US 12,438,526 B2
Analysis device
Hiroto Akatsuka, Chiyoda-ku (JP); and Masayuki Terada, Chiyoda-ku (JP)
Assigned to NTT DOCOMO, INC., Chiyoda-ku (JP)
Appl. No. 17/996,367
Filed by NTT DOCOMO, INC., Chiyoda-ku (JP)
PCT Filed Mar. 24, 2021, PCT No. PCT/JP2021/012303
§ 371(c)(1), (2) Date Oct. 17, 2022,
PCT Pub. No. WO2021/229921, PCT Pub. Date Nov. 18, 2021.
Claims priority of application No. 2020-085283 (JP), filed on May 14, 2020.
Prior Publication US 2024/0048127 A1, Feb. 8, 2024
Int. Cl. H03H 17/02 (2006.01)
CPC H03H 17/0257 (2013.01) 8 Claims
OG exemplary drawing
 
1. An analysis device that inputs observation data related to an observation target and outputs estimated observation data, the analysis device comprising:
processing circuitry configured to
input the observation data which is data obtained from mobile phone operation data that provides a distribution of people at a certain time, wherein the observation data is repeatedly input after a predetermined unit of time elapses;
generate first sequence data by applying a first conversion to the observation data;
obtain a current state variable by performing state estimation using a Kalman filter based on a state variable, the state variable being generated for the first sequence data, a state space model, and each element included in the first sequence data based on the state space model, the state space model being obtained by modeling a time-series variation of each element included in the first sequence data;
generate second sequence data by applying an observation process using an observation matrix of the state space model to the current state variable;
generate the estimated observation data by performing an elaboration process of correcting each element included in the second sequence data under a predetermined condition so that a value of each cell of the estimated observation data does not become a negative value and the number of cells having a value other than 0 included in the estimated observation data is suppressed;
apply a second conversion that is an inverse conversion of the first conversion; and
output the estimated observation data.