Apparatus and method for low-pass equivalent processing

Abstract

Low pass equivalent processing is a method for use with a network analyzer, that gethers a sequence of NDATA samples, each sample including real and imaginary components of a signal, over a frequency range from F1 to f2 in frequency steps of (f2-f1)/(NDATA-1), where f1 is greater than zero. The sequence of samples is processed to estimate a characteristic of the signal with respect to a selected discontinuity in the signal. The sequence of samples is converted to time domain, to generate time domain data over a range from zero to t. The selected discontinuity is identified as lying within a range of t1 to t2. Next, the phase of the time domain data is unwound to correct a portion of the phase error resulting from the one sided spectrum that is not dependent on the distance to the discontinuity through the network. Next, the magnitude peak JMAX of the data samples between time range t1 to t2, is determined. The phase angles of the data at two points removed from the magnitude JMAX by a number of data points dependent on the type of biasing window used in measurement of the data are used to determine a slope through JMAX. From this information, an estimate is made of the location of the discontinuity in the network. The phase of the time domain data is unwrapped in response to that estimate of location according to an empirically derived equation.