US 11,989,965 B2
Cross-correlation system and method for spatial detection using a network of RF repeaters
Alireza Tarighat Mehrabani, Los Angeles, CA (US)
Assigned to AR & NS Investment, LLC, Newport Coast, CA (US)
Filed by AR & NS Investment, LLC, Newport Coast, CA (US)
Filed on Jun. 24, 2020, as Appl. No. 16/910,537.
Prior Publication US 2021/0405759 A1, Dec. 30, 2021
Int. Cl. G06V 40/10 (2022.01); G06F 3/01 (2006.01); G06F 18/25 (2023.01); G06V 40/20 (2022.01); H04W 4/70 (2018.01); H04W 8/00 (2009.01); H04W 64/00 (2009.01)
CPC G06V 40/103 (2022.01) [G06F 3/017 (2013.01); G06F 18/25 (2023.01); G06V 40/28 (2022.01); H04W 4/70 (2018.02); H04W 8/005 (2013.01); H04W 64/003 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A cross-correlation system, comprising:
control circuitry configured to:
input sensor data from a plurality of sensors into an input layer of a deep neural network;
train the deep neural network to:
cross-correlate the sensor data from the plurality of sensors at an input sampling stage;
solve for a relationship between the cross-correlated sensor data and identity of one or more users represented in the cross-correlated sensor data, and
track subsequent movements of the one or more identified users based on the cross-correlated sensor data;
obtain first sensor data of a first user from a radio detection and ranging system, wherein the first sensor data is a spatial point cloud data of a body of the first user;
detect a first portable device carried by the first user based on the first sensor data of the first user;
obtain second sensor data from the first portable device based on the detection of the first portable device of the first user, wherein the second sensor data includes device identity, location information, and signal strength information of the first portable device;
determine location coordinates of the first portable device based on the second sensor data, wherein the location coordinates include at least one of an angle and a distance of the first portable device;
utilize the trained deep neural network to cross-correlate the first sensor data and the second sensor data to:
obtain cross-correlated information of the first user,
identify the first user based on the cross-correlated information of the first user, and
determine a position of the first user based on the cross-correlated information of the first user, wherein coordinates of the spatial point cloud data are merged with the location coordinates of the first portable device;
recognize a first gesture specific to the first user based on the cross-correlated information of the first user;
identify a first controllable device from a plurality of controllable devices and a first action that is to be executed at the identified first controllable device, based on the first gesture; and
control the identified first controllable device to execute the first action based on the first gesture.