Built-in three-dimensional accelerometer based identity authentication method of handheld device

Abstract

The invention provides a built-in three-dimensional accelerometer based identity authentication method of a handheld device. The method comprises the steps of continuously performing the same action at least four times by a user handholding a built-in three-dimensional accelerometer sensor type device in order to acquire data; performing matrix singular value decomposition for the acquired data; performing feature enumeration for the decomposed data to obtain 20 features; analyzing the 20 features based on the maximum correlation and minimum redundancy method; training a user identity authentication model through a primary support vector machine based on the selected optimal features; determining the model expires before each authentication after modeling; if not, authenticating the identity; if so, updating and then authenticating. Compared with the prior art, the identity authentication method has the advantages that the authentication accuracy is high, the misjudgment rate is small, and the peering and emulation attack can be resisted, so that the privacy of a user can be protected well; the calculation for the extracted feature costs a little, and is quickly performed; the operation is convenient and can be performed by one hand; the way that the user handholds a mobile phone, the location and the gesture lasting time are not depended; in addition, additional hardware devices are saved.

Description

technical field [0001] The invention relates to an identity authentication method based on behavioral characteristics, in particular to an identity authentication method based on a handheld device embedded with a three-dimensional accelerometer, and belongs to the technical field of human-computer interaction. Background technique [0002] At present, more and more handheld devices are embedded with 3D accelerometers, such as smart phones, smart bracelets, smart watches, and portable GPS. Many, such as e-mail, electronic payment, and various social networking applications. These applications store a large amount of private and sensitive information, so smartphones are becoming more and more private, and the identity authentication of mobile phone users is becoming more and more important. [0003] Currently there are many identity authentication methods on smart phones, and these methods can be divided into three categories: knowledge-based authentication technology, biomet...

AI Technical Summary

Problems solved by technology

This method selects and enumerates 20 statistical features that can authenticate users from the original acceleration data, has high authentication accuracy, and effectively solves the problem that traditional mobile phone unlocking methods are easy to be peeped and imitated. These gestures are relatively simple and convenient for users to use. and memory, and does not require additional hardware devices, relatively cheap

Images