US 12,244,507 B2
Intelligent data forwarding in edge networks
Francesc Guim Bernat, Barcelona (ES); Ned M. Smith, Beaverton, OR (US); Kshitij Arun Doshi, Tempe, AZ (US); Suraj Prabhakaran, Aachen (DE); Timothy Verrall, Pleasant Hill, CA (US); Kapil Sood, Portland, OR (US); and Tarun Viswanathan, El Dorado Hills, CA (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Sep. 25, 2020, as Appl. No. 17/033,140.
Prior Publication US 2021/0021533 A1, Jan. 21, 2021
Int. Cl. H04L 47/28 (2022.01); H04L 47/74 (2022.01); H04L 47/78 (2022.01); H04L 49/15 (2022.01)
CPC H04L 47/28 (2013.01) [H04L 47/746 (2013.01); H04L 47/788 (2013.01); H04L 49/15 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A system for intelligent data forwarding in edge networks comprising:
at least one processor; and
memory including instructions that, when executed by the at least one processor, cause the at least one processor to perform operations to:
receive a request from an edge user device for a service via a first endpoint;
calculate a time value using a timestamp of the request;
train a movement prediction model using edge user device tracking data, wherein the trained movement prediction model produces a first probability that the edge user device is in a vicinity of the first endpoint and a second probability that the edge user device will move into a vicinity of a second endpoint, the trained movement prediction model based on historical location and time data for the edge user device;
predict, using the trained movement prediction model, motion characteristics for the edge user device using the time value;
evaluate the motion characteristics for the edge user device using the trained movement prediction model to determine the first probability that the edge user device will be in a vicinity of the first endpoint and the second probability that the edge user device will move into a vicinity of the second endpoint;
transmit a response to the request to the first endpoint and to the second endpoint based on the second probability being above a first endpoint probability threshold, the second endpoint determined based on the second probability.