Energy-Adaptive WDF Notch Filter for Fast Transient Suppression and Narrowband Interference Rejection

Diagnostic inaccuracies in biomedical systems are often caused by power-line interference, which significantly decreases the accuracy of electrocardiogram (ECG) data. While traditional IIR notch filters work well for steady-state noise reduction, they distort important ECG components like QRS complexes due to their large transient responses. Further, the traditional IIR filter structures are sensitive to coefficient quantization. In this work, a low coefficient sensitive wave digital structure based notch filter is presented with rapid transient suppression. The undesired transient duration can be reduced by taking pole radius of notch filter close to unity but this will increase the notch bandwidth resulting in suppression of other frequency components. This work proposes an adaptive Wave Digital Filter-based Notch Filter using an energy-gated pole radius scheduling technique to address this constraint. The approach changes the pole radius in real time based on the filter's output energy at that moment. To quickly suppress transients, a smaller pole radius is employed at first. As the signal energy diminishes, the pole radius is slowly raised to provide a narrow steady-state notch. When compared to fixed-radius notch filters, the proposed filter greatly improves denoising performance in terms of mean square error and signal-to-noise ratio. This work directly supports United Nations Sustainable Development Goal 3 promoting Good Health and Well-Being by enhancing signal reliability of biomedical devices.