Cyber-Physical Network Vulnerabilities Stem From Weakly Connected Nodes, Study Finds

Academic researchers studying the security of cyber-physical systems have uncovered critical vulnerabilities in how networked infrastructure responds to targeted cyberattacks, according to a newly published study in Nature Communications.

Led by Francesco Sorrentino, a mechanical engineering professor at the University of New Mexico School of Engineering, and recent UNM doctorate graduate Amirhossein Nazerian, the research evaluates how hardware and software interactions destabilize when security breaches occur within interconnected networks.

Cyber-physical systems integrate digital code with physical components such as sensors, actuators, and hardware controllers. Common operational applications include load-balancing power grids, cooperative drone formations, and autonomous vehicle fleets that rely on inter-vehicle communication to manage traffic flow and prevent collisions.

The study focused on intruder attacks, a scenario in which an adversary compromises a system's software layer while leaving its physical machinery undamaged. While conventional cybersecurity protocols typically prioritize safeguarding high-density hubs—nodes maintaining the highest concentration of connections—the research team found that destabilization occurs most rapidly when attackers compromise nodes with minimal connections.

Using mathematical modeling, the investigators demonstrated that agents with sparse incoming connections possess fewer valid data streams to neutralize corrupted or malicious inputs. Consequently, the lack of redundant, authentic signals allows compromised data to overwhelm the targeted node and propagate operational failures throughout the broader network structure.

The findings carry direct implications for critical energy infrastructure, particularly power grids that rely on automated load-shedding protocols to drop electrical loads during unexpected disruptions and avert regional blackouts. Identifying these low-degree structural weaknesses provides grid operators with specific parameters for reinforcing system defenses against targeted digital intrusions.

The project was conducted alongside co-authors Sahand Tangerami of K. N. Toosi University of Technology, Malbor Asllani of Florida State University, David Phillips of UNM, and Hernán Makse of City College of New York, as first reported by TechXplore. Building on earlier 2026 work demonstrating that biological networks transmit signals more effectively than synthetic configurations, the team plans to design architectural frameworks that maintain system stability under hostile conditions.