LMS Test.Lab Principal Component Analysis helps reduce complex noise and vibration problems to more manageable, independent, uncorrelated problems with multiple partially correlated references. These problems range from structure-borne road-noise to the combined effects of air-conditioning compressor and engine noise. By applying singular value decomposition techniques on a multi-reference crosspower matrix, this matrix is decomposed into its principal components, yielding as many single reference sets of crosspowers as there are references. The solution also features other functions such as virtual coherence and virtual spectra. Principle component analysis results can be used for further transfer path analysis.