llewelld / harbour-contrac

An implementation of the Apple/Google Exposure Notification specification for Sailfish OS.
GNU General Public License v2.0
28 stars 5 forks source link

"Attenuation calibration" download does not work #65

Open dryo opened 3 years ago

dryo commented 3 years ago

Hi there!

I'm following the development since the beginning and now installed the app for the first time.

I really don't know what I have to enter for attenuation calibration settings and the download button does not work (for me). I entered the given URI https://developers.google.com/android/exposure-notifications/files/ into my desktop browser and it showed a 404.

How is this supposed to work?

BTW: A big thank you for the app! I hope to install this on my parent's phones.

llewelld commented 3 years ago

Thanks for flagging this issue up, and for the kind words. It seems Google updates the info periodically and the filename changes as a result. The current location (as I write this) is:

https://developers.google.com/android/exposure-notifications/files/en-calibration-2020-08-12.csv

If the above link doesn't work, you should be able to get the latest file on the following page (under the "Device calibration list" link):

https://developers.google.com/android/exposure-notifications/ble-attenuation-computation#device-list

I'll update the link in the next version of the app. Here are some values for common SFOS phones taken from the spreadsheet:

Phone              | RSSI correction | Transmit power
-------------------+-----------------+---------------
Xperia X   (F5121) |               8 |            -34
Xperia XA2 (H3113) |               5 |            -30
Xperia 10  (I3113) |               5 |            -30
cintema commented 3 years ago

Unfortunately there is no calibration data for the Gemini PDA...

llewelld commented 3 years ago

Unfortunately there is no calibration data for the Gemini PDA...

I'm afraid not (and as a Gemini owner myself, I appreciate that it's a shame). I can only advise that you should use similar values to other devices. The average RSSI correction across all devices is around 3.5, while the average transmit power is around -23. In case it helps, as explained in the overview, the way the attenuation is calculated is as follows:

Attenuation = TX_power - (RSSI_measured + RSSI_correction)

Roughly, the higher the attenuation, the greater the distance to the other user. So if you want to increase your chance of getting a high risk warning you should increase the RSSI correction. If you want to increase the chance of others getting a high risk warning, you should decrease the transmit power.