The latest research shows that real-world tests indicate that gait authentication may be a viable way to protect smartphones and other mobile devices from cybercrime.
A study led by the University of Plymouth required smartphone users to perform daily activities while the motion sensors in their mobile devices capture data about their stride patterns.
The results showed that the average accuracy of the system in identifying individuals’ gait was about 85%, and when they walked normally and fast, this number rose to nearly 90%.
There are currently more than 6.3 billion smartphone users worldwide who use their devices to provide a wide range of services and store sensitive and confidential information.
Despite the existence of authentication mechanisms (such as passwords, PINs, and biometrics), research has shown that the levels of security and availability of such methods vary widely.
Write Computer and SecurityAccording to the researchers, the study shows that-within the appropriate framework-gait recognition may be a viable technology to protect individuals and their data from potential crimes.
Scholars at the Plymouth Cyber Security, Communication and Network Research Center have been focusing on developing a series of innovative authentication mechanisms to provide more secure and usable solutions.
This research builds on previous work by evaluating a multi-algorithm gait recognition system, and is the first research to apply this using real-world data.
In this study, 44 participants between the ages of 18 and 56 were asked to carry a global smartphone device for 7 to 10 days.
They were required to put their smartphones in their pockets to record sensor data captured by the device’s gyroscope and accelerometer during different physical activities.
During the test, each participant produced an average of 4,000 sample activities, which were divided into records showing normal and fast walking, as well as climbing and descending stairs.
This shows that the potential error rates of normal and fast walking are 11.38% and 11.32%, respectively, and when participants go down and go upstairs, this number rises to 24.52% and 27.33%, respectively.
The researchers said that this emphasizes the need to further improve the ability to automatically distinguish a wider range of walking activities so that multiple algorithm recognition methods can target specific walking characteristics.
Nathan Clark, a professor of cybersecurity and digital forensics at the University of Plymouth and recently a researcher at the Chartered Institute of Information Security, said: “With the development of smartphones, security controls must be significantly improved. This has led to a significant increase in user authentication. Users need to repeatedly verify their devices and the numerous applications contained in them. Gait authentication has become a non-invasive way to capture the necessary level of personal information, but-until now-all testing it takes place in a controlled environment.
“Gait recognition alone cannot solve the problem of usable and convenient identity verification, but it can form a vital tool in the cyber arsenal, helping to enhance the understanding of user identity. This research shows for the first time that, in addition to “Under laboratory control conditions, what level of performance can actually be achieved. Obviously, the performance level is affected; but research also shows that for most users, these problems can be overcome to an acceptable level. “
Story source:
Material Provided by the University of Plymouth. The original work was written by Alan Williams. Note: The content can be edited according to style and length.
