US 12,341,664 B2
Network local connectivity demand prediction method and device
Etienne Coulon, Claix (FR); and Jean-Michel Cambot, Castelnau-le-Lez (FR)
Filed by LYTN SA, Montpellier (FR)
Filed on Apr. 25, 2023, as Appl. No. 18/306,406.
Prior Publication US 2024/0364597 A1, Oct. 31, 2024
Int. Cl. H04L 41/147 (2022.01); H04L 41/046 (2022.01); H04L 41/16 (2022.01); H04L 43/04 (2022.01); H04L 43/0811 (2022.01)
CPC H04L 41/147 (2013.01) [H04L 41/046 (2013.01); H04L 41/16 (2013.01); H04L 43/04 (2013.01); H04L 43/0811 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A network local connectivity demand prediction method, comprising:
providing a local agent with an initial prediction of a local connectivity demand during a duration, said local agent is associated to a physical network device that belongs to a computer network comprising at least a physical network device,
performing within a time period equivalent to the duration, more than one iteration of:
a step of aggregating data during an increment of time within said time period, said data includes values representative of, at least:
a connectivity demand of the computer network comprising at least a physical network device,
a purpose of use of the computer network comprising at least a physical network device,
a local connectivity demand prediction for the physical network device,
a step of training with said data a machine learning device to produce a machine learning model configured to provide a local connectivity demand prediction,
a step of deploying the machine learning model upon at least one local agent, and
a step of providing an updated prediction by updating the initial prediction in accordance with the deployed machine learning model operated by a local agent.