Aliasing errors and dealiasing techniques for bandwidth limited signals

Aliasing errors occur in a sanpling process if the requirements of the sampling theorem are not met tar a bend limited signal. All efforts to this date in attacking this problem have been directed toward the design of filters placed at the sailer's input so as to eliminate the high frequencies of the input spectrum in order to prevent aliasing. Such filtering is generally unsatisfactory due to distortions introduced in the data. A more desirable method of handling aliasing would be some technique idiich uhra applied to any spectrum (aliased or non-aliased) would determine the actual irput spectrum. A dealiasing technique for bandwidth limited signals is presented. It requires that the sampling process be operated at two sanple rates. The lowest sampling rate should be at least slightly higher in frequency than the highest frequency present at the ixput spectrum, A digital computer program is included that performs the dealiasing operation and yields a dealiased spectrum.