0
$\begingroup$

I'm not sure if this is right platform to post . I am researching on WiFi Localization algorithms using Machine learning and was wondering if anyone knows any dataset which is publicly available. Some companies do post them online - > https://www.deepsig.io/datasets (datasets here are generated on GNURadio )

The dataset I'm looking can be like this

AP_location | Rssi of Client | Client_location | PhaseValues(if avaialable)

Thanks

$\endgroup$
  • 1
    $\begingroup$ what would "phase value" be? Makes little sense, since "phase" is a relative thing; and in the case of OFDM systems like Wifi, even less sense, since these systems are employed exactly when the signal is wider than the coherency bandwidth of the channel, and thus you get different phases on different parts of your spectrum. $\endgroup$ – Marcus Müller Mar 20 at 16:59
  • 1
    $\begingroup$ also, since this all happens due to multipath, that dataset needs to contain a map of obstacles and reflectors, too, otherwise there's just too many degrees of freedom your ML needs to learn... this is a bit of an old topic, and numerous publications of the past have shown that, to little surprise, completely blind location estimation using ML doesn't work very well in indoor environments. $\endgroup$ – Marcus Müller Mar 20 at 17:02
  • $\begingroup$ ieeexplore.ieee.org/search/… $\endgroup$ – Marcus Müller Mar 20 at 17:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.