Question: Should FCC-region 6 GHz entries use the India PSD methodology?

[Chen-Yu Tsai]

Hi,

On Sun, Apr 12, 2026 at 6:38 AM Louis Kotze <loukot at gmail.com> wrote:
>
> Hi Chen-Yu,
>
> Following the patch series I just submitted, I have a question about
> the FCC-region 6 GHz entries.
>
> Currently 12 countries (US, CA, BR, AR, CO, PE, SV, MX, CL, KY, HN,
> PY) use 12 dBm for 6 GHz LPI. I understand this was chosen as the
> absolute minimum across all device types and channel widths.
>
> However, the India (IN) entry (commit 5db6ce73) uses a different
> methodology — PSD at 20 MHz minimum channel width:
>
>   India: 11 dBm/MHz × 20 MHz = 24 dBm
>
> Applying the same approach to FCC (5 dBm/MHz PSD):
>
>   FCC: 5 dBm/MHz × 20 MHz = 18 dBm

The PSD limit is actually -1 dBm/MHz. The device class we consider is
6XD (indoor low power client). This translates to 12.01 dBm at 20 Mhz
channel width. So the current rule in the database is correct.

> 18 dBm is compliant at every standard WiFi channel width (>=20 MHz)
> for both AP and client devices:
>
>   Width   PSD limit   FCC client max   18 dBm   Legal?
>   20 MHz    18 dBm       24 dBm        18 dBm   Yes
>   40 MHz    21 dBm       24 dBm        18 dBm   Yes
>   80 MHz    24 dBm       24 dBm        18 dBm   Yes
>   160 MHz   27 dBm       24 dBm        18 dBm   Yes
>   320 MHz   30 dBm       24 dBm        18 dBm   Yes
>
> The reason I ask: during my work on the South Africa (ZA) fix, the
> 9 dBm gap between the regdb (14 dBm) and the AP's advertised power
> (23 dBm) caused repeated "regulatory prevented using AP config,
> downgraded" errors and complete 6 GHz association failure. The FCC
> gap is 18 dBm (12 vs 30), which would cause the same or worse
> symptoms for any Linux user connecting to a standard FCC LPI access
> point on 6 GHz.
>
> Would the India PSD-at-20-MHz methodology be appropriate for FCC
> countries as well?

As mentioned above, they are using the same methodology. It's just that
the FCC PSD limit is very low.


Thanks
ChenYu