Controlling proximity-malware infection in diverse tactical mobile networks using K-distance pruning

Proximity-based malware, such as cyber cellphone worms and trojans, has been identified to be one of the serious open threats to wireless networks operating at the tactical edge. The ability of malware to exploit the vulnerabilities of a wireless mobile platform is often a function of the mix of software and hardware which define its platform. In this paper, we utilize a proactive strategy which takes advantage of platform diversity metrics to defend against cyber worm attacks in tactical networks. This strategy uses dynamic K-Distance pruning to minimize the total number of shared edges between nodes with similar platforms while retaining access to mission-critical services. To achieve this we introduce a new neighborhood diversity metric and show how it can be used to limit the spread of malware in a multi-platform mobile network by reducing contact between machines with similar vulnerabilities. We test our assumptions via an agent-based simulation of epidemic malware in a synthetic tactical network.