US 12,260,284 B2
Probabilistic contextual inference using RFID tag-interactions
Cagri Tanriover, Portland, OR (US); Rahul C. Shah, San Francisco, CA (US); and Chieh-Yih Wan, Beaverton, OR (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Dec. 16, 2020, as Appl. No. 17/124,091.
Prior Publication US 2021/0103708 A1, Apr. 8, 2021
Int. Cl. G06K 7/10 (2006.01); G06K 7/00 (2006.01); G06N 7/01 (2023.01)
CPC G06K 7/10366 (2013.01) [G06K 7/00 (2013.01); G06N 7/01 (2023.01)] 24 Claims
OG exemplary drawing
 
1. A computing device, comprising:
a communication interface configured to receive a series of time-varying receive signal strength indicator (RSSI) data values associated with each respective radio-frequency identification (RFID) tag from among a set of RFID tags located within a monitored environment; and
processing circuitry configured to, for the series of time-varying RSSI data values associated with each respective RFID tag from among the set of RFID tags:
calculate an energy of frequency components of the series of time-varying RSSI data values;
calculate an RFID tag energy summation characterization based upon a summation of the energy of the frequency components of the series of time-varying RSSI data values that are less than a threshold frequency value that indicates that variations in the time-varying RSSI data values correspond to physical movement of the respective RFID tag;
calculate, based upon the RFID tag energy summation characterization, a motion confidence level metric indicative of a probability that the respective RFID tag was subjected to physical movement during a span of time corresponding to the series of time-varying RSSI data values; and
provide, to a data analytics platform, the motion confidence level metric as part of an analytical dataset to enable the data analytics platform to determine which of the set of RFID tags were subjected to physical movement due to user interaction.