Keystroke transcription from acoustic emanations using continuous wavelet transform

Download

ThesisAbdullahOzkan.pdf

Date

2022-9

Author

Özkan, Abdullah

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

319
views

66
downloads

One of the most common methods of communication is written communication. Written communication has been found in various forms over the years and has changed shape with technical and technological developments. Today, written communication has shifted to digital media and keyboards have become one of the most frequently used entry points. This makes keyboards a critical node in the flow of information. There are several ways in which information entered through the keyboard can leak. Acoustic propagation is one of these leakage pathways. In the literature, various approaches have been proposed for this attack type, which aim to process the keystroke sounds and capture the information. However, the usage of continous wavelet transform for this purpose has not been explored before. Continous wavelet transform provides better resolution in both time and frequency for impulse-like signals. Therefore, this transformation is better suited to the analysis of keystroke sounds than other conventional transforms. In this thesis, we propose a method based on the continuous wavelet transform for transcription of keystrokes from the acoustic emanations of a keyboard, using wave files as input and recovering the written text as output with up to 57.2% accuracy. Initially a dataset was generated by recording keystroke sounds of 38 keys pressed multiple times in different ways. These were then analysed using the continuous wavelet transform in order to detect, segregate and obtain features of keystroke sounds. Various classification algorithms were tested and performances were recorded. Recommendations were made for improving the system output by using dictionaries and language models, as well as the information obtained from the confusion matrix itself.

Subject Keywords

Acoustic Propagation, Text Extraction, Continuous Wavelet Transform, Keyboard, Machine Learning

URI

https://hdl.handle.net/11511/99509

Collections

Graduate School of Informatics, Thesis

Suggestions

OpenMETU
Core

COS Phi: Artificial Pheromone System for Robotic Swarms Research Arvin, Farshad; Krajnik, Tomas; Turgut, Ali Emre; Yue, Shigang (2015-10-02) Pheromone-based communication is one of the most effective ways of communication widely observed in nature. It is particularly used by social insects such as bees, ants and termites; both for inter-agent and agent-swarm communications. Due to its effectiveness; artificial pheromones have been adopted in multi-robot and swarm robotic systems for more than a decade. Although, pheromone-based communication was implemented by different means like chemical (use of particular chemical compounds) or physical (RFID...
Message Scheduling for the FlexRay Protocol: The Dynamic Segment Schmidt, Şenan Ece; Schmidt, Klaus Verner (Institute of Electrical and Electronics Engineers (IEEE), 2009-06-01) The FlexRay communication protocol is expected to be the de facto standard for high-speed, in-vehicle communication. In this paper, we formally investigate the scheduling problem for the dynamic segment (DS) of FlexRay. We take the bounds on the generation times and the timing requirements of the signals into consideration to propose a reservation-based scheduling approach that preserves the flexible medium access of the DS. To obtain efficient schedules, we formulate a nonlinear integer program- ming probl...
Message Scheduling for the FlexRay Protocol: The Static Segment Schmidt, Klaus Verner; Schmidt, Şenan Ece (Institute of Electrical and Electronics Engineers (IEEE), 2009-06-01) In recent years, time-triggered communication protocols have been developed to support time-critical applications for in-vehicle communication. In this respect, the FlexRay protocol is likely to become the de facto standard. In this paper, we investigate the scheduling problem of periodic signals in the static segment of FlexRay. We identify and solve two subproblems and introduce associated performance metrics: 1) The signals have to be packed into equal-size messages to obey the restrictions of the FlexRa...
Performance evaluation of real-time noisy speech recognition for mobile devices Yurtcan, Yaser; Günel Kılıç, Banu; Department of Information Systems (2019) Communication is important for people. There are many available communication methods. One of the most effective methods is through the use of speech. People can comfortably express their feelings and thoughts by using speech. However, some people may have a hearing problem. Furthermore, understanding spoken words in a noisy environment could be a challenge even for healthy people. Speech recognition systems enable real-time speech to text conversion. They mainly involve capturing of the sound waves and con...
Wearable tactual communicators: designing products with technology-mediated touch Şener Pedgley, Bahar; Pedgley, Owaın Francıs (2021-08-01) In some situations, it is not possible or desirable for communications to take place through voice- or screen-based interaction. People may want to act discreetly around other people; it may not be physically possible to interact with a display-based product; or it may be inappropriate or ineffective to talk out loud. Furthermore, in challenging circumstances such as the COVID-19 pandemic, social touch has been diminished and discouraged. In these cases, alternative means of communication can be made throug...

Citation Formats

A. Özkan, “Keystroke transcription from acoustic emanations using continuous wavelet transform,” M.S. - Master of Science, Middle East Technical University, 2022.