"Do You Hear What I Hear? A Cyberattack."
Cybersecurity analysts work with a significantly large amount of data, especially in the performance of activities such as monitoring network traffic. Yang Cai, a senior systems scientist at CyLab, stresses that important patterns often get buried by a lot of trivial or normal patterns. For years, Cai has been working to develop ways to make it easier to spot abnormalities in network traffic. A few years ago, Cai and his research group developed a data visualization tool that can allow network traffic patterns to be seen. Now, he has developed a way to hear network traffic patterns. Cai and two co-authors demonstrated how cybersecurity data can be heard in the form of music in a new study recently presented at the Conference on Applied Human Factors and Ergonomics. They showed a change in music when there is a change in the network traffic. Cai said they wanted to articulate normal and abnormal patterns through music. Although the process of sonification in which audio is used to perceptualize data is not a new concept, sonification to make data more appealing to the human ear is new. The researchers experimented with various sound mapping algorithms to transform numeral datasets into music with different melodies, harmonies, time signatures, and tempos. They made music using network traffic data from a real malware distribution network and presented it to non-musicians. The non-musicians were found to be able to accurately recognize changes in pitch when played on different instruments. An individual is not required to be a trained musician to hear changes in the music. In the future, Cai's vision is that an analyst will be able to explore cybersecurity data using virtual reality goggles presenting the visualization of the network space. As the analyst moves closer to a data point or cluster of data, music representing that data would become more audible. This article continues to discuss the new study on the transformation of cybersecurity data into music to make abnormalities easier to detect.