Redundant-dictionary based adaptive sampling for transient ECG signal measurement

This paper presents a novel adaptive sampling method for transient electrocardiogram (ECG) signal detection. The main contribution of our approach is to implement an efficient adaptive sampling scheme with dictionary-based redundant match processing that enhances the power and memory capacity performance in the electrocardiogram sensor nodes. In our scheme, the architecture of the adopted redundant dictionary is based on the Lempel-Ziv-Welch (LZW) algorithm. Compared with other traditional adaptive sampling methods, our design computation logic is simple, and easy to implement in sensor nodes. According to our test results, the tested ECG signal can be reconstructed at a deep optimized sampling rate. Throughout experimentation with the uniform sampling strategies, the proposed redundant dictionary-based sampling algorithm significantly decreased sampling rate down to less than 18% of a fixed-rate sampling, and samples space down to approximate 80%, while maintaining the integral reconstruction for ECG signal detecting.