US 11,856,425 B2
Automatic characterization of AP behaviors
Javier Cruz Mota, Assens (CH); Jean-Philippe Vasseur, Saint Martin d'Uriage (FR); Pierre-André Savalle, Rueil-Malmaison (FR); and Grégory Mermoud, Venthône (CH)
Assigned to Cisco Technology, Inc., San Jose, CA (US)
Filed by Cisco Technology, Inc., San Jose, CA (US)
Filed on Jan. 20, 2021, as Appl. No. 17/152,917.
Application 17/152,917 is a continuation of application No. 15/620,109, filed on Jun. 12, 2017, granted, now 10,917,803.
Prior Publication US 2021/0144572 A1, May 13, 2021
Int. Cl. H04W 24/08 (2009.01); H04W 24/02 (2009.01); H04W 88/08 (2009.01)
CPC H04W 24/08 (2013.01) [H04W 24/02 (2013.01); H04W 88/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, at a device, observed access point (AP) features of one or more APs in a monitored network;
clustering, by the device, the observed AP features within a multi-dimensional latent space defined by a model to form AP feature clusters, wherein the observed AP features include data regarding an observed state of one or more wireless APs in the monitored network and the AP feature clusters are indicative of a plurality of APs that have similar AP behavioral patterns, and wherein the clustering of the observed AP features within the latent space comprises representing sequences of consecutive points in the latent space, which correspond to a single AP of the one or more APs, as trajectories;
applying, by the device, user defined labels to the AP feature clusters within the latent space; and
using, by the device, the user defined labels to the AP feature clusters to describe future behavioral patterns of the one or more APs in the monitored network.