Abstract: A resonance tuner receives and digitizes an analog signal in response to a resonance of a structure thereby creating a plurality of time samples. A series of the time samples are buffered upon burst detection. A power spectrum is estimated by computing a Time-To-Frequency-Transform of the series of time samples and a magnitude of each of the resulting frequency samples is squared. At least one subset associated with at least one spectral peak is selected from the frequency samples. Each spectral peak has at least one sample with a sufficient magnitude and being spectrally adjacent to any other sample in another spectral peak by less than a threshold. A fundamental spectral peak is determined in a fundamental subset including a spectral peak with a sample at the lowest frequency greater than zero. The fundamental spectral peak has the sample with the largest magnitude within the fundamental subset.