Simulation Experiment for Signal Frequency Estimation

Three complex sinusoidal signals with SNR values of SNR1=30dB, SNR2=30dB, and SNR3=27dB were analyzed using an MVDR method based on singular value decomposition (SVD). The simulation employed 1000 signal samples and a 4-tap FIR filter to estimate the power spectral density function. The implementation involves covariance matrix construction, eigenvalue decomposition, and adaptive weight calculation for spectral estimation.

Signal-to-Noise Ratio Calculation Tool (snr2.m)

Custom-developed SNR calculation tool snr2.m with function signature: snr2(s2ana, fs, f0, fl, fh, bl, bh). Parameters: s2ana - input signal for analysis, fs - sampling frequency, f0 - fundamental signal frequency, [fl, fh] - signal energy distribution band, [bl, bh] - noise bandwidth range. The algorithm separates signal and noise components through frequency-domain analysis.