Google's New IoT Vision with Advanced Machine Learning capabilities
By apacciooutlook | Monday, December 03, 2018
FREMONT, CA: Google unveils new Internet of Things (IoT) vision with open source code and designs for proto type and pilot projects, enabling developers and manufacturers to offer better consumer products, reports Steven Max Patterson for Network World.
Google’s vision embodies IoT standards, open source frameworks and cloud services that are highly competitive in the market.
Machine Learning for IoT Networks
If This Then That (ITTT) rule engine controls and integrates IoT devices, and Google’s Tensorflow, an open source machine learning project, provides necessary tools to developers in building large and intelligent IoT networks. Tensorflow software library utilizes data flow graphs for numerical computation and its flexiblearchitecture allows deploying computation to one or more CPUs or GPUs in a desktop, server, or mobile device with a single Application Program Interface (API).
Bluetooth Light Energy (LE) beacons and cloud for Geolocation
Google has invested in outdoor generation research for more than a decade to build Application Program Interface (API) that creates apps with geographic position for the purpose of navigation and geofencing to locate people and things.
Google’s open source LE beacon, Eddystone offers information to everyone in the range and uses Bluetooth for direct and point-to-point communication. Eddystone increases the resolution in longitude and latitude or relative scale, from meters to centimeters within the building, hence bringing the indoors of a geolocation.
Eddystone uses broadcast –only radios and sends small packets of data with unique IDs, URLs, or application-specific data. Google provides a cloud based registry of information of each Eddystone beacon that can be used by app developers to store the meaning of each beacon, managing innumerable unique devices with Google’s large-scale systems experience and IoT experience.