Abstract:
Electronic healthcare (e-health) networks are increasingly popular, particular during pandemics such as COVID-19. This reinforces the importance of ensuring security and ...View moreMetadata
Abstract:
Electronic healthcare (e-health) networks are increasingly popular, particular during pandemics such as COVID-19. This reinforces the importance of ensuring security and privacy for data-in-transit. One such solution is steganography-based schemes that utilize biological signals (e.g., ECG) as cover signals to preserve the privacy of patient personal information without affecting the diagnostic features. There are various limitations in existing steganography-based schemes, and in this study we present an effective privacy protection scheme leveraging both multidimensional steganography and shared keys. To enhance security and accelerate signal processing in our design, the Fast Walsh-Hadamard transform (FWHT) is employed to decompose ECG signals into a set of coefficients, of which the less-significant coefficients are used to construct the multidimensional space. The negotiated shared keys facilitate the embedding of encrypted data in the constructed space. We then evaluate the proposed scheme using different categories of ECG signals in the MIT-BIH database. It is observed that the signal distortion is minimal (i.e., less than 1%), even if the embedded data reaches the maximum embedding capacity. The security analysis also demonstrates that unauthorized retrieval of hidden information is not practical, within a short period of time.
Published in: IEEE Transactions on Network Science and Engineering ( Volume: 8, Issue: 2, 01 April-June 2021)