A double comb down-conversion spectral processing method based on time-frequency analysis

Abstract

The application discloses a double comb down-conversion spectrum processing method based on time-frequency analysis. The processing method comprises the following steps: performing time-frequency transformation on two groups of sampling data obtained after sampling of a to-be-detected electromagnetic signal through a double comb with two different repetition frequencies, to obtain time-frequency distribution data of the two groups of sampling data; extracting start and end time and spectrum of the down-converted to-be-detected electromagnetic signal in the time-frequency distribution data of the two groups of sampling data respectively; matching the down-converted to-be-detected electromagnetic signal in the two groups of time-frequency distribution data according to the start and end time and the spectrum shape of the down-converted to-be-detected electromagnetic signal, to find two down-converted to-be-detected electromagnetic signals corresponding to the same to-be-detected electromagnetic signal; calculating spectrum offset values of the two down-converted to-be-detected electromagnetic signals corresponding to the same to-be-detected electromagnetic signal; calculating a down-conversion multiple according to the spectrum offset values and the double comb repetition frequency difference; and determining the original spectrum position of the to-be-detected electromagnetic signal according to the down-conversion multiple and the spectrum of the down-converted to-be-detected electromagnetic signal. The application can determine the original spectrum position of the to-be-detected electromagnetic signal through the double comb down-conversion spectrum processing method based on time-frequency analysis, and is not limited by the electromagnetic signal speed and bandwidth.

Description

Technical Field

[0001] This invention relates to the field of signal processing, and more particularly to a dual-comb down-conversion spectrum processing method based on time-frequency analysis. Background Technology

[0002] In the past, electromagnetic frequencies and electromagnetic spectrums were typically measured using electronic devices. Electromagnetic measurements using purely electronic methods offer high resolution and flexibility. However, in situations requiring electromagnetic measurements of broadband and high-speed signals, limitations in speed and bandwidth may prevent the use of electronic devices for such measurements.

[0003] Microwave photonics technology has attracted widespread attention due to its advantages such as ultra-wide operating bandwidth, large dynamic range, low loss, light weight, and inherent anti-electromagnetic interference capability. It is considered an effective way to characterize and process electromagnetic signals and provides a new solution to overcome the limitations of electronic technology.

[0004] An optical frequency comb light source with two repetition frequencies can generate two sets of combs with different frequencies. Each comb consists of a series of pulse sequences with fixed time intervals. In the frequency domain, it appears as a series of frequency combs with a fixed frequency interval of frep. An electromagnetic signal can be mixed with the spectral line of these frequency combs that is closest to it, i.e., the frequency is m times frep, thereby downconverting it to a frequency band below frep / 2. The downconversion frequency is an integer multiple of the repetition frequency m, where m is the downconversion factor.

[0005] If we need to deduce the spectral position information of the electromagnetic signal under test from this information, we need to know the down-conversion factor m. If we can simultaneously mix multiple optical pulse signals with different repetition frequencies with the electromagnetic signal under test, we can obtain multiple mixed signals, and then calculate the value of m, thereby obtaining the relevant information of the electromagnetic signal under test.

[0006] Because optical frequency combs have the advantages of large bandwidth and fixed repetition frequency, using two sets of optical frequency combs with different repetition frequencies can downconvert single-tone electromagnetic signals over a wide frequency range and recover the frequency of the single-tone electromagnetic signal through frequency inversion. However, current signal processing methods cannot invert and recover downconverted electromagnetic signal data with a certain spectral width.

[0007] To address the aforementioned problems, this invention provides a dual-comb down-conversion spectrum processing method based on time-frequency analysis. The method includes: performing time-frequency transformation on two sets of sampled data obtained after sampling the electromagnetic signal under test using dual-combs with two different repetition frequencies to obtain time-frequency distribution data of the two sets of sampled data; extracting the start and end times and spectrum of the down-converted electromagnetic signal under test from the time-frequency distribution data of the two sets of sampled data respectively; matching the down-converted electromagnetic signals under test in the two sets of time-frequency distribution data according to the start and end times and spectral shape of the down-converted electromagnetic signals under test to find two down-converted electromagnetic signals under test corresponding to the same electromagnetic signal under test; calculating the spectral offset value of the two down-converted electromagnetic signals under test corresponding to the same electromagnetic signal under test; calculating the down-conversion factor based on the spectral offset value and the difference in repetition frequencies of the dual-combs; and determining the original spectral position of the electromagnetic signal under test based on the down-conversion factor and the spectrum of the down-converted electromagnetic signal under test. This invention enables the determination of the original spectral position of the electromagnetic signal under test through a dual-comb downconversion spectrum processing method based on time-frequency analysis, without being limited by the speed and bandwidth of the electromagnetic signal. Summary of the Invention

[0008] This invention provides a dual-comb downconversion spectrum processing method based on time-frequency analysis.

[0009] This invention discloses a dual-comb down-conversion spectrum processing method based on time-frequency analysis, comprising:

[0010] Step 1: Perform time-frequency transformation on the two sets of sampled data obtained after the electromagnetic signal under test is sampled by a dual comb with two different repetition frequencies to obtain the time-frequency distribution data of the two sets of sampled data;

[0011] Step 2: Extract the start and end times and spectrum of the down-converted electromagnetic signal from the time-frequency distribution data of the two sets of sampled data respectively;

[0012] Step 3: Match the down-converted electromagnetic signals in the two sets of time-frequency distribution data according to the start and end times and spectral shape of the down-converted electromagnetic signals to find two down-converted electromagnetic signals that correspond to the same electromagnetic signal.

[0013] Step 4: Calculate the spectral offset values ​​of two down-converted electromagnetic signals corresponding to the same electromagnetic signal under test;

[0014] Step 5: Calculate the downconversion factor based on the spectral offset value and the difference in repetition frequencies between the two combs;

[0015] Step 6: Determine the original spectral position of the electromagnetic signal under test based on the downconversion factor and the spectrum of the downconverted electromagnetic signal under test.

[0016] In one example, in step 1, the electromagnetic signal to be measured can be a combination of one or more electromagnetic signals, such as a single-tone electromagnetic signal or an electromagnetic signal with multiple frequency components.

[0017] In one example, in step 1, the method for performing time-frequency transformation on the two sets of sampled data includes, but is not limited to, one or more of the following methods: short-time Fourier transform, wavelet transform, Wigner-Ville transform, Hilbert-Huang transform, Cohen-type bilinear transform, Affine-type transform, and Wigner-Hough transform.

[0018] In one example, in step 2, the method for extracting the start and end times and spectrum of the down-converted electromagnetic signal in the time-frequency distribution data of the two sets of sampled data includes, but is not limited to, a combination of methods for performing edge detection on the down-converted electromagnetic signal in the two time-frequency distribution data, using the minimum and maximum values ​​of the time axis coordinates of the edge detection results as the start and end times of the down-converted electromagnetic signal, and extracting the spectrum of the down-converted electromagnetic signal based on the minimum and maximum values ​​of the frequency axis coordinates of the edge detection results.

[0019] In one example, in step 3, the method of matching the down-converted electromagnetic signals in two sets of time-frequency distributions to find two down-converted electromagnetic signals that correspond to the same electromagnetic signal includes, but is not limited to, a combination of methods of matching down-converted electromagnetic signals with similar start and end times and similar spectral shapes to find two down-converted electromagnetic signals that correspond to the same electromagnetic signal.

[0020] In one example, in step 4, the method for calculating the spectral offset of two down-converted electromagnetic signals corresponding to the same electromagnetic signal under test includes, but is not limited to, a combination of one or more of the following methods: calculating the difference between the frequency corresponding to the maximum spectral amplitude of the down-converted electromagnetic signal under test in the second set of matched sampled data and the frequency corresponding to the maximum spectral amplitude of the down-converted electromagnetic signal under test in the first set of sampled data; calculating the difference between the frequency corresponding to the centroid of the spectral envelope of the down-converted electromagnetic signal under test in the second set of matched sampled data and the frequency corresponding to the centroid of the spectral envelope of the first set of spectra; and calculating the frequency offset between the spectrum of the down-converted electromagnetic signal under test in the second set of matched sampled data and the maximum cross-correlation function of the spectrum of the down-converted electromagnetic signal under test in the first set of sampled data.

[0021] In one example, in step 5, the method for calculating the downconversion factor based on the spectral offset value and the difference in repetition frequencies of the dual combs is to calculate the ratio of the spectral offset value of the downconverted electromagnetic signal under test to the difference in the two repetition frequencies of the dual comb light source, and then round the ratio to obtain the downconversion factor.

[0022] In one example, step 6, the method for determining the original spectral position of the electromagnetic signal under test based on the down-conversion factor and the spectrum of the down-converted electromagnetic signal under test, includes the following steps:

[0023] Step 1: Calculate the repetition frequency difference between the second set of comb repetition frequency and the first set of comb repetition frequency, and calculate the ratio of the offset value of the spectrum of the electromagnetic signal under test after downconversion to the repetition frequency difference.

[0024] Step 2: If the ratio of the frequency difference between the down-converted electromagnetic signal spectrum offset and the double-comb repetition frequency is negative, shift the spectrum of the down-converted electromagnetic signal to the right of the frequency axis by the amount of shifting, which is the product of the repetition frequency and the down-conversion factor, thereby determining the original spectral position of the electromagnetic signal under test.

[0025] Step 3: If the ratio of the frequency offset value of the down-converted electromagnetic signal to the frequency difference of the double comb repetition frequency is positive, the spectrum of the down-converted electromagnetic signal to be tested can be mirrored about the vertical axis of the spectrum. Then, the mirrored spectrum is shifted to the right of the frequency axis by the product of the repetition frequency and the down-conversion ratio, thereby determining the original spectral position of the electromagnetic signal to be tested.

[0026] This invention discloses a dual-comb down-conversion spectrum processing method based on time-frequency analysis. The method includes: performing time-frequency transformation on two sets of sampled data obtained after sampling the electromagnetic signal under test using dual-comb sampling with two different repetition frequencies to obtain time-frequency distribution data of the two sets of sampled data; extracting the start and end times and spectrum of the down-converted electromagnetic signal under test from the time-frequency distribution data of the two sets of sampled data respectively; matching the down-converted electromagnetic signals under test in the two sets of time-frequency distribution data according to the start and end times and spectral shape of the down-converted electromagnetic signals under test to find two down-converted electromagnetic signals under test corresponding to the same electromagnetic signal under test; calculating the spectral offset value of the two down-converted electromagnetic signals under test corresponding to the same electromagnetic signal under test; calculating the down-conversion factor based on the spectral offset value and the difference in repetition frequencies of the dual-comb; and determining the original spectral position of the electromagnetic signal under test based on the down-conversion factor and the spectrum of the down-converted electromagnetic signal under test. This invention can determine the original spectral position of the electromagnetic signal under test through a dual-comb down-conversion spectrum processing method based on time-frequency analysis without being limited by the speed and bandwidth of the electromagnetic signal. Attached Figure Description

[0027] The present invention will now be described in further detail with reference to the accompanying drawings, wherein:

[0028] Figure 1 A flowchart of a dual-comb downconversion spectrum processing method based on time-frequency analysis;

[0029] Figure 2 This is the first set of sampled data obtained after dual-comb sampling;

[0030] Figure 3 This is the second set of sampling data obtained after dual-comb sampling;

[0031] Figure 4 This is a time-frequency distribution data graph of the first set of sampled data;

[0032] Figure 5 This is a time-frequency distribution data graph of the second set of sampled data;

[0033] Figure 6 This is the spectrum diagram of the down-converted electromagnetic signal under test in the first set of sampled data;

[0034] Figure 7 This is the spectrum diagram of the down-converted electromagnetic signal under test in the second set of sampled data;

[0035] Figure 8 This is the original spectral position diagram of the electromagnetic signal under test. Detailed Implementation

[0036] Example 1

[0037] In the process of determining the original spectral position of the electromagnetic signal under test, the electromagnetic signal under test is subjected to repetition frequencies of f. rep1 =53.751901MHz and f rep2 Two sets of sampling data were obtained after dual-comb sampling at 53.753023MHz. The sampling data are as follows: Figure 2 , Figure 3 As shown, a short-time Fourier transform is performed on the two sets of sampled data to obtain the time-frequency distribution data of the two sets of sampled data, as shown in Figure 4. Figure 5 As shown; Sobel edge detection was used to perform edge detection on the time-frequency distribution data of the two sets of sampled data. The minimum and maximum values ​​of the time axis coordinates of the edge detection results were 2 microseconds and 9 microseconds, respectively. Therefore, the start and end times of the down-converted electromagnetic signal under test were 2 microseconds and 9 microseconds, respectively. The minimum and maximum values ​​of the frequency axis coordinates of the edge detection results were 1 MHz and 20 MHz, respectively. The spectrum of the down-converted electromagnetic signal under test was then extracted within the range of 1 MHz and 20 MHz. The extracted spectrum is shown below. Figure 6 , Figure 7As shown. The spectrum shapes of the down-converted electromagnetic signals under test with start and end times ranging from 2 microseconds to 9 microseconds are similar. Therefore, the down-converted electromagnetic signals under test with start and end times ranging from 2 microseconds to 9 microseconds and spectrum distributions at 1 MHz and 20 MHz in the two sets of time-frequency distribution data are matched to become two down-converted electromagnetic signals under test corresponding to the same electromagnetic signal under test. The frequency difference corresponding to the centroid of the spectral envelope in the spectrum of the two down-converted electromagnetic signals under test after matching is used to calculate the spectrum of the down-converted electromagnetic signal under test in the second set of time-frequency distribution data compared with the first set of time-frequency distribution data. The offset value of the spectrum of the down-converted electromagnetic signal under test in the data set is 51600 Hz; the repetition frequency difference between the second comb repetition frequency and the first comb repetition frequency is 1122 Hz, and the ratio of the spectrum offset value to the repetition frequency difference is 45.9. Taking the absolute value after rounding, the down-conversion factor is 46. Since the ratio of the spectrum offset value to the repetition frequency difference is positive, the spectrum of the down-converted electromagnetic signal under test in the first set of sampling data is mirrored about the vertical axis of the spectrum. Then, the mirrored spectrum is shifted to the right of the frequency axis by an amount equal to the repetition frequency f of the first comb. rep1 The product of 53.751901MHz and the downconversion factor m = 46 is used to determine the original spectral position of the electromagnetic signal under test. The original spectral position of the electromagnetic signal under test is as follows: Figure 8 As shown.

[0038] Example 2

[0039] In the process of determining the original spectral position of the electromagnetic signal under test, the electromagnetic signal under test is subjected to repetition frequencies of f. rep1 =53.751901MHz and f rep2Two sets of sampling data were obtained after dual-comb sampling at 53.753023MHz. Short-time Fourier transform was performed on the two sets of sampling data. The Laplacian edge detection method was used to perform edge detection on the time-frequency distribution data of the two sets of sampling data respectively. The start and end times of the down-converted electromagnetic signal under test were found to be 3 microseconds and 7 microseconds. The minimum and maximum values ​​of the frequency axis coordinates of the edge detection results were 2 MHz and 20 MHz respectively. The spectrum of the down-converted electromagnetic signal under test was then extracted in the range of 10 MHz and 20 MHz. The extracted spectrum is shown in the figure. The down-converted electromagnetic signals under test with start and end times ranging from 2 to 9 microseconds have similar spectral shapes. Therefore, the down-converted electromagnetic signals under test with start and end times ranging from 2 to 9 microseconds and spectral distributions at 10 MHz and 20 MHz in the two sets of time-frequency distribution data are matched to become two down-converted electromagnetic signals under test corresponding to the same electromagnetic signal under test. The down-converted electromagnetic signal under test in the first set of time-frequency distribution data is calculated using the method of calculating the frequency difference corresponding to the centroid of the spectral envelope in the spectrum of the two matched down-converted electromagnetic signals under test. The spectrum offset compared to the down-converted electromagnetic signal spectrum in the second set of time-frequency distribution data is -37010 Hz. The repetition frequency difference between the second comb repetition frequency and the first comb repetition frequency is 1000 Hz. The ratio of the spectrum offset to the repetition frequency difference is -37.01. Taking the absolute value after rounding, the down-conversion factor is 37. Since the ratio of the spectrum offset to the repetition frequency difference is negative, the spectrum of the down-converted electromagnetic signal in the second set of sampling data is shifted to the right of the frequency axis by the amount of shifting the repetition frequency f of the second comb. rep2 The original spectral position of the electromagnetic signal under test is determined by multiplying the frequency of 53.753023MHz by the downconversion factor m=37.

Claims

1. A dual-comb down-conversion spectrum processing method based on time-frequency analysis, characterized in that, The specific steps are as follows: Step 1: Perform time-frequency transformation on the two sets of sampled data obtained after the electromagnetic signal under test is sampled by a dual comb with two different repetition frequencies to obtain the time-frequency distribution data of the two sets of sampled data; Step 2: Extract the start and end times and spectrum of the down-converted electromagnetic signal from the time-frequency distribution data of the two sets of sampled data respectively; Step 3: Match the down-converted electromagnetic signals in the two sets of time-frequency distribution data according to the start and end times and spectral shape of the down-converted electromagnetic signals to find two down-converted electromagnetic signals that correspond to the same electromagnetic signal. Step 4: Calculate the spectral offset values ​​of two down-converted electromagnetic signals corresponding to the same electromagnetic signal under test; Step 5: Calculate the downconversion factor based on the spectral offset value and the difference in repetition frequencies between the two combs; Step 6: Determine the original spectral position of the electromagnetic signal under test based on the down-conversion ratio and the spectrum of the down-converted electromagnetic signal under test. This includes the following sub-steps: Calculate the repetition frequency difference between the second comb repetition frequency and the first comb repetition frequency; calculate the ratio of the offset value of the spectrum of the down-converted electromagnetic signal under test to the repetition frequency difference; if the ratio of the offset value of the spectrum of the down-converted electromagnetic signal under test to the frequency difference of the double comb repetition frequencies is negative, shift the spectrum of the down-converted electromagnetic signal under test to the right of the frequency axis by the product of the repetition frequency and the down-conversion ratio, thereby determining the original spectral position of the electromagnetic signal under test; if the ratio of the offset value of the spectrum of the down-converted electromagnetic signal under test to the frequency difference of the double comb repetition frequencies is positive, first mirror the spectrum of the down-converted electromagnetic signal under test about the vertical axis of the spectrum, and then shift the mirrored spectrum to the right of the frequency axis by the product of the repetition frequency and the down-conversion ratio, thereby determining the original spectral position of the electromagnetic signal under test.

2. The dual-comb down-conversion spectrum processing method based on time-frequency analysis as described in claim 1, characterized in that, The method for time-frequency transformation of the two sets of sampled data described in step 1 includes one or more of the following methods: short-time Fourier transform, wavelet transform, Wigner-Ville transform, Hilbert-Huang transform, Cohen-type bilinear transform, Affine-type transform, and Wigner-Hough transform.

3. The dual-comb down-conversion spectrum processing method based on time-frequency analysis as described in claim 1, characterized in that, The method for extracting the start and end times and spectrum of the down-converted electromagnetic signal in the time-frequency distribution data of the two sets of sampled data in step 2 includes performing edge detection on the down-converted electromagnetic signal in the two time-frequency distribution data, taking the minimum and maximum values ​​of the time axis coordinates of the edge detection results as the start and end times of the down-converted electromagnetic signal, and extracting the spectrum of the down-converted electromagnetic signal based on the minimum and maximum values ​​of the frequency axis coordinates of the edge detection results.

4. The dual-comb down-conversion spectrum processing method based on time-frequency analysis as described in claim 1, characterized in that, The method described in step 3 for matching the down-converted electromagnetic signals in two sets of time-frequency distribution data to find two down-converted electromagnetic signals that correspond to the same electromagnetic signal includes matching down-converted electromagnetic signals with similar start and end times and similar spectral shapes to find two down-converted electromagnetic signals that correspond to the same electromagnetic signal.

5. The dual-comb down-conversion spectrum processing method based on time-frequency analysis as described in claim 1, characterized in that, The method for calculating the spectral offset of two down-converted electromagnetic signals corresponding to the same electromagnetic signal under test in step 4 includes a combination of one or more of the following methods: calculating the difference between the frequency corresponding to the maximum spectral amplitude of the down-converted electromagnetic signal under test in the second set of sampled data after matching and the frequency corresponding to the maximum spectral amplitude of the down-converted electromagnetic signal under test in the first set of sampled data; calculating the difference between the frequency corresponding to the centroid of the spectral envelope of the down-converted electromagnetic signal under test in the second set of sampled data after matching and the frequency corresponding to the centroid of the spectral envelope of the first set of spectra; and calculating the frequency offset between the spectrum of the down-converted electromagnetic signal under test in the second set of sampled data after matching and the maximum cross-correlation function of the spectrum of the down-converted electromagnetic signal under test in the first set of sampled data.

6. The dual-comb down-conversion spectrum processing method based on time-frequency analysis as described in claim 1, characterized in that, The method described in step 5 for calculating the downconversion factor based on the spectral offset value and the difference in repetition frequencies of the dual combs is to calculate the ratio of the spectral offset value of the downconverted electromagnetic signal to the difference in repetition frequencies between the two dual comb light sources, and then round the ratio to obtain the downconversion factor.

Images