I left a Kindle Paperwhite with my mom a month ago and it disappeared in the house, probably in between books somewhere. I know it still connects to wifi and receives documents through the kindle email at least once a day, so it has some battery left.

Is there any call home, lost beep, or similar feature I could use to help focus the search?

At worst I thought of setting up some laptops/phones sniffing packages and broadcast strength coming from the kindle's mac address (it shows up every once in a while in the router logs) and triangulate that to have a gross estimation of the location. Any idea on which software would be the most appropriate for it (ideally for ubuntu/android/windows)? Thank you for your answers before the battery runs out

Found the Kindle using the sniffing technique described below. However I'd still like to know if there's a way to find it that's less overkill.

I discovered that the Kindle updates its contents daily while in deep sleep around 5:30-6:30AM. So I set up the delivery of a biggish pdf file through the kindle email so I'd ensure a long enough connection, end set up three laptops with a Kali Linux running live with airodump sniffing packages (modified from here: http://www.androidauthority.com/capture-data-open-wi-fi-726356/ )

The following procedure can be used to physically find any active wifi device connected to a network:

Three lines of code in the terminal in at least three Kali Linux running devices (2D location, at least 4 receivers out of plane for 3D):
To start monitor mode on a wifi card (in this case wlan0, check on iwconfig if needed):
airmon-ng start wlan0
Airmon usually changes the card name to wlan[n]mon after this. Check the wifi interface name on iwconfig and modify accordingly if the next line returns a related error.
Then run airodump to find your wifi network and its channel:
airodump-ng wlan0mon
Enter Ctrl-C to end the capture once you note the desired network's channel Run the following line to run airodump again only at the specified channel and record the data to the "allthedata.csv" file in the terminal directory (usually the root for a live cd/usb if you didn't change it). Replace the square brackets by the channel number:
airodump-ng -c [channel number 1, 6 or 11 usually] -w allthedata wlan0mon

I started the pcs before going to sleep and as I woke up a bit past the usual update time, I hit ctrl-C to stop the capture and looked up the device on the allthedata.csv files by its MAC number which can be obtained at the router logs. There are in fact 4 allthedata files with different extensions per capture, and on the xml you can find the manufacturer of the device in the logs, so even if you don't know the mac address by looking up an Amazon branded device connected to your router you may find the Kindle entry.

At the Kindle entry, find the signal strength value and note it for all the receiver computers. Than convert the values from dbm to mW using the following formula: mW=10^(dBm/10)*1000.

To triangulate the signal (for a 2D location), draw a chart with the position of the laptop wifi receivers and for every pair of receivers, draw a line perpendicular to the line going through the points in a point inversely proportional to the square root of the mW power values, as the signal drops of proportionally to the square of the distance (for 3D use at least 4 receivers and a surface between the points).

The lines should cross at the source of the emission - the Kindle was at a closet in this case. The margin of error was about 1m/3ft even with several concrete/brick walls in between the receivers and the emission.

[edit 2] -This simplified location approach is only possible if the emitter is located in between the receivers, that is, a three-set of receivers (2D) form a triangle that has the emitter inside it. -Different receivers might have different antenna gains and correcting for it will greatly increase accuracy. A quick and dirty way of doing so is positioning the receivers at the same position from a constant power emitter (wifi router, for example) and using the signal strength value difference between them as a calibration factor to be subtracted it from the signals received afterwards.

  • 2
    Instead of editing your question with your solution, you should post it as an answer to your own question (and you might consider already accepting your answer) And Kudos for finding your Kindle! Nice approach. – Daniel Aug 18 '17 at 10:28

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.