ECG Signal Adaptive Arithmetic Coding Based On Wavelet Transform

ECG is an important indicators of the clinical diagnosis.The realization of real-time and efficient ECG data compression has practical significance in the aspect of remote medical services.ECG data compression system not only consider the efficiency of the system,but also take into account the Real-time of ECG signal compression encoding.Under the premise of efficient compression of the ECG data,in order to reduce the time delay, we put forward a real-time ECG signal compression encoding scheme.ECG is similar to the periodic signal,there is a certain correlation between inside and outside the heart beat,then creat a multi-cycle probability statistical model can greatly improve the compression ratio of the compression coding system,to use this correlation we should segment the ECG signalFirst of all,according to the structural features of the ECG and then recognize the R waves by the max value detection and the adaptive repeated matching,we use the R-wave’s position to split the signal cycle.And each cycle of ECG signal is processed with discrete wavelet transform,the coefficients from which are done through a Dead-zone quantizer.Then the coefficients are decomposed into important position quantization coefficient stream, symbol stream,PMSB,RB.Finally,we make use of the correlation between the ECG signal to establish a multi-cycle probability statistical model for the realization of adaptive arithmetic coding. Then we can process the ECG signals like above until the end.The test ECG records are from MIT-BIH arrhythmia database.Experimental results show that within a certain error range,the reconstructed ECG signal data is coincident with the original signal,the experimental program not only ensure the ECG signal compression coding efficiency at the same time, but it also raises a very good compression in real time.