A sampling criterion enabling signal reconstruction with specified limit error

by Dennis James Wilkins

Institution: Oregon State University
Department: Electrical and Electronics Engineering
Degree: MS
Year: 1968
Keywords: Electronic data processing
Record ID: 1532613
Full text PDF: http://hdl.handle.net/1957/46386


A sampling criterion which provides a numerical relationship between sampling rate and worst-case peak error for linear interpolation of sample points is presented. The criterion, based upon the second derivative of a waveform, is derived, its properties are observed for a sine wave, and its applicability to complex signals is discussed. An approximate measure of the second derivative for an amplitude-time function is implemented using a linear analog circuit, and this device in conjunction with an analog computer is used to confirm the validity of the sampling criterion. Possible application to on-line variable-rate sampling control for data compression is discussed in the conclusion.