What is the Internet of Things?
The internet we are most familiar with is the Internet of People. You and I interacting, either anonymously or personally via web pages, social networks, or whatever mechanism suites our purpose. The Internet of Things is primarily concerned with interactions between devices of various sorts. Examples include medical devices (pacemakers), domestic devices (heating and refrigeration for example), infrastructure devices (traffic light controllers, water metering, video surveillance), manufacturing devices, environmental – and so on. It can be characterized as a machine-to-machine Internet, instead of a person-to-person network.
Businesses are particularly interested in the Internet of Things (IoT) because it may provide them with more information on customers, which in turn will allow them to offer more relevant products in a more timely manner. A prototype is already being tested in a collaboration between Microsoft and Miele, the luxury white goods manufacturer. A refrigerator with sensors may detect the contents of a fridge, suggest replenishment, and possibly automatically create orders for a delivery from a preferred supermarket. An oven might have sensors to detect cooking patterns and suggest more suitable heat settings and duration. The applications are only limited by the imagination.
To achieve all of this it will be necessary to integrate several layers of communication. The devices need to communicate with some form of centralized data collection facility, and to achieve this a layer called The Web of Things is needed (WoT). This will provide the software infrastructure necessary for the communication to take place, just as the World Wide Web allows people to communicate via web pages.
Changes will also be needed in the structure of resource locator codes. While there are billions of people, there will be trillions of devices, and the current structure of resource location codes will not cater for these numbers.
Once a device can communicate with the Internet, and its data collected, organizations need to put in place processing facilities that can handle the data volumes and analyze them for meaning. One mechanism is streaming data, where the streams of data coming into a centralized data facility are analyzed in real-time for almost immediate responses. And so a sensor in a manufacturing plant that is signaling an alarm needs immediate response, as does a signal from a motor vehicle that indicates an imminent failure. The other genre of processing is data mining, where patterns of behavior are identified so that a more optimal response can be made. A traffic regulation system for example might learn that the optimal configuration of traffic light behavior on Saturday morning can be improved by giving preference to traffic flowing in a particular direction.
So there are two aspects to the Internet of Things. The first is the communication between devices and centralized data processing, and the other is the act of analysis once the data is received.
At the present time there is a great deal of confusion in this area (inevitably), with various communication standards being proposed, and new technology vendors providing their own infrastructure as a proprietary solution. New languages and tools for application developers are also appearing, and so it’s something of a brave new world.
The Internet of Things is inevitable, but concerns have been raised over security and personal freedoms. Nonetheless IoT will happen and various pundits have given estimates in the hundreds of billions of devices and sensors connected during the coming decade. The Internet of Things is going to be much bigger that the Internet of People.