US 12,155,725 B2
Sensor node, server apparatus, identification system, method and program
Yasue Kishino, Tokyo (JP); Yoshinari Shirai, Tokyo (JP); Shin Mizutani, Tokyo (JP); Takuma Otsuka, Tokyo (JP); Takayuki Suyama, Tokyo (JP); and Futoshi Naya, Tokyo (JP)
Assigned to NIPPON TELEGRAPH AND TELEPHONE CORPORATION, Tokyo (JP)
Appl. No. 17/424,463
Filed by NIPPON TELEGRAPH AND TELEPHONE CORPORATION, Tokyo (JP)
PCT Filed Jan. 15, 2020, PCT No. PCT/JP2020/001108
§ 371(c)(1), (2) Date Jul. 20, 2021,
PCT Pub. No. WO2020/153202, PCT Pub. Date Jul. 30, 2020.
Claims priority of application No. 2019-009586 (JP), filed on Jan. 23, 2019.
Prior Publication US 2022/0109727 A1, Apr. 7, 2022
Int. Cl. H04L 67/12 (2022.01); G06N 20/00 (2019.01); G16Y 40/10 (2020.01); G16Y 40/35 (2020.01)
CPC H04L 67/12 (2013.01) [G06N 20/00 (2019.01); G16Y 40/10 (2020.01); G16Y 40/35 (2020.01)] 5 Claims
OG exemplary drawing
 
1. A sensor node connected to a server device through a communication network, the sensor node comprising a processor configured to execute operations comprising:
acquiring a piece of sensor data from a sensor in the sensor node;
determining, based on an identification model, a class of the piece of sensor data, wherein the class is a part of a plurality of classes to identify occurrence of a predetermined event using the identification model created by the server device in advance;
calculating a degree of certainty of the piece of sensor data being in the determined class;
selecting the piece of sensor data among a plurality of sensor data from a transmission buffer, wherein the degree of certainty of the piece of sensor data indicates a minimum degree of certainty of being identified in the class of the plurality of classes among the plurality of sensor data stored in the transmission buffer, and the transmission buffer stores the sensor data and the degree of certainty of the sensor data in association with each other; and
transmitting the selected piece of sensor data as training data to the server device for training the identification model.