The giant business model of the Internet of Things I)Introduction to the IoT, political requirements

Since years I follow up the continuous development of the increasing application of computer control in the things we are surrounded with. In the start of my professional career beginning of the nineties when I was working with a design team for white goods we implemented simple chip computers into heating devices to perform complex control procedures. However the cost of these solutions superseded all mechanical control options. It was perceived modern by the customer to have a stepless temperature control or a simple led-display showing the actual preset value of the device. In these times as today the architectures applied were just older versions of actual chip-families now built cheap in mass production. There was no problem during the times where every device controller was kept as a single application.

I learned first about IoT application when I read an article about a fridge that was able to order missing content by itself. So a new revolutionary application dropped in: the network connection. Even older chip families offer today the (cheap) access to the internet. Be it by radio connection or wired. A new option was born: Things were enabled to communicate with other things. I was fascinated about all the options and started to build own small applications in my home. To not overly stress my budget, I focused on getting lots of functionality cheaply. Challenged me in learning all about new buses and radio protocols that could increase functionality or ease application of the things in my home. With this kind of personal hobby, I achieved a great insight to the upcoming challenge in the future. Recently I bought a complete computer with new architecture for $8. This preconfigured fantastic gadget is just as big as a thumb but offers the possibilities of a desktop computer, with Network access, radio access, low energy consumption, eprom for code storage and so on.

Let me tell a small story to show the basic options in this new world:

Some years ago I implemented an electric door opener into my home system. I glued a China sourced RFID receiver into the kitchen door and connected it to the door opener via my home-software. My youngest just needed to carry a chip when she went to school. Should it be lost – no problem she would just take a new one, the old one could be deleted from the access list. My home software also offers a web front end where all the applied functions can be displayed graphically.

One day she forgot to carry her key and found that nobody was in the house when she returned from school. She sent a message into our families chatgroup, that she was sitting outside in the cold. I received her message, opened the VPN and let her in. I even registered that she made her way in, because the movement sensor in the living room activated. I need to tell, that the meeting was happening in Shanghai China. None of my colleagues noticed my activity and the meeting was not interrupted. Learning out of this the next development of my home system will be a mobile phone detection system. Once a phone is in the range it activates the option of door opening within an applet-program that opens automatically in the phone.

As this kind of steps – implemented within a nice and easy programming environment – are set easily, a fast and steady system-specific development makes its way. Especially the fact that those systems are developed by combination of very different sensors and actors offers tremendous options.

So let’s now go a little further. All these little connected computers sitting inside whatever products are individually programmed or run based on very different operation systems. Any and Each of this devices are individually controllable and accessible. But how do we get these things into a big synchronised service oriented orchestra?

Actually this is the problem. There are no real standards set so far. The range of possible interactions is so wide, that it is really hard to find a way of standardization. In subsystems there are some standards defined today but proprietary behaviour and isolating marketing structures are still obstacles to overcome.

To simplify a little the IoT is a combination of sensors and actors. Now every device can have multiple sensors and actors. Ideally every sensor and every actor are now a freely accessible entity inside the network. In consequence we are in the world of sensing and acting. To design the ideal system any sensor and actor in the world are accessible in a very standard way. There will be an authorization system layer necessary to protect data and avoid intrusions. So every THING would offer a common interface to the world. As there is no physical limitation the service systems an input of a sensor could be used in an unlimited manner (like i.E. a datastream from a webcam) and in an unlimited amount of applications. So the signal from the local energy counter is used by the home system to control energy flow whilst in the same time it is used to monitor the consumption of the energy in the whole village and by the stock market for energy to settle pricing. So why to spend lots of money on many redundant sensors – let’s just put a sensor everywhere. This could be one of the endless business options of the future

Today we are far from this vision. Political will needs to be applied to regulate and develop the service-layer of the system in a way similar to the disconnection of infrastrucural networks from a supplier. Each supplier gets the same chances to offer services to the network. The service layer represents the network. The content (the bots the sensors the actors) is supplied thru various business models. So the service layer authority has to organize the interface standards – approve new applications and to keep the service levels of data supply. Clearly defined rules and boundaries would offer a gigantic business option for a society that is fast enough to focus the policy of IoT.

Every second article I read about IoT handles the issue of security. And indeed this gets probably the biggest challenge in the next future. We may smile when we read about hackers changing the lightning arrangement on a façade, or accessing lots of IoT LED bulbs to tint a whole quarter in pink or green. But once we are admitting to influence many things in our lives by such systems this is rather scary.

Because of the fact, that the communication access is not regulated many of our systems are today open to various attacks or discontinuities. Especially cost critical applications in home appliances or industrial control contain old hardware combined with old software. This can be changed by creating a special service layer interface with a need of regular update and technical approval. Like in public infrastructure networks, the service level layer needs strict regulation. A newly created service layer authority will have a lot of work to do by creating and approving devices safety standards, common criticality layers and certifiying data trust levels i.E. for sensor data. Once implemented this development could be an enormous driver for a completely new industry.

So my vision is quite clear now: We need to provide a politically set basis of regulation and standards for the service layer. To set a business environment why not making Germany a hotspot for this technology?

Bernhard Rau   Altenlotheim, 16.2.2017