Real-time notch filtering based on the modified subspace-based high-resolution frequency estimator

This paper presents the real-time notch filtering based on a modified subspace-based high-resolution frequency estimator. The modified subspace-based frequency estimator is proposed based on the propagator using total least squares (TLS) to enhance the robustness to noise, which can be extended to the recursive implementation using data projection method (DPM). In addition, a subspace-based amplitude estimation method is proposed based on the available frequency estimates. For notch filtering, a second-order infinite impulse response (IIR) notch filter is directly designed in the discrete-time domain with given notch frequency and notch bandwidth. The performance of the proposed method is validated through numerical simulations and a real-world example.