Security, Privacy, and Anonymity in Computation, Communication, and Storage

In recent years, there have been many studies using the data generated by the built-in sensors of mobile phones for authentication and the selection of features is involved in the use of sensor data. This article discusses the method of biological feature selection by taking the mobile phone acceleration sensor as an example. 30 participants were invited to walk with their mobile phones for data collection to obtain data set 1. Several characteristics were evaluated from multiple aspects to select a number of effective features. 15 participants were invited to collect data set 2 under the condition of simulating dialy life. A featurebased authentication method is proposed and a success rate of 93.6% is obtained on data set 1. On the data set 2, 91.90% of the recognition success rate was obtained.