BioHashing for Human Acoustic Signature Based on Random Projection

Nowadays, advanced forgery and spoofing techniques are threatening the reliability of conventional biometric modalities. This has been motivating the recent investigation of a novel yet promising modality transient evoked otoacoustic emission (TEOAE), which is an acoustic response generated from cochlea after a click stimulus. Unlike conventional modalities that are easily accessible or captured, TEOAE is naturally immune to replay and falsification attacks as a physiological outcome from the human auditory system. The corresponding biometric system where sensitive template data are stored consequently becomes the main intrusion target. To secure the template data, BioHashing based on random projection for TEOAE is proposed in this paper. In particular, feature sets are projected onto random subspaces according to randomly generated pseudomatrices, which ensure nonlinkability of hash templates across various systems; binary quantization after projection prevents recovery of original biometric data from the distorted templates. Quantitative analysis and experimental results verify that the proposed method fulfills robustness criteria, irreversibility, and diversity, and meanwhile guarantees decent recognition performance.