A Cloud-Based Cheat Detection Method for Smartphones by Analyzing Motion Big Data from Accelerometers
The objective of this invention is to automatically detect and prevent cheating acts of tampering with information relating to tap operations on a smartphone. The mainstream conventional anti-cheat methods have been methods of detecting tampering of binary data for applications and methods of detecting tampering of storage or memory areas used in read/write operations for applications. These methods have not been effective in the case where the tampering detection method has become known or in the case where program code for detection itself has been tampered with. In view of this situation, this invention realizes general tap event validation by way of cheat detection utilizing data that is naturally entered by a legitimate user and that cannot be easily entered by a cheater. A technical feature of this invention is that vibrations that occur when taps are performed on a smartphone are recorded by using an acceleration sensor on the smartphone, and time-series information about tap events is compared with time-series information obtained from the acceleration sensor on the server side, whereby a legitimate tap event physically entered by a human user is distinguished from a tap event automatically generated or tampered with by a program. The vibration values output from the acceleration sensors are obtained by observing physical phenomena, and thus it is difficult to accurately and automatically generate the vibration values by using a program. Therefore, it is possible to detect cheating acts effectively even in the event that the technology of this invention is made open to the public.