MEASURE THE MUTUAL PHASE RELATIONSHIP OF A SET OF SPECTRAL COMPONENTS GENERATED BY A SIGNAL GENERATOR

Abstract

A novel method is described that enables the measurement of the phase of the spectral components generated by an harmonic phase reference generator without using a calibrated sampling oscilloscope. First one uses a vector network analyzer to measure the output reflection coefficient of an harmonic phase transfer standard and of the harmonic phase reference generator that needs to be characterized. Next one connects the transfer standard to a microwave receiver and one measures the spectral components that are generated by the transfer standard. One then connects the harmonic phase reference generator to be characterized to the microwave receiver. The spectral components of the harmonic phase reference generator to be characterized are calculated by using the spectrum of the transfer standard as measured by the receiver. The spectrum of the harmonic phase reference generator to be characterized as measured by the receiver and the known spectrum of the transfer standard.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

None

FEDERALLY SPONSORED RESEARCH

None

SEQUENCE LISTING

Not Applicable.

BACKGROUND

1. Field of the Invention

This invention relates to the characterization of harmonic phase reference generators. An harmonic phase reference generator is a repetitive electrical pulse generator with an adjustable repetition rate, also called the fundamental frequency. An harmonic phase reference generator is used as the harmonic phase standard for a “large-signal network analyzer”, also known as LSNA. An LSNA, typically operating in the microwave frequency range, measures time varying electrical quantities like the current waveforms and voltage waveforms at the terminals of an electrical device like for example a diode or a transistor. The characterization of an harmonic phase reference generator itself is time consuming and can take more than a whole day. An harmonic phase reference generator is completely characterized when one knows its output reflection coefficient as a function of frequency across its full bandwidth of operation and when one knows the amplitude and the phase of all the spectral components which are generated by the harmonic phase reference generator, and this for a multitude of fundamental frequencies.

2. Description of the Related Art

In the prior art the characterization of an harmonic phase reference generator is done as follows. The output reflection coefficient of the harmonic phase reference generator is measured by a vector network analyzer. Next a microwave sampling oscilloscope is characterized by using the nose-to-nose calibration method as described in “Individual Characterization of Broadband Sampling Oscilloscopes with a ‘Nose-to-Nose’ Calibration Procedure”, by Jan Verspecht and Ken Rush, published in IEEE Transactions on Instrumentation and Measurement, Vol. 43, No. 2, pp. 347-354, April 1994. The harmonic phase reference generator to be characterized is then connected to the input of the oscilloscope and the repetitive electrical pulse which appears at the output of the harmonic phase reference generator is being measured and digitized by the oscilloscope. This process is illustrated in FIG. 1. After being digitized by the oscilloscope the measured waveforms are processed by a computer which calculates the phase and amplitude of all spectral components generated by the harmonic phase reference generator. The spectral components are calculated using the discrete Fourier transform algorithm. The whole process is explained in “Large-Signal Network Analysis—‘Going beyond S-parameters’” by Jan Verspecht, 62nd ARFTG Conference Short Course Notes, USA, December 2003. The method explained above is the only method I know of that is used to characterize harmonic phase reference generators and has significant disadvantages. A first disadvantage is that the method requires a measurement lab which is equipped with at least three microwave sampling oscilloscope modules and two oscilloscope mainframes in order to be able to perform a ‘nose-to-nose’ calibration procedure. A second disadvantage of the method is that it takes a long time to perform. Even after one has performed a nose-to-nose calibration, which by itself takes one day, it still takes at least several hours for the characterization of one harmonic phase reference generator for a multitude of fundamental frequencies. This is a real problem when a whole series of harmonic phase reference generators needs to be calibrated. With the invented method and measurement set-up one does not need sampling oscilloscopes and the characterization of one harmonic phase reference generator is possible in less than 15 minutes.

BRIEF SUMMARY OF THE INVENTION

This invention is a method and a set of measurement setups to characterize harmonic phase reference generators much faster than the prior art and which does not require the repeated use of calibrated sampling oscilloscopes. The new method is based on using one harmonic phase reference generator as a transfer standard. The harmonic phase reference generator transfer standard is the only harmonic phase reference generator which needs to be characterized by the method described in the prior art. All other harmonic phase reference generators are characterized by comparing them to the harmonic phase reference generator transfer standard using the invented method and measurement setups.

The preferred embodiment of the new method requires a harmonic phase reference generator transfer standard, a vector network analyzer and a broadband microwave receiver which may be uncalibrated. A systematic linear distortion may be present in measurements performed with the broadband microwave receiver. First one uses the vector network analyzer to measure the output reflection coefficient of the harmonic phase reference generator transfer standard and of all the harmonic phase reference generators to be characterized. Next one connects the harmonic phase reference generator transfer standard to the microwave receiver and one measures the phase and the amplitude of the spectral components which are generated by the harmonic phase reference generator transfer standard for a multitude of fundamental frequencies. The measurements described above are sufficient to characterize the linear transfer characteristic of the broadband microwave receiver. One then disconnects the harmonic phase reference generator transfer standard from the receiver and one connects the harmonic phase reference generators to be characterized to the microwave receiver. One then measures the phase and the amplitude of the spectral components which are generated by the harmonic phase reference generator to be characterized for a multitude of fundamental frequencies. The phases and amplitudes of the output spectra of the harmonic phase reference generators to be characterized can then accurately be determined by making calculations based on the harmonic phase reference generator transfer standard output spectrum measurement, the output spectrum measurement of the harmonic phase reference generators to be characterized, the knowledge of the phase and amplitude of the spectral components of the harmonic phase reference generator transfer standard (which has been determined a priori using prior art), and the measured output reflection coefficients of the microwave receiver, the harmonic phase reference generators to be characterized and the harmonic phase reference generator transfer standard.

An alternative embodiment of the method uses a calibrated LSNA, where the harmonic phase reference generator transfer standard has been used as the harmonic phase standard of the LSNA. All harmonic phase reference generators to be characterized are then measured by connecting them to the calibrated LSNA and by using the calibrated LSNA to accurately measure the phase and amplitude of the spectral components of the harmonic phase reference generators to be calibrated. While running in vector network analyzer mode, the LSNA can also be used to characterize the output reflection coefficients of the harmonic phase reference generators to be characterized.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram showing the measurement set-up as it is described in the prior art using a calibrated oscilloscope.

FIG. 2 is a block diagram illustrating the part of the new measurement method where the VNA is connected with the harmonic phase reference generator to be characterized.

FIG. 3 is a block diagram illustrating the part of the new measurement method where the VNA is connected with the harmonic phase reference generator transfer standard.

FIG. 4 is a block diagram illustrating the part of the new measurement method where the VNA is connected with the broadband microwave receiver.

FIG. 5 is a block diagram illustrating the part of the new measurement method where the transfer standard harmonic phase reference generator is connected with the broadband microwave receiver.

FIG. 6 is a block diagram illustrating the part of the new measurement method where the harmonic phase reference generator to be characterized is connected with the broadband microwave receiver.

FIG. 7 is a block diagram illustrating the new measurement set-up where the harmonic phase reference standard to be characterized is connected to one of the test ports of a calibrated LSNA.

REFERENCE NUMERALS

1 sampling oscilloscope that has been calibrated by using the “nose-to-nose” procedure

2 harmonic phase reference generator to be characterized

3 vector network analyzer

4 harmonic phase reference generator transfer standard

5 broadband microwave receiver

6 large-signal network analyzer

DETAILED DESCRIPTION

The preferred embodiment of the invented method is described in what follows. The problem that is solved by the new method is the efficient measurement of the phase and amplitude of the output spectrum, noted S_HPR, of an harmonic phase reference generator 2, and this for a multitude of fundamental frequencies. The method determines the output spectrum S_HPR by using an a priori characterized harmonic phase reference generator 4 as a transfer standard, and this for the said multitude of fundamental frequencies. The amplitude and phase of the output spectral components generated by transfer standard 4 has been characterized by using prior art, e.g. by using an oscilloscope 1 which has been calibrated by the nose-to-nose calibration procedure. This illustrated in FIG. 1. The a priori known output spectrum of transfer standard 4 is noted S_TRA. The first step of the new method is shown in FIG. 2. Reference generator 4 is connected to a network analyzer 3 in order to measure its output reflection coefficient, which is noted R_HPR. If the output reflection coefficient of transfer standard 4 is a priori unknown, it is measured by connecting transfer standard 4 to the network analyzer 3, as shown in FIG. 3. The output reflection coefficient of transfer standard 4 is noted R_TRA. The new method uses a broadband microwave receiver 5 which has a linear transfer characteristic which is noted T_REC. The linear transfer characteristic T_REC introduces a linear distortion in any spectrum measured by receiver 5. The new method does not require T_REC to be known. The output reflection coefficient of receiver 5 is measured by connecting the input connector of receiver 5 to the test port of the network analyzer 3 as shown in FIG. 4. The output reflection coefficient of receiver 5 is noted R_REC. Next one connects transfer standard 4 to receiver 5 as shown in FIG. 5 and one uses receiver 5 to measure the amplitude and the phase of the spectral components generated by transfer standard 4, and this for the said multitude of fundamental frequencies. The corresponding measured output spectrum is noted S_MTRA. S_MTRA is given by $\begin{array}{cc}{S}_{\mathrm{MTRA}}=\frac{T_{\mathrm{REC}}{S}_{\mathrm{TRA}}}{1-R_{\mathrm{REC}}{R}_{\mathrm{TRA}}}& \mathrm{Eq}.1\end{array}$

Note that S_MTRA is different from the actual output spectrum S_TRA, since it has been linearly distorted by the transfer characteristic T_REC of receiver 5 as well as by the mismatch effects represented by R_REC and R_TRA. Next one connects reference generator 2 to receiver 5, as shown in FIG. 6, and one uses receiver 5 to measure the phase and amplitude of the spectral components generated by reference generator 2, and this for the said multitude of fundamental frequencies. The corresponding measured output spectrum is noted S_MHPR. S_MHPR is given by $\begin{array}{cc}{S}_{\mathrm{MHPR}}=\frac{T_{\mathrm{REC}}{S}_{\mathrm{HPR}}}{1-R_{\mathrm{REC}}{R}_{\mathrm{HPR}}}& \mathrm{Eq}.2\end{array}$

Note that S_MHPR is different from the actual output spectrum S_HPR, since it has been linearly distorted by the unknown transfer characteristic T_REC as well as by the mismatch effects represented by R_REC and R_HPR. In Eq.1 and Eq.2 the only unknowns are T_REC and S_HPR. The other quantities S_MTRA, S_MHPR, T_REC, S_TRA, R_REC and R_TRA are known. The quantity of interest S_HPR can be calculated for the said multitude of fundamental frequencies and is given by: $\begin{array}{cc}{S}_{\mathrm{HPR}}=\frac{S_{\mathrm{MHPR}}{S}_{\mathrm{TRA}}\left(1-R_{\mathrm{REC}}{R}_{\mathrm{HPR}}\right)}{S_{\mathrm{MTRA}}\left(1-R_{\mathrm{REC}}{R}_{\mathrm{TRA}}\right)}& \mathrm{Eq}.3\end{array}$

With the new method as described above one characterizes reference generator 2 without the use of an oscilloscope measurement as it is always done with the method described in the prior art. This is achieved by using the a priori characterized transfer standard 4. The different measurement steps as illustrated in FIG. 2, FIG. 3, FIG. 4, FIG. 5 and FIG. 6 can be performed in any arbitrary order without having an influence on the result.

From Eq.1 one concludes that the measured quantities S_MHPR, R_REC, R_TRA and the a priori known S_TRA can also be used to determine the amplitude and the phase of the a priori unknown transfer characteristic T_REC of receiver 5. This is done through the application of the following equation. $\begin{array}{cc}{T}_{\mathrm{REC}}=\frac{S_{\mathrm{MTRA}}\left(1-R_{\mathrm{REC}}{R}_{\mathrm{TRA}}\right)}{S_{\mathrm{TRA}}}& \mathrm{Eq}.4\end{array}$

An alternative embodiment of the invention is illustrated in FIG. 7. With this method one first calibrates a large-signal network analyzer 6. Note that this calibration requires the use of transfer standard 4, which in this case is a part of the whole LSNA 6. After LSNA 6 has been calibrated, one connects reference generator 2 to LSNA 6 and one uses LSNA 6 to measure the spectrum noted S_LSNA for the said multitude of fundamental frequencies. When properly calibrated S_LSNA will equal S_HPR, such that one has a simple method to directly measure S_HPR for the said multitude of fundamental frequencies. This method requires a lab which is equipped with a large-signal network analyzer. LSNA 6 can be used in vector network analyzer mode to measure R_HPR.

Claims

1-6. (canceled)

7. A method comprising the steps of:

connecting a first signal generator that generates a first set of spectral components to a broadband microwave receiver whereby the mutual phase relationship of said first set of spectral components is accurately known and whereby said broadband microwave receiver has a phase distortion error;

using said broadband microwave receiver to generate a measurement of the mutual phase relationship of said first set of spectral components;

connecting a second signal generator that generates a second set of spectral components to said broadband microwave receiver;

using said broadband microwave receiver to generate a measurement of the mutual phase relationship of said second set of spectral components;

eliminating said phase distortion error of said broadband microwave receiver from said measurement of the mutual phase relationship of said second set of spectral components by using said measurement of the mutual phase relationship of said first set of spectral components and said measurement of the mutual phase relationship of said second set of spectral components and said accurately known mutual phase relationship of said first set of spectral components.

8. The repetition of the method described in claim 7 for a multitude of fundamental frequencies.

9. The method of claim 7 whereby one uses the measured output reflection coefficients of said microwave receiver, said first signal generator and said second signal generator to correct said second measurement of the mutual phase relationship of said second set of spectral components for distortion caused by mismatch effects.