US 11,696,130 B2
Mobile phone authentication method using implicit authentication
Dae Seon Choi, Daejeon (KR)
Assigned to FOUNDATION OF SOONGSIL UNIVERSITY-INDUSTRY COOPERATION, Seoul (KR)
Appl. No. 17/256,230
Filed by FOUNDATION OF SOONGSIL UNIVERSITY-INDUSTRY COOPERATION, Seoul (KR)
PCT Filed Dec. 31, 2018, PCT No. PCT/KR2018/016966
§ 371(c)(1), (2) Date Dec. 28, 2020,
PCT Pub. No. WO2020/111377, PCT Pub. Date Jun. 4, 2020.
Claims priority of application No. 10-2018-0148513 (KR), filed on Nov. 27, 2018.
Prior Publication US 2021/0266740 A1, Aug. 26, 2021
Int. Cl. H04W 12/06 (2021.01); G06N 20/00 (2019.01); H04W 12/63 (2021.01); H04W 12/68 (2021.01); H04W 12/65 (2021.01); H04W 4/02 (2018.01)
CPC H04W 12/06 (2013.01) [G06N 20/00 (2019.01); H04W 4/025 (2013.01); H04W 12/63 (2021.01); H04W 12/65 (2021.01); H04W 12/68 (2021.01)] 7 Claims
OG exemplary drawing
 
1. A terminal authentication method using implicit authentication, the method comprising:
sending, by a server, an authentication number and a request for transmission of behavior data and environment information data to the user terminal;
by a server, receiving behavior data and environment information data from a user terminal for when a user checks an authentication number for user authentication of the user terminal;
by the server, detecting a start point of a behavior of the user terminal for checking the authentication number by performing peak detection in the received behavior data, and storing the behavior data from the detected start point; and
by the server, extracting feature data from the received environment information data and learning the extracted feature data to build a learning model,
wherein the behavior data is received and stored according to each posture of the user terminal classified in advance, wherein the detecting the start point of the behavior and the storing the behavior data from the detected start point comprises:
by the server, normalizing the received behavior data; and
calculating a mean value for a plurality of the received behavior data for each posture of the user terminal;
by the server, calculating a degree of accordance or discordance using dynamic time warping by comparing the behavior data from the detected start point with the behavior data according to the determined posture of the user terminal among behavior data stored in advance;
wherein the calculating the degree of accordance or discordance using dynamic time warping comprises by the server, comparing the behavior data from the detected start point with a mean value for a plurality of behavior data stored in advance for the determined posture of the user terminal;
by the server, calculating reliability by inputting the received environment information data to a learning model built in advance; and
by the server, determining whether the user authentication is successful based on the calculated degree of accordance or discordance and the calculated reliability.