Some doubt
Opened this issue · 4 comments
When I introduced this dataset to my colleagues as utilising the preamble part from the device in LoRa , everyone was confused. Because the preamble part is not the same for different devices, we can directly use the preamble part for device identification, so why do we need to train the model to identificate.
I would like to clarify that the preamble part is the same for all LoRa packets with the same SF/BW configuration. (Preamble here refers to eight repeated unmodulated LoRa symbols that do not contain the MAC address).
The MAC address and cryptographic keys can be tempered/stolen, while hardware characteristics-based RF fingerprints are relatively reliable, as the attacker needs to make hardware-level modifications to mimic legitimate transmitters.
I would like to clarify that the preamble part is the same for all LoRa packets with the same SF/BW configuration. (Preamble here refers to eight repeated unmodulated LoRa symbols that do not contain the MAC address).
The MAC address and cryptographic keys can be tempered/stolen, while hardware characteristics-based RF fingerprints are relatively reliable, as the attacker needs to make hardware-level modifications to mimic legitimate transmitters.
非常感谢您的解答
hi , I want to know the difference between Closed_set_RFFI and Openset_RFFI_TIFS
Dear Author, please can you give me the details of the platforms(for example python version, conda version, pip version, etc. ) to successfully install packages and run the codes without encountering any difficulties?
Also, how do I install this ''vs2017_win-64==19.16.27038=h2e3bad8_2''?