Mother wavelet selecting method for selective mapping technique ECG compression

Abstract

The electrocardiogram compression method presented in this research processes the residual signal which is the difference between the original signal and the reference signal. The residual signal is transformed to wavelet domain and then the redundant information is eliminated in wavelet domain. The selection of mother wavelet is one of main factor to maintain the important data in wavelet domain. The difference type of mother wavelet has its own shape and its own characteristic. Therefore this affects to the performance of compression. This work compares the efficiency of compression algorithm by using difference type of Mother Wavelet. The test shows that no mother wavelet which is the best for all ECG. Thereby, reducing the time consuming for selection the proper mother wavelet, the Best of Four Method is introduced. This algorithm uses four types of mother wavelet to be competitors, `db1', `db2', `db9' and `bior2.4'. The result shows that the selected mother wavelet types have the good performance on overall tested signals. Moreover, the `db1' mother wavelet has the best performance on more than a half of all signals.