Data is the new oil for the car industry. Cars generate data about how they\nare used and who's behind the wheel which gives rise to a novel way of\nprofiling individuals. Several prior works have successfully demonstrated the\nfeasibility of driver re-identification using the in-vehicle network data\ncaptured on the vehicle's CAN (Controller Area Network) bus. However, all of\nthem used signals (e.g., velocity, brake pedal or accelerator position) that\nhave already been extracted from the CAN log which is itself not a\nstraightforward process. Indeed, car manufacturers intentionally do not reveal\nthe exact signal location within CAN logs. Nevertheless, we show that signals\ncan be efficiently extracted from CAN logs using machine learning techniques.\nWe exploit that signals have several distinguishing statistical features which\ncan be learnt and effectively used to identify them across different vehicles,\nthat is, to quasi "reverse-engineer" the CAN protocol. We also demonstrate that\nthe extracted signals can be successfully used to re-identify individuals in a\ndataset of 33 drivers. Therefore, not revealing signal locations in CAN logs\nper se does not prevent them to be regarded as personal data of drivers.\n