US 10,257,678 B2 · Scalable data discovery in an internet of things (IoT) system

Abstract

Data discovery for sensor data in an M2M network uses probabilistic models, such as Gaussian Mixing Models (GMMs) to represent attributes of the sensor data. The parameters of the probabilistic models can be provided to a discovery server (DS) that respond to queries concerning the sensor data. Since the parameters are compressed compared to the attributes of the sensor data itself, this can simplify the distribution of discovery data. A hierarchical arrangement of discovery servers can also be used with multiple levels of discovery servers where higher level discovery servers using more generic probabilistic models.

Claims18

Dependent claims

In a communications network connected to at least one group of sensors, a method comprising:

storing, at a gateway of a plurality of gateways of the communications network, sensor data from the at least one group of sensors in the communications network;
generating, at the gateway of the communications network, a probabilistic model comprising one or more parameters which describe one or more attributes of the sensor data;
and
sending, to a discovery server of the communications network, the one or more parameters of the probabilistic model for use in data discovery,
wherein the discovery server, upon receiving a request for sensor data, is configured to determine, based on the one or more parameters of the probabilistic model, a probability that one or more of the gateways of the communications network has the sensor data stored thereon.

2 more elements

Dependencies are inferred from the claim text.

Scalable data discovery in an internet of things (IoT) system

Life of this patent

Scalable data discovery in an internet of things (IoT) system

Life of this patent

Abstract

Claims18