A data-driven compressive sensing framework tailored for energy-efficient wearable sensing
We propose a new ECG data compression algorithm based on a learned overcomplete dictionary to exploit the correlation between signals in adjacent heart beats. The learned overcomplete dictionary is constructed by K-SVD dictionary learning algorithm, after preprocessing and normalization of length and magnitude. Using the overcomplete dictionary, the proposed algorithm can find sparse estimation, which can represent the ECG signal effectively. Experimental results on MIT-BIH arrhythmia database confirms that our proposed algorithm has high compression ratio while minimizing data distortion.