antennas can have significantly higher actual signal strength than one with 2dBi antennas as both would have the same transmitted radio power cap. Of course one using 12dBi antennas would still have a higher RSS than one with 6dBi if both used their different transmit power caps (the 12dBi one just being 3dBm lower transmit power than the 6dBi one).

Hopefully that makes some sense to you.

The FCC is in large part concerned with "fair neighbor" usage of unlicensed bands, which is good and valid. Me personally, I have no neighbors close enough to be impacted by my network (a hundred feet plus the walls of our houses seperate my closest neighbor and I don't know that she has a wifi network. My next closet neighbor is 200ft away and also has trees between me and them). So I am not terribly worried about EIRPs and that kind of stuff. 9dBi antennas without reducing transmit power? Fair game. Heck I'd go with 12dBis on my outdoor AP if it didn't look so darned ridiculous sticking down from the edge of my roof.

 

[thiggins]

Your own article

That's what I thought you might be referring to.

ASUS may very well have reduced some transmit levels. But looks like they missed a few.

 

[thiggins]

If I'm understanding this correctly, the FCC just increased the allowable power in the U-NII-1 band: http://www.fcc.gov/document/5-ghz-u-nii-ro

Looks like it is at least in progress. Trying to find out when the change becomes official.
http://www.revolutionwifi.net/2014/03/fcc-eases-restrictions-on-5ghz-unii-1.html

 

[RMHC]

Looks like it is at least in progress. Trying to find out when the change becomes official.
http://www.revolutionwifi.net/2014/03/fcc-eases-restrictions-on-5ghz-unii-1.html

It became effective June 2: https://www.federalregister.gov/art...nfrastructure-u-nii-devices-in-the-5-ghz-band

Referring back to the report and order linked to previously, the commission discusses the changes at various points, including the part you quoted earlier in this thread, but the actual change affecting channels 36-48 is to section 15.407 of the Part 15 rules (it starts on p. 49 of the PDF or Word version of the R&O, which is easier to read than the plain text or the HTML).

The indoor restriction is lifted and the power limits are raised to be the same as UNII-3.

 

[thiggins]

It became effective June 2: https://www.federalregister.gov/art...nfrastructure-u-nii-devices-in-the-5-ghz-band

Referring back to the report and order linked to previously, the commission discusses the changes at various points, including the part you quoted earlier in this thread, but the actual change affecting channels 36-48 is to section 15.407 of the Part 15 rules (it starts on p. 49 of the PDF or Word version of the R&O, which is easier to read than the plain text or the HTML).

The indoor restriction is lifted and the power limits are raised to be the same as UNII-3.

Thank you very much. I'll amend the article and post news.

 

[htismaqe]

The FCC is in large part concerned with "fair neighbor" usage of unlicensed bands, which is good and valid. Me personally, I have no neighbors close enough to be impacted by my network (a hundred feet plus the walls of our houses seperate my closest neighbor and I don't know that she has a wifi network. My next closet neighbor is 200ft away and also has trees between me and them). So I am not terribly worried about EIRPs and that kind of stuff. 9dBi antennas without reducing transmit power? Fair game. Heck I'd go with 12dBis on my outdoor AP if it didn't look so darned ridiculous sticking down from the edge of my roof.

My closest neighbor is over a mile away.

 

[azazel1024]

KHAN!!!!!!!!

Sigh, if only.