Internet of Things – Latest news and information about the IoT

We provide a brief overview of the most relevant developments in the field of IoT

At IOTIQ, we carry IoT in our name. Accordingly, it is close to our hearts to share our enthusiasm for the Internet of Things and to make the topic more easily accessible to a larger audience. That's why today we're giving you a little update from the IoT sphere: What's new? What's coming in 2023 and beyond?

 

AI and IoT go hand in hand

 

It's estimated that more than 14.4 billion IoT devices will be connected to the Internet in 2023, and by 2025, it's estimated that there will be more than 25 billion. This means that the world of IoT is also generating more and more data. However, the mere collection and storage of this data does not create any particular added value. This can only be said to be the case if this data is also analyzed and converted into a form from which benefits can be derived. This is where AI comes into play. The combination of AI and IoT is also known as AIoT. AI technology combined with IoT infrastructure can contribute significantly to the analysis and utility of data and help improve the functionality of IoT devices. In a private context, this could mean, for example, that robotic vacuum cleaners learn how to better navigate their environment, thus virtually optimizing themselves on an ongoing basis.

For companies, AIoT offers promising opportunities to automate and optimize their processes. In addition, new products and services can be developed that are better tailored to the needs of customers. According to DMEXCO, intelligent robots could take on a central role in areas such as e-commerce, security and traffic and logistics in the future.

 

Digital twins

 

Have you ever heard of digital twins? This is a virtual model of a physical object. This physical object is equipped with sensors that measure various aspects of performance. The resulting data is transferred to the digital copy so that it reflects the behavior of the object. For example, simulations can be run to see how certain changes would affect the object. It can also be used to investigate performance issues and identify optimization opportunities. All of this can be figured out without having to change the object's physical state. What distinguishes digital twins from simple simulations is that they can themselves examine any number of simulations and, in the process, significantly more processes from a wide variety of angles. Their potential to deliver new valuable insights is thus significantly higher.


One should always be aware of what private data and information is collected by IoT devices and, in the worst case, could fall into the wrong hands.


New cybersecurity challenges

 

Like so many things, these developments unfortunately have their downsides. Bringing more and more devices into our homes and factories also means that there are more and more points of attack in our lives that could potentially be exploited by cybercriminals. This ranges from tapping user data to invading via a poorly protected device and infiltrating the entire network. Here, it is actually more of a disadvantage that more and more devices are being connected to the Internet. This is because a single weak link can put the entire network at risk. Depending on which IoT devices are then affected, this can have devastating consequences. Think of self-driving cars, for example, which can pose a major security risk to passengers and third parties, or the healthcare sector, where particularly sensitive data is handled.

But it goes without saying that this can also have very unpleasant consequences in private life. For example, criminals can use hacked devices in 'smart homes' to monitor the homes of their victims. Only recently, pictures of private moments recorded by an iRobot Roomba J7 caused a stir here and subsequently ended up on the Internet. The mission of these models from a special developer version was to collect and transmit recordings from their everyday work for training the AI used. To do this, vacuuming robots took footage of obstacles that got in their way on their cleaning rounds. The video footage was then analyzed by a service provider on behalf of iRobot. To help the vacuuming robots learn how to navigate their surroundings, employees were asked to label objects in the footage. One of these 'objects' was a woman on the toilet, photos of which later appeared in private Facebook groups. Apparently, the commissioned employees had screen-shot and published these and other scenes.

Strictly speaking, iRobot is not guilty of any wrongdoing. The test users were supposedly informed that the devices could take pictures at any time. The manufacturer also states that contracts were signed with the service provider to prevent such privacy violations.

 

The incident clearly shows that even precautions such as contractual agreements cannot guarantee complete protection against data privacy violations. At the same time, one should always be aware of what private data and information is collected by IoT devices and, in the worst case, could fall into the wrong hands. In this case, this point has particular significance, because the recordings were not made through unauthorized access, but to all appearances even with the consent of the test users. Since some users deny that they knew about the data collection, it seems questionable how explicitly this quite important function of the vacuum robots was communicated. Since you cannot always rely on all essential information being communicated transparently and in detail, it is unfortunately essential to inform yourself in detail.

 

The Internet of Things - Fun and Dangerous

 

As the IoT space grows, so does the need for legal control and regulation. After all, these days almost any device that has an on/off switch can theoretically be connected to the Internet and thus become a security risk. That's an armada of devices whose data could fall into the wrong hands.

Among other things, the EU is expected to introduce regulations in 2023 that will impose stricter rules on the collection and storage of data for manufacturers and operators of smart devices.

 

In addition to legal regulations, measures that users can take themselves will also help to improve cybersecurity and data security in the IoT. Of course, none of these can guarantee one hundred percent security. But every protective measure helps to minimize risks. For example, it's important to regularly update devices and software, as this is where security vulnerabilities are fixed. Another important point is changing default passwords on IoT devices. Smart home devices come with a factory-set password that is used to log into the network for the first time and, accordingly, is usually very simple. While this is convenient in further handling as well, it can create a serious security risk. This is because these simple passwords are very vulnerable to brute force attacks. High computing power is used to try out a large number of possible user password variants. With particularly simple passwords, the probability of hitting the bull's eye at some point is naturally much higher than with complex passwords.

The Wi-Fi network should also be protected by a strong encryption method. If you often have visitors who also use the Internet at home, you should consider setting up guest access. And as always, be careful when dialing into publicly accessible Wi-Fi. A VPN helps minimize the risks. (We've written more in-depth about how to protect yourself and your data in the digital world in this article).


Technologies such as edge computing and chiplets will accompany the growth of the IoT and play a crucial role in managing the huge amounts of data and making it usable.


Edge computing

 

As digitalization continues and the IoT evolves, larger and larger amounts of data are being created. One solution to handle these masses is edge computing. Here, the processing of data takes place close to its source - the edge of the network (the edge). In other words, the data is processed by the devices that aggregate it themselves, or at least in close proximity. The need for transfer to more distant data centers is minimized. As a result, data can be processed in near real time by avoiding the time it takes to transmit it. This enables near-delay-free transmission for augmented reality applications, for example. Another use case is autonomous driving. Here, data must be processed within fractions of a second and decisions made on the basis of this data, which in the most dramatic case could mean the difference between life and death. There must be no delays in this process because, for example, the Internet connection is inconsistent or the servers operate at a slower pace when they are very busy. Edge computing can work around these problems because the data doesn't have to be transported away from the vehicle at all for processing. (You can learn more about edge computing here).

 

Chiplets

 

Chiplets are another promising development. Processors are traditionally manufactured on a single piece of silicon. This is called a monolithic design. Moore's Law (an observational rule from 1965) states that the number of transistors in a silicon integrated circuit doubles about every two years. For about 50 years, this prediction was also true. During this time, an increasing number of functions were incorporated into the uniformly planned design. Gradually, however, the capabilities of silicon reached their limits. The manufacturing processes became more and more error-prone as the design became more complex. As a result, even the smallest defects can mean that chips either have to be sold with fewer cores or are even discarded.

Chiplets make it possible to produce several smaller chips with defined functions instead of a CPU on a single piece of silicon with the targeted number of cores. These functional units can then be combined and assembled 'LEGO-like'. The advantage is that a defective chiplet can be easily replaced and the entire design does not have to be disposed of. In addition, the material and manufacturing process of the chiplets can be optimized for their specific functions. Compared to classic SoC (system-on-chip), not only the performance but also the power consumption of the system improves.

An open industry standard for the interfaces through which chiplets can be connected makes it possible to combine components from different suppliers. (You can read more about chiplets here).

 

Conclusion

 

The world of IoT is growing every day. With it, exciting new opportunities for both businesses and individuals are constantly being added. Each new development promises not only increased efficiency and process optimization, but also greater convenience and enjoyment. At the same time, there is no question that every new territory also harbors dangers - cybersecurity and data security issues are therefore topics of fundamental importance for the further development, security and general acceptance of the IoT.

 

Are you looking for modern solutions for working in the digital age?