Cybersecurity researchers have created a new type of cyberattack—not to cause harm, but to help protect some of the world’s most important computer systems.
By understanding how attackers might secretly exploit hidden weaknesses, the team hopes to make technologies such as aircraft navigation, medical devices and vehicle safety systems much more secure.
The study, published in ACM Transactions on Cyber-Physical Systems, was led by researchers from Washington State University, working with collaborators from the University of Colorado Colorado Springs and Metro State University.
The project introduces a proof-of-concept attack called NosyNeighbor, which was designed to reveal vulnerabilities that could otherwise remain unnoticed.
Unlike many cyberattacks that directly target software or attempt to break into a computer system, NosyNeighbor works in a much quieter way.
It observes what is happening from the outside and gathers clues about the system’s internal activity without interfering with its normal operation. By studying these clues, it can estimate what tasks the computer is performing and use that information to plan a future attack.
The researchers say this type of testing is important because many modern technologies depend on highly reliable, precisely timed computer systems.
Aircraft flight controls, anti-lock braking systems in cars, autonomous vehicles and implanted medical devices such as pacemakers all rely on software that must perform critical tasks within extremely short time limits.
Even delays of a few thousandths of a second can sometimes lead to serious problems.
Many of these systems are built using separate software sections, known as partitions. Each partition performs a different job while working together with the others to complete a larger task.
For example, an aircraft may have separate partitions for navigation, engine control, flight systems and passenger entertainment. Although these sections are isolated from one another, they still operate on the same overall system.
The researchers wanted to know whether information leaking from one partition could reveal what was happening in another. Their experiments showed that it could.
NosyNeighbor uses what cybersecurity experts call a side-channel attack. Rather than attacking the software itself, it watches indirect signals, particularly how long certain computer tasks take to finish. Small differences in timing can reveal valuable information about which programs are running and when.
During testing, the attack correctly identified the task being performed about 73% of the time under normal operating conditions. While that number may not sound especially high, the researchers say it is more than enough to create serious risks. An attacker often needs only one successful opportunity to interfere with a safety-critical system.
For example, if a hacker can predict exactly when an autonomous vehicle’s braking system is active, they could attempt to disrupt that function at a dangerous moment, such as when the vehicle approaches an intersection.
The researchers emphasize that NosyNeighbor was created to strengthen cybersecurity, not weaken it. By exposing vulnerabilities before criminals discover them, engineers can develop stronger defenses and improve the design of future safety-critical systems.
According to the team, thinking like an attacker is one of the best ways to build better protection.
As computers become more complex and increasingly rely on software from multiple vendors, understanding these hidden risks will become even more important for keeping transportation, healthcare and other critical technologies safe.
