Internet of Things, IoTs, is a ubiquitous presence in today’s world, Google for it and you get numbers ranging from 20 billions to over 50 billions as of 2020. 30 billion difference in the estimate is no peanuts and it reflects both the difficulty in counting them as well as the fuzziness in their definition.
When you say that a “Thing” is a part of the IoT it implies that it is connected to the Internet. However, how do you define “connected”? Does it need to have a direct link, plus a unique address or a unique identity will be sufficient? Is a device like a smartphone an IoT? It has connectivity, of course, and it has a unique address. What about the accelerometer in the smartphone, or its microphone or the temperature sensor, the compass, … they all are connected to the internet, although via the smartphone and they can all be uniquely identified. We have over 3 billion active smartphones in the world as of 2020, each of them with at least 10 sensors. Should these be counted as IoT? That will make for over 30 billion IoT just by themselves…
As you see the landscape is fuzzy but the overall message is clear: IoTs are playing a significant role in todays’ and, even more so, tomorrow’s network and service areas. What is notable in IoT is their huge variety in terms of communications requirements. You have billions of them that will send just a few bytes once in a while, other billions that are streaming, continuously, high-def video, you got billions that can wait for quite a bit of time to send their data and others that need to get reaction within few ms.
There are IoTs that connect directly to the network (and to a data processing centre) other that process data locally and only send polished up information, with plenty in between that form local cluster where raw data are processed.
There are IoTs engaged in one way communications (like sensors) others that are entertaining two ways communications, partly asynchronously and partly via hand-shaking.
5G with its mm waves can use very small antennas (the size of an antenna is related to the wavelength it has to manage) hence it can be ideal for small size IoTs, 6G with sub-mm wavelength can be even better! Additionally, the smaller the cell, the less power required for data transmission, an important bonus in many IoT applications (notice that 4G, and of course later evolutions, supports native IP, thus reducing the need to process a more complex stack, hence decreasing power demand on the IoT side).
IoT will keep growing becoming a “dense” presence in any ambient. Part of them (a significant part actually) will be embedded in products (as they are today in smartphones and in vehicles) and the overall trend is towards open data framework and public API allowing third party services to access the data in the IoT and “created/harvested” by IoT. With 6G one can assume that part of these IoT will become points (sources) forming a massively distributed data centre at the edges (ambient data centre). Processing power and local artificial intelligence are likely to create ambient awareness and support fully autonomous systems (see next).