Keyboard Acoustic Emanations: An Evaluation of Strong Passwords and Typing Styles
The sounds resulting from keyboard typing can reveal information about the input data. In this paper, the authors revisit such keyboard acoustic emanations for the purpose of eavesdropping over "Strong" (random) passwords. Prior work that capitalized on dictionary and HMM language models is not applicable in this scenario; rather the attacker can only utilize the raw acoustic information which has been recorded. They evaluate the use of signal processing techniques to eavesdrop over 6-character strong passwords. In doing this evaluation, they carefully examine the effect of typing style a crucial variable largely ignored by prior research on the detection accuracy.