Implementation of a DNA-based anomaly identification system utilizing associative string processor (ASP)

The genetic material that encodes the unique characteristics of each individual, such as gender, eye color, and other human features is the well-known DNA. In this work, we introduce an anomaly intrusion detection system, built on the notion of a DNA sequence or gene, which is responsible for the normal network traffic patterns. Subsequently, the system detects suspicious activities by searching the “normal behavior DNA sequence” through string matching. Conversely, string matching is a computationally intensive task and can be converted into a potential bottleneck without high-speed processing. Furthermore, conventional software implemented string matching algorithms have not kept pace with the ever increasing network speeds. As a result, we adopt a monitoring phase that is hardware implemented with the intention that DNA pattern matching is performed at wire-speed. Finally, we provide the details of our FPGA implementation of the bioinformatics-based string matching technique. The associative string processor (ASP) is an associative memory-based micro-architecture with long fixed-length words that can be partially searched. We show that the proposed micro-architecture can handle fixed-length patterns at a rate of more than one character per cycle.