Researchers at the Indian Institute of Technology (IIT), Madras, have developed an innovative algorithm-driven strategy to reduce disruption to critical networks such as air traffic control and power distribution during a target attack.
The work has been reviewed and published in the famous international journal “Physics A: Statistical Mechanics and Its Applications”.
Different types of technological networks are the backbone of infrastructure in the modern world, and it is essential to create security measures to protect these networks from both failure and targeted attacks. The strategy developed by the researchers works to strengthen networks through rational re-wiring of a given network to reduce their vulnerabilities.
Air traffic, road traffic, power distribution infrastructure and even social media platforms are examples of highly connected networks and therefore extremely vulnerable to target attacks.
As Internet of Things (IoT) -based technologies are widely applied in society, it is important to build resilient networks for such attacks.
The terrorist attacks of September 11, 2001, targeted a single country in a single day, causing the collapse of the entire airline industry. Such threats are a reminder that in today’s highly interconnected world, there is a high risk of adverse events that disrupt entire networks.
The strategy developed by IIT Madras has been tested on two infrastructure networks of aviation and power distribution.
It has been found that the algorithm extends the strength of these networks to target attacks. The algorithm was also highly effective in strengthening the ‘canonical scale-free network’, which represents more real-world networks than existing strategies for mitigating target attacks on these networks.
Simply put, this extra capacity can be compared to the extra tires kept in the car. Passive spare tires are used if four tires (nodes) of the vehicle become lazy while traveling.
The algorithm takes a network whose excess capacity is determined as an input and provides a modified network with additional capacity, additional power consumption for the network, etc. Importantly, the algorithm also optimizes the costs associated with adding extra.
