An artificial immunity-enhancing module for internet servers against cyberattacks

摘要

This study focused on an artificial immunity-enhancing module for high-availability servers against cyberattacks on the internet. Similar to the human immune system, an artificial immunity-enhancing module consists of innate and adaptive immune functions. The innate immune function detects cyberattacks on a known or unknown vulnerability of a server application, although this function causes to restart the server application to recover its execution control. The adaptive immune function adaptively learns the cyberattacks detected by the innate immune function using a random forest classifier. In addition, the adaptive immune function prevents subsequent cyberattacks without restarting the server application before the innate immune function detects the cyberattacks. Performance tests showed that the detection accuracy of a prototype was 92.16%, achieving a true negative rate of 99.13% by adaptively acquiring immunity against cyberattacks. Moreover, the overhead of the prototype had little effect on the performance of the server application.