Secure and efficient networks

An early version of this paper appeared at the 23rd Workshop on the Economics of Information Security 2023 (WEIS 2023) conference.

44 Pages Posted: 3 Jul 2024

Date Written: June 29, 2024

Abstract

More connected networks are often regarded as more efficient (e.g. Lindenlaub & Prummer, 2021), yet an increased connectivity can also elevate security risks (e.g. Goyal & Vigier, 2014), suggesting that sparsely connected networks might be preferable for security. However, security and efficiency often cannot be considered at the expense of one another-this trade-off is central to our study. We develop a model in which two players, the Defender and the Attacker, engage in a network design and defense game over two stages. First, the Defender creates a network topology from a set of nodes by establishing costly links between them. Then, the Defender chooses multiple nodes to protect, and the Attacker simultaneously selects one node to attack. Our study yields three main findings: firstly, the cost of a single link significantly influences the Defender's topology choice when the network is under attack; secondly, we characterize optimal network topologies, revealing that structures like the centrally protected star, often deemed most secure, emerge only as extremes when the cost of a single link is sufficiently high; thirdly, we identify a new network topology, the 'maxi-core network, ' which extracts the maximum possible payoff for the Defender in situations where defensive resources are scarce. Our findings can guide managerial and cybersecurity decisions within organizations under the threat of cyberattacks.

Keywords: Strategic Network Formation, Conflict, Cybersecurity, Network Defense

JEL Classification: D74, D85

Suggested Citation

Stupak, Oleh, Secure and efficient networks (June 29, 2024). An early version of this paper appeared at the 23rd Workshop on the Economics of Information Security 2023 (WEIS 2023) conference., Available at SSRN: https://ssrn.com/abstract=4880791 or http://dx.doi.org/10.2139/ssrn.4880791

Oleh Stupak (Contact Author)

University of Cambridge ( email )

Trinity Ln
Cambridge, CB2 1TN
United Kingdom

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
50
Abstract Views
127
PlumX Metrics