Implicit protection method and system based on multi-sensor data fusion

Abstract

The invention discloses an implicit protection method based on multi-sensor data fusion, which comprises the following steps: dividing the data collected by an APP-oriented sensor into user screen touch operation-related data, behavior action posture-related data and position information-related data; generating a screen touch operation feature library, a behavior action posture feature library and a position information library through the data, respectively and correspondingly and further calculating the matching probability and trust coefficient of the user with a legal owner of a mobile phone, respectively; abstracting the obtained trust coefficient and the matching probability as a weak matcher, synthesizing a strong matcher through an AdaBoosting algorithm and further calculating a comprehensive matching probability of the user with the legal owner of the mobile phone; and setting a safety matching threshold according to the type of the APP and performing protective processing according to a comparison result of the comprehensive matching probability with the safety matching threshold. The invention also discloses an implicit protection system based on multi-sensor data fusion. The method and the system in the invention can implicitly identify behaviors of illegal use of the mobile phone without the authorization of the owner and discover the situation that the mobile phone is stolen and out of control timely.

Description

technical field [0001] The invention relates to the technical field of safety protection, in particular to an implicit protection method and system based on multi-sensor data fusion. Background technique [0002] With the rapid development of the mobile Internet, mobile terminals represented by Android phones play an increasingly important role in our daily lives. While mobile phones provide convenience to people, they also store a large amount of personal privacy data such as mobile phone address books, short messages, etc. More importantly, the mobile phone is also tightly bound to the user's personal property, such as online banking APP, Alipay, WeChat payment, etc. This makes once the mobile phone is out of control, it will bring huge risks such as privacy leakage and property loss to users. [0003] Mobile phone out of control can be roughly divided into the following three situations: [0004] ①Mobile phone remote attack: By writing malicious apps such as Trojan hor...

AI Technical Summary

Problems solved by technology

[0004] ①Mobile phone remote attack: By writing malicious apps such as Trojan horses, tricking users into clicking download and running, thereby stealing private information such as user accounts;

[0005] ② Unauthorized use of mobile phones: Mobile phone users may be peeping at the lock screen password due to carelessness. After being taken and unlocked by others, they directly impersonate the user to cheat around in WeChat and other APPs, or browse and spy on private information, etc.

[0006] ③Loss of mobile phone: The mobile phone may be stolen or lost accidentally in life. After the person with bad intentions gets the mobile phone, although he cannot use the mobile phone directly, he can take the mobile phone and leave the scene first to crack the password and other behaviors that violate the privacy of the legal owner of the mobile phone.

At present, the recognition accuracy of this type of method is easily affected by many factors such as the environment, and usually requires the cooperation of traditional explicit authentication methods

At the same time, this type of certification protection method still has problems such as over-reliance on specific sensors and limited scope of application.

[0010] It can be seen that the existing authentication protection methods cannot well deal with various mobile phone out-of-control situations

Images