“Smart City” systems quickly and steadily penetrate in all spheres of human life and most actively into the traffic infrastructure, including manned and unmanned vehicles, smart traffic lights, bus stops, tableau with bus and train schedules, and much more. Researchers at the St. Petersburg Federal Research Center of the Russian Academy of Sciences have developed specialized software to protect such systems. The proposed approach allows for timely detecting hacker attacks, as well as for determining what interfaces have failed or shown false results. Many published articles summarize the proposed and implemented scientific results, the one entitled "Monitoring the situation in the car to detect potential threats based on smartphone sensors" and is published in the journal Sensors (https://www.mdpi.com/1424-8220/20/18/5049), deserves special mentioning.
“Today, large cities all over the world are actively implementing “smart” systems in traffic and logistics to facilitate human life. First of all, the manned and unmanned vehicles are considered, in turn they include “smart” traffic lights, turnstiles, and other. The proposed project, assumes the development of a system working prototype as aimed at such interfaces protection. The project, ultimately, intends for human security ensuring,” says Andrey Chechulin, Leading Researcher at the Laboratory of Computer Security Problems of SPC RAS.
The algorithms of the “smart city” security system are aimed at solving two main tasks. Firstly, the interfaces have to detect attacks and security violations on a person’s part - such violations can be either a fake user card on the turnstile in the subway, or sick, drunk and sleepy persons who try to control the "smart city" elements, trying to hide their condition from the security system.
Secondly, the security program should recognize attacks from malicious programs - this refers to the disruption of the operation of “smart city” services, like traffic lights, whose incorrect operation can cause serious traffic jams and paralyze the movement of public and freight transport. The unmanned vehicle’s orientation system in space may also be disrupted, thus, leading to accidents. To solve the above problems, St. Petersburg scientists have trained the security system to use and analyze data from as many “smart city” sensors as possible.
“For instance, when the developed security system is used at an unmanned vehicle, then it makes decisions based on information received from the distance sensor to objects, as well as from navigation systems, and from cameras. So, it is important to understand that the developed algorithm does not simply summarize the information; it permanently compares it, as necessary in order to indirectly find a weak link in the interface system attacked by hackers or has failed. Then the drone will be able to prevent an accident, and the system operator will see the malfunction and try to fix it," Chechulin explained.
In the experiments run, through developed prototypes and use of the driver's personal smartphone sensors, cases of violations on the driver’s part of the manned vehicle (drivers’ distraction or falling asleep) were successfully detected. Besides, the developed systems allowed for identifying various types of computer attacks aimed at elements of the "smart city". This research was supported by the Grant from the Russian Foundation for Basic Research (RFBR) under Research Project No. 19-29-06099.