MP4 video made by Blackberry

Not much info here but I can try to take a swing at this..

You'll want the original mp4 video or close to it as possible. I'm assuming the metadata does not contain a nice serial number of the phone so you'll have to take a different route to uniquely identify. (mediainfo is the good at looking though exif data)

If you still have access to the phone, ask if you can create tests recordings and determine consistencies for structure on a global and local level. PRNU can be useful tool if you have the time and means of preforming an experiment.

From there you'll just have to determine if the original video is consistent, inconsistent, or inconclusive with your tests.

There's some good information on (source camera identification) out there on the web if needed.

Best of luck,
ZG

Thanks so much for the reply Zac. Regarding the exif data - I have used exiftool to view that and there's nothing distinguishable here. What can I use to generate the PRNU data and can this been done for video files ? Are there any limitations and how reliable is this. Aside from research papers, I cannot find any tools to assist with this. Are there any commercial or open source ones ?

Thanks

Do you have the device you suspect the video is from?

Yes I do have the device.

Maybe of interest, google for "Sensor Pattern Noise".

jaclaz

Blackberry handsets store thumbs.dat or .db files (can't remember which exactly) and within these you will find thumbnail representations of the images/videos, these also contain dates and times for videos and file names. Obviously this doesn't doesn't verify that the video was actually created by that handset but if you confirm it's presence within the .dat/.db's then you can say the video was at least present on the device, if it is in a non-standard location on the device then this points towards knowledge of the video. Sorry if this is a little vague but we haven't got a great deal of info to go on.