TY - CPAPER AB - This paper presents a hardware realisation of a novel ECG baseline drift removal that preserves the ECG signal integrity. The microcontroller implementation detects the fiducial markers of the ECG signal and the baseline wander estimation is achieved through a weighted piecewise linear interpolation. This estimated drift is then removed to recover a “clean” ECG signal without significantly distorting the ST segment. Experimental results using real data from the MIT-BIH Arrhythmia Database (recording 100 and 101) with added baseline wander (BWM1) from the MIT-BIH Noise Stress Database show an average root mean square error of 34.3uV (mean), 30.4u V (median) and 18.4uV (standard deviation) per heart beat. AU - Guven,O AU - Eftekhar,A AU - Kindt,W AU - Constandinou,TG DO - 10.1109/ISCAS.2017.8050663 EP - 1574 PB - IEEE PY - 2017/// SP - 1571 TI - Low-power real-time ECG baseline wander removal: hardware implementation UR - http://dx.doi.org/10.1109/ISCAS.2017.8050663 UR - http://hdl.handle.net/10044/1/46111 ER -