Abstract
A technique is presented by which the power density spectrum of an analog computer waveform may be meas ured to a specified amplitude precision and frequency resolution. The procedure involves sampling the wave form, storing the samples in a digital computer, and dig itally computing the autocorrelation function and the cosine transform of the autocorrelation function to yield the power spectrum. The theory required for selection of the parameters of sampling and computation is presented as are the required methods of data smoothing and proces sing. A description is given of an application of the tech nique to the design of a solid-state frequency multiplier.
The computational system is designer-controlled and is capable of giving rapid indications of the effect of changes in the process being simulated on the analog computer.
Get full access to this article
View all access options for this article.