Higher Power Now Allowed For 5 GHz Wi-Fi

[yetollerey]

FCC Increases 5GHz Spectrum for Wi-Fi
http://www.fcc.gov/document/fcc-increases-5ghz-spectrum-wi-fi-other-unlicensed-uses

Higher Power Now Allowed For 5 GHz Wi-Fi
http://www.smallnetbuilder.com/wireless/wireless-news/32435-higher-power-now-allowed-for-5-ghz-wi-fi

Does it mean that new generation of routers are near? maybe some of the present routers will be able to apply the new spectrum via firmware upgrade?

 

[azazel1024]

Based on the changes, I'd expect at least a very small handful of routers to have new firmware and possibly some client end driver changes to.

Its an open ended question of how much of this is controlled in the router firmware (upgradable) or in the radio firmware (generally not upgradable). I know of at least a couple of routers where it is strictly on the radio firmware setting hard limits, so you wouldn't be able to update it.

As for new stuff, I'd be suprised if router manufacturers weren't taking it in to account now. Heck, might even be taken in to account with new revs of existing products.

 

[sfx2000]

Not to hijack this thread -- there's been a fair amount of discussion on this very topic here... lots of good info there

http://forums.smallnetbuilder.com/showthread.php?t=17710

sfx

 

[thiggins]

FCC Recert needed

The problem isn't where the limits are baked in. It's that any router wanting to implement this change must be resubmitted to FCC for re-certification.

This ain't cheap and there is more $ to be made selling new stuff. So don't expect many (any?) routers to bump up the power.

Another issue that has been pointed out to me is that the new rule did not relax the restricted band requirement. So unless devices have been specifically designed to support higher transmit power levels at those frequencies, they may not be able to control spurious emissions to meet the restricted band requirements at the full 30 dBm.

Bottom line: expect to see higher power in Channels 36 - 48 in new producrs and no upgrades for existing APs / routers.

 

[azazel1024]

I didn't even think about the recert requirements.

 

[speedingcheetah]

So....the higher channels are left untouched then?

The lower channels can now use the same power as the higher?...

 

[thiggins]

So....the higher channels are left untouched then?

The lower channels can now use the same power as the higher?...

That is correct.

 

[sm00thpapa]

If the lower channels can run at the same speeds as the upper channels what's the deal with the AC3200 routers than? Both 5GHz bands can run at the same speed.

 

[thiggins]

If the lower channels can run at the same speeds as the upper channels what's the deal with the AC3200 routers than? Both 5GHz bands can run at the same speed.

The FCC rule changes power level on the lower band. Supported MCS / link rates remain the same.

The change is about power, not "speed".

 

[azazel1024]

The FCC rule changes power level on the lower band. Supported MCS / link rates remain the same.

The change is about power, not "speed".

Yes, but with the higher power levels it means that at the same distance, other than odd path propogation issues, you'd have the same MCS/link rates for both the upper and lower 5GHz band now. Previously because of the power limitations, you'd likely have lower MCS/link rates at the same distance on the lower 5GHz band (other than being maybe same room).

The actual maximum link rates are unchanged though.

 

[KenAF]

Is there a list of wireless routers that currently implement the new higher power levels at 5Ghz?

Tim - This (implementation of higher power at 5Ghz) might be something you mention in future wireless router reviews.

 

[thiggins]

Is there a list of wireless routers that currently implement the new higher power levels at 5Ghz?

Tim - This (implementation of higher power at 5Ghz) might be something you mention in future wireless router reviews.

Any product that raises the power level on the low 5 GHz band must be resubmitted to the FCC for certification. Not gonna happen since it takes time and is not inexpensive.

Manfs will focus on new routers as they always do. The NETGEAR R8000 is the first router to support higher power levels in the low band. I think I mentioned this in the review. I will ask ASUS about the AC87

 

[azazel1024]

I would guess it is going to be a crap shoot for the next 2-4 months with new routers if they support the higher power levels. At a guess, most new products 4+ months from now probably will support the higher power levels.

Out of curiosty, I assume the answer is "yes", but what about the client side of things? I assume that would require recert as well, which means most likely since clients lag routers, we are looking at 6-12 months before we see clients that are supporting higher power levels in the UNI-I band?

 

[thiggins]

I would guess it is going to be a crap shoot for the next 2-4 months with new routers if they support the higher power levels. At a guess, most new products 4+ months from now probably will support the higher power levels.

Out of curiosty, I assume the answer is "yes", but what about the client side of things? I assume that would require recert as well, which means most likely since clients lag routers, we are looking at 6-12 months before we see clients that are supporting higher power levels in the UNI-I band?

Any changes to product designs that fall under FCC rules require recertification if it changes RF characteristics.

I'd say it's unlikely you'll see clients support higher power levels due to power consumption and thermal concerns.

 

[sfx2000]

Any changes to product designs that fall under FCC rules require recertification if it changes RF characteristics.

I'd say it's unlikely you'll see clients support higher power levels due to power consumption and thermal concerns.

There's enough issues with the FCC's new rules, along with international concerns as 5 GHz is very inconsistent, that we might not see many products that are intended for multiple markets to leverage into the new rules as proposed by the FCC.

However, the US Market is large enough that vendors on the AP side likely will develop products - the challenge is the clients...

 

[azazel1024]

Any changes to product designs that fall under FCC rules require recertification if it changes RF characteristics.

I'd say it's unlikely you'll see clients support higher power levels due to power consumption and thermal concerns.

I dunno. The original 50mw limitation is rather low power. I am sure there are plenty of phones that don't exceed that on any band, but I'd wager most laptops probably exceed 50mw of power, where allowed.

*edit*

Or possibly not. I just tried to do some poking around. It obviously might not be indicative of all clients, but I found some evidence that the Intel 7260ac caps around 40mw of Tx power, though not sure if that is 2.4GHz, low UNI-I, UNI-III or all of the above. Some later driver versions actually cap it at 25mw. Anecdotal evidence only.

What I found was that the 7260ac appears capped at 16dBm Tx, but some driver versions cap it at 14/15dBm, oddly. Which works out to around 40mw and 32/25mw respectively.

*edit* I'll choose to believe the FCC certification documents. You can apparently set the 7260ac to 16dBm in the drivers, however the FCC certs claim 15dBm for 2.4GHz and 13.5dBm for 5GHz operation, or 32/21mW.

 

[stevech]

Most all WiFi chipsets I'm familiar with use 30mW RMS in the highest few modulation orders. At lower speeds, the need for high linearity in the output amplifiers is reduced. Therefore some chipsets go up to 100mW in the lower speeds, or the non-OFDM speeds.

Remember though, that the difference between 30 and 50mW is tiny in proportion to the signal loss in the path from TX to RX. So 30 vs. 50 won't be noticed in most situations. And, 50mW vs 100mW is not much difference either, in the RF world where the inverse square law applies to attenuation vs. distance.

 

[azazel1024]

Sure, but there is a big difference between 32mW and 250mW. Its a difference in almost 3x line of sight range. Even with path loss from interior obstructions, that is likely to the difference between having a usable signal two rooms over and having a usable signal 3 rooms over. Or, another way of looking at it is likely 1-2MCS levels higher at medium and long range.

 

[sfx2000]

Most all WiFi chipsets I'm familiar with use 30mW RMS in the highest few modulation orders. At lower speeds, the need for high linearity in the output amplifiers is reduced. Therefore some chipsets go up to 100mW in the lower speeds, or the non-OFDM speeds.

Remember though, that the difference between 30 and 50mW is tiny in proportion to the signal loss in the path from TX to RX. So 30 vs. 50 won't be noticed in most situations. And, 50mW vs 100mW is not much difference either, in the RF world where the inverse square law applies to attenuation vs. distance.

Not only does one lose Tx Power on the higher order modulations - this is offset by coding gain as one steps down from 64QAM to 16QAM to 8PSK, etc...

in 802.11n - Space Time Block Codes and Low Density Parity-Check Codes (STBC/LDPC) are optional, but when implemented, actually add about 3-5 dB of coding gain on top of the RF link budget - newer chipsets implement both, so if anything, this is a good reason to upgrade AP's

 

[KGB7]

FCC should have done this years ago. This is what happens when you hire wrong people for the job and pay them too much money. Bureaucracy at its best.

I've seen first hand how government works at the bottom level. The employees do bare minimum that is required of them, collect a pay check and go home. Because they have a job security. 90% of the same people would get fired after few month in private sector.