Turning network traffic data into music

Cybersecurity analysts woody with an tremendous magnitude of data, particularly erstwhile monitoring web traffic. If 1 were to people the information successful substance form, a azygous day's worthy of web postulation whitethorn beryllium akin to a heavy phonebook. In different words, detecting an abnormality is similar uncovering a needle successful a haystack.

"It's an water of data," says Yang Cai, a elder systems idiosyncratic successful CyLab. "The important patterns we request to spot go buried by a batch of trivial oregon mean patterns."

Cai has been moving for years to travel up with ways to marque abnormalities successful network postulation easier to spot. A fewer years ago, helium and his probe radical developed a information visualization instrumentality that allowed 1 to spot web postulation patterns, and present helium has developed a mode to perceive them.

In a caller survey presented this week astatine the Conference connected Applied Human Factors and Ergonomics, Cai and 2 co-authors amusement however cybersecurity information tin beryllium heard successful the signifier of music. When there's a alteration successful the web traffic, determination is simply a alteration successful the music.

"We wanted to articulate mean and abnormal patterns done music," Cai says. "The process of sonification—using audio to perceptualize data—is not new, but sonification to marque information much appealing to the quality receptor is."

The researchers experimented with respective antithetic "sound mapping" algorithms, transforming numeral datasets into euphony with assorted melodies, harmonies, clip signatures, and tempos. For example, the researchers assigned circumstantial notes to the 10 digits that marque up immoderate fig recovered successful data: 0, 1, 2, 3, 4, 5, 6, 7, 8, and 9. To correspond the 3rd and 4th digits of the mathematical changeless Pi—4 and 1—they modified the clip signature of 1 measurement to 4/4 and the pursuing measurement to 1/4.

While this each whitethorn dependable reasonably complicated, 1 doesn't request to beryllium a trained instrumentalist to beryllium capable to perceive these changes successful the music, the researchers found. The squad created euphony utilizing network traffic information from a existent malware organisation web and presented the euphony to non-musicians. They recovered that non-musicians were capable to accurately admit changes successful transportation erstwhile played connected antithetic instruments.

"We are not lone making music, but turning abstract information into thing that humans tin process," the authors constitute successful their study.

Cai says his imaginativeness is that someday, an expert volition beryllium capable to research cybersecurity information with virtual world goggles presenting the visualization of the web space. When the expert moves person to an idiosyncratic information point, oregon a clump of data, euphony representing that information would gradually go much audible.

"The thought is to usage each of humans' sensory channels to research this cyber analytical space," Cai says.

Co-authors Jakub Polaczyk (left) and Katelyn Croft (right) were some students of Cai's and are alumni of Carnegie Mellon's College of Fine Arts.

While Cai himself is not a trained musician, his 2 co-authors connected the survey are. Jakub Polaczyk and Katelyn Croft were erstwhile students successful Carnegie Mellon University's College of Fine Arts. Polaczyk obtained his Artist Diploma successful Composition successful 2013 and is presently an award-winning composer based successful New York City. Croft obtained her master's grade successful harp show successful 2020 and is presently successful Taiwan studying the power of Western euphony connected Asian music.

Before graduating successful 2020, Croft worked successful Cai's laboratory connected a virtual recital project. Polaczyk took Cai's University-wide course, "Creativity," successful 2011 and the 2 person collaborated ever since.

"It has been a precise bully collaboration," Cai says. "This benignant of cross-disciplinary collaboration truly exemplifies CMU's strengths."

---

---

More information: Jakub Polaczyk et al, Compositional Sonification of Cybersecurity Data successful a Baroque Style, Advances successful Artificial Intelligence, Software and Systems Engineering (2021). DOI: 10.1007/978-3-030-80624-8_38