DFS ETSI conformance for weather channels (again)

DFS ETSI conformance for weather channels (again)

[Jean-Pierre TOSONI]

Hi list,

I just stumbled on this patch that IMHO was rejected for a bad reason:

  [PATCH v3 4/4] cfg80211: DFS use 10 minutes CAC when weather channels
  At http://www.spinics.net/lists/linux-wireless/msg115296.html

The reason given is the "unachievable" requirement in the ETSI norm of
99.99% detection success. It was even considered disabling the weather
channels.

*But the norm (ETSI EN 301 893 V1.7.1) does not say this.* It states
that 99.99% is the target performance:

 - during conformance tests only, (5.3.8.1.1 paragraph 5)
 - specifically *not* a requirement with specific real life radars,
(5.3.8.1.1 paragraph 5)
 - on a limited series of 20 tests, (5.3.8.2.1.2 item g)
 - with a pulse power at +10dB above other pulse detection tests,
(5.3.8.2.1.2 item f)
 - with some patterns excluded from the test. (5.3.8.2.1.2 item g)

Hence it is very possible that existing products can achieve the
conformance test target.

That being said, would it be possible to reexamine that dropped patch
and apply it?

Best regards,

Jean-Pierre Tosoni

Re: DFS ETSI conformance for weather channels (again)

[Kalle Valo]

"Jean-Pierre TOSONI" <jp.tosoni@acksys.fr> writes:

> Hi list,
>
> I just stumbled on this patch that IMHO was rejected for a bad reason:
>
>   [PATCH v3 4/4] cfg80211: DFS use 10 minutes CAC when weather channels
>   At http://www.spinics.net/lists/linux-wireless/msg115296.html
>
> The reason given is the "unachievable" requirement in the ETSI norm of
> 99.99% detection success. It was even considered disabling the weather
> channels.
>
> *But the norm (ETSI EN 301 893 V1.7.1) does not say this.* It states
> that 99.99% is the target performance:
>
>  - during conformance tests only, (5.3.8.1.1 paragraph 5)
>  - specifically *not* a requirement with specific real life radars,
> (5.3.8.1.1 paragraph 5)
>  - on a limited series of 20 tests, (5.3.8.2.1.2 item g)
>  - with a pulse power at +10dB above other pulse detection tests,
> (5.3.8.2.1.2 item f)
>  - with some patterns excluded from the test. (5.3.8.2.1.2 item g)
>
> Hence it is very possible that existing products can achieve the
> conformance test target.
>
> That being said, would it be possible to reexamine that dropped patch
> and apply it?

Don't we support this by getting the CAC time in the regulatory
database?

commit 31559f35c5724976fd975e5d7e90cdb693b8dd27
Author: Janusz Dziedzic <janusz.dziedzic@tieto.com>
Date:   Fri Feb 21 19:46:13 2014 +0100

    cfg80211: DFS get CAC time from regulatory database
    
    Send Channel Availability Check time as a parameter
    of start_radar_detection() callback.
    Get CAC time from regulatory database.
    
    Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com>
    Signed-off-by: Johannes Berg <johannes.berg@intel.com>

-- 
Kalle Valo

RE: DFS ETSI conformance for weather channels (again)

[Jean-Pierre Tosoni]

Thanks Kalle,

I will check the wireless-regdb list instead.
But it looks like the current python script that generates regulatory.bin
was not updated for CAC time yet :-(

Cheers,
Jean-Pierre

-----Message d'origine-----
De : Kalle Valo [mailto:kvalo@codeaurora.org] 
Envoyé : jeudi 26 mars 2015 17:06
À : Jean-Pierre TOSONI
Cc : linux-wireless@vger.kernel.org
Objet : Re: DFS ETSI conformance for weather channels (again)

"Jean-Pierre TOSONI" <jp.tosoni@acksys.fr> writes:

> Hi list,
>
> I just stumbled on this patch that IMHO was rejected for a bad reason:
>
>   [PATCH v3 4/4] cfg80211: DFS use 10 minutes CAC when weather channels
>   At http://www.spinics.net/lists/linux-wireless/msg115296.html
>
> The reason given is the "unachievable" requirement in the ETSI norm of 
> 99.99% detection success. It was even considered disabling the weather 
> channels.
>
> *But the norm (ETSI EN 301 893 V1.7.1) does not say this.* It states 
> that 99.99% is the target performance:
>
>  - during conformance tests only, (5.3.8.1.1 paragraph 5)
>  - specifically *not* a requirement with specific real life radars,
> (5.3.8.1.1 paragraph 5)
>  - on a limited series of 20 tests, (5.3.8.2.1.2 item g)
>  - with a pulse power at +10dB above other pulse detection tests,
> (5.3.8.2.1.2 item f)
>  - with some patterns excluded from the test. (5.3.8.2.1.2 item g)
>
> Hence it is very possible that existing products can achieve the 
> conformance test target.
>
> That being said, would it be possible to reexamine that dropped patch 
> and apply it?

Don't we support this by getting the CAC time in the regulatory database?

commit 31559f35c5724976fd975e5d7e90cdb693b8dd27
Author: Janusz Dziedzic <janusz.dziedzic@tieto.com>
Date:   Fri Feb 21 19:46:13 2014 +0100

    cfg80211: DFS get CAC time from regulatory database
    
    Send Channel Availability Check time as a parameter
    of start_radar_detection() callback.
    Get CAC time from regulatory database.
    
    Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com>
    Signed-off-by: Johannes Berg <johannes.berg@intel.com>

--
Kalle Valo

Re: DFS ETSI conformance for weather channels (again)

[Kalle Valo]

"Jean-Pierre Tosoni" <jp.tosoni@acksys.fr> writes:

> I will check the wireless-regdb list instead. But it looks like the
> current python script that generates regulatory.bin was not updated
> for CAC time yet :-(

IIRC CAC patches for CRDA were not accepted or something. But it should
work with kernel internal regdb.

-- 
Kalle Valo