Abstract
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. Usually there is 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 dataset 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.
Highlights
This paper deals with the problem of microphone identification using physical features.Identification is the process of using claimed or observed attributes to single out one entity among others in a set of identities [1]
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
The results show that the application of selected entropy measures is able to achieve a very high identification accuracy, the approach is quite robust to the presence of noise and it provides the benefit of an high dimensionality reduction, which makes it suitable to practical applications
Summary
This paper deals with the problem of microphone identification using physical features.Identification is the process of using claimed or observed attributes to single out one entity among others in a set of identities [1]. Technologies (ICT) to implement identification is to use cryptographic algorithms, where the identity of an electronic device is verified using a cryptographic key stored in the device itself Even if this approach is predominantly used in modern ICT systems, it has some weaknesses, which are well known in literature [2]. Complex key management processes and systems must be implemented to distribute the cryptographic material (e.g., private and public keys) and support their renewal in case of obsolescence or migration from one cryptographic system to another In some cases, these requirements cannot be implemented in a cost-effective way, and the research community has proposed new approaches, including the one proposed in this paper, which is based on physical features
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
