US 11,755,886 B2
Passive positioning with radio frequency sensing labels
Simone Merlin, San Diego, CA (US); Jamie Menjay Lin, San Diego, CA (US); Brian Michael Buesker, San Diego, CA (US); Rui Liang, San Diego, CA (US); Daniel Hendricus Franciscus Dijkman, Haarlem (NL); Farhad Ghazvinian Zanjani, Almere (NL); Ilia Karmanov, Amsterdam (NL); Vamsi Vegunta, San Diego, CA (US); Harshit Joshi, San Diego, CA (US); Bibhu Mohanty, Del Mar, CA (US); and Raamkumar Balamurthi, San Diego, CA (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Apr. 13, 2021, as Appl. No. 17/229,805.
Prior Publication US 2022/0327360 A1, Oct. 13, 2022
Int. Cl. G06N 3/00 (2023.01); G06N 3/045 (2023.01); H04B 17/318 (2015.01); H04W 16/20 (2009.01); H04W 64/00 (2009.01)
CPC G06N 3/045 (2023.01) [H04B 17/318 (2015.01); H04W 16/20 (2013.01); H04W 64/006 (2013.01)] 56 Claims
OG exemplary drawing
 
1. An apparatus for training one or more sensing models, the apparatus comprising:
at least one network interface;
at least one memory; and
at least one processor coupled to the at least one memory, the at least one processor being configured to:
receive, via the at least one network interface, radio frequency data associated with at least one wireless device, the radio frequency data being based on radio frequency signals reflected from a first object and received by the at least one wireless device;
obtain training label data, the training label data being based at least in part on the first object and input data; and
generate a sensing model based on the radio frequency data and the training label data, wherein the training label data is received in real-time from a labeling device, the labeling device comprises a camera and a simultaneous localization and mapping (SLAM) system to simultaneously generate the training label data with the camera and to determine a location of the labeling device, wherein the camera is further configured to track orientation of the labeling device and provide the location and the orientation of the labeling device to generate the training label data.