| Summary: | 碩士 === 國立成功大學 === 電機工程學系碩博士班 === 101 === With the aging population in the world and people’ awareness of health improved, “Remote Home Care System” is gradually taken seriously. To avoid the burden of bandwidth and memory capacity, which are sprung from the ECG holter system, we present the lossless and lossly compression algorithms with the method of QRS-wave difference and apply to the Wireless Holter System. With different application requirements, users can choose high-compression rate or perfect reconstructed algorithms to obtain higher signal quality and to reduce more storage memory size.
For the proposed lossless compression algorithm, we use backward difference and QRS-wave difference to process ECG signal, and huffman encoding to produce the compressed data. In the lossy compression algorithm respect, we take fast fourier transform (FFT) as the kernel function which belongs to the lossy compression. After the procedure of QRS-wave difference, we partition off the RR interval and interpolate to 256 points. And then, we can get the compressed data by using FFT coefficients to the procedure of backward difference, run length encoding and Huffman coding.
The proposed algorithms were evaluated by using all patterns from MIT-BIH arrhythmia database and the test pattern which we measure the ECG signal from our ECG read-out circuit. For the proposed lossless algorithm, the average CR is 2.66 and 1.22. Furthermore, for the proposed lossy algorithm, the analytic results such as CR, PRD, PRDB, PRDN, RMS, SNR, and QS’s value is 6.69, 0.13, 1.37, 1.90, 1.20, 35.07 and 53.69. By using the test pattern which we measure the ECG signal, the analytic results such as CR, PRD, PRDB, PRDN, RMS, SNR, and QS’s value is 5.44, 0.14, 0.99, 3.46, 37.28, 30.01 and 40.21.
|
|---|
