Introduction to Internet of Things

Since the late 1980s, the Internet has evolved dramatically. The latest stage is the use of this global network for communication with objects or between objects, a development known as the Internet of Things (IoT). The evolution of the IoT is rapid: since 2014, the number of connected objects has exceeded the number of connected humans, and it is predicted that 50 billion objects will be connected by 2020.

This article presents the essential characteristics of the Internet of Things.

It begins with an overview of the many types of application in which the IoT introduces intelligence: networks, healthcare, home, city, industry, TV, automotive, and so on. Then, the general architecture is detailed with its different levels: detection and identification of information, network construction, information processing and actions required for the application, to which corresponds the layered diagram: component level, network level, support level (service and application) and application level.

The range of fields involved is broad enough to innovate and create value. This article provides an overview of all the elements in the value chain: multi-technology gateways between objects and networks to connect as many objects as possible, telecoms infrastructures dedicated to the IoT, specific platforms, technology integration for customized solutions, connected object service operators.

Hardware components and their development environment are illustrated. They include :

• object-level identifiers, sensors, displays, actuators ;

• microcontrollers or low-cost processors and boards with or without light OS for simple objects, processors with OS such as Linux, Android, IOS and cloud resources for complex objects;

• networks, which can be "short-distance", "long-distance" or "cellular". The various network technologies that can be used are described in detail, according to constraints such as range, throughput, cost and security.

Managing an application comprising hundreds, thousands or even more objects requires a high-performance development medium (platform). Both proprietary and open-source platforms are available. They must satisfy a number of requirements:

• manage components and integration support ;

• information security ;

• define the data collection protocol;

• enable data analysis to obtain relevant information (Big Data)