CC BY
5INTERNATIONAL SCIENTIFIC AND TECHNICAL CONFERENCE “DIGITAL TECHNOLOGIES: PROBLEMS AND SOLUTIONS OF PRACTICAL IMPLEMENTATION IN THE INDUSTRY”
APRIL 27-28, 2023
PROVISION OF THE RELIABILITY OF IoT SENSORS Muradova A. A.1, Gulturaev F. N.2
1 (TUIT named after Muhammad al-Khwarizmi, PhD, associate professor)
2(TUIT named after Muhammad al-Khwarizmi, 4th-year student)
https://doi.org/10.5281/zenodo.7851456
Abstract. The article presents the main problems of ensuring the reliability of the Internet of things IoT. The concepts of Internet of Things sensors, the reasons for the need to ensure the reliability of IoT sensors are given. The main results are shown and examples are given in which the requirements for the reliability of these elements in various fields of human activity are not met.
Keywords: IoT sensors, IoT, sensor reliability, reliability parameters, fault tolerance problem, redundancy, failures.
IoT (Internet of Things) or the Internet of Things is a system of physical objects (“things”) interconnected using built-in sensors, software, and/or other technologies. This connection is needed to transfer data to other devices in the system or to other systems over the Internet. Simply put, physical objects go online to send or receive information. The main advantage in the IoT market is cost. Therefore, priority is given to cheap but unreliable components. Unreliable devices break, make mistakes, freeze, and require maintenance. "If something doesn't work, it's already outdated." This is a quote from Canadian philosopher Marshall McLuhan that accurately describes the state of the art. Everything fails: computers freeze, smartphones slow down, elevators stop between floors, space probes go astray, and people make mistakes [1].
First mistakes. The topic of reliability, especially part of it - fault tolerance, is as big as security. The letter S in the term IoT is responsible for Security, and the letter R is responsible for Reliability - reliability. Human factor. It is more or less clear with hardware, errors are described and reproducible, but the biggest source of errors in IoT systems is a person.
On January 13, 2018, all residents of Hawaii received an alert on their mobile phones about a missile threat and that they needed to hide in a bomb shelter. It is not clear who made the mistake: the operator or the person who designed the interface. But if you look at the picture, the answer suggests itself. What to click on to trigger a test, not combat, missile threat alert? If you don't know the answer, you're wrong. The operator pressed the wrong button, and the mass mailing started. The system did not have any parameters by which it would be possible to prevent or confirm the sending: “Are you sure you want to warn about a missile threat?”. It took 30 minutes for the center's employees to realize what had happened and send a message stating that the attack was false.
Resiliency problem. The problem of fault tolerance for the Internet of things is very important because the number of devices is growing. According to a report by the consulting company McKinsey, in 2013 there were 10 billion IoT devices in the world, and by 2023 this number will grow to 50 billion. We simply cannot physically repair all these meters - there is simply not enough time. Systems that were designed to be serviced by people will not help us, instead, we will fix them [2,3].
Redundancy. IoT devices operate on the basis of information theory: there is a signal source, a receiver, an encoder, a modulator, a propagation medium, and an error source that interferes and distorts the real situation. A good way to reduce interference is to add redundancy,
INTERNATIONAL SCIENTIFIC AND TECHNICAL CONFERENCE “DIGITAL TECHNOLOGIES: PROBLEMS AND SOLUTIONS OF PRACTICAL IMPLEMENTATION IN THE INDUSTRY”
APRIL 27-28, 2023
with which we can detect a critical situation and level its effect of it: notify the operator or correct the error. An example of redundancy is the “STRIJ” network. Most devices on the network transmit without acknowledgment: the device emits a signal and the base station receives it.
Let's imagine a situation. We have a zone with interference, in which the probability of delivering a message to the base station is 90%, and the presentation is required to show no more than 1% loss. There seems to be a lot of work: fix protocols, and reduce range, but the quick and easy solution is redundancy. Next to the station that receives a signal with a delivery probability of 0.9, we put a second one with the same delivery probability, and the probability of failure of both stations at the same time is 0.01. The probability multiplication theorem applies here: the probability of failure of each station individually is 0.1, and the failure of both at once is just 1%, provided that the base stations are independent. This area between base stations will have the highest probability of reception [4,5].
Another way to demonstrate the principle of redundancy is the Watchdog Timer. This is a physical device that is built in by most processor manufacturers. If the Watchdog Timer does not receive a signal from the computer after a certain period of time, the device restarts the computer.
When using WT, it is not reliability that increases, but availability. The computer detects the problem, takes control action, and restarts the computer. This is very fond of NASA and knows many different ways to use the Watchdog Timer.
Below is an example of a multi-stage Watchdog Timer: when certain events occur, it sends NMI - a hardware interrupt that will be mandatory in the work on the processor. When some event occurs, Watchdog tells the computer: "Try to reboot itself, otherwise we will turn off the power." If the first timer fails, the second timer will work [6-8].
Redundancy works well within the operating system. This is how our base station works. It consists of various independent modules. The autonomy of modules prevents an error from getting from one module to another - a "pool" with errors is created, which we block. Higher in the hierarchy is a set of supervisors: scripts that monitor the situation according to certain parameters. For example, if the process is in the operating system, it is not a Zombie and does not flow from memory. The root element is the scheduler, such as cron.
The hierarchical structure creates good parameters in terms of system availability: if a module crashes, the supervisor sees it and reloads it, there is some redundancy in modules, and some modules perform the function of others [9, 10].
We realized that reliability is important because both people and equipment fail. You need to think about reliability the same way you think about security. For pilot presentations, turn on Watchdog, add redundancy, and simplify to make it impossible to make mistakes. Think about how to move into conditions where the system is guaranteed to work. Now let's move on to the last method, which is different from the rest, and techies often ignore it.
Reliability and cost of error
Home internet of things. If the kettle cannot be turned on remotely, the user will have to turn it on manually - this is not so comfortable, but not too scary. At home, there are far fewer critical tasks and there are rarely situations in which a breakdown will lead to disaster. The cost of a mistake in domestic use is low, so reliability in IoT, although important, is not as critical as in IIoT.
INTERNATIONAL SCIENTIFIC AND TECHNICAL CONFERENCE “DIGITAL TECHNOLOGIES: PROBLEMS AND SOLUTIONS OF PRACTICAL IMPLEMENTATION IN THE INDUSTRY”
APRIL 27-28, 2023
Industrial Internet of Things. If the sensor does not work in production, the business can lose millions: the machine will break, the goods in the warehouse will deteriorate, and valuable cargo will be lost. Therefore, in industrial IoT platforms, first of all, attention is paid to reliability: they develop fault-tolerant systems, duplicate important sensors, and install backup power supplies in case of a power outage. It can be argued that due to the breakdown of home smart devices, a person can flood neighbors or lose property in a fire. But even such losses are incomparable with the catastrophes that can be caused by the disruption of the IIoT system: damage in the billions of rubles or environmental disasters.
Complexity of algorithms
Home internet of things. His IoT devices are rarely truly smart - they usually respond to commands and simple "annoyances" like rising temperatures. And the platforms that unite them usually just give commands - complex data analytics is not needed in everyday life. There are exceptions, for example, smart speakers that can support dialogue. But even they are far from the complex systems of artificial intelligence, analytics and machine learning that are used in industry.
Industrial Internet of Things. Production tasks require more complex solutions to model the production chain, identify risks, collect and analyze information from many different sensors. IIoT systems use complex technologies to store and analyze information: big data, artificial intelligence, and machine learning algorithms, so it’s more correct to call them smart. The sensors themselves in IIoT are also often more complex: they can analyze several different physical indicators and compare them with each other. For example, a dew point sensor must be able to analyze pressure, humidity and temperature [11].
Information processing speed. Home internet of things. Home smart things respond to commands in a few seconds, in domestic conditions this is acceptable. If you gave the command to boil the kettle, you will hardly notice if it turned on in one or three seconds - this is not essential for household chores. Industrial Internet of Things. In industry, technological processes often require responses in milliseconds. Sensors must constantly capture up-to-date information, devices must instantly respond to commands.
Internet of Things Security. Home internet of things. Smart home devices access the Internet through regular channels. The information they transmit requires only basic protection against viruses and cyber threats. Even if a massive cyber attack happens, it will not lead to a catastrophe. In addition, home devices are usually produced in large quantities, and excessive protection will increase the cost of devices too much - no one will buy them.
Industrial Internet of Things. In business and industry, the situation is much more serious - attackers can try to gain access to the smart devices of an enterprise, which threatens millions in losses, and sometimes an industrial disaster. Therefore, IIoT networks pay increased attention to security.
The following methods are used for protection: Isolation from the Internet, work only in the network inside the office or factory. Encryption of industrial data, for example, the AES-256 algorithm. To crack it, you need to pick up a key with a length of 256 bits - even a modern supercomputer will take more time than the estimated age of our universe. Protection of software and operating systems. Intrusion detection and prevention systems - they notify if someone else tries to gain access to the system [12].
INTERNATIONAL SCIENTIFIC AND TECHNICAL CONFERENCE “DIGITAL TECHNOLOGIES: PROBLEMS AND SOLUTIONS OF PRACTICAL IMPLEMENTATION IN THE INDUSTRY”
APRIL 27-28, 2023
The amount of data collected. Home internet of things. Home devices generate very small amounts of information: in a year it will not accumulate even several gigabytes. Storage and processing of this information do not require bulk storage and special complex tools. Industrial Internet of Things. Sensors on equipment can generate gigabytes of information every day. All this must be quickly collected, recorded in databases, stored and, if necessary, retrieved for analysis. For this, enterprises build entire systems for working with Big Data and use special approaches to data analytics. Often home smart devices store data directly inside themselves or in the memory of the smartphone from which you control the smart home. For IIoT, separate storages of huge volumes are needed. Companies often use the cloud for this.
Conclusions
The future of the Internet of things is unreliability: IoT is aimed at mass markets, and for the mass market, price is the decisive factor. This means that the Internet of Things will consist of many cheap and unreliable devices. As the number of devices grows, so will the number of failures. We simply do not have enough hands to correct all the mistakes. Therefore, the only way - the devices must independently deal with the consequences of failures. These are autonomous systems that must learn to fix themselves. We propose to tackle the topic of reliability and learn how to show cool pilots using three methods: simplify everything you can, add redundancy, and create conditions under which the pilot is guaranteed to work. Do not forget that we are all people and are guided not by logic, but by feelings, so create beautiful projects. To delve into the topic of reliability, let's start with an article on operability, reliability, safety, and then we will study the experience of the NASA Jet Propulsion Laboratory. They created Voyager and Curiosity and know everything about reliability. Let's be inspired by the greats.
REFERENCES
1. R. Agarwal, and L. Das, M. “RFID Security in the Context of “Internet of Things,” First International Conference on Security of Internet of Things, Kerala, 17-19 August 2012, pp. 51-56. http://dx.doi.org/10.1145/2490428.2490435.
2. E. Biddlecombe, “UN Predicts “Internet of Things,” July 6, 2009.
3. R. Lombreglia, “The Internet of Things,” Boston Globe, October, 2010.
4. A. Reinhardt, “A Machine-to-Machine Internet of Things,” 2004.
5. M. Gigli, and S.Koo, “Internet of Things, Services and Applications Categorization. Advances in Internet of Things,” 1, 2011, pp. 27-31.
6. ITU Internet Reports, International Telecommunication Union. The Internet of Things: 7th Edition, 2005.
7. B.A. Li, and J.J. Yu, “Research and Application on the Smart Home Based on Component Technologies and Internet of Things,” Procedia Engineering, 15, 2011, pp. 2087-2092, doi.org/10.1016/j.proeng.2011.08.390
8. F. Razzak, “Spamming the Internet of Things: A Possibility and its probable Solution,” Procedia Computer Science, 10, 2011, pp.658-665, doi.org/10.1016/j.procs.2012.06.084
9. W. Shao, and L. Li, “Analysis of the Development Route of IoT in China,” Perking: China Science and Technology Information, 24, 2009, pp. 330-331.
10. C.Sun, “Application of RFID Technology for Logistics on Internet of Things,” 2012.
INTERNATIONAL SCIENTIFIC AND TECHNICAL CONFERENCE “DIGITAL TECHNOLOGIES: PROBLEMS AND SOLUTIONS OF PRACTICAL IMPLEMENTATION IN THE INDUSTRY”
APRIL 27-28, 2023
11. X.-Y. Chen, and Z.-G. Jin, “Research on Key Technology and Applications for the Internet of Things,” PhysicsProcedia, 33, 2012, pp. 561-566, doi.org/10.1016/j.phpro.2012.05.104.
12. J.Naveen, “3 Ways In Which IoT Reliability Can Be Improved,” Computer Vision and AR. Web 3.0 , Hyperautomation, September, 2022.
