Mobile intelligent device sensor fingerprint protection method

Abstract

The invention discloses a mobile intelligent equipment sensor fingerprint protection method. The method comprises the following steps: step 1, collecting sensor data: obtaining original data of a motion sensor of mobile intelligent equipment; Step 2, data preprocessing: processing the acquired original data of the motion sensor to enable the data lengths of the data to be consistent and uniformlydistributed on a time axis; Step 3, fingerprint protection data confrontation training based on the neural network: performing fingerprint protection confrontation training on the fingerprint recognition neural network model through the training data; And step 4, data output: preprocessing the sensor original data acquired in real time according to the step 2, performing fingerprint protection processing on the preprocessed sensor original data through the trained fingerprint recognition neural network model, and outputting sensor data without fingerprint information according to an output scene after processing. The method can successfully protect the fingerprint information of different equipment sensors, can control the utility loss, and does not influence the use of related sensors.

Description

technical field [0001] The invention relates to the security field of mobile smart devices, in particular to a method for protecting sensor fingerprints of mobile smart devices. Background technique [0002] With the prosperity of the mobile Internet, the security and privacy of mobile devices has become an issue of increasing concern. Accurately identifying users and their mobile devices enables various mobile services (personalized cross-platform recommendations, advertising cross-device services, etc.) And data processing (data fusion, data transaction, etc.) becomes possible, but at the same time it also brings about the leakage of users' potential private information. Existing methods for obtaining device fingerprints, such as browsers obtaining cookies, APP requesting the IMEI of Android devices and the IDFA of Apple devices, etc., due to people's growing concerns about privacy, these methods have been more and more clarified. Therefore, some novel methods have been p...

AI Technical Summary

Problems solved by technology

However, sensor fingerprint extraction is based on hardware, which takes advantage of the heterogeneity of the sensor due to the limitation of the processing technology in the manufacturing process. These sensor differences are almost constant and difficult to modify during the life cycle of the device. Some existing methods to protect the privacy of sensor fingerprints Methods, such as correction and noise addition, are difficult to ensure that the recognition accuracy is reduced at the same time and the loss of utility is controllable

However, there is currently no good way to protect sensor fingerprints. How to prevent these methods from exposing user privacy without loss of data utility is an urgent problem to be solved.

Images