Research On MEMS-based Cryptographic Primitives For Smart Phones

With the development of mobile information technology,smartphone has become the most commonly used data processing terminal.Authentication methods like password,PIN-code are not safe,and the loss and leak of personal privacy information will bring great potential safety hazard to the users,so the smartphone needs a more secure way of identification.The thesis focus on existing problems of smartphone identification methods.Based on the low cost micro-electromechanical systems(MEMS)which are extensive integrating in the mobilephones,we propose a dynamic gesture identification method,and realize the application of the proposed method on Android smartphone,namely the unlock software based on dynamic gestures.Besides,we firstly put forward a key generation method based on dynamic gestures,which combines biological recognition technology and cryptography together.The main research includes:(1)A smartphone identification method based on gesture.We presented MEDTW that controls the difference between the latitudes contributing to the total Euclidean distance by smartphone morphological characteristics to enhance identification accuracy and introduces early termination to guarantee the efficiency.Meanwhile,we proposed restricted areas touch trigger gesture acquisition scheme and authentication gesture length selection scheme to improve the accuracy and efficiency.Experimental results prove the effectiveness of the method.(2)A gestures unlock system based on Android.Users can unlock smartphone by swiping gestures in the air.The APP has good user experience and precision.The main work includes the APP interface design and implementation,system data storage and remote database,the realization of function module(including sensor control,gestures registration and unlock).The APP has been online,and we collected 2117 gestures data from 94 smartphones based on this APP and analyses the data.(3)A smartphone key generation method based on gesture.We presented a key generation scheme that uses fuzzy vault algorithm based on dynamic gesture.The proposal combines the fuzziness of biological characteristics and the accuracy of the cryptography,which can effectively make up for the biological feature extraction of internal differences.This article is the first case which proposes using dynamic gesture to generate key.The simulation results prove the effectiveness.