Re: [wireless-regdb] wireless-regdb: Update regulatory rules for the US on 6 GHz band

[Seth Forshee <sforshee@kernel.org>]

On Sat, Dec 18, 2021 at 04:34:13PM +0900, Sungbo Eo wrote:
> On 2021-12-18 06:55, sforshee@kernel.org wrote:
> > On Mon, Dec 13, 2021 at 09:19:50AM +0000, Asura Liu (asuliu) wrote:
> > >  From 3db25ea674232fea6a5efca292f6ed3fd8eba7a2 Mon Sep 17 00:00:00 2001
> > > From: Asura Liu <asuliu@cisco.com>
> > > Date: Mon, 13 Dec 2021 16:46:28 +0800
> > > Subject: [PATCH] wireless-regdb: Update regulatory rules for the US on 6 GHz
> > > band
> > > MIME-Version: 1.0
> > > Content-Type: text/plain; charset=UTF-8
> > > Content-Transfer-Encoding: 8bit
> > > 
> > > According to FCC 20-51, FCC adopts rules to unlicensed use of the 6 GHz band:
> > > "59. Third, the Commission limits the low-power indoor access points to lower power levels than the standard-power access points that operate under the control of an AFC. Consistent with the Commission's approach for the existing U-NII bands, the Commission specifies both a maximum power spectral density and an absolute maximum transmit power, both in terms of EIRP. Specifically, the Commission allows a maximum radiated power spectral density of 5 dBm per 1 megahertz and an absolute maximum radiated channel power of 30 dBm for the maximum permitted 320-megahertz channel (or 27 dBm for a 160-megahertz channel). In addition, to ensure that client devices remain in close proximity to the indoor access points, the Commission limits their PSD and maximum transmit power to 6 dB below the power permitted for the access points."
> > > See https://www.federalregister.gov/documents/2020/05/26/2020-11236/unlicensed-use-of-the-6-ghz-band
> > > 
> > > And 47 CFR § 15.407 describe this as following:
> > > (a) (5) For an indoor access point operating in the 5.925-7.125 GHz band, the maximum power spectral density must not exceed 5 dBm e.i.r.p. in any 1-megahertz band. In addition, the maximum e.i.r.p. over the frequency band of operation must not exceed 30 dBm.
> > > (a) (8) For client devices operating under the control of an indoor access point in the 5.925-7.125 GHz bands, the maximum power spectral density must not exceed −1 dBm e.i.r.p. in any 1-megahertz band, and the maximum e.i.r.p. over the frequency band of operation must not exceed 24 dBm.
> > 
> > Thanks for the patch. A couple of quick notes about the patch
> > description. It's preferred to wrap lines in the body at around 75
> > characters, and it's required that you include a Signed-off-by tag
> > indicating your agreement to the DCO for your contribution (see
> > CONTRIBUTING).
> > 
> > Additional comments below.
> > 
> > > ---
> > > db.txt | 6 ++++++
> > > 1 file changed, 6 insertions(+)
> > > 
> > > diff --git a/db.txt b/db.txt
> > > index b898799..c6ef9b6 100644
> > > --- a/db.txt
> > > +++ b/db.txt
> > > @@ -1606,6 +1606,12 @@ country US: DFS-FCC
> > > 	# https://www.federalregister.gov/documents/2021/05/03/2021-08802/use-of-the-5850-5925-ghz-band
> > > 	# max. 33 dBm AP @ 20MHz, 36 dBm AP @ 40Mhz+, 6 dB less for clients
> > > 	(5850 - 5895 @ 40), (27), NO-OUTDOOR, AUTO-BW, NO-IR
> > > +	# 6ghz band
> > > +	# https://www.federalregister.gov/documents/2020/05/26/2020-11236/unlicensed-use-of-the-6-ghz-band
> > > +	# https://docs.fcc.gov/public/attachments/FCC-20-51A1_Rcd.pdf
> > > +	# max. 30 dBm AP @ 320MHz, 27 dBm AP @ 160MHz, 6 dB less for clients
> > > +	(5925 - 7125 @ 320), (30), NO-OUTDOOR, AUTO-BW
> > > +	(5925 - 7125 @ 320), (24), NO-OUTDOOR, AUTO-BW, NO-IR
> > 
> > The kernel doesn't currently support multiple rules with different flags
> > for the same range. This is an issue that's come up several times, but
> > so far nothing has been done about it.
> > 
> > Even ingoring that, I don't think these rules accomplish the intended
> > purpose. There's nothing that would require a client device to use the
> > NO-IR rule, so they could end up using the higher power limit and
> > transmitting before detecting transmission from an AP.
> > 
> > I also suspect that we should be able to express the AFC requirement in
> > the database before permitting AP operation in this range.
> > 
> > Currently I think the best we're able to do is to use the lowest common
> > denominator, which is the 24 dBm rule with NO-IR.
> > 
> > Thanks,
> > Seth
> 
> Hi, sorry for intruding.
> 
> I thought max EIRP rule for larger bandwidth also applies to smaller
> bandwidth, that means we would use 24 dBm tx power with 20 MHz channel as
> well. But the regulation limits PSD to -1 dBm/MHz and it leads to 12 dBm for
> 20 MHz. Shouldn't we use 12 dBm rule to fit the smallest bandwidth?
> 
> I still don't understand exactly how the rules work, my apologies if I am
> mistaken.

Before I respond, let me be upfront about my limitations. I'm not an RF
engineer. My day job no longer has anything to do with wireless and has
not for a number of years now. I haven't had time to really keep up with
recent developments either, so my impressions are based on some light
research and what I remember of information I rarely use nowadays.

After doing a little bit of reading, I suspect that you are probably
right (in principle at least, I didn't check the math). It seems there's
been a movement towards specifying PSD so that larger bandwidths can
effectively have a higher max EIRP than lower bandwitdths to overcome
diminishing SNR with increasing bandwidth.

That leaves us in the unfortunate position of needing to specify an EIRP
suitable for a bandwidth of 20 MHz when larger bandwidths could be using
a higher power limit. It seems like the Linux regulatory framework needs
to evolve to support PSD limits. I don't currently have the time or the
inclination to do this work though, and until someone does we're stuck
with the status quo.

Thanks,
Seth