Resilient systems and network solutions
Malicious cyber activity is growing at an unprecedented rate. A leading Internet security firm reported there were more than 317 million new malicious code signatures in 2014. Additionally, attacks are increasing in sophistication as authors create malware that circumvents standard signature-based antivirus defense systems.
“The best defense—especially in cyberspace—is anticipating and preparing for your adversary’s attack. Endowing IT system administrators with the ability to act preemptively against potential malware attacks instead of acting after the fact is a paradigm shift that tilts the advantage toward IT infrastructure defenders.”
Program Manager, Science and Technology IMAM
To reverse this troubling trend, Charles River Analytics developed PMD, a program that will use advanced machine learning techniques to predict significant new malware attacks and generate preemptive defenses. This innovative program uses statistical models focused on features extracted from malware families to predict possible courses of malware evolution. Once implemented, PMD will advance malware detection and defense capabilities beyond those offered by current antivirus resources.
PMD alerts and empowers information technology (IT) administrators to fend off an impending cyberattack. PMD’s technology can predict next-stage developments in malware evolution, making it possible for administrators to anticipate and block malware intrusions.
The Charles River Analytics Predictive Malware Defense (PMD) contract was awarded through the DHS S&T Cyber Security Division’s (CSD) Long Range Broad Agency Announcement DHSST-LRBAA14-02. The award is part of CSD’s larger Internet Measurement and Attack Modeling (IMAM) project that is working with cybersecurity researchers to develop solutions in the areas of resilient systems and networks, modeling of Internet attacks as well as network mapping and measurement.
For more information about CSD’s IMAM project, visit https://www.dhs.gov/science-and-technology/anms