Effective Reversible Data Hiding in Electrocardiogram Based on Fast Discrete Cosine Transform

摘要

Based on the fast discrete cosine transform (FDCT), the authors present an effective reversible data hiding method for electrocardiogram (ECG) signal. First, an input ECG data is transformed into a series of nonoverlapping bundles by one-dimensional (1-D) FDCT. The FDCT bundles are subsequently attributed into two disjoint subsets according to a simple classification rule. Then, two pieces of data bits in different length are separately embedded in the selected coefficients of the classified bundles via the least significant bit (LSB) technique. Simulations confirmed that the hidden message can be extracted without distortion while the original ECG signal can be fully recovered. In addition, the perceived quality of the proposed method is good while the hiding capacity is superior to existing techniques. Since computational complexity is simple, the proposed method is feasible to be applied in real-time applications, or to be installed in the health care (or wearable) devices.