摘要: This paper presents a method of detecting periodicities in data that exploits series projections onto "periodic subspaces". The algorithm finds its own set nonorthogonal basis elements (based on the data), rather than assuming fixed predetermined as Fourier, Gabor, and wavelet transforms. A major strength approach is it linear-in-period linear-in-frequency or linear-in-scale. derived analyzed, output compared to Fourier transform number examples. One application finding grouping rhythms musical score, another separation periodic waveforms with overlapping spectra, third patterns astronomical data. Examples demonstrate both strengths weaknesses method.