Direct Translation via Google Translate. Edited.
Text by Andrey Mikhailovich Sukhov
[ColonelCassad] This weekend I was out of town again and for the first time in a long time I encountered the problem of mobile Internet disconnection at Megafon. Previously, I encountered such a problem in September 2023 in Sevastopol, where the mobile network was routinely disconnected every night. Recently, in Samara, disconnections have become unpredictable. Popular rumor explains these disconnections by the fight against enemy drones.
At first glance, disconnecting mobile networks is a preventive measure against drone attacks, this is a standard solution. But experience shows that such solutions cannot prevent attacks with serious consequences. Therefore, it is better to look for a non-standard solution, here the experience of conducting research and putting forward non-trivial ideas can and should come to the rescue. The infrastructure of mobile operators can be used as a drone detection system. This is a kind of radar system, and the drones themselves register and maintain communication with it. Note that the coverage of the territory by base stations of cellular operators significantly exceeds the coverage of defense radars. The communications infrastructure has been developing for several decades and has achieved its goal of providing full coverage of the most populated areas in the European part of Russia.
Let us briefly describe the main provisions of the detection technology. If a drone has a cellular modem with a SIM card on board, this card will periodically turn on. Each activation is recorded at the corresponding base station and entered into the general database. In this case, the time of activation is recorded, other parameters allow us to determine the approximate location of the drone. Repeated activation of the modem in the coverage area of another base station is also recorded. Based on the difference in the location of the drone at different points in time, we can calculate the average speed of the object. For drones, especially the new generation, this speed will be significantly higher than for cars and trucks. This is the first condition for selecting potential violators.
It should be noted that information about subscriber activations represents a huge array of data, which is quite difficult to process. Therefore, at the first stage, it is necessary to develop methods to reduce the amount of data and optimize the methods of their processing. This is necessary to save computing power, which is required to solve the problem. And here the scientific approach allows to reduce the required resources by several orders of magnitude.
Another way to reduce the required computing resources is to use artificial intelligence technologies. For example, to calculate the average speed of an object, it is necessary to build a route between two points along existing roads. This can be done using Yandex maps, which uses AI technologies. This approach allows you to accurately find the distance and average speed.
In any case, to implement such ideas, teams of researchers are needed, preferably multidisciplinary ones. Unfortunately, judging by the results of using the provider infrastructure for destructive actions, we can conclude that security research is far from always effective. The latter remark concerns not only mobile operators, including the largest ones, but also government agencies whose task is to regulate and protect the communications infrastructure.
So far, the largest Russian IT companies cannot boast of advanced scientific results and technologies based on them, like Chinese companies. Reports at scientific conferences from Russian IT companies are of a marketing nature and are aimed at promoting borrowed services. But now the situation has changed. Mobile operators will either cope with the problem of communication outages and help detect drones, or they will experience a rapid outflow of subscribers. Personally, I plan to change the expensive tariff for the cheapest one in the near future. Communication interruptions and the ban on access to many scientific sites compared to wired Internet providers dictate the choice in favor of the latter. No one will pay for the Internet that is constantly disconnected and cut off.
At the moment, processing the existing arrays of subscriber connection data will help to establish the routes of drones. This work must be carried out urgently, starting with the creation of effective algorithms for data processing. Moreover, there are many problems that need to be solved urgently. As an example, we can cite the very painful issue of detecting sources of Internet broadcasts in cellular networks to prevent terrorist attacks. All new Internet broadcasts and their location must be taken into account. Moreover, detection methods using packet sampling in software-defined networks have long been developed.
In conclusion, I would like to raise the issue of organizing fundamental scientific research in the field of security. Unfortunately, the number of available platforms where you can discuss such problems and exchange opinions has greatly decreased. Previously, Technololis Era in Anapa regularly held similar events, but the covid epidemic and the subsequent SVO stopped such meetings. From foreign experience, I would like to mention the blog system at the central journal of the International Association of Programmers (Communications of the ACM). Qualified ACM members can publish their ideas and receive feedback in the form of discussion. Nothing similar has been created in Russia yet.
(c) Andrey Mikhailovich Sukhov, professor of the Department of Software Systems at Samara University, professor of the Department of Computer Engineering and Modeling at Crimean Federal University, Doctor of Technical Sciences
|
