We investigate the robustness of a network defense isolation technique called Software Defined Perimeter (SDP). SDP tries to limit security risk by isolating users access based on user accounts, rules and resources. We show how a competitive co-evolutionary framework can be used to evaluate different SDP configurations. It employs a simulation to test the strength of different SDP configurations against different attackers. It uses a co-evolutionary algorithm to search through the action spaces of SDP configurations and possible attackers. These results enable the comparison of different SDP configurations based on the objective of minimizing the number of potentially compromised high-value resources.
Paper
Full text
Securing the software defined perimeter with evolutionary co-optimization
Semantic Scholar · Computer Science · 2020
Abstract
We investigate the robustness of a network defense isolation technique called Software Defined Perimeter (SDP). SDP tries to limit security risk by isolating users access based on user accounts, rules and resources. We show how a competitive co-evolutionary framework can be used to evaluate different SDP configurations. It employs a simulation to test the strength of different SDP configurations against different attackers. It uses a co-evolutionary algorithm to search through the action spaces of SDP configurations and possible attackers. These results enable the comparison of different SDP configurations based on the objective of minimizing the number of potentially compromised high-value resources.