| Abstrakt: |
This work involves the design of a VLSI (Very Large-Scale Integration) architecture for mathematical morphology using the FLC (Fuzzy Logic Controller) for ECG (Electrocardiogram) QRS detection. The morphology filtering process is used to remove impulsive noise in ECG signals and suppress the baseline drift in the ECG. In the morphological filter, the FLC steps—Fuzzifier, Fuzzy decoding, Decision inference logic, and Defuzzifier—can be utilized to select either erosion or dilation. Then, modulus accumulation improves the complexity of the QRS, and finally, the threshold provides details regarding the functional status of the heartbeat rate. Based on the MIT-BIH arrhythmia database, the suggested technique is fabricated with TSMC 90 nm technology, occupying an area of 0.64 mm², consuming 2.04 µW of power, and operating at a computational frequency of 0.4 kHz. The sensitivity (SEN) achieves 99.82%, Positive Predictivity (+ P) reaches 99.87%, and the Detection Error Rate (DER) is 1.08%. |
Full text from SpringerLink