Microphones recordings from the built-in microphone from 34 mobile phones (each equipped with a single microphone) stimulated by a pneumatic hammer sound. The recordings are in PCM format at 44.1 KHz. No SIM was inserted in the mobile phone and no personal data was recorded. The recording took place in a laboratory. The data set is composed by 800 repetitions of the recording based on the repetition of the pneumatic hammer sound for each microphone. The dataset is useful for studies on multi-factor authentication of mobile phones and fight against distribution of counterfeit electronic products.
The data set is a three dimensional matrix with three dimensions: the first is the identifier of the phone, the second is the audio recording and the third is the repetition of the sound.
How to cite
European Commission, Joint Research Centre (JRC) (2020): Audio recordings microphones dataset (pneumatic hammer). European Commission, Joint Research Centre (JRC) [Dataset] PID: http://data.europa.eu/89h/c69d1d91-fa54-46f8-8d37-1414c89dbb95
- MDPI, BASEL, SWITZERLAND
Research findings have shown that microphones can be uniquely identified by audio recordings since physical features of the microphone components leave repeatable and distinguishable traces on the audio stream. This property can be exploited in security applications to perform the identification of a mobile phone through the built-in microphone. The problem is to determine an accurate but also efficient representation of the physical characteristics, which is not known a priori.
There is usually a trade-off between the identification accuracy and the time requested to perform the classification. Various approaches have been used in literature to deal with it ranging from the application of handcrafted statistical features to the recent application of deep learning techniques.
This paper evaluates the application of different entropy measures (Shannon Entropy, Permutation Entropy, Dispersion Entropy, Approximate Entropy, Sample Entropy and Fuzzy Entropy) and their suitability for microphone classification. The analysis is validated against an experimental data set of built-in microphones of 34 mobile phones, stimulated by three different audio signals. The findings show that selected entropy measures can provide a very high identification accuracy in comparison to other statistical features and that they can be robust against the presence of noise. This paper performs an extensive analysis based on filter features selection methods to identify the most discriminating entropy measures and the related hyper-parameters (e.g., embedding dimension). Results on the trade-off between accuracy and classification time are also presented.
- Published by
- European Commission, Joint Research Centre
- Created date
- Modified date
- Issued date
- Data theme(s)
- Science and technology
- Update frequency