Illustration: Benny Box

The Internet of Things requires a more stable network

Monday 12 Aug 19
|
by Morten Andersen

Contact

Lars Dittmann
Professor, Group Leader
DTU Fotonik
+4545 25 38 51

Narrowband-Internet of Things

The Watch4Life project is based on the NarrowBand-the Internet of Things (NB-IoT) standard, which several telecoms have started providing in Denmark.
The standard is tailored to support communication between machines.


In contrast to the 4G network, the NB-IoT is not suitable for the transmission of video and sound, which requires a large bandwidth.


With NB-IoT, you can stay within a single frequency bandwidth of 200 kHz. In turn, the assurance that the data will reach its destination is greatly enhanced.


Among other things, the emergence of a specific network for communication between appliances opens up for more patients to be treated at home. This benefits both quality of life and the economy.

Vital communication from measuring equipment on, e.g., patients who are treated at home but monitored by the hospital, risks being lost in between the streaming of movies and updates of operating systems on mobile phones. This is why things like telemedicine should run on a separate network.

A collaborative project between DTU researchers and three companies will develop equipment that sends health data from the patient’s home to the hospital with a very high level of security regarding the prevention of failures.

“It’s an illusion to think that we can run all the services we want on the mobile network. Telemedicine and other forms of Internet of Things tighten the requirements for the network’s quality significantly,” says Lars Dittmann, professor at DTU Fotonik. He is chairman of the steering committee behind the project, which has been named Watch4Life.

Telemedicine is good for the patients that can be treated in their own home. And it’s good for the economy, because you save a lot of money every time you prevent a hospital admission. It’s with good reason then, that many tests and experiments are carried out in this field. However, the quality of the network hasn’t been given enough attention, says Lars.

“Most projects develop equipment that can measure various biological parameters of the chronically ill patients in their own home. But so far, almost all solutions have been offline. I understand why that’s the case, because the overall network quality is actually too bad to justify going online—but we want to change that.”

More failures after transition to 4G

For most people it might come as a surprise that the common mobile network’s quality is too low for telemedicine—many have the idea that the network gets better each year. But the uptime on the network actually decreased after transitioning to the 4G network, which we have today.

Previously, all the telecommunication base stations had battery back-ups, which made them able to run for a few hours in situations where the power grid failed. 4G, however, requires a closer network of base stations, and as the new base stations were installed, large parts of the previous battery back-ups were discontinued.

“In so doing, the companies saved money, but it has resulted in less uptime,” says Lars Dittmann.

"Telemedicine and other forms of Internet of Things tighten the requirements for the network’s quality significantly."
Professor Lars Dittmann, DTU Fotonik

With the earlier 2G and 3G networks, the system would generally be online 99.999 per cent of the time. Today, the objective is that the system must be online 99.9 per cent of the time. This seemingly small difference means that the overall failures of a full year now amount to several hours rather than the earlier approx. five minutes.

“It’s actually a dramatic deterioration. The companies are likely pleased that our memories are generally quite short,” says Lars Dittmann and adds: “Maybe it’s good enough for Netflix and other entertainment, but when it comes to telemedicine, it can be disastrous if the network is down at an unfortunate time. At the same time, it is currently unclear who would be responsible if a patient is injured because the network has failed.”

Add to this the consequences for the economy:

“Even if you’re lucky and there are no consequences for the patient’s health, a network failure will naturally lead to uncertainty. We must remember that a patient who feels insecure can at any time ask to be admitted to a hospital. If this happens frequently, the economic rationale for telemedicine will collapse.”

A FIFA World Cup supplier

The actual development of the equipment for the telemedicine patients will be a collaboration between researchers at DTU Fotonik and the young Danish company Leikr.

The company manufactures electronics that help professional athletes. For instance, it was Leikr who supplied the technology that made it possible for the 2018 FIFA World Cup judges to know whether a ball was over the line. All the judges wore a bracelet with a display that would immediately show them the word GOAL when the ball was over the line.

So far, Leikr has produced systems that don’t run on public networks, but instead establish their own closed network—a new network called NarrowBand-Internet of Things (NB-IoT, see fact box). 

“There are so many active mobile phones and other communication going on in a football stadium, and we didn’t dare trust that we could use the mobile network without interruptions. But the emergence of the NB-IoT network means that in the future, we can offer solutions such as telemedicine without having to establish separate networks. That’s our motivation for entering the project,” says Lars Møller, Managing Director of Leikr.

Smartphone unsuitable

The new standard provides enhanced assurance that the data will reach its destination. It also uses far less power than the 4G network.

“In addition, it’s interesting that NB-IoT has a significantly better reach into buildings than the mobile network does. The patients who carry the telemedical equipment will more often than not be indoors where ordinary mobile phones can experience temporary losses of connection,” says Lars Møller.

Patients with the pulmonary disease COPD can benefit from monitoring their own body temperature, for example. If the temperature rises, it may be due to an infection—and if you realize it early, you can easily prevent the condition worsening.

Leikr develops the equipment the patients will wear, which is reminiscent of a digital wristwatch.

“Right now, it’s very hyped to develop solutions as apps for smartphones, but in the United States, the authorities are realizing that it may not be the best way to go when people’s health is at stake. The risk that the telemedical functions will fail when the neighbour’s daughter streams a movie over the mobile network, or when the manufacturer of your smartphone chooses to send you a control system update, is too great. Many apps simmer in the background on an ordinary smartphone, and you’re not always in control of what’s going on. The advantage of NB-IoT is that the traffic runs on a separate network, and with our device, we have 100 per cent control over the software on the device. An additional benefit is that no third-party app can listen in and intercept health information,” explains Lars Møller.

Quickly past the teething troubles

Leikr’s prototypes will be developed in close collaboration with researchers at DTU Fotonik. The research group uses both theoretical means, such as mathematical modelling and simulation, and testing of prototypes in the laboratory and in the field.

“Before NB-IoT will be able to be a part of applications such as telemedicine, where time and reliability are crucial, it is important to examine the technology’s robustness and scalability. We examine how well the technology works in challenging conditions, such as in basements, where common mobile signals are usually poor,” says associate professor Sarah Ruepp, who heads the project’s work package on IoT infrastructure for reliable communication.

In addition to Leikr and the researchers from DTU Fotonik, the project group comprises two large companies: TDC Group Denmark and Huawei Finland.

This adds an extra dimension, says Lars Møller, Director of Leikr:

“Technically, we could have one or more products ready to go within six months to a year. But telemedicine is a new market for us. There are many factors in relation to regulatory approvals and, not least, sales channels that can be difficult to overcome for a small business. We are therefore pleased that we can team up with larger companies through the Watch4Life project.”

Autonomous cars will require a better network

The perspectives of the Watch4Life project go much further than telemedicine.


“The Internet of Things will spread to a great many areas, and it will be crucial that the network is able to support the new applications. So far, it’s been the entertainment industry in particular that has driven the development of the Internet. And entertainment consumers may accept the fact that it sometimes takes a bit longer to download a film, but that won’t do with the Internet of Things,” says Lars Dittmann, professor at DTU Fotonik.


Today, the delay of the 4G system’s signal is approx. 200 milliseconds. That’s way too much for Internet of Things applications, emphasizes Lars Dittmann:


“You can imagine that two self-driving cars meet on the road. They’ll have to exchange a couple of messages in order to figure out which of them should give way. Before they finish, they have long since crashed into each other. Everyone will probably be able to understand this example, but it’s no less true for industry processes; if the different tools and systems must give way and wait for each other, the economy will quickly go west. You can also think of a wind farm, where a wind turbine runs out of control. Any delay increases the risk of the error spreading to the entire wind farm and perhaps leading to a major blackout. For this reason, we have to ‘slice’ the Internet, in order to ensure a higher quality for the most important types of applications.”

News and filters

Get updated on news that match your filter.