Abstract | ||
---|---|---|
We examine the problemof acoustic emanations of printers. We present a novel attack that recovers what a dot-matrix printer processing English text is printing based on a record of the sound it makes, if the microphone is close enough to the printer. In our experiments, the attack recovers up to 72 % of printed words, and up to 95 % if we assume contextual knowledge about the text, with a microphone at a distance of 10cmfrom the printer. After an upfront training phase, the attack is fully automated and uses a combination of machine learning, audio processing, and speech recognition techniques, including spectrum features, Hidden Markov Models and linear classification; moreover, it allows for feedback-based incremental learning. We evaluate the effectiveness of countermeasures, and we describe how we successfully mounted the attack in-field (with appropriate privacy protections) in a doctor's practice to recover the content of medical prescriptions. |
Year | Venue | Keywords |
---|---|---|
2010 | USENIX Security Symposium | hidden markov models,english text,feedback-based incremental learning,acoustic side-channel attack,appropriate privacy protection,machine learning,novel attack,contextual knowledge,audio processing,dot-matrix printer,attack in-field,side channel attacks |
Field | DocType | Citations |
Computer security,Computer science,Incremental learning,Speech recognition,Side channel attack,Audio signal processing,Linear classifier,Hidden Markov model,Microphone | Conference | 46 |
PageRank | References | Authors |
2.51 | 18 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Michael Backes | 1 | 2801 | 163.28 |
Markus Dürmuth | 2 | 423 | 25.28 |
Sebastian Gerling | 3 | 183 | 10.12 |
Manfred Pinkal | 4 | 1116 | 69.77 |
Caroline Sporleder | 5 | 453 | 31.84 |