US 12,394,295 B2
Systems and methods for low power consumption geofencing
Shyam T Shyamalan, Hillsborough, NJ (US)
Assigned to Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed by Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed on Oct. 16, 2023, as Appl. No. 18/380,555.
Prior Publication US 2025/0124777 A1, Apr. 17, 2025
Int. Cl. G08B 21/02 (2006.01); G06F 30/27 (2020.01)
CPC G08B 21/0261 (2013.01) [G06F 30/27 (2020.01); G08B 21/0269 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
generating, during a learning phase, a model to determine whether a device is within a geofence, wherein the generating the model comprises:
in response to receiving a first lookup location request, determining a first device location using location identification logic, wherein the location identification logic includes a relatively high-power location determination facility and a relatively low-power location determination facility, and wherein the relatively high-power location determination facility is used to determine the first device location,
determining whether the device is within the geofence based on the first device location,
receiving first radio signal information associated with the relatively low-power location determination facility, the first radio signal information not used by the relatively high-power location determination facility, and
recording the first radio signal information and whether the device is within the geofence in a training data structure, and
performing a training operation using the training data structure to create the model;
in response to receiving, after the learning phase, a second lookup location request, utilizing the model to determine whether the device is within the geofence by:
using second radio signal information measured by the device as model inputs to generate a model output that indicates whether the device is within the geofence, the second radio signal information not used by the relatively high-power location determination facility, and
determining a confidence level associated with the model output; and
if the confidence level is below a threshold, determining a second device location using the location identification logic, and determining whether the device is within the geofence based on the second device location.