A method and system for measuring frequency stability

Abstract

The present application relates to a frequency stability measurement method and system in the field of electronic technology. The method includes: respectively performing isolation amplification and frequency division processing on the acquired reference frequency signal and the frequency signal to be measured; doubling the frequency division signal of the reference frequency signal as the DDS clock source, generating a DDS frequency control word by roughly measuring the frequency of the frequency signal to be measured, thereby generating two orthogonal DDS signals, taking the frequency division result of the frequency signal to be measured and the two orthogonal DDS signals as the inputs of quadrature double mixing for quadrature double mixing processing, performing low-pass filtering, AD sampling, phase detection, phase unwrapping processing, and filtering on the double mixing processing result, and then calculating the phase difference between the measured frequency signal and the new reference frequency signal, and calculating the frequency stability according to the phase difference. This method can achieve the measurement of the frequency stability of the reference frequency signal and the frequency signal to be measured for both homologous and non-homologous cases, and at the same time, the measurement of the frequency stability of heterogenous frequency signals is also relatively accurate.

Description

Technical Field

[0001] The present application relates to the field of electronic technology, and particularly to a method and system for measuring frequency stability. Background Art

[0002] A measurement system for frequency stability mainly measures the frequency accuracy (frequency deviation) and frequency stability of a reference frequency source and a frequency source to be calibrated. The frequency accuracy of a frequency source reflects the degree of consistency between the frequency value of the frequency source and the nominal value, and the measurement requires a comparison measurement with an absolutely accurate reference source; the frequency stability reflects the ability of the frequency value of the frequency source to remain within a frequency deviation over a period of time, regardless of the magnitude of the frequency deviation, and is a quantitative description of the unstable components of the frequency value. The more commonly used method for characterizing stability is the Allan variance. Although the standard variance can also achieve statistical functions, since the frequency instability of the frequency source signal is mainly due to various random noise fluctuations, the standard variance often does not obtain a convergent result. Therefore, the standard variance is not suitable for statistical frequency stability. In addition to the Allan variance, which can be used to characterize frequency stability, there are also many methods such as the Hadamard variance and the modified Allan variance, and each method has its own advantages for different purposes.

[0003] Common methods for measuring frequency stability generally include:

[0004] ① Beat method. The frequency signal to be measured and the reference frequency signal are simultaneously sent to a double-balanced mixer to beat out a low-frequency signal with a frequency equal to the difference between the two signals. After low-pass filtering and limiting amplification, it is fed to an electronic counter. The counter repeatedly measures the period or frequency value (M times) of this beat signal at the selected sampling time τ, and then estimates the time-domain frequency stability.

[0005] ② Dual-mixer time-difference method. The two signals are respectively sent to the corresponding ports of a pair of double-balanced mixers, and are beat out with a common conversion oscillator to generate low-frequency signals. The time difference between the two beat signals is measured by a counter, and then the time-domain frequency stability is calculated using the estimation formula of σ y (τ).

[0006] ③ Frequency-difference multiplication method: First, the frequency fluctuation of the frequency signal to be measured is multiplied by a frequency-difference multiplier, and then the frequency of the multiplied signal is measured by a counter. The time-domain frequency stability is calculated according to the estimation formula of σ y (τ). This method is applicable to the measurement of the time-domain frequency stability of precision frequency sources with integer fixed-point frequencies.

[0007] ④ Phase-comparison method: The phase-comparison method is generally a digitally implemented dual-mixer time-difference method. This method uses two-channel high-speed AD acquisition technology to simultaneously sample the frequency signal to be measured and the reference frequency signal respectively. Using digital mixing and digital filtering technologies, the final real-time phase difference (time difference) is calculated, and finally converted into frequency stability.

[0008] Among them, the phase comparison method is often used for the comparison of frequency standard accuracy and long-term indicators. The double-mixing frequency difference method and the frequency difference multiplication method are mainly used for the comparison of short-term stability. The phase comparison method is widely used in the measurement of frequency accuracy and long-term stability, and can achieve a very high comparison accuracy. The relative error of the comparison is of the order of E-15 (second stability). Thus, it can be used to measure frequency stability, long-term and short-term indicators of frequency, etc. However, the time response of the phase comparison method is relatively slow and it is difficult to be used in the comparison of short-term indicators. The double-mixing frequency difference measurement has a relatively high accuracy and has low requirements for the measurement of time intervals. However, its measurement accuracy is limited by the noise of amplifiers and mixers. The beat method has a relatively high frequency measurement resolution, a simple scheme and high accuracy. After beating, the period is measured in real time by a counter. After beat frequency, the measurement accuracy is increased by the beat factor times. And the counter only needs to detect low-frequency signals, which is beneficial to the implementation of the counter. However, the time scale accuracy of the counter and the errors introduced by signal shaping and the background noise of the equipment limit the improvement of the measurement accuracy. The frequency difference multiplication method has complex hardware, many implementation links, and is also prone to introducing other interference signals, and its application scenario is limited.

[0009] The measurement accuracy of the existing frequency stability measurement methods for heterodyne signals drops severely, and the processing links increase significantly. The introduced noise and interference seriously affect the accuracy of the measurement results. For the measurement methods of heterodyne signals, the National Time Service Center of the Chinese Academy of Sciences and Xidian University in China have conducted in-depth research on time-frequency measurement and proposed a series of new measurement concepts, such as: the application of edge effect in time-frequency measurement, the phase coincidence algorithm of the group phase quantum characteristics, to solve the measurement of frequency stability and phase noise between heterodyne signals. But mainly it is still research at the theoretical level, and it is difficult to guarantee the actual implementation indicators. Summary of the Invention

[0010] Based on this, it is necessary to provide a frequency stability measurement method and system for the above technical problems.

[0011] A frequency stability measurement method, the method includes:

[0012] Isolate and amplify the obtained reference frequency signal and the frequency signal to be measured respectively.

[0013] Perform frequency division processing on the isolated and amplified reference frequency signal and the frequency signal to be measured respectively to obtain three reference frequency signals and three frequency signals.

[0014] Measure the frequencies of the third reference frequency signal and the third frequency signal to obtain the DDS frequency control word.

[0015] After frequency doubling the second reference frequency signal, it is used as the reference clock source of the direct digital frequency synthesizer, and the direct digital frequency synthesizer is controlled by the DDS frequency control word to obtain two orthogonal new reference frequency signals;

[0016] According to the two orthogonal new reference frequency signals, the first reference frequency signal, and the second reference frequency signal, orthogonal double mixing processing is performed to obtain a mixing processing result.

[0017] After sequentially performing low-pass filtering, analog-to-digital sampling, phase detection, phase unwrapping, and filtering on the mixing processing result, the phase difference between the frequency signal to be measured and the new reference frequency signal is calculated.

[0018] According to the phase difference, the frequency stability is calculated using the Allan variance.

[0019] According to the phase difference and the DDS frequency control word, the frequency difference between the frequency signal to be measured and the reference frequency signal is calculated.

[0020] In one embodiment, after frequency doubling the second reference frequency signal, it is used as the reference clock source of the direct digital frequency synthesizer, and the direct digital frequency synthesizer is controlled by the DDS frequency control word to obtain two orthogonal new reference frequency signals, including:

[0021] After frequency doubling the second reference frequency signal, it is used as the reference clock source of the direct digital frequency synthesizer, and the direct digital frequency synthesizer is controlled by the DDS frequency control word to obtain two orthogonal new reference frequency signals; the new reference frequency signals include: a sine new reference frequency signal and a cosine new reference frequency signal; the expressions of the sine new reference frequency signal and the cosine new reference frequency signal are respectively:

[0022] y s =sin(Ω' X t+φ1)

[0023] y c =cos(Ω' X t+φ1)

[0024] where y s is the sine new reference frequency signal, y c is the cosine new reference frequency signal, φ1 is the initial phase, Ω' X is the new reference frequency, and t is time.

[0025] In one embodiment, the two orthogonal new reference frequency signals include a sine new reference frequency signal and a cosine new reference frequency signal.

[0026] According to the two-way orthogonal new reference frequency signals, the first-way frequency signal, and the second-way frequency signal, orthogonal double mixing processing is performed to obtain a mixing processing result, including

[0027] Mix the sine new reference frequency signal with the first-way frequency signal to obtain a sine mixing result.

[0028] Mix the cosine new reference frequency signal with the second-way frequency signal to obtain a cosine mixing result.

[0029] In one embodiment, the mixing processing result includes: a sine mixing result and a cosine mixing result.

[0030] After successively performing low-pass filtering, analog-to-digital sampling, phase detection, phase unwrapping, and filtering processing on the mixing processing result, calculate the phase difference between the frequency signal to be measured and the new reference frequency signal, including:

[0031] Perform low-pass filtering on the sine mixing result and the cosine mixing result respectively to obtain a sine low-pass filtering result and a cosine low-pass filtering result.

[0032] Use an analog-to-digital acquisition module to perform data acquisition on the sine low-pass filtering result and the cosine low-pass filtering result respectively to obtain a sine digital mixing signal and a cosine digital mixing signal; the analog-to-digital acquisition module samples using a sampling clock with a preset frequency.

[0033] According to the sine digital mixing signal and the cosine digital mixing signal, perform phase detection using a phase detector and then perform phase unwrapping to obtain a continuous phase signal.

[0034] After filtering the phase signal, calculate the phase difference between the frequency signal to be measured and the new reference frequency signal.

[0035] In one embodiment, according to the sine digital mixing signal and the cosine digital mixing signal, perform phase detection using a phase detector and then perform phase unwrapping to obtain a continuous phase signal, including:

[0036] According to the sine digital mixing signal and the cosine digital mixing signal, perform an arctangent transformation to obtain a phase signal in the range of (-π / 2, π / 2).

[0037] After performing phase unwrapping on the phase signal in the range of (-π / 2, π / 2), obtain a continuous phase signal.

[0038] In one embodiment, before the step of successively performing low-pass filtering, analog-to-digital sampling, phase detection, phase unwrapping, and filtering processing on the mixing processing result and calculating the phase difference between the frequency signal to be measured and the new reference frequency signal, further include:

[0039] Convert the first reference frequency signal into a square wave signal, perform frequency division processing on the square wave signal to obtain a 1 kHz clock signal, and use the 1 kHz clock signal as the sampling clock of the preset frequency of the analog-to-digital acquisition module.

[0040] In one embodiment, the second reference frequency signal is frequency-multiplied and used as the reference clock source of the direct digital frequency synthesizer, and the direct digital frequency synthesizer is controlled by the DDS frequency control word to obtain two orthogonal new reference frequency signals, including:

[0041] The second reference frequency signal is frequency-multiplied by means of harmonic frequency multiplication and used as the reference clock source of the direct digital frequency synthesizer, and the direct digital frequency synthesizer is controlled by the DDS frequency control word to obtain two orthogonal new reference frequency signals.

[0042] In one embodiment, the sine low-pass filtering result and the cosine low-pass filtering result are respectively collected by the analog-to-digital acquisition module to obtain a sine digital mixing signal and a cosine digital mixing signal, including:

[0043] The sine low-pass filtering result and the cosine low-pass filtering result are respectively collected by an integral analog-to-digital acquisition module to obtain a sine digital mixing signal and a cosine digital mixing signal.

[0044] A frequency stability measurement system, the system includes: a signal input module, a frequency multiplication module, a frequency measurement module, a direct digital frequency synthesizer, a double orthogonal mixing module, a filtering and sampling module, a phase detection module, and a data processing module.

[0045] The signal input module includes two input branches composed of an isolation amplifier and a frequency division module; the first input branch is used to isolate and amplify the reference frequency signal by the isolation amplifier, and then perform frequency division processing by the frequency division module to obtain three reference frequency signals, and transmit the second reference frequency signal to the frequency multiplication module and the third reference frequency signal to the frequency measurement module; the second input branch is used to isolate and amplify the frequency signal to be measured by the isolation amplifier, and then perform frequency division processing by the frequency division module to obtain three frequency signals, and transmit the second frequency signal and the third frequency signal to the double orthogonal mixing module and the first frequency signal to the frequency measurement module.

[0046] The frequency multiplication module is used to frequency-multiply the second reference frequency signal, use the obtained frequency-multiplied signal as the reference clock source of the direct digital frequency synthesizer, and transmit it to the direct digital frequency synthesizer.

[0047] The frequency measurement module is used to measure the frequency of the received third-path frequency signal and the third-path reference frequency signal, obtain the DDS frequency control word, and transmit the DDS frequency control word to the direct digital frequency synthesizer and the data processing module.

[0048] The direct digital frequency synthesizer is used to generate a new reference frequency signal according to the received reference clock source and the DDS frequency control word. The new reference frequency signal includes a sine new reference frequency signal and a cosine new reference frequency signal.

[0049] The quadrature mixing module is used to mix the sine new reference frequency signal and the first-path frequency signal to obtain a sine mixed signal, and transmit the sine mixed signal to the first sampling branch of the filter sampling module. The cosine new reference frequency signal and the second-path frequency signal are mixed to obtain a cosine mixed signal, and the cosine mixed signal is transmitted to the second sampling branch of the filter sampling module.

[0050] The filter sampling module includes two sampling branches composed of a low-pass filter and an AD module. The two sampling branches are used to filter and sample the sine mixed signal and the cosine mixed signal by using the low-pass filter and the AD module respectively to obtain a sine digital frequency signal and a cosine digital frequency signal;

[0051] The phase detection module is used to perform phase detection on the sine digital frequency signal and the cosine digital frequency signal by using a phase detector based on the arctangent conversion method, and then perform processing by using a phase unwrapping module to obtain a continuous phase signal, and transmit the continuous phase signal to the data processing module.

[0052] The data processing module is used to filter the received continuous phase signal, calculate the phase difference between the frequency signal to be measured and the new reference frequency signal according to the obtained filtering result, calculate the frequency stability by using the Allan variance according to the phase difference, and is also used to calculate the frequency difference between the frequency signal to be measured and the reference frequency signal according to the phase difference and the DDS frequency control word.

[0053] In one embodiment, the AD module includes a sampling clock with a preset frequency. The sampling clock is a 1 kHz square wave obtained by converting the third reference frequency signal into a square wave and then performing frequency division processing.

[0054] The above frequency stability measurement method and system. The method includes: separately performing isolation amplification and frequency division processing on the acquired reference frequency signal and the frequency signal to be measured; doubling the frequency division signal of the reference frequency signal as the DDS clock source, generating a DDS frequency control word by roughly measuring the frequency of the frequency signal to be measured, thereby generating two orthogonal DDS signals. The frequency of the DDS signal is close to the frequency of the signal to be measured. Taking the frequency division result of the frequency signal to be measured and the two orthogonal DDS signals as the inputs of orthogonal double mixing for orthogonal double mixing processing. After the double mixing processing result is subjected to low-pass filtering, AD sampling, phase detection, phase unwrapping, and filtering processing, the phase difference between the signal to be measured frequency signal and the new reference frequency signal is calculated, the frequency stability is calculated according to the phase difference, and the phase difference between the signal to be measured frequency signal and the reference frequency signal is calculated according to the phase difference and the frequency control word. This method can achieve the measurement of the frequency stability of the reference frequency signal and the frequency signal to be measured with the same source and different sources, and at the same time, the measurement of the frequency stability of different frequency signals is also relatively accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a schematic flowchart of the frequency stability measurement method in an embodiment;

[0056] Figure 2 is the schematic diagram of the beat method in the prior art;

[0057] Figure 3 is the schematic diagram of the improved beat method in the prior art;

[0058] Figure 4 is the block diagram of the frequency stability measurement system in an embodiment;

[0059] Figure 5 is the block diagram of the frequency stability measurement system in another embodiment;

[0060] Figure 6 is the phase difference after 5M / 10M same-source de-skewing in another embodiment;

[0061] Figure 7 is the 5M / 10M same-source frequency stability curve in another embodiment;

[0062] Figure 8 is the measurement result of non-same-source 5M / 10M in another embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0063] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0064] In one embodiment, asFigure 1 As shown, a frequency stability measurement method is provided, and the method includes the following steps:

[0065] Step 100: Isolate and amplify the acquired reference frequency signal and the frequency signal to be measured respectively.

[0066] Specifically, the reference frequency signal and the frequency signal to be measured are sine signals. The expression of the reference frequency signal is: sin(Ω R t + φ0), and the expression of the frequency signal to be measured is: sin(Ω X t + φ X ); where Ω R is the frequency of the reference frequency signal, φ0 is the initial phase of the reference frequency signal, Ω X is the frequency of the frequency signal to be measured, φ X is the initial phase of the frequency signal to be measured, φ X varies with time t and is the phase fluctuation of the frequency signal to be measured, and t is time.

[0067] The reference frequency signal and the frequency signal to be measured can be the same-source frequency signals or different-source frequency signals.

[0068] Step 102: Perform frequency division processing on the isolated and amplified reference frequency signal and the frequency signal to be measured respectively to obtain three reference frequency signals and three frequency signals.

[0069] Step 104: Measure the frequencies of the third reference frequency signal and the third frequency signal to obtain the DDS frequency control word.

[0070] Step 106: After doubling the frequency of the second reference frequency signal, use it as the reference clock source of the Direct Digital Synthesizer (DDS for short), and control the direct digital frequency synthesizer through the DDS frequency control word to obtain two orthogonal new reference frequency signals.

[0071] Specifically, use the doubled frequency of the reference frequency signal as the reference clock source of the DDS, generate the DDS frequency control word by roughly measuring the frequency of the frequency signal to be measured, and thus generate two orthogonal DDS signals. Use the two orthogonal DDS signals as the new reference frequency signals.

[0072] The frequencies of the two orthogonal new reference frequency signals output by the DDS are close to the frequency of the frequency signal to be measured.

[0073] The reference frequency signal is: sin(Ω R t + φ0), and the frequency signal to be measured is: sin(Ω X t + φ X ), where φX is a quantity that changes with time t, and is the phase fluctuation of the frequency signal to be measured. If the phase fluctuation φ of the frequency signal to be measured can be measured X , then the frequency stability - Allan variance value of the frequency signal to be measured can be calculated. The method of obtaining phase information generally adopts the method of quadrature demodulation and down-conversion. However, since the frequency of the frequency signal to be measured is inconsistent with the frequency of the reference frequency signal, it is impossible to directly down-convert. At this time, the frequency of the reference frequency signal can be changed to achieve frequency change. The conventional methods of frequency change are PLL and DDS, both of which have their own advantages and disadvantages. The PLL method requires an external source (VCO) to lock. When the requirement for frequency stability is very high, it is difficult to meet the requirements of the locked VCO index. There are two factors affecting the output frequency stability. The first factor is that the background stability of the phase-locked loop itself will affect the stability of the output frequency; the second factor is the influence of the stability of the VCO. Generally speaking, the stability of the VCO output can follow the input reference frequency signal in the long-term stability, but the short-term stability is often determined by itself. The demarcation point between the long-term stability and the short-term stability is generally determined by the loop bandwidth of the phase-locked loop. In actual engineering, it is very difficult to achieve a source with frequency stability meeting the requirements of high-precision measurement through this method; the third factor is that the PLL has limited accuracy in changing the frequency output. The integer-divided phase-locked loop chip has too large a step, and the fractional phase-locked loop chip can have a smaller step, but it is also very difficult to meet very fine frequency adjustment through the method of integer division and fractional ratio. The DDS method generally uses the frequency of the reference frequency signal after frequency multiplication as its main frequency, so as to control the output frequency signal through the DDS frequency control word.

[0074] In this method, the reference frequency signal of the beat method is changed to the frequency signal output by the DDS. The clock of the DDS is obtained by multiplying the reference frequency signal. Theoretically, the frequency stability of the DDS output signal is the same as that of the reference frequency signal. Of course, there is a certain loss in actual engineering. As long as the loss is controlled within a certain range, it will not become the main factor affecting the frequency stability index of the output signal. The advantage of the DDS is that the frequency adjustment is accurate and has high precision, and the adjustment is very convenient. The output frequency can be easily adjusted through the DDS frequency control word, and two orthogonal signals can be output simultaneously.

[0075] Step 108: According to the two orthogonal new reference frequency signals, the first frequency signal, and the second frequency signal, perform quadrature double mixing processing to obtain the mixing processing result.

[0076] Specifically, in this direction, the mixing in the beat method is changed to quadrature double mixing. The inputs of the quadrature double mixing respectively come from the signals after frequency division of the frequency signal to be measured and the two orthogonal signals of the DDS.

[0077] Step 110: After the mixing processing result is successively subjected to low-pass filtering, analog-to-digital sampling, phase detection, phase unwrapping, and filtering processing, calculate the phase difference between the signal with the frequency to be measured and the new reference frequency signal.

[0078] Specifically, after double orthogonal mixing and signal processing of the mixing processing result, calculate the phase difference between the signal with the frequency to be measured and the new reference frequency signal output by the DDS. The frequency stability can be calculated based on the phase difference.

[0079] The mixing processing result obtained by orthogonal double mixing is subjected to low-pass filtering and then AD acquisition. The clock signal for AD acquisition is a square wave signal with a frequency of 1 kHz. In this way, the sampling rate is 1 kHz and the sampling interval is 1 millisecond, meeting the requirement for measuring millisecond stability. The I and Q signals after AD acquisition are converted into phase signals through arctangent transformation. Since the phase signal can only represent the range of -π / 2 to π / 2, considering the continuity of the phase, a phase unwrapping module is added later. The output after phase unwrapping is a continuous phase signal.

[0080] Step 112: Calculate the frequency stability using the Allan variance based on the phase difference.

[0081] Specifically, the frequency stability index of the signal output by the DDS (new reference frequency signal) can be considered approximately equal to the frequency stability index of the reference frequency signal. It can be considered that the output of the DDS signal is a new reference frequency signal: sin(Ω' X t + φ1) and cos(Ω' X t + φ1). Where φ1 is the initial phase. For the reference frequency signal, we generally consider it an ideal signal, so φ1 can be considered a constant. It should be noted here that for the measurement of Ω X , we do not necessarily need to be very precise. However, considering that the sampling rate of the subsequent AD acquisition is 1 kHz, the measurement accuracy should be at least within 100 Hz (the requirement can be fully met using the equal-precision measurement method). Then Ω X - Ω' X < 100 Hz to ensure that the frequency signal after the output low-pass filter satisfies the Nyquist sampling theorem. After arccotangent and phase unwrapping, the output phase can be obtained as (Ω X - Ω' X )t + φ X - φ1. Among them, the first term (Ω X - Ω' X )t is a quantity with a fixed slope, and the third term φ1 is a constant, which can be easily eliminated. These two terms do not affect the calculation of the Allan variance. The term that finally plays a role in calculating the Allan variance is φ X , thus meeting the final measurement requirement.

[0082] Step 114: Calculate the frequency difference between the frequency signal to be measured and the reference frequency signal according to the phase difference and the DDS frequency control word.

[0083] Specifically, the frequency difference result between the frequency signal to be measured and the reference frequency signal can be obtained by combining the DDS frequency control word calculated by the frequency measurement module with the phase difference data. That is: Calculate Ω X -Ω' X and combine the value with the previous DDS frequency control word (to obtain the current frequency value Ω' X ) to calculate the relative frequency deviation.

[0084] In the above method for measuring frequency stability, the method includes: respectively isolating and amplifying, and performing frequency division processing on the obtained reference frequency signal and the frequency signal to be measured; doubling the frequency division signal of the reference frequency signal as the DDS clock source, generating a DDS frequency control word by roughly measuring the frequency of the frequency signal to be measured, thereby generating two orthogonal DDS signals. The frequency of the DDS signal is close to the frequency of the signal to be measured. Use the frequency division result of the frequency signal to be measured and the two orthogonal DDS signals as the inputs of quadrature double mixing for quadrature double mixing processing. After the double mixing processing result is subjected to low-pass filtering, AD sampling, phase detection, phase unwrapping, and filtering processing, calculate the phase difference between the frequency signal to be measured and the new reference frequency signal, calculate the frequency stability according to the phase difference, and calculate the phase difference between the frequency signal to be measured and the reference frequency signal according to the phase difference and the frequency control word. This method can realize the measurement of the frequency stability of the reference frequency signal and the frequency signal to be measured with the same source and different sources, and at the same time, the measurement of the frequency stability of the heterodyne signal is also relatively accurate.

[0085] In one embodiment, step 106 includes: doubling the second reference frequency signal and using it as the reference clock source of the direct digital frequency synthesizer, and controlling the direct digital frequency synthesizer through the DDS frequency control word to obtain two orthogonal new reference frequency signals; the new reference frequency signals include: a sine new reference frequency signal and a cosine new reference frequency signal; the expressions of the sine new reference frequency signal and the cosine new reference frequency signal are respectively:

[0086] y s =sin(Ω' X t+φ1)

[0087] y c =cos(Ω' X t+φ1)

[0088] Where y s is the sine new reference frequency signal, y c is the cosine new reference frequency signal, φ1 is the initial phase, and Ω' Xis the new reference frequency, and t is the time.

[0089] In one embodiment, the two orthogonal new reference frequency signals include a sine new reference frequency signal and a cosine new reference frequency signal; step 108 includes: mixing the sine new reference frequency signal with the first frequency signal to obtain a sine mixing result; mixing the cosine new reference frequency signal with the second frequency signal to obtain a cosine mixing result.

[0090] In one embodiment, the mixing processing results include: a sine mixing result and a cosine mixing result; step 110 includes: respectively performing low-pass filtering on the sine mixing result and the cosine mixing result to obtain a sine low-pass filtering result and a cosine low-pass filtering result; respectively using an analog-to-digital acquisition module to perform data acquisition on the sine low-pass filtering result and the cosine low-pass filtering result to obtain a sine digitized mixing signal and a cosine digitized mixing signal; the analog-to-digital acquisition module samples using a sampling clock with a preset frequency; according to the sine digitized mixing signal and the cosine digitized mixing signal, after performing phase detection using a phase detector and through phase unwrapping, a continuous phase signal is obtained; after filtering the phase signal, the phase difference between the signal with the frequency to be measured and the new reference frequency signal is calculated.

[0091] In one embodiment, according to the sine digitized mixing signal and the cosine digitized mixing signal, after performing phase detection using a phase detector and through phase unwrapping, a continuous phase signal is obtained, including: according to the sine digitized mixing signal and the cosine digitized mixing signal, performing an arctangent transformation to obtain a phase signal in the range of (-π / 2, π / 2); after phase unwrapping the phase signal in the range of (-π / 2, π / 2), a continuous phase signal is obtained.

[0092] In one embodiment, before step 110, it further includes: converting the first reference frequency signal into a square wave signal, performing frequency division processing on the square wave signal to obtain a 1 kHz clock signal, and using the 1 kHz clock signal as the sampling clock with the preset frequency of the analog-to-digital acquisition module.

[0093] In one embodiment, step 106 includes: doubling the second reference frequency signal by means of harmonic frequency doubling and using it as the reference clock source of the direct digital frequency synthesizer, and controlling the direct digital frequency synthesizer through the DDS frequency control word to obtain two orthogonal new reference frequency signals.

[0094] Specifically, for high-precision measurement requirements, if the clock frequency multiplication method is selected by PLL, the bottleneck mentioned above will be encountered again. In practical applications, the reference frequency signal is often an atomic clock and a high-performance crystal oscillator, and its output frequency is basically a conventional frequency, generally 5MHz or 10MHz. Then the harmonic frequency multiplication method can be used. The harmonic frequency multiplication method can achieve integer frequency multiplication, such as multiplication to 100MHz. For a 10MHz reference frequency signal, a 2×5 frequency multiplication method can be used (first distort the signal, take the second harmonic filter amplification output, and then distort it again, take the 5th harmonic filter amplification output), and a 2×5×2 method can be used for a 5MHz reference frequency signal. For a 100MHz main frequency DDS, a sinusoidal signal output within 30MHz can be generated, so it can meet the frequency stability measurement of the frequency signal to be measured within 30MHz.

[0095] If the fixed reference frequency signal is 10MHz or 5MHz, the 10MHz or 5MHz signal is multiplied to 100MHz. The frequency multiplication module adopts harmonic frequency multiplication to minimize the frequency stability loss after frequency multiplication. In this way, the frequency stability characteristics of the two orthogonal signals of DDS and the frequency stability of the reference frequency signal maintain good consistency. The measurement range can meet the measurement requirements of 1MHz-30MHz. After orthogonal dual mixing filtering, the I / Q amplitude and phase calibration is performed. Since the AD acquisition rate is 1kHz, a very high-precision integral ADC can be used. The overall index of the measurement result is high, and the measurement of heterofrequency signals is also relatively accurate.

[0096] In one of the embodiments, the sine low-pass filtering result and the cosine low-pass filtering result are respectively collected by analog-to-digital acquisition modules to obtain a sine digitized mixing signal and a cosine digitized mixing signal, including: the sine low-pass filtering result and the cosine low-pass filtering result are respectively collected by integral analog-to-digital acquisition modules to obtain a sine digitized mixing signal and a cosine digitized mixing signal.

[0097] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0098] Frequency is a statistic. Generally, the characteristics of frequency signals are generally measured by frequency stability and frequency accuracy indicators. The frequency stability parameter is generally represented by the Allan variance. Frequency measurement generally requires a reference frequency and a frequency signal to be measured. When the reference frequency and the frequency to be measured are of the same frequency value, it is relatively easy to measure. Generally, the beat method (as shown in Figure 2 ), or the improved beat method (as shown in Figure 3 ) can be used for measurement. If the dual-mixer time-difference method is used, higher measurement accuracy can be obtained. However, the above methods cannot measure the problem of non-homogeneous frequency signals. If non-homogeneous frequency signals need to be measured, the system needs to be improved to a certain extent.

[0099] In one embodiment, as shown in Figure 4 , a frequency stability measurement system is provided. The system includes: a signal input module 10, a frequency multiplication module 20, a frequency measurement module 30, a direct digital frequency synthesizer 40, a double quadrature mixing module 50, a filtering and sampling module 60, a phase detection module 70, and a data processing module 80.

[0100] The input end of the isolation amplifier 101 of the first input branch of the signal input module 10 receives the reference frequency signal. The output end of the isolation amplifier 101 of the first input branch is connected to the input end of the frequency division module 103 of the first input branch. The second output end of the frequency division module 103 of the first input branch is connected to the input end of the frequency multiplication module 20. The output end of the frequency multiplication module 20 is connected to the clock interface of the direct digital frequency synthesizer. The first output end of the frequency division module 103 of the first input branch is connected to the first input end of the frequency measurement module 30.

[0101] The input end of the isolation amplifier 101 in the second input branch of the signal input module 10 receives the frequency signal to be measured. The output end of the isolation amplifier 101 in the second input branch is connected to the input end of the frequency division module 103 in the second input branch. The first output end of the frequency division module 103 in the second input branch is connected to the second input end of the frequency measurement module 30. The output end of the frequency measurement module 30 is connected to the control input end of the direct digital frequency synthesizer 40 and the first input end of the data processing module 80. The output end of the direct digital frequency synthesizer 40, the second output end and the third output end of the frequency division module 103 in the second input branch are all connected to the input end of the double quadrature mixing module 50. The first output end of the output end of the double quadrature mixing module 50 is connected to the input end of the low-pass filter 601 in the first sampling branch of the filter sampling module 60. The output end of the low-pass filter 601 in the first sampling branch is connected to the input end of the AD module 602 in the first sampling branch. The output end of the AD module 602 in the first sampling branch is connected to the first input end of the phase detector 701. The output end of the phase detector 701 is connected to the input end of the phase unwrapping module 702. The second output end of the double quadrature mixing module 50 is connected to the input end of the low-pass filter 601 in the second sampling branch. The output end of the low-pass filter 601 in the second sampling branch is connected to the input end of the AD module 602 in the second sampling branch. The output end of the AD module 602 in the second sampling branch is connected to the second input end of the phase detector 701. The output end of the phase unwrapping module 702 is connected to the second input end of the data processing module 80.

[0102] Among them, the signal input module 10 includes two input branches composed of an isolation amplifier 101 and a frequency division module 103. The first input branch is used to isolate and amplify the reference frequency signal by using the isolation amplifier 101, and then perform frequency division processing by using the frequency division module 103 to obtain three reference frequency signals, and transmit the second reference frequency signal to the frequency doubling module 20 and the third reference frequency signal to the frequency measurement module 30. The second input branch is used to isolate and amplify the frequency signal to be measured by using the isolation amplifier 101, and then perform frequency division processing by using the frequency division module 103 to obtain three frequency signals, and transmit the second frequency signal and the third frequency signal to the double quadrature mixing module 50, and transmit the first frequency signal to the frequency measurement module 30.

[0103] The frequency doubling module 20 is used to perform frequency doubling processing on the second reference frequency signal, and use the obtained frequency doubling signal as the reference clock source of the direct digital frequency synthesizer 40 and transmit it to the direct digital frequency synthesizer.

[0104] A frequency measurement module is used to measure the frequencies of the received third frequency signal and the third reference frequency signal, obtain a DDS frequency control word, and transmit the DDS frequency control word to a direct digital frequency synthesizer and a data processing module.

[0105] A direct digital frequency synthesizer 40 is used to generate a new reference frequency signal according to the received reference clock source and the DDS frequency control word. The new reference frequency signal includes: a sine new reference frequency signal and a cosine new reference frequency signal.

[0106] A quadrature mixing module 50 is used to mix the sine new reference frequency signal and the first frequency signal to obtain a sine mixed signal, and transmit the sine mixed signal to the first sampling branch of a filtering and sampling module 60. It mixes the cosine new reference frequency signal and the second frequency signal to obtain a cosine mixed signal, and transmits the cosine mixed signal to the second sampling branch of the filtering and sampling module 60.

[0107] The filtering and sampling module 60 includes two sampling branches composed of a low-pass filter 601 and an AD module 602. The two sampling branches are used to filter and sample the sine mixed signal and the cosine mixed signal respectively by using the low-pass filter 601 and the AD module 602 to obtain a sine digital frequency signal and a cosine digital frequency signal.

[0108] A phase detection module 70 is used to perform phase detection on the sine digital frequency signal and the cosine digital frequency signal by using a phase detector 701 based on the arctangent conversion method, and then process it by using a phase unwrapping module 702 to obtain the phase difference between the signal to be measured frequency and the new reference frequency signal, and transmit the phase difference to the data processing module 80.

[0109] The data processing module 80 filters the received continuous phase signals, calculates the phase difference between the signal to be measured frequency and the new reference frequency signal according to the obtained filtering result, calculates the frequency stability by using the Allan variance according to the phase difference, and is also used to calculate the frequency difference between the signal to be measured frequency and the reference frequency signal according to the phase difference and the DDS frequency control word.

[0110] In one of the embodiments, the AD module includes a sampling clock with a preset frequency. The sampling clock is a 1 kHz square wave obtained by converting the third reference frequency signal into a square wave and then performing frequency division processing. The frequency stability measurement system is as Figure 5 shown.

[0111] In a specific embodiment, taking the 5 MHz signal output by 8607 as the reference frequency signal, and taking the 10 MHz signal which is 2 times the frequency of the 5 M output by 8607 as the signal to be measured frequency, and using a 50 Hz filter, the measurement result of the system is as Figure 6 、 Figure 7As shown, where Figure 6 is the phase difference curve after 5M / 10M homologous de-skewing. The abscissa is time, and the ordinate is the relative phase time difference, where the relative phase time difference refers to the relative phase difference between the signal to be measured and the DDS signal (reflecting: φ X / Ω X ); Figure 7 is the 5M / 10M homologous frequency stability curve calculated by stable32 software. Among them, the abscissa is the average time, and the ordinate is the Allan variance (the Allan variance is the result calculated based on the phase difference). Figure 7 The coordinates of the key points on the frequency stability curve in are: (1.00E-03, 6.26E-12), (2.00E-03, 1.23E-11), (4.00E-03, 2.82E-11), (1.00E-02, 3.38E-11), (2.00E-02,,1.65E-11), (4.00E-02, 8.20E-12), (1.00E-01, 3.37E-12), (2.00E-01, 1.68E-12), (4.00E-01, 8.84E-13), (1.00E+00, 3.40E-13), (2.00E+00, 1.73E-13), (4.00E+00, 8.24E-14), (1.00E+01, 3.65E-14), (2.00E+01, 1.84E-14), and (4.00E+01, 6.40E-15). Among them, (1.00E-03, 6.26E-12) represents the abscissa of 10 -3 , and the ordinate of 6.26×10 -12 . It can be seen from the results that there are good measurement results for the measurement of non-homogeneous frequency signals. The 1-second stability is 3.4E-13, and the stability curve is basically linearly related to the average time τ in theory. There is an inflection point at the 10ms stability because a 50Hz filter is used.

[0112] In a specific embodiment, the 5MHz signal input by 8607 is used as the reference frequency signal, and the free-running (free-running means not being controlled, that is, not changing the control amount. Generally, when a control amount is fixed without adjustment, its output is allowed) 10MHz oven-controlled crystal oscillator is used as the signal to be measured frequency. 50Hz filtering is adopted, and the measurement results of the non-homologous 5M / 10M frequency stability and frequency difference are as Figure 8 shown, Figure 8The coordinates of the key points on the middle curve are: (1.00E-03, 8.93E-12), (2.00E-03, 1.75E-11), (4.00E-03, 3.25E-11), (1.00E-02, 4.83E-11), (2.00E-02, 2.42E-11), (4.00E-02, 1.46E-11), (1.00E-01, 4.98E-12), (2.00E-01, 3.16E-12), (4.00E-01, 2.34E-12), (1.00E+00, 2.22E-12), (2.00E+00, 2.43E-12), (4.00E+00, 2.52E-12), (1.00E+01, 3.23E-12), (2.00E+01, 3.34E-12), and (4.00E+01, 3.81E-12).

[0113] Since the reference frequency signal uses an 8607 oven-controlled crystal oscillator with a second stability index of 2E-13, which is exactly one order of magnitude higher than that of the oven-controlled crystal oscillator to be measured (with a second stability index of about 2E-12). Judging from the measurement results, the 1-second stability index is 2.2E-12, which fully meets the second stability index of the oven-controlled crystal oscillator. Moreover, the long-term stability of the free-running oven-controlled crystal oscillator is not very good. It can be seen from the curve that it slightly rises above 10-second stability, which fully conforms to the characteristic curve of the oven-controlled crystal oscillator.

[0114] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0115] The above-described embodiments merely represent several implementation manners of the present application. Their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A frequency stability measurement method, characterized in that The method includes: Separately isolating and amplifying the obtained reference frequency signal and the frequency signal to be measured; Separately performing frequency division processing on the isolated and amplified reference frequency signal and the frequency signal to be measured to obtain three reference frequency signals and three frequency signals; Measuring the frequencies of the third reference frequency signal and the third frequency signal to obtain the DDS frequency control word; Doubling the second reference frequency signal through harmonic frequency doubling as the reference clock source of the direct digital frequency synthesizer, and controlling the direct digital frequency synthesizer through the DDS frequency control word to obtain two orthogonal new reference frequency signals, namely the sine new reference frequency signal and the cosine new reference frequency signal, which are respectively expressed as: where y s is a sine new reference frequency signal, and y c is a cosine new reference frequency signal, is the initial phase, and Ω ' X is the new reference frequency, and t is time; Mixing the sine new reference frequency signal with the first frequency signal to obtain a sine mixing result, and mixing the cosine new reference frequency signal with the second frequency signal to obtain a cosine mixing result; After successively performing low-pass filtering, analog-to-digital sampling using an integral ADC, phase detection, phase unwrapping, and filtering on the mixing processing result, calculating the phase difference between the frequency signal to be measured and the new reference frequency signal; Calculating the frequency stability using the Allan variance according to the phase difference; Calculating the frequency difference between the frequency signal to be measured and the reference frequency signal according to the phase difference and the DDS frequency control word.

2. The method according to claim 1, wherein The mixing processing result includes: a sine mixing result and a cosine mixing result; After successively performing low-pass filtering, analog-to-digital sampling, phase detection, phase unwrapping, and filtering on the mixing processing result, calculating the phase difference between the frequency signal to be measured and the new reference frequency signal includes: Separately performing low-pass filtering on the sine mixing result and the cosine mixing result to obtain a sine low-pass filtering result and a cosine low-pass filtering result; Separately collecting data on the sine low-pass filtering result and the cosine low-pass filtering result using an analog-to-digital acquisition module to obtain a sine digitalized mixing signal and a cosine digitalized mixing signal; the analog-to-digital acquisition module samples using a sampling clock with a preset frequency; According to the sine digitalized mixing signal and the cosine digitalized mixing signal, performing phase detection using a phase detector and then through phase unwrapping to obtain a continuous phase signal; After filtering the phase signal, calculating the phase difference between the frequency signal to be measured and the new reference frequency signal.

3. The method according to claim 2, characterized in that, According to the sine digitalized mixing signal and the cosine digitalized mixing signal, performing phase detection using a phase detector and then through phase unwrapping to obtain a continuous phase signal, including: According to the sine digitalized mixing signal and the cosine digitalized mixing signal, performing an arctangent transformation to obtain a phase signal in the range of (-π / 2, π / 2); After unwrapping the phase signal in the range of (-π / 2, π / 2), obtaining a continuous phase signal.

4. The method according to claim 1, wherein Before the step of calculating the phase difference between the frequency signal to be measured and the new reference frequency signal after successively performing low-pass filtering, analog-to-digital sampling, phase detection, phase unwrapping, and filtering on the mixing processing result, Convert the first reference frequency signal into a square wave signal, perform frequency division processing on the square wave signal to obtain a 1 kHz clock signal, and use the 1 kHz clock signal as the sampling clock for the preset frequency of the analog-to-digital acquisition module.

5. A frequency stability measurement system, characterized in that, The system includes: a signal input module, a frequency multiplication module, a frequency measurement module, a direct digital frequency synthesizer, a dual quadrature mixing module, a filtering and sampling module, a phase detection module, and a data processing module; The signal input module includes two input branches composed of isolation amplifiers and frequency division modules; the first input branch is used to isolate and amplify the reference frequency signal using an isolation amplifier, and then perform frequency division processing using a frequency division module to obtain three reference frequency signals, and transmit the second reference frequency signal to the frequency multiplication module and the third reference frequency signal to the frequency measurement module; the second input branch is used to isolate and amplify the frequency signal to be measured using an isolation amplifier, and then perform frequency division processing using a frequency division module to obtain three frequency signals, and transmit the second frequency signal and the third frequency signal to the dual quadrature mixing module, and transmit the first frequency signal to the frequency measurement module; The frequency multiplication module is used to perform frequency multiplication processing on the second reference frequency signal by means of harmonic frequency multiplication, and use the obtained frequency multiplication signal as the reference clock source of the direct digital frequency synthesizer and transmit it to the direct digital frequency synthesizer; The frequency measurement module is used to measure the frequency of the received third frequency signal and the third reference frequency signal to obtain a DDS frequency control word, and transmit the DDS frequency control word to the direct digital frequency synthesizer and the data processing module; The direct digital frequency synthesizer is used to generate a new reference frequency signal according to the received reference clock source and DDS frequency control word. The new reference frequency signal includes: a sine new reference frequency signal and a cosine new reference frequency signal, which are respectively expressed as: Among them, y s is a sine new reference frequency signal, and y c is a cosine new reference frequency signal, is the initial phase, and Ω ' X is the new reference frequency, and t is the time; The dual quadrature mixing module is used to mix the sine new reference frequency signal and the first frequency signal to obtain a sine mixed signal, and transmit the sine mixed signal to the first sampling branch of the filtering and sampling module, mix the cosine new reference frequency signal and the second frequency signal to obtain a cosine mixed signal, and transmit the cosine mixed signal to the second sampling branch of the filtering and sampling module; The filtering and sampling module includes two sampling branches composed of a low-pass filter and an AD module. The two sampling branches are used to filter and sample the sine mixed signal and the cosine mixed signal using the low-pass filter and the AD module respectively to obtain a sine digital frequency signal and a cosine digital frequency signal. Among them, an integrating ADC is applied in the AD module; The phase detection module is used to perform phase detection on the sine digital frequency signal and the cosine digital frequency signal using a phase detector based on the arctangent conversion method, and then perform processing using a phase unwrapping module to obtain a continuous phase signal, and transmit the continuous phase signal to the data processing module; The data processing module is used to filter the received continuous phase signals, calculate the phase difference between the signal with the frequency to be measured and the new reference frequency signal according to the obtained filtering result, calculate the frequency stability using the Allan variance based on the phase difference, and is also used to calculate the frequency difference between the signal with the frequency to be measured and the reference frequency signal according to the phase difference and the DDS frequency control word.

6. The system according to claim 5, wherein The AD module includes a sampling clock with a preset frequency, and the sampling clock is a 1 kHz square wave obtained by converting a third reference frequency signal into a square wave and then performing frequency division processing.

Images