Ten years ago, the Internet of Things was symbolised by an intelligent fridge that can automatically order milk when you run out. What does it mean today?
Pratik Prajapati: Today, we are no longer talking about the Internet of Things but rather the Internet of Everything. Starting with simple appliances such as a fridge, right through to complex end devices such as robots or machines in the context of Industry 4.0 that are able to optimise their own processes. Most recently, the Internet of Things played a key role in stopping the spread of Covid-19. The wireless networking of the devices is mostly achieved with wide coverage on account of its especially energy-saving technology. Wireless IoT technology also helps us to save on resources in more remote regions and densely built-up innercity areas – for example, through sensor systems and intelligent control technology.
What do you need for an Internet of Things?
Claus-Robert Ziegahn: The basic components are devices that are connected to a cloud via a wired or a wireless connection. When we speak of wireless connectivity, we distinguish between cellular transmission technology, through which devices log on to controlled wireless cells (4G/5G) by means of SIM technology, and non-cellular transmission technology, such as Wi-Fi or Bluetooth. The transmitted data is usually processed centrally. However, there is currently a trend towards the wireless Edge solution, through which data is processed on the device itself by means of machine learning.
For example?
Pratik Prajapati: This could be a voice-controlled computer, such as Siri or Alexa, that responds to voice commands – and keeps learning whilst doing so. Or a camera in the jungle that can identify an elephant. After carrying out many observations, Artificial Intelligence can recognise irregularities that might, for example, indicate that an animal is sick or injured. Only these results are then sent to the cloud or to the internet.
What is the advantage of the Edge solution?
Claus-Robert Ziegahn: Only very little data needs to be sent. This means that the devices use less electricity and have a longer service life. With so-called low-power wide-area networks (LPWANs), data can be collected from hundreds of measurement points such as cameras or sensors via a gateway – the crucial point is the small quantity of data needed. There is a wide range of different transmission technologies, and the infrastructure required will be determined by the use case in question. We provide help with needs assessment, with the selection of the right components, set-up, testing and managing the roll-out.
What project are you working on at the moment?
Pratik Prajapati: As mentioned earlier, we are actually working on a project in Africa at the moment. With Accenture Development Partnership and Liquid Studios, umlaut has developed some highly advanced IoT sensors for “okala” which enable the monitoring of wild animals and ecosystems. This application requires that the sensors can be used in harsh, tropical conditions where there are only limited options in terms of electricity supply and connectivity. The solution we have developed is cost-effective and highly scalable. We want to measure the behaviour of wild animals in stressful situations, for example when there is a wildfire or when they are threatened by poachers. The vision: to use sensor technology and machine learning to detect danger before it becomes acute – and in the best case, to prevent this from happening.
What role does trust play here? How safe is information in the Internet of Things?
Claus-Robert Ziegahn: We authenticate the technology before proceeding and we are familiar with the encryption standards of the devices being used. Of course all the data that is measured and transmitted has to be interpreted by software and checked by the user. Ensuring faultless functioning is a part of our testing procedure. Yet trust can also be abused. If a sensor measures a silo level, for example, the operator must be able to trust in the fact that the data has not been manipulated. Blockchain technology could perhaps be helpful in this area in the future – to make sensitive data forgery-proof.
What areas will IoT be used for in the future?
Claus-Robert Ziegahn: We are currently focusing on applications that use wireless LPWANs. This is an area that is getting very interesting in relation to the smart city, in particular with the measurement of traffic volumes, air pollution and noise levels. Here, the software issues a warning in good time – allowing those responsible to take action to counteract the problem before the system is in danger of collapsing. There have also been some exciting developments with the tracking of consignments in the field of global logistics. Highly advanced chip technology and wireless data networks form the basis for applications that will considerably improve our lives and the way we live in this world.
Are there no white spots, i.e. areas with no network coverage for IoT?
Pratik Prajapati: Not for much longer. Telecommunications companies are already developing ways to allow LPWANs to run via satellite as well as via 5G. This will see the last of the white spots in network coverage disappearing. We estimate that the technology will be licensed and commercialised by the year 2025. The Internet of Things will then become the Internet of Everyone and Everything. In a current project for a customer, together with our colleagues from Accenture, we are already working on switching the transmission technology to satellite IoT as soon as this becomes available.
Many thanks for talking to us!
