Method and measuring device for removing an unwanted signal component from a measurement signal

Abstract

A method for removing at least one unwanted signal component from a measurement signal at the output (310) of a device under test (DUT) (320), comprising the following method step: - Performing a first measurement of the measurement signal with a measuring unit at the output (310) of the device under test (DUT) (320), wherein the device under test (320) is supplied with a measurement signal with a first frequency at its input (330), characterized by the further process steps: - Performing a second measurement of the measurement signal at the output (310) of the object under test (320) with the measuring unit, wherein the object under test (320) is supplied at its input (330) with a measurement signal with a second frequency which differs from the first frequency, and - Subtracting the measurement results of the second measurement from the measurement results of the first measurement.

Description

[0001] The invention relates to a measuring method and a measuring device for removing at least one unwanted signal component from a measurement signal at the output of a device under test (DUT).

[0002] The measurement of the noise figure F, also known as the noise number F or noise figure NF, is an essential part of any characterization of components such as amplifiers, mixers, or receivers. The noise figure F is a quality factor that describes the amount of excessive noise present in a system. The noise figure F is defined as the quotient of the input signal-to-noise ratio and the output signal-to-noise ratio. The noise figure NF is then calculated as the base-10 logarithm of the noise figure F multiplied by a factor of 10 (NF = 10 * log(F)).

[0003] Accurate measurement of the noise figure F, or the determination of the noise figure NF, is crucial for improving the correlation between simulations and measurements and for refining circuit models more quickly. In particular, a precise noise figure allows for the optimization of transmit / receive systems, such as those used in radar applications. Improved noise characteristics enable a reduction in transmit power, allowing for the use of less expensive and lighter transmitters.

[0004] There are essentially two noise figure measurement methods. The most common is the so-called Y-factor or hot / cold source measurement method. The second noise figure measurement method is the so-called cold source method, which is preferably performed with a vector network analyzer (VNA). However, unwanted signal components in a measurement signal at the output of a device under test, from which, for example, the noise figure is to be measured, distort the measurement results and do not allow for a sufficiently accurate measurement of the noise figure F and thus a sufficiently accurate determination of the noise figure NF.

[0005] From DE 10 2005 008 734 A1, a method and a system for identifying and / or eliminating an unknown number of sinusoidal interference signals in a generally colored noise signal are known. The method according to this publication includes a Fast Fourier Transform with practical frequency resolution and additionally employs an efficient numerical method. To detect sinusoidal interference signals in a noise signal, the entire frequency range of the noise signal to be measured is first divided into several frequency bands using a Fast Fourier Transform filter bank. In each of these bands, the noise signal is approximately white and contains only a limited, comparatively small number of spectral lines belonging to sinusoidal interference signals.Subsequently, the frequencies and power levels of the sinusoidal interference signals can be determined using a method for eigenvalue analysis of autocorrelation matrices.

[0006] Furthermore, DE 10 2017 206 081 A1 discloses a measuring device for determining the noise figure of a band-limited device under test. It comprises a vector network analyzer configured to measure the transfer function of the device under test, a power meter configured to measure the noise power of the device under test, and a compensation device. The compensation device is configured to determine the noise figure from the transfer function of the device under test, the noise power of the device under test, and a known transfer function of filters used in the vector network analyzer.

[0007] Furthermore, DE 10 2011 079 086 A1 relates to a measuring device for suppressing an interference signal, which includes a mixer, an analog-to-digital converter, a normalizing device, and a replacement device. The mixer is configured to mix a first signal with a first intermediate frequency and a second signal with a second intermediate frequency. The analog-to-digital converter is configured to digitize the first signal mixed with the first intermediate frequency to form a first measurement signal and to digitize the second signal mixed with the second intermediate frequency to form a second measurement signal. The normalizing device is configured to shift the frequency of at least the first measurement signal and / or the second measurement signal such that they have a common center frequency.The replacement device is designed to combine the first measurement signal and the second measurement signal into a single overall measurement signal by removing interference signal peak value ranges that correspond to the interference signal.

[0008] A disadvantage of the method and system described in DE 10 2005 008 734 A1 is the computationally intensive and therefore time-consuming detection of the sinusoidal interference signals. Furthermore, the use of powerful transformation filter banks and processing units to perform the necessary mathematical operations increases the manufacturing costs of such a system.

[0009] The object of the present invention is therefore to provide a method and a measuring instrument for removing at least one unwanted signal component from a measurement signal at the output of a device under test, thereby enabling a sufficiently accurate determination of the noise figure F without requiring a complex measurement setup with expensive measuring instruments and additional components. Furthermore, such a measurement should achieve significantly reduced measurement times compared to known methods for measuring noise figures.

[0010] The measurement setup realized with the method according to the invention is also intended to serve to measure various relevant parameters of a measurement object, such as S-parameters and noise figure, without having to change the measurement setup.

[0011] The method is particularly well suited for measuring low noise figures and is ideally suited for use in conjunction with a vector network analyzer.

[0012] The problem is solved with respect to the method for removing at least one unwanted signal component from a measurement signal at the output of a device under test (DUT) by the features of claim 1 and 11, respectively, and with respect to the measuring instrument according to the invention by the features of claim 10 and 18, respectively. Advantageous embodiments of the method according to the invention for removing at least one unwanted signal component from a measurement signal at the output of a device under test are specified in the dependent claims. The advantageous embodiments of the method according to the invention specified in the dependent claims also apply to the measuring instrument according to the invention.

[0013] The inventive method for removing at least one unwanted signal component from a measurement signal at the output of a device under test (DUT) first comprises a first measurement of the measurement signal with a measuring unit at the output of the DUT, wherein the DUT is supplied with a measurement signal at its input with a first frequency. A second measurement of the measurement signal at the output of the DUT is then performed with the measuring unit, wherein the DUT is supplied with a measurement signal at its input with a second frequency. This second frequency differs from the first frequency. Finally, the measurement results of the second measurement are subtracted from the measurement results of the first measurement.

[0014] Advantageously, two measurements are performed using an unchanged measurement setup. Applying a first frequency to the input of the device under test (DUT) generates an output signal at the DUT's output. This output signal contains both the desired and unwanted signal components. In a second measurement, the DUT's input is subjected to a second frequency, chosen such that the unwanted signal components generated in the first measurement are produced at the DUT's output. Subtracting the second measurement, which contains the unwanted signal components, from the first measurement, which contains both the desired and unwanted signal components, yields a measurement result that includes only the desired signal components.

[0015] Preferably, during the first measurement, the device under test is supplied with a measurement signal at its input containing first frequencies from a first frequency range, and during the second measurement, it is supplied with a measurement signal at its input containing second frequencies from a second frequency range. The second frequency range is different from the first.

[0016] Advantageously, in a first measurement, not just one initial frequency is applied to the input of the device under test (DUT), but several initial frequencies within a first frequency range. For example, the first frequency range is between 1 GHz and 3 GHz. Starting at 1 GHz, the initial frequency is increased in equidistant frequency steps Δf until 3 GHz is reached. It is conceivable that the measurement signals at the output of the DUT for each initial frequency applied to the input of the DUT are stored in a memory so that they are available for further processing. In a second measurement, several secondary frequencies within a second frequency range are similarly applied to the input of the DUT. For example, the second frequency range is between 3 GHz and 9 GHz. Starting at 3 GHz, the secondary frequency is increased in equidistant frequency steps Δf until 9 GHz is reached.Analogous to the first measurement, the measurement signals at the output of the object under test are stored in a memory for the second frequency applied to the input of the object under test.

[0017] Preferably, both the first and second measurements at the output of the device under test are noise figure measurements. The measuring unit used is advantageously a vector network analyzer (VNA).

[0018] Advantageously, the method according to the invention is suitable for performing a noise figure measurement at the output of a device under test. Determining the noise figure of a device under test serves to refine circuit models and to optimize transmit and receive systems. The use of a vector network analyzer (VNA) is particularly advantageous, since with a VNA, both the scattering parameters (S-parameters) and the noise figure can be determined with only one measurement setup.

[0019] Preferably, at least one unwanted signal component is caused by a harmonic oscillation and / or a subharmonic oscillation in the object under test.

[0020] Advantageously, the method according to the invention is particularly useful for removing unwanted signal components caused by harmonic vibrations in the object being measured. Especially in amplifiers or frequency-converting components, harmonic overtones of a fundamental frequency of an applied measurement signal arise due to nonlinearities of individual components and also during the frequency conversion process itself.

[0021] Preferably, at least one unwanted signal component is caused by the third harmonic overtone.

[0022] Advantageously, the method according to the invention allows unwanted signal components caused by the third harmonic to be removed. It is precisely these signal components, generated by the third harmonic, that significantly influence the measurement of a signal at the output of a device under test.

[0023] Preferably, the device under test (DUT) to be measured by the method is a frequency-converting device under test.

[0024] Preferably, the second frequency is an integer multiple of the first frequency. Particularly preferably, the second frequency is three times the first frequency. It is also conceivable that, instead of a measurement signal with a first frequency, a measurement signal with a first frequency range containing first frequencies is applied to the input of the device under test (DUT). Similarly, in a second measurement, instead of a measurement signal with a second frequency, a measurement signal with a second frequency range containing second frequencies is applied to the input of the DUT. The second frequencies preferably correspond to an integer multiple of the first frequencies, and particularly preferably to three times the first frequencies. The frequency range can, for example, also be implemented by a comb signal consisting of many discrete frequencies at constant intervals.

[0025] Advantageously, during a second measurement, the input of the object under test is subjected to three times the frequency of the first frequency, since the third harmonic overtones in particular cause unwanted signal components in a measurement signal at the output of the object under test.

[0026] Preferably, before performing the first measurement, a first broadband measurement signal is applied to the input of the device under test (DUT) and the corresponding measurement signals are measured broadband at the output of the device under test (DUT).

[0027] Preferably, measurement signals at the output of the object under test are identified as unwanted signal components when a defined threshold, in particular a power threshold, is exceeded.

[0028] Advantageously, signal components that exceed a defined power threshold can be easily identified as unwanted signal components and suppressed without high computational effort.

[0029] Preferably, a measuring device for measuring a measurement signal at the output of a device under test (DUT) connected to the measuring device and for removing at least one unwanted signal component from the measurement signal at the output of a device under test (DUT) connected to the measuring device comprises at least one processor, one signal generator, and one analysis unit. The processor is configured to perform a first measurement of the measurement signal at the output of the device under test, wherein the device under test is supplied with a first frequency at its input by the measuring device. The processor is further configured to perform a second measurement of the measurement signal at the output of the device under test (DUT), wherein the device under test (DUT) is supplied with a measurement signal at its input with a second frequency. This second frequency differs from the first frequency.Finally, the processor is designed to subtract the measurement results of the second measurement from the measurement results of the first measurement.

[0030] Another method according to the invention for removing at least one unwanted signal component from a measurement signal at the output of a device under test (DUT) comprises, first, a first measurement of the measurement signal with a measuring unit at the output of the DUT. In this first measurement, the DUT is supplied with a measurement signal at a first frequency at its first input. Subsequently, a second measurement of the measurement signal at the output of the DUT is performed with the measuring unit, wherein the DUT is supplied with noise at its first input. This measurement is performed in a first frequency range. A third measurement of the measurement signal at the output of the DUT is then performed with the measuring unit, wherein the DUT remains supplied with noise at its first input. This measurement is performed in a second frequency range. This second frequency range differs from the first frequency range.Finally, the measurement results of the third measurement are subtracted from the measurement results of the second measurement, taking into account the results of the first measurement.

[0031] Advantageously, the second and third measurements are performed with an unchanged measurement setup. The measurement in a first frequency range at the output of the device under test (DUT) generates an intermediate frequency signal in the measuring unit. This signal contains both the desired and unwanted signal components. In a third measurement, the DUT's output is measured in a second frequency range, selected such that the unwanted signal components generated in the second measurement are detected at the output of a mixer integrated into the measuring unit. Subtracting the third measurement, which contains the unwanted signal components, from the second measurement, which contains both the desired and unwanted signal components, yields a measurement result that includes only the desired signal components.

[0032] Preferably, a fourth measurement of the measurement signal is performed at the output of the device under test using the measuring unit. In this measurement, the device under test is subjected to noise at its first input. This fourth measurement is performed in a third frequency range.

[0033] Advantageously, in the fourth measurement, the output of the device under test is measured in a third frequency range, which is selected such that the unwanted signal components generated in the second measurement are detected at the output of a mixer integrated into the measuring unit. The additional subtraction of the fourth measurement, which contains the unwanted signal components, from the second measurement, which contains both the desired and unwanted signal components, leads to a further improved measurement result that includes only the desired signal components.

[0034] Preferably, in the first and second measurements, a local oscillator integrated into the measuring unit is set to a first local oscillator frequency. In the third measurement, the local oscillator integrated into the measuring unit is set to a second local oscillator frequency.

[0035] Preferably, the second internal local oscillator frequency is set to be twice as high as the first local oscillator frequency.

[0036] Advantageously, the method according to the invention allows unwanted signal components caused by the second harmonic to be removed. These unwanted signal components, which contain the second harmonic, significantly influence the measurement result of a test object. By setting the local oscillator integrated in the measuring unit to twice the frequency of the first and second measurements, these unwanted signal components are measured reliably and accurately.

[0037] Preferably, in the fourth measurement, the local oscillator integrated in the measuring unit is set to a third local oscillator frequency.

[0038] Preferably, the third local oscillator frequency is set to be three times higher than the first local oscillator frequency.

[0039] Advantageously, the method according to the invention allows unwanted signal components caused by the third harmonic to be removed. The third harmonic also negatively affects the measurement result of a test object. By setting the local oscillator integrated in the measuring unit to three times the frequency of the local oscillator frequency of the first and second measurements, these unwanted signal components are also measured reliably and accurately.

[0040] Preferably, the method is also intended for a frequency-converting device under test. To measure a frequency-converting device under test, a measurement signal with a second frequency is applied to its second input.

[0041] Advantageously, signal components that exhibit a defined frequency offset in the case of a frequency-converting device under test can be measured in such a way that this frequency offset is taken into account in the previously described procedure steps. This allows unwanted signal components to be identified and suppressed even for frequency-converting devices under test.

[0042] Another preferred embodiment of a measuring device comprises, for measuring a measurement signal at the output of a device under test (DUT) connected to the measuring device and for removing at least one unwanted signal component, caused by an internal mixer and an internal local oscillator of the measuring unit, from the measurement signal at the output of a device under test (DUT) connected to the measuring device. The processor is configured such that it performs a first measurement of the measurement signal at the output of the device under test, wherein the device under test is supplied with a first frequency at its first input by the measuring device.

[0043] The processor is further configured to perform a second measurement of the measurement signal at the output of the device under test (DUT), whereby the DUT is subjected to noise at its input and the measurement is performed in a first frequency range. The processor is further configured to perform a third measurement, whereby the DUT is again subjected to noise at its input and the measurement is performed in a second frequency range. This second frequency range differs from the first. Finally, the processor is configured to subtract the measurement results of the third measurement from the measurement results of the second measurement, taking the first measurement into account.

[0044] Preferably, the processor is configured to perform a fourth measurement of the measurement signal at the output of the device under test. In this process, the device under test is subjected to noise at its first input. This fourth measurement, initiated by the processor, is performed in a third frequency range.

[0045] Advantageously, the processor performs a fourth measurement in a third frequency range, which is selected such that the unwanted signal components generated during the second measurement are detected at the output of a mixer integrated into the measuring device. Subtracting the measurement result of the fourth measurement, which includes the unwanted signal components, from the second measurement, which includes both the desired and unwanted signal components, results in a further improved measurement result that contains only the desired signal components.

[0046] Preferably, in the first and second measurements, the processor sets the integrated local oscillator in the measuring device to a first local oscillator frequency. In the third measurement, the integrated local oscillator is set to a second local oscillator frequency.

[0047] Preferably, the second internal local oscillator frequency is set to be twice as high as the first local oscillator frequency.

[0048] Advantageously, the measuring device according to the invention allows unwanted signal components caused by the second harmonic to be removed from the measurement results. This is achieved by the processor setting the local oscillator integrated in the measuring device to twice the frequency of the first measurement and performing a second measurement using the measuring unit. Thus, the unwanted signal components are measured reliably and accurately and can therefore be removed.

[0049] Preferably, in the fourth measurement, the local oscillator integrated in the measuring device is set to a third local oscillator frequency.

[0050] Preferably, the third local oscillator frequency of the local oscillator integrated in the measuring device is set to be three times higher than the first local oscillator frequency.

[0051] Advantageously, the measuring device according to the invention allows unwanted signal components caused by the third harmonic to be removed. The processor sets the local oscillator integrated into the measuring device to three times the frequency of the local oscillator frequency of the first measurement. With this setting, the processor uses the measuring unit to measure the unwanted signal in the frequency range of the third harmonic.

[0052] Preferably, the measuring device is also designed for use with a frequency-converting device. To measure a frequency-converting device, a measurement signal with a second frequency is applied to its second input.

[0053] Advantageously, signal components that exhibit a defined frequency offset in the case of a frequency-converting device under test can be measured in such a way that this frequency offset is taken into account by the processor. This allows unwanted signal components to be identified and suppressed, even for frequency-converting devices.

[0054] Various embodiments of the invention are described below by way of example with reference to the drawing. The reference numerals in Fig. Numbers 1-5 are all three digits and each begins with the number corresponding to the figure number. The tens and units digits of the reference symbols are the same for identical elements in different figures. For example, the unit of analysis in Fig. 3, designated with reference number 303, while the analysis unit in Fig. 4 is designated with the reference number 403. The corresponding figures in the drawing show in detail: Fig. 1. A general representation of a measurement object to explain the terms noise figure and noise measure; Fig. 2 an example of a test object which includes a frequency mixer; Fig. 3 an embodiment of a measuring device according to the invention for removing unwanted signal components; Fig. 4A a block diagram to explain in more detail the first measurement of the method according to the invention; Fig. 4B a block diagram to further explain the second measurement of the method according to the invention; Fig. 5A an example of a measurement signal with a first frequency at the input of a device under test and the resulting measurement signal at the output of the device under test; Fig. 5B an example of a measurement signal with a second frequency at the input of a device under test and the resulting measurement signal at the output of the device under test; Fig. 6 an embodiment of a flowchart for an embodiment of the inventive method for removing unwanted signal components; Fig. 7 a block diagram to further explain the first measurement of a second embodiment of the method according to the invention; Fig. 8 a block diagram to further explain the second measurement of a second embodiment of the method according to the invention; Fig. 9A an example of a measurement signal with intended and unintended noise components; Fig. 9B an example of a measurement signal with a frequency shift of the unwanted noise components; Fig. 9C is an example of a measurement signal with resulting error-prone noise signals; Fig. 9D is an example of the intended and unintended noise components; Fig. 9E is an example of a frequency shift of the unwanted noise components; Fig. 9F is an example of the resulting corrected noise signal; Fig. 10 a block diagram to further explain the first measurement of a third embodiment of the method according to the invention; Fig. 11 a block diagram to further explain the second measurement of a third embodiment of the method according to the invention; and Fig. 12 an embodiment of a flowchart for a second embodiment of the inventive method for removing unwanted signal components.

[0055] Fig. Figure 1 shows a device under test 120, with an input 130 and an output 110. To explain the terms noise figure, also called noise factor, and noise measure, an electronic amplifier with a gain factor G is chosen as the device under test 120.

[0056] Amplifier 120 receives a signal power S1 and a noise power N1 at its input 130. The ratio of the signal power S1 to the noise power N1 represents the signal-to-noise ratio (SNR). The signal-to-noise ratio at the input is SNR. ein =S1 / N1.

[0057] At the output 110 of amplifier 120, a signal power S2 and a noise power N2 can be measured. The ratio of the signal power S2 to the noise power N2 represents the signal-to-noise ratio at the output of amplifier 120. The signal-to-noise ratio at the output is SNR. aus =S2 / N2.

[0058] The noise figure F, or noise factor F, is derived from the ratio of the signal-to-noise ratio at the input to the signal-to-noise ratio at the output, F=SNR. ein / SNR ausTaking into account the gain factor G of the electronic amplifier 120, we get F = 1 / G * (N2 / N1). The noise figure F is often given logarithmically in decibels (dB) and expressed as the noise measure F. dB or also referred to as NF (Engl.: Noise Figure, NF), where F dB =NF= 10*log ( F).

[0059] Within an amplifier, fluctuations in the gain factor occur. A non-constant gain factor leads to non-linear distortion of the signal being amplified. For example, a sine wave with frequency f1 is distorted, resulting in a frequency mixture consisting of the original sine wave with frequency f1 and the harmonic overtones of the sine wave. These harmonic overtones are integer multiples of the original frequency f1.

[0060] Fig. Figure 2 shows an example of a frequency-converting device under test (DUT) 220, in this case a mixer, on which noise figure measurements are also frequently performed. The mixer includes a local oscillator (LO) that generates an oscillator frequency f0 or has an input that can be connected to an external oscillator with the frequency f0. A mixer typically consists of electronic components such as diodes and transistors and serves to convert the frequency of electrical signals.

[0061] In general, a mixer can be used to convert a specific frequency band with a defined bandwidth, applied to the mixer's input, into a higher or lower frequency band. The frequency f0 of the local oscillator determines the center frequency of the mix. In the example according to... Fig. In step 2, a high-frequency signal (RF signal) is applied to input 230 of mixer 220, which is transformed into a low-frequency intermediate frequency signal and can be accessed at the output of mixer 220 as an intermediate frequency signal (IF signal). Such a mixer is called a step-down mixer.

[0062] Particularly with down-converters, it frequently occurs that the output of the mixer contains not only the desired received signal with frequency f HF = f LO + / - f ZF is present, but also a mirror signal with the mirror frequency f HF,Sp = f LO - / + f ZF .

[0063] Therefore, when measuring a signal at the output of a mixer, especially in the case of step-down mixers, unwanted signal components are also measured. Fig. 1 and Fig. Figure 2 shows examples of devices under test that exhibit measurement signals with unwanted signal components at their outputs. These unwanted signal components must be removed to increase the measurement accuracy of the measurement signal at the device's output and to achieve acceptable measurement accuracy, particularly when measuring the noise figure.

[0064] Fig. Figure 3 now shows a measurement setup for implementing the inventive method for removing at least one unwanted signal component from a measurement signal at the output of a device under test (DUT).

[0065] The measurement setup consists of a measuring device 300, which includes a signal generator 302, a processor 301 (Central Processing Unit, CPU), and an analysis unit 303. The CPU 301 is connected to the signal generator 302 and also to the analysis unit 303 in order to perform the process steps according to the invention. The signal generator 302 is connected to an output 342 of the measuring device 300. The analysis unit 303 is connected to an input 344 of the measuring device 300.

[0066] The output 342 of the measuring instrument 300 is connected via a measuring line 341 to an input 330 of the device under test (DUT) 320. This allows the input 330 of the DUT 320 to be supplied with a measurement signal generated by the signal generator 302. The input 344 of the measuring instrument 300 is connected via a measuring line 343 to an output 310 of the DUT 320. This allows the measurement signal from the output 310 of the DUT 320 to be fed to the analysis unit 303, and thus the measurement of the signal present at the output 310 of the DUT 320 to be measured.

[0067] The in Fig. The measuring device 300 shown, which is connected to a measuring object 320, allows, with the help of the analysis unit 303, a first measurement of a measurement signal, which is present at the output 310 of the measuring object 320, to be carried out and the input 330 of the measuring object 320 to be supplied with a first frequency using the signal generator 302.

[0068] Subsequently, a second measurement is taken at output 310 of the object under test 320 using the measurement setup shown. Fig. 3 is carried out by the signal generator 302 now generating a second frequency, with which the input 330 of the object under test 320 is applied.

[0069] Finally, using the analysis unit 303, the measurement result of the second measurement is subtracted from the measurement result of the first measurement in order to remove the unwanted signal component that is present in the measurement result of the first measurement.

[0070] The control of the process steps is handled by the CPU 301, which is connected to both the signal generator 302 and the analysis unit 303.

[0071] Fig. Figure 4A shows a block diagram to further explain the first measurement of the method according to the invention. The noise figure F or the noise factor NF in a first frequency range of 1 GHz to 3 GHz is to be determined at the output 410 of the device under test 420, for example, an electronic amplifier. Due to amplifier characteristics already described, the third harmonics of the measurement signal applied to its input 430 are generated in the amplifier 420. The measurement signal at the output 410 of the amplifier 420 therefore contains unwanted signal components caused by the third harmonics of the first frequencies at the input 430 of the amplifier 420. In a first frequency range of, for example, 1 GHz to 3 GHz, the third harmonics generated in the amplifier 420 lie in the frequency range of 3 GHz to 9 GHz.

[0072] The CPU is not shown. However, it is assumed that the CPU performs and monitors the corresponding control steps in order to carry out the method according to the invention.

[0073] The signal generator 402 generates a measurement signal of the first frequency of 1GHz in a first measurement and increases the frequency in defined time steps, preferably in equidistant time steps Δt, by defined frequency steps, preferably by equidistant frequency steps Δf up to a frequency of 3GHz.

[0074] To clearly illustrate the procedure, it is described using a measurement signal with a first frequency of 1 GHz. For a first frequency increased by one frequency step n*Δf, the procedure is analogous, where n is an integer.

[0075] The measurement signal generated by signal generator 402, with a frequency of 1 GHz, is fed to amplifier 420 at its input 430. Due to non-linearities within amplifier 420, third harmonics with a frequency of 3 GHz are generated within amplifier 420, resulting in unwanted signal components at amplifier output 410. Therefore, the output 410 of amplifier 420 contains a measurement signal with a 1 GHz component and an unwanted signal component caused by the third harmonics at 3 GHz.

[0076] This measurement signal at output 410 of amplifier 420 is fed to the analysis unit 403, which is capable of suppressing the signal component caused by the 1 GHz measurement signal generated by signal generator 402. To suppress the measurement signal generated by signal generator 402, it is not only fed to input 430 of the device under test 420, but is also fed directly to the analysis unit 403. For example, a cross-correlation function within the analysis unit 403 is able to filter out or suppress the measurement signal generated by signal generator 402 from the measurement signal at output 410 of the device under test 420. A measurement signal remains, which contains the unwanted signal component caused by the harmonic overtones in amplifier 420. This remaining measurement signal is then fed to the storage unit 425.

[0077] Fig. Figure 4B shows a block diagram to further illustrate the second measurement of the method according to the invention. The CPU is not shown. However, it is assumed that the CPU performs and monitors the corresponding control steps in order to carry out the method according to the invention.

[0078] In a second measurement, the signal generator 402 generates a measurement signal with the second frequency of 3GHz and increases the frequency in defined time steps, preferably in equidistant time steps Δt, by defined frequency steps, preferably by equidistant frequency steps Δf up to a frequency of 9GHz.

[0079] To clearly explain the second measurement, the procedure is described analogously to the first measurement only for a second frequency of 3 GHz; the same procedure applies to each subsequent second frequency.

[0080] The 3 GHz measurement signal generated by signal generator 402 is fed to the device under test 420, in this example the amplifier, at its input 430. Due to non-linearities within amplifier 420, third harmonics with a frequency of 9 GHz are generated in amplifier 420, which lead to unwanted signal components at the output 410 of amplifier 420. This unwanted signal component can be disregarded in the second measurement, as it falls outside the usable bandwidth of amplifier 420 anyway.

[0081] The 3 GHz measurement signal at the output of amplifier 420 is fed to the analysis unit 403 in this second measurement without being suppressed by any corresponding function within the analysis unit. A measurement signal remains, which includes a 3 GHz signal component. This remaining measurement signal is then fed to the storage unit 425.

[0082] Finally, the measurement result of the second measurement at the second frequency of 3 GHz, which is stored in memory unit 425, is subtracted from the measurement result of the first measurement at the first frequency of 1 GHz, which is also stored in memory unit 425. By calculating the difference, at least one unwanted signal component, namely that caused by the third harmonic of the 1 GHz signal, is removed from the measurement signal at output 410 of the device under test 420.

[0083] This method is to be carried out analogously for each first frequency within a first frequency range and for each second frequency within a second frequency range. In the first measurement, several first frequencies in a first frequency range can be applied to the input 430 of the amplifier 420, and the corresponding measurement signals at the output 410 of the amplifier 420 can be determined and stored in the memory unit 425. Similarly, the second measurement is then carried out with several second frequencies in the second frequency range, the corresponding measurement signals at the output 410 of the amplifier 420 are determined and stored in the memory unit 425. The measurement results of the second measurement are then subtracted from the measurement results of the first measurement.

[0084] Fig. In diagram 500A, 5A shows a measurement signal 550, which, according to the measurement setup, Fig. 4A is generated in a first measurement using the signal generator 402. The diagram shows a sine wave with a frequency f1 in the spectral representation.

[0085] Fig. Figure 5A shows the measurement signal at output 410 of the device under test 420, also in spectral representation. The measurement signal includes a signal 551 with frequency f1, which represents the amplified input signal 550 and also has frequency f1. Signal 551 is shown with a dashed line because it is suppressed in the analysis unit 403 and is not considered further. Additionally, signal 552 is shown with frequency 3f1. Due to non-linearities generated in the amplifier by components that are not ideal, the third harmonic of the measurement signal applied to input 430 of amplifier 420 is generated in the amplifier, thus causing unwanted signal components 552 in the measurement signal at output 410 of amplifier 420. Noise 553, generated in amplifier 420, is superimposed on the measurement signal at output 410.These unwanted signal components 552 lead to an inaccuracy in the measurement of the noise figure or noise measure and are therefore undesirable.

[0086] Fig. Figure 5B shows a measurement signal 560 in diagram 500B, which corresponds to the measurement setup in Fig. 4B is generated in a second measurement using signal generator 402. The diagram shows a sine wave 560 with a frequency of 3f1 in the spectral representation. This measurement signal with the second frequency 3f1 corresponds to three times the frequency of the first frequency f1 and thus to the third harmonic overtone, which is generated in amplifier 420 during the first measurement.

[0087] In Fig. Figure 5B shows the measurement signal at output 410 of the device under test 420, also in spectral representation. The measurement signal includes a signal 561 with a frequency of 3f1, which represents the amplified input signal 560, which also has a frequency of 3f1. If the measurement result of the second measurement, as shown in Figure 501B, is now subtracted from the measurement result of the first measurement, as shown in Figure 501A, then the unwanted signal components 552 are removed. Thus, only the noise 553 remains, and a sufficiently accurate determination of the noise figure can be carried out.

[0088] It should be noted that further harmonics are usually generated within the device under test, e.g., in the mixer or amplifier, resulting in additional unwanted signal components at the device's output. However, these additional unwanted signal components, compared to the unwanted signal component caused by the third harmonic, can usually be disregarded. Therefore, removing the unwanted signal components caused by the third harmonic allows for a sufficiently accurate measurement of the noise figure without requiring expensive measuring equipment or long measurement times.

[0089] Fig. Figure 6 shows an embodiment of a flowchart for the inventive method for removing at least one unwanted signal component from a measurement signal at the output 310 of a device under test (DUT) 320.

[0090] In a first process step 600, a first measurement of a measurement signal is performed with a measuring unit at the output 310 of the device under test (DUT) 320. Preferably, the measurement performed is a noise figure measurement and the measuring unit is a vector network analyzer (VNA). A measurement signal with a first frequency is applied to the input 330 of the DUT 320. It is also conceivable that the first frequency forms the lower frequency of a first frequency range, and that this first frequency is incrementally increased by a defined frequency value Δf at defined time intervals Δt until an upper frequency of the first frequency range is reached. For example, the lower frequency of a first frequency range between 1 GHz and 3 GHz is 1 GHz. The lower frequency is increased by, for example, 0.1 GHz at defined time intervals Δt to 1.1 GHz, 1.2 GHz, 1.3 GHz, etc., until finally the upper frequency of 3 GHz is reached.The frequency steps Δf and the time intervals Δt can be selected depending on the required accuracy.

[0091] In a second process step 601, a second measurement of the measurement signal at output 310 of the device under test 320 is performed using the measuring unit. During this second measurement, the device under test 320 is supplied with a measurement signal at its input 310 with a second frequency. The second frequency differs from the first frequency, which is applied to input 310 of the device under test 320 during the first measurement.

[0092] Analogous to the first measurement, it is also conceivable that the second frequency forms the lower frequency of a second frequency range, and that this second frequency is incrementally increased by a defined frequency value Δf at defined time intervals Δt until an upper frequency of the second frequency range is reached. Preferably, the second frequency is an integer multiple m of the first frequency, in particular three times (m=3) the first frequency. For example, the lower frequency of the second frequency range (m*1GHz-m*3GHz) based on a first frequency range (1GHz-3GHz) is m*1GHz. The lower frequency is increased by, for example, m*0.1GHz at a defined time interval Δt. Assuming that m=3 is chosen, this increases to 3.3GHz, 3.6GHz, 3.9GHz, etc., until finally the upper frequency of 9GHz is reached.

[0093] The frequency steps Δf and the time intervals Δt can be selected depending on the required accuracy.

[0094] Finally, in a last step of the process (602), the measurement results of the second measurement are subtracted from the measurement results of the first measurement. If measurements are taken within a defined range, for example, between 1 GHz and 3 GHz, a second measurement is always performed for each initial measurement at, for example, 1 GHz, 1.1 GHz, 1.2 GHz, etc., at, for example, m=3, 3 GHz, 3.3 GHz, 3.6 GHz, etc. The measurement result from, for example, 3 GHz is subtracted from the measurement result at 1 GHz, the measurement result at 3.3 GHz is subtracted from the measurement result at 1.1 GHz, the measurement result at 3.6 GHz is subtracted from the measurement result at 1.2 GHz, and so on.

[0095] Fig. Figure 7 shows a block diagram to further explain the first measurement of the second embodiment of the method according to the invention. The gain of the object 720 is to be determined at a first frequency of, for example, 1 GHz in this first measurement.

[0096] Due to the previously described properties of mixers, mixers used in a measuring instrument generate harmonic overtones of the local oscillator signal. Furthermore, as already described, mixing products are generated in the measurement signals applied to input 744. The measurement signal processed in the measuring instrument therefore contains unwanted signal components caused by the harmonics of the instrument's internal local oscillator signal, which, combined with frequencies at input 744, generate unwanted mixing products. The aforementioned interference products are largely suppressed by the following configuration of the measuring instrument 700 used in the first measurement.

[0097] A signal generator, which is an integral part of the measuring device 700, generates a measurement signal 741 of the first frequency of 1GHz in the first measurement.

[0098] The first measurement signal 741 is fed via a first connection 742 of the measuring device 700 to a first input 730 of the object being measured 720. The measurement signal 743, present at the output 710 of the device under test 720, is received by the measuring instrument 700 at its measurement input 744. The measuring instrument 700 is designed to perform the measurement selectively. A gain or attenuation of the device under test 720 at the aforementioned frequency is determined from the level difference between input 744 and output 742 of the measuring instrument 700. The gain / attenuation value thus determined as a function of frequency is preferably represented as an S-parameter. Other possible representations include X-parameters, Z-parameters, or a scalar representation of the gain. The measurement results are stored in a memory unit, which is not shown for the sake of simplicity, for further processing.

[0099] Fig. Figure 8 shows a block diagram to further explain the second, third and fourth measurements of the second embodiment of the method according to the invention.

[0100] In the second measurement, the first input 730 of the device under test 720 is terminated with a termination element 770. Generally, a termination resistor with an impedance of 50 ohms, 75 ohms, or an impedance matched to the characteristic impedance is used. Besides providing optimal impedance termination of the first input 730 of the device under test 720, this termination element 770 also serves as a noise source. The noise signal 753 generated by the termination element 770 is fed into the first input 730 of the device under test 720.

[0101] In the case of a 50 Ohm termination element, the signal power of the noise signal is as follows: P N= (-174 + 10 * log(B)) dBm, where B is the bandwidth of the signal. This results in a noise level of -81 dBm for a bandwidth of 2 GHz.

[0102] The output of the device under test (DUT) 710 and the input 744 of the measuring instrument 700 remain connected as in the first measurement. The receiver unit of the measuring instrument 700, which is not shown in the block diagram for the sake of simplicity, is tuned to the frequency to be measured in this second measurement. A mixer at the input of the receiver unit, in conjunction with the receiver unit's local oscillator, produces a harmonic oscillation at an internal intermediate frequency of the measuring instrument. In this example, the set frequency is 1 GHz. Starting from this setting, the output signal 743 of the DUT is now measured in a first frequency range. This measured signal 743 from the second measurement now includes the noise signal in the 1 GHz range. Thus, the noise signals around the desired fundamental frequency and the noise signals of the odd harmonics are contained in a noise signal mixture.The measured noise signals from the second measurement are also stored in the memory unit for further processing.

[0103] This noise signal mixture is heavily corrupted by the noise components described above, caused by the odd harmonics. Therefore, a third measurement is performed. The setup for this third measurement remains unchanged from the second measurement; see [reference]. Fig. 8.

[0104] For the third measurement, however, the receiver unit of the measuring device 700 is set so that its local oscillator oscillates at twice the frequency set in the first measurement. The measurement is then performed in a second frequency range. In this example, the set frequency is 2 GHz. The output signal 743 of the device under test is measured. This measured signal 743 from the second measurement now includes the noise signal above 2 GHz. The noise signal around the desired fundamental frequency is no longer contained within the measured noise signal. The even harmonics are completely contained within the measured noise signal mixture. The measured noise signals from the third measurement are also stored in the memory unit for further processing.

[0105] For the fourth measurement, the receiver unit of the measuring device 700 is set so that its local oscillator oscillates at three times the frequency set in the first measurement. The measurement is then performed in a third frequency range. In this example, the set frequency is 3 GHz. The output signal 743 of the device under test is measured. This measured signal 743 from the third measurement now includes the noise signal around 3 GHz and around the subsequent multiples of the third harmonic. The noise signal is not present around the desired fundamental frequency, nor around the second and fourth harmonics of the fundamental frequency. The noise signals of the odd harmonics are completely contained within the measured noise signal mixture. The measured noise signals from the third measurement are also stored in the memory unit for further processing.

[0106] In the evaluation unit of the measuring device 700, the stored measurement results of the third measurement in the second frequency range and the fourth measurement in the third frequency range are subtracted from the stored measurement results of the second measurement in the first frequency range, in this example from 1 GHz upwards. Any frequency offset that may occur in the third measurement is adjusted so that the frequencies of the second and third measurements correspond to each other. By calculating the difference, at least one unwanted signal component in the harmonic overtone range is removed.

[0107] The evaluation unit also compensates for the systematic errors in the measurement results obtained from the difference calculation, which are present due to the amplification of the measured object 720. The amplification-dependent deviations must be compensated because the noise figure to be determined is a parameter independent of the amplification.

[0108] Fig. Figure 9A shows an example of a diagram of a measurement signal 910 from the first measurement. In addition to the measurement signal 910, which is amplified or attenuated by the object under test, the output signal of the object under test 710 exhibits noise components 920, 930, and 940. The first noise component 920 originates in the region of the fundamental frequency f. cThis first noise component 920 consists of the noise signal of the device under test 720 and a noise component of the device under test 720 at the image frequency of the mixer integrated in the measuring instrument. A further noise component lies in the range of the first overtone, or the second harmonic, of the fundamental frequency f. C . Another noise component lies in the range of the second overtone, or the third harmonic, of the fundamental frequency f. c .

[0109] Fig. Figure 9B shows the shift of the noise components to the fundamental frequency f by the mixer integrated into the measuring device 700. c The measurement signal 910. The noise signals overlap with the shift in frequency.

[0110] Fig. Figure 9C shows the signal measured in the first measurement of the method according to the invention. In addition to the measurement signal 910, the measuring instrument 700 detects a noise signal 950. This noise signal, as described, consists of the noise generated by the object under test 720 and additional unwanted noise signals. This signal mixture cannot be separated at the output of the mixer integrated in the measuring instrument 700.

[0111] Fig. Figure 9D shows an example diagram of a noise signal distribution at the output 710 of the device under test 720 with a connected noise source 770. In addition to the noise signal 960, which is amplified or attenuated by the device under test, the output signal 710 of the device under test 720 exhibits further noise components 970 and 980. The first noise component 960 is in the region of the fundamental frequency f. CThis first noise component 960 consists of a noise signal from the object under test 720 and a noise component that arises from the reflection of the mixer integrated in the measuring instrument 700 at the intermediate frequency. A further noise component 970 lies in the range of the first overtone, or the second harmonic, of the fundamental frequency f. C . Another noise component, 980, lies in the range of the second overtone, or the third harmonic, of the fundamental frequency f. C .

[0112] Fig. Figure 9E shows the shift of the noise components to the fundamental frequency f by the mixer integrated into the measuring device 700. Cof the measurement signal 910. Due to the frequency shift, the noise signals superimpose to form a noise signal 990 measured in the second measurement. This signal mixture cannot be separated by conventional means. Therefore, the method according to the invention is used. For this purpose, in a third measurement, the noise component 970 is measured in the region of the second harmonic, as already described. In addition, in a fourth measurement, the noise component 980 is measured in the region of the third harmonic. Now, the noise signal components 970 and 980 from the third and fourth measurements are subtracted from the noise signal 990 of the second measurement.

[0113] Fig. Figure 9F shows the noise signal 990, which is measured by the method according to the invention. As can be seen, the unwanted noise components were removed by the method, so that only the desired noise component is measured.

[0114] Fig. Figure 10 shows a block diagram to further explain the first measurement of the third embodiment of the method according to the invention. The gain or attenuation of a frequency-converting test object 720 is to be determined in this first measurement at a first frequency of, for example, 1 GHz.

[0115] As already described in the second embodiment, mixers used in a measuring device 700 generate harmonic overtones of the local oscillator signal. The noise components transferred to the noise signal being measured by these overtones distort the measurement result and must therefore be suppressed.

[0116] A signal generator, which is an integral part of the measuring device 700, generates a measurement signal 741 of the first frequency of 1GHz in the first measurement.

[0117] The first measurement signal 741 is fed to a first input 730 of the device under test 720 via a first connection 742 of the measuring instrument 700. The second input 751 of the device under test 720 is connected to a second output 752 of the measuring instrument 700. A measurement signal 750 with a second fixed frequency, an external local oscillator signal 750, is supplied to the device under test via this connection. This external local oscillator signal 750 is generated by another signal generator integrated into the measuring instrument 700. The supplied measurement signal 750 has a level suitable for adequately driving the frequency-converting device under test 720, in this example a mixer, according to the specifications.

[0118] The measurement signal 743, present at the output 710 of the device under test 720, is received by the measuring instrument 700 at its measurement input 744. The measuring instrument 700 is designed to perform the measurement selectively. A gain or attenuation of the device under test 720 at the aforementioned frequency is determined from the level difference between input 744 and output 742 of the measuring instrument 700. The gain / attenuation value thus determined as a function of frequency is preferably represented as an S-parameter. Other possible representations include X-parameters, Z-parameters, or a scalar representation of the gain. The measurement results are stored in a memory unit, which is not shown for the sake of simplicity, for further processing.

[0119] Fig. Figure 11 shows a block diagram to further explain the second, third, and fourth measurements of the third embodiment of the method according to the invention. In the second measurement, the first input 730 of the device under test 720 is terminated with a termination element 770. Generally, a termination resistor is used here, e.g., with an impedance of 50 ohms, 75 ohms, or an impedance adapted to the characteristic impedance, as explained in the second embodiment. The second input 751 of the device under test remains connected to the second output 752 of the measuring device 700, and the second measurement signal 750 with the same parameters from the first measurement is fed in.

[0120] The output of the device under test 710 and the input 744 of the measuring instrument 700 remain connected as in the first measurement. The receiver unit of the measuring instrument 700, which is not shown in the block diagram for the sake of simplicity, is set to harmonic oscillation at an internal intermediate frequency in this second measurement. In this example, the set frequency is 1 GHz. Starting from this setting, the output signal 743 of the device under test is now measured in a first frequency range. This measured signal 743 from the second measurement now includes the noise signal in the 1 GHz range. Thus, the noise signals around the desired fundamental frequency and the noise signals in the range of further harmonics are contained in a noise signal mixture. The measured noise signals from the second measurement are also stored in the memory unit for further processing.

[0121] The noise components in the wide harmonic range cause errors in the noise measurement. Therefore, a third measurement is performed. The setup for this third measurement remains unchanged from the second measurement; see [reference]. Fig. 10.

[0122] For the third measurement, the receiver unit of the measuring device 700 is set so that its local oscillator oscillates at twice the frequency set in the first measurement. The measurement is then performed in a second frequency range. In this example, the set frequency is 2 GHz. The output signal 743 of the device under test is measured. This measured signal 743 from the second measurement now includes the noise signal above 2 GHz. The noise signal around the desired fundamental frequency is no longer contained within the measured noise signal. The even harmonics are completely contained within the measured noise signal mixture. The measured noise signals from the third measurement are also stored in the memory unit for further processing.

[0123] For the fourth measurement, the receiver unit of the measuring device 700 is set so that its local oscillator oscillates at three times the frequency set in the first measurement. The measurement is then performed in a third frequency range. In this example, the set frequency is 3 GHz. The output signal 743 of the device under test is measured. This measured signal 743 from the third measurement now includes the noise signal around 3 GHz and around the subsequent multiples of the third harmonic. The noise signal is not present around the desired fundamental frequency, nor around the second and fourth harmonics of the fundamental frequency. The noise signals of the odd harmonics are completely contained within the measured noise signal mixture. The measured noise signals from the third measurement are also stored in the memory unit for further processing.

[0124] In the evaluation unit of the measuring device 700, the stored measurement results of the third measurement in the second frequency range and the fourth measurement in the third frequency range are subtracted from the stored measurement results of the second measurement in the first frequency range, in this example from 1 GHz upwards. Any frequency offset that may occur in the third measurement is adjusted so that the frequencies of the second and third measurements correspond to each other. By calculating the difference, an unwanted signal component in the harmonic overtone range is removed.

[0125] The evaluation unit also compensates for the systematic errors in the measurement results obtained from the difference calculation, which are present due to the amplification of the measured object 720. The amplification-dependent deviations must be compensated because the noise figure to be determined is a parameter independent of the amplification.

[0126] Fig. Figure 12 shows a flowchart for the second embodiment of the inventive method for removing at least one unwanted signal component from a measurement signal at the output 710 of a test object (DUT) 720.

[0127] In a first process step 1000, a first measurement of a measurement signal is performed with a measuring unit at the output 710 of the device under test (DUT) 720. Preferably, the measurement performed is a noise figure measurement and the measuring unit is a vector network analyzer (VNA). A measurement signal 741 with a first frequency is applied to the first input 730 of the device under test. During the measurement, a measurement value is recorded and stored for each set first frequency. These measurements are preferably a forward S-parameter. Gain measurements and other parameter representations can also be used.

[0128] It is also conceivable, for example, that the first frequency forms the starting frequency of a first frequency response of the object under test, and that this first frequency is incrementally increased by a defined frequency value Δf at defined time intervals Δt until an upper stop frequency of the first frequency range is reached. The starting frequency could be defined as 1 GHz and the stop frequency as 3 GHz, for example. This results in a first frequency response in the range of 1 GHz to 3 GHz. The frequency step size determines the frequency resolution in a measured curve. The time intervals between frequency changes Δt are determined depending on the settling times of the measuring instrument 700 and the object under test 720. Signal generators and analysis units contained within the measuring unit are particularly relevant here.

[0129] In a second process step 1001, a second measurement of the measurement signal at output 710 of the device under test 720 is performed using the measuring unit. During this second measurement, the device under test 720 is subjected to a noise signal 753 at its input 730. To generate this noise signal, the input 730 of the device under test 720 is disconnected from the connection to the measuring unit and terminated with a termination element 770. No additional noise source is required other than the inherent noise of the termination element 770.

[0130] The receiver integrated into the measuring unit, which receives the measurement signal 743 via the input of the measuring unit, is tuned to a first frequency to be measured. This first frequency to be measured corresponds to the first harmonic of a frequency to be received. In this example, 1 GHz is selected as the first frequency to be measured. The receiver integrated into the measuring unit measures the noise power in a first frequency range starting at 1 GHz. The upper limit of the first frequency range is higher than the third harmonic, preferably higher than the fifth harmonic, of the first frequency to be measured. The measurement results are then stored in a memory unit of the measuring unit.

[0131] In a third process step 1002, the measurement setup remains unchanged. The receiver integrated into the measuring unit, which receives the measurement signal 743 via the input of the measuring unit, is tuned to a second frequency to be measured. In this embodiment, this second frequency to be measured corresponds to the third harmonic of a frequency to be received. In this example, starting from the first harmonic frequency of 1 GHz, the second frequency to be measured is set to 3 GHz. The receiver integrated into the measuring unit measures the noise power in a second frequency range starting at 3 GHz. The upper limit of the second frequency range is higher than the fifth harmonic frequency of the first frequency to be measured. The measurement results of the third measurement are again stored in a memory unit of the measuring unit.

[0132] Finally, in a last step 1003, the results from the third step 1002 are subtracted from the results of the second measurement in the second step 901. The measurement result, in this example with a third harmonic frequency at 3 GHz, is subtracted from the measurement result with a first harmonic frequency at 1 GHz. Level fluctuations and offset values ​​caused by the amplification or attenuation by the device under test 720 are also compensated for. These level fluctuations and offset values ​​were determined in the first step 1000 by subtracting the measurement results from the first step 1000. The final step 1003 yields the noise figure required by the user, which has been corrected for unwanted signal components.

[0133] Within the scope of the invention, all described and / or drawn features and / or claimed features can be combined with one another as desired. Instead of measuring amplifiers, the invention is particularly suitable for measuring frequency-converting objects such as mixers.

Claims

[1] A method for removing at least one unwanted signal component from a measurement signal at the output (310) of a device under test (DUT) (320), comprising the following method step: - Performing a first measurement of the measurement signal with a measuring unit at the output (310) of the device under test (DUT) (320), wherein the device under test (320) is supplied with a measurement signal with a first frequency at its input (330), characterized by , the further procedural steps: - Performing a second measurement of the measurement signal at the output (310) of the object under test (320) with the measuring unit, wherein the object under test (320) is supplied at its input (330) with a measurement signal with a second frequency which differs from the first frequency, and - Subtracting the measurement results of the second measurement from the measurement results of the first measurement. [2] Method according to claim 1, characterized by, that the object being measured (320) is supplied with a measurement signal with first frequencies from a first frequency range at its input (330) during the first measurement and with a measurement signal with second frequencies from a second frequency range at its input (330) during the second measurement, wherein the second frequency range is different from the first frequency range. [3] Method according to claim 1 or 2, characterized by , that the first and second measurement at the output (310) of the device under test (320) is a noise figure measurement, and / or that the measuring unit is a vector network analyzer (VNA). [4] Method according to any one of claims 1 to 3, characterized by , that at least one unwanted signal component is caused by a harmonic oscillation and / or a subharmonic oscillation in the object under test (320). [5] Method according to any one of the preceding claims, characterized by, that at least one unwanted signal component is caused by the third harmonic overtone. [6] Method according to any one of the preceding claims, characterized by , that the device under test (DUT) (320) is a frequency-converting device under test. [7] Method according to claim 6, characterized by , that the second frequency is an integer multiple of the first frequency, in particular that the second frequency is three times the frequency of the first frequency. [8] Method according to any one of claims 1 to 7, characterized by , that the following procedural steps are carried out before the first measurement is performed: - Applying an initial broadband measurement signal to the input (330) of the device under test (DUT) (320), and - Performing a broadband measurement of the measurement signals at the output (310) of the device under test (DUT) (320). [9] Method according to claim 8, characterized by, that measurement signals at the output (310) of the object under test (320) are identified as unwanted signal components when a defined threshold, in particular a power threshold, is exceeded. [10] Measuring device (300) comprising at least one processor (301), a signal generator (302) and an analysis unit (303) for measuring a measurement signal at the output (310) of a device under test (DUT) (320) connected to the measuring device (300) and for removing at least one unwanted signal component from the measurement signal at the output (310) of a device under test (DUT) (320) connected to the measuring device (300), wherein the processor (301) is configured to perform the following steps: - Performing a first measurement (600) of the measurement signal at the output (310) of the device under test (DUT) (320), wherein the device under test (DUT) (320) can be subjected to a first frequency at its input (330) by the measuring instrument (300), - Performing a second measurement (601) of the measurement signal at the output (310) of the device under test (DUT) (320), wherein the device under test (DUT) (320) can be supplied at its input (330) with a measurement signal with a second frequency that differs from the first frequency, and - Subtracting the measurement results (602) of the second measurement from the measurement results of the first measurement. [11] A method for removing at least one unwanted signal component from a measurement signal at the output (710) of a device under test (DUT) (720), comprising the following method steps: - Performing a first measurement of the measurement signal (743) with a measuring unit at the output (710) of the object under test (720), wherein the object under test (720) is supplied with a measurement signal (741) at its first input (730) with a first frequency, - Performing a second measurement of the measurement signal (743) at the output (710) of the object under test (720) with the measuring unit, wherein the object under test (720) is subjected to noise (753) at its first input (730), and wherein the measurement is performed in a first frequency range, - Performing a third measurement of the measurement signal at the output (710) of the object under test (720) with the measuring unit, wherein the object under test (720) is subjected to noise (753) at its first input (730), and wherein the measurement is performed in a second frequency range, and - Subtracting the measurement results of at least one subsequent measurement from the measurement results of the second measurement, taking into account the measurement results of the first measurement. [12] Procedure Claim 11, characterized by , that after the third measurement has been carried out, the following procedural step is performed: - Performing a fourth measurement of the measurement signal at the output (710) of the object under test (720) with the measuring unit, wherein the object under test (720) is subjected to noise (753) at its first input (730), and wherein the measurement is performed in a third frequency range. [13] Method according to one of claims 11 or 12, characterized by , that in the first measurement and in the second measurement a local oscillator integrated in the measuring unit is set to a first local oscillator frequency, and that in the third measurement the local oscillator integrated in the measuring unit is set to a second local oscillator frequency. [14] Method according to claim 13, characterized by , that the second local oscillator frequency is twice as large as the first local oscillator frequency. [15] Method according to any one of claims 11 to 14, characterized by, that in the fourth measurement the local oscillator integrated in the measuring unit is set to a third local oscillator frequency. [16] Method according to claim 15, characterized by , that the third local oscillator frequency is set to be three times as high as the first local oscillator frequency. [17] Method according to any one of claims 11 to 16, characterized by , that the object under test (720) is a frequency-converting object under test, and that the object being measured (720) is supplied with a measurement signal with a second frequency (750) at its second input (751). [18] Measuring device (760) comprising at least one processor (761), a first signal generator (762), a second signal generator (764) and an analysis unit (763) for measuring a measurement signal at the output (710) of a device under test (DUT) (720) connected to the measuring device (700) and for removing at least one unwanted signal component from the measurement signal at the output (710) of a device under test (DUT) (720) connected to the measuring device (700), wherein the processor (761) is configured to perform the following steps: - Performing a first measurement of the measurement signal (743) with a measuring unit at the output (710) of the object under test (720), wherein the object under test (720) can be supplied with a measurement signal (741) with a first frequency at its first input (730), wherein the object under test (720) can be supplied with a measurement signal with a second frequency (750) at its second input (751), and - Performing a second measurement of the measurement signal (743) at the output (710) of the object under test (720) with the measuring unit, wherein the object under test (720) can be subjected to noise (753) at its first input (730), - Performing a third measurement of the measurement signal at the output (710) of the object under test (720) with the measuring unit, wherein the object under test (720) can be subjected to noise (753) at its first input (730), and wherein the measurement is performed in a second frequency range, and - Subtracting the measurement results of at least one subsequent measurement from the measurement results of the second measurement, taking into account the measurement results of the first measurement. [19] Measuring device claim 18, characterized by , that the processor (761) is designed to perform the following steps: - Performing a fourth measurement of the measurement signal at the output (710) of the object under test (720) with the measuring unit, wherein the object under test (720) is subjected to noise (753) at its first input (730), and wherein the measurement is performed in a third frequency range. [20] Measuring device according to one of claims 18 or 19, characterized by , that a local oscillator integrated into the measuring instrument is set to a first local oscillator frequency in the first measurement and in the second measurement, and that the local oscillator integrated into the measuring instrument is set to a second local oscillator frequency in the third measurement. [21] Measuring device according to claim 20, characterized by , that in the local oscillator integrated into the measuring device, the second local oscillator frequency is twice as large as the first local oscillator frequency. [22] Measuring device according to any one of claims 18 to 21, characterized by, that a local oscillator integrated into the measuring device is set to a third local oscillator frequency in the fourth measurement. [23] Measuring device according to claim 22, characterized by , that in the local oscillator integrated into the measuring device, the third local oscillator frequency is three times as large as the first local oscillator frequency. [24] Measuring device according to any one of claims 18 to 23, characterized by , that the object being measured (720) is a frequency-converting object being measured, and that the object being measured (720) is supplied with a measurement signal with a second frequency (750) at its second input (751).

Images