Digital correlator and amplitude-phase dynamic calibration method applied to nulling interferometer

Abstract

The application provides a digital correlator and amplitude-phase dynamic calibration method applied to a nulling interferometer and relates to the technical field of digital signal processing, and comprises a digital acquisition module, a correlation processing module and a data transceiving module connected in sequence; the digital acquisition module performs analog-digital conversion on two-channel intermediate frequency analog signals to obtain two-channel time domain digital signals; the correlation processing module performs complex correlation processing on the two-channel time domain digital signals; and the data transceiving module uploads the complex correlation results to an upper computer through a PCIe protocol; the application does not need quadrature demodulation, avoids errors caused by quadrature demodulation, solves the problem of high-speed broadband signal cross-correlation in a parallel pipeline mode, improves the overall cross-correlation calculation efficiency, can be developed into a device module specially used for digital signal cross-correlation processing requirements, provides technical support for related applications of microwave signal cross-correlation processing, and is suitable for solar radio nulling interferometers, solar image instruments and other applications requiring digital signal cross-correlation processing requirements.

Description

Technical Field

[0001] This invention belongs to the field of digital signal processing technology, and particularly relates to a digital correlator and a dynamic amplitude-phase calibration method for a zero-elimination interferometer. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Solar flares generate a large number of high-energy electrons, which threaten satellite equipment, space activities, and personnel safety within the Earth's space system, making them one of the most dangerous space weather hazards. Precisely determining the radio spectrum of solar flares in the millimeter-wave region will provide powerful clues to understanding the distribution of high-energy electrons accelerated by flares. Although millimeter-wave solar eruptions are of extremely high academic importance, observations of them have been limited to date; this is because atmospheric absorption in the millimeter-wave region hinders observation of solar flares.

[0004] From the centimeter wave band to the millimeter wave band, atmospheric absorption fluctuations are significant, and solar ejections are very quiet. Due to the large fluctuations in the emitted steady-state radio waves, it is difficult to detect the bursts through the output level. To solve this problem, there are three methods:

[0005] (1) Increase the antenna aperture to observe parts of the sun.

[0006] If the antenna's field of view is reduced, only a portion of the sun can be observed. The quiet solar component will also decrease proportionally with the width of the field of view, thus reducing the amount of fluctuation caused by atmospheric absorption of vibrations.

[0007] (2) Compensation of atmospheric radiation for absorption by the quiet sun

[0008] The method was used in the 35 GHz polarimeter at the Nobeyama Solar Radio Observatory in Japan.

[0009] (3) Interference is performed using the outputs of two small-aperture antennas to eliminate fluctuations in the quiescent sun.

[0010] This method is employed by the 80 GHz intensity meter at the Nobeyama Solar Radio Observatory in Japan. On one hand, the Fourier component of the quiescent sun is zero at specific spatial frequencies (approximately 140, 230, and 350 times the wavelength). By appropriately spacing the two antennas, they become insensitive to the quiescent sun, virtually eliminating atmospheric fluctuations. On the other hand, at lower spatial frequencies, since the bursts occur in a region smaller than the entire solar region, the correlation outputs of the two antennas are not lost, resulting in the cancellation of the quiescent sun component caused by atmospheric fluctuations, thus enabling the detection of the bursts. For example, when the spacing between the two antennas is 230 wavelengths, for the 80 GHz intensity meter... In a static Sun, the correlation output of these two antennas is zero, while for bursts originating from regions much smaller than the entire Sun, it possesses almost perfect sensitivity. If each antenna is installed separately, the effective antenna spacing varies with the Sun's hour angle, so both antennas must be mounted on a coplanar equatorial support. If the two antennas are driven on separate supports, the antenna spacing, viewed from the Sun's angle, will vary with the Sun's hour angle, making it impossible to guarantee that it is a specific multiple of the wavelength. To overcome this challenge, placing the two antennas on a coplanar equatorial support is crucial. Furthermore, by extracting the output of the two antennas using a correlator, bursts occurring anywhere on the Sun's surface can be detected.

[0011] The instrument that detects solar flares using the third method is usually called a radio null interferometer. Currently, the correlators used in millimeter-wave null interferometers are only the 90GHz device at the University of Bern in Switzerland and the 80GHz device at Nobeyama in Japan. The above devices only have a few observation frequencies and no interference results within the frequency band. Moreover, their correlators use analog devices, which are basically composed of in-phase / quadrature power dividers, analog multipliers and low-pass filters. The structure is complex, the system noise figure is large, and the system was built early and gradually no longer meets the needs of radio observation.

[0012] Compared to the two interferometers mentioned above, domestic radio null-elimination interferometers have also been developed. The 39.5-40GHz solar radio null-elimination interferometer at Shandong University is one such example. It solves the problem that previous null-elimination interferometers only had a few observation frequencies and no interference results within the frequency band. It consists of an antenna, an analog front-end, a digital correlator, and a host computer, with a structure as follows: Figure 1 As shown, the digital correlator is the core component of the interferometer, which performs acquisition, filtering, and cross-correlation calculations on the intermediate frequency signal output from the analog front end.

[0013] There is a great deal of research on digital correlators, but most of them are applied to microwave radiometers and heliographs. There is currently no digital correlator specifically for millimeter-wave solar radio nulling interferometry. The correlators in existing nulling interferometers process analog signals, which results in a limited range of applications. In addition, they require orthogonal demodulation, which leads to related calculation errors and efficiency problems. Summary of the Invention

[0014] To overcome the shortcomings of the prior art, this invention provides a digital correlator and dynamic amplitude-phase calibration method for null-eliminating interferometers, eliminating the need for orthogonal demodulation and avoiding errors caused by orthogonal demodulation. It employs a parallel pipeline approach to solve the problem of cross-correlation of high-speed broadband signals, improving overall cross-correlation calculation efficiency. This can be developed into a dedicated device module for digital signal cross-correlation processing, providing technical support for microwave signal cross-correlation processing applications. It is applicable to solar radio null-eliminating interferometers, heliographs, and other applications requiring digital signal cross-correlation processing.

[0015] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0016] The first aspect of this invention provides a digital correlator applied to a zero-elimination interferometer;

[0017] A digital correlator for a zero-elimination interferometer includes a digital acquisition module, a correlation processing module, and a data transceiver module connected in sequence.

[0018] The digital acquisition module performs analog-to-digital conversion on the two incoming intermediate frequency analog signals to obtain two time-domain digital signals.

[0019] The correlation processing module performs complex correlation processing on the two-channel time-domain digital signals to obtain the complex correlation result;

[0020] The data transceiver module uploads the correlation results to the host computer via the PCIe protocol;

[0021] The complex correlation is obtained by filtering with a Hanning window, real-time parallel FFT processing, and three-level butterfly operation to obtain frequency domain data. The frequency domain data is then multiplied by conjugate to obtain the visibility function. Finally, the cross power spectrum and phase difference of the two radio signals are obtained based on the visibility function as the complex correlation result.

[0022] Furthermore, the digital acquisition module is connected to the analog front end of the zero-elimination interferometer and inputs the two-channel intermediate frequency analog signals of the analog front end of the zero-elimination interferometer.

[0023] Furthermore, the digital acquisition module, including an ADC chip, processes the intermediate frequency analog signal of each channel as follows:

[0024] The ADC chip performs digital sampling at a sampling rate of 1.25 Gsps and a resolution of 14 bits.

[0025] The signal from each sampled channel is channelized into 8 channels to obtain a time-domain digital signal.

[0026] Furthermore, the digital acquisition module and the related processing module are connected via an FMC physical interface and use the JESD204B high-speed protocol to transmit signals.

[0027] Furthermore, the related processing module, together with the digital acquisition module, is located below the antenna of the zero-elimination interferometer, and uses optical fiber to transmit the related processing results to the data transceiver module.

[0028] Furthermore, the relevant processing module consists of an FPGA board equipped with an XCKU115 chip, and the specific processing steps are as follows:

[0029] Each of the eight data streams in each channel is windowed using a Hanning window.

[0030] Real-time parallel FFT processing is performed on the windowed data of each channel, and two-channel frequency domain data is obtained through three-level butterfly operation;

[0031] The visibility function is obtained by multiplying the two channel frequency domain data by their conjugates.

[0032] The power and phase of the visibility function are calculated and accumulated over a preset time to obtain the cross power spectrum and phase difference of the two radio signals.

[0033] Furthermore, the sampling rate of the FFT processing is 1.25 Gsps, the number of FFT points is 8192, the channel bandwidth is 153 kHz, and the frequency resolution is 153 kHz.

[0034] Furthermore, the process of uploading the complex correlation results to the host computer via the PCIe protocol specifically involves: packaging and framing the complex correlation results and then transmitting them to the host computer via PCIe. Each frame of data is 128KB in size and consists of 1025×4 points, with the first point of each 1025 points being the frame header data.

[0035] The second aspect of the present invention provides a dynamic amplitude-phase calibration method for a digital correlator applied to a zero-elimination interferometer, which is used for the dynamic amplitude-phase calibration of a digital correlator applied to a zero-elimination interferometer provided in the first aspect.

[0036] A dynamic amplitude-phase calibration method for a digital correlator applied to a zero-elimination interferometer involves iteratively executing the following steps within the acceptable power range of the analog front-end and the ADC of the digital correlator, using a continuously varying power from low to high as the preset power.

[0037] A noise signal at a preset power level is generated using a noise source. After passing through an amplifier and an adjustable attenuator, the noise signal is split into two paths by a power divider and input to two analog front-end channels respectively.

[0038] In the analog front end, the noise signal is amplified, mixed, and filtered, and two channels of intermediate frequency analog signals are output to the digital correlator.

[0039] In the digital correlator, the cross power spectrum and phase difference of the two channel signals are calculated at a preset power level and sent to the host computer;

[0040] The host computer generates binary coefficients based on the cross power spectrum and phase difference of the two channel signals, and feeds them back to the digital correlator for amplitude and phase consistency compensation.

[0041] Furthermore, after calculating the cross power spectrum and phase difference, the digital correlator also sends control commands to the adjustable attenuator to adjust the attenuation factor of the adjustable attenuator by controlling the output level.

[0042] The above one or more technical solutions have the following beneficial effects:

[0043] This invention eliminates the need for orthogonal demodulation, avoiding errors introduced by it. It employs a parallel pipeline approach to solve the problem of cross-correlation in high-speed broadband signals, improving overall cross-correlation calculation efficiency. It can be developed into a dedicated device module for digital signal cross-correlation processing, providing technical support for microwave signal cross-correlation applications. It is applicable to solar radio null-elimination interferometers, heliographs, and other applications requiring digital signal cross-correlation processing. Specifically:

[0044] (1) The digital correlator of the present invention uses a 14-bit ADC chip, which can effectively improve the accuracy of the system and make the system more sensitive; it uses real-time parallel FFT processing, which makes the digital correlator consume less resources and the system has better real-time performance; it uses digital devices instead of traditional analog devices, which can flexibly calculate the visibility function and obtain the cross power spectrum and phase.

[0045] (2) This invention proposes a dynamic calibration method for amplitude and phase consistency, which can reduce the amplitude and phase error of the analog front end under different power signals and improve the correlation.

[0046] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0047] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0048] Figure 1 This is a structural diagram of the interferometer.

[0049] Figure 2 This is a structural diagram of the digital correlator in the first embodiment.

[0050] Figure 3 This is a flowchart of the signal processing for the digital correlator in the first embodiment.

[0051] Figure 4 This is a flowchart of the time-frequency conversion process in the first embodiment.

[0052] Figure 5 This is a block diagram illustrating the implementation of the second embodiment. Detailed Implementation

[0053] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0054] Example 1

[0055] This embodiment discloses a digital correlator applied to a zero-elimination interferometer. Its function is to perform analog-to-digital conversion, filtering, FFT, digital correlation and other processing on the 680MHz-1180MHz intermediate frequency signals output from the two-channel analog front-end of the zero-elimination interferometer, and to obtain the cross power spectrum and phase of the two-channel radio signals, which are then sent to the host computer for storage and display.

[0056] Figure 2 This is a block diagram of a digital correlator. Figure 3 This is a flowchart of the signal processing of a digital correlator, such as... Figure 2 , Figure 3 As shown, a digital correlator applied to a zero-elimination interferometer consists of three modules: a digital acquisition module, a correlation processing module, and a data transceiver module.

[0057] Digital acquisition module

[0058] It is used to input two channels of 680MHz-1180MHz intermediate frequency analog signals and to perform analog-to-digital conversion, channelization, packetization and other processing on the intermediate frequency analog signals.

[0059] The digital acquisition module mainly includes an ADC chip and a clock chip. The ADC has a dynamic range of ≥60dB and an effective bit depth of 9 bits. The input is an analog bandwidth signal from the two-channel analog front end of the zero-canceling interferometer. Each channel is digitally sampled at a sampling rate of 1.25Gsps and a resolution of 14 bits to obtain ADC channel 1 data and ADC channel 2 data. The signal of each channel is channelized into 8 channels. After obtaining the channelized time-domain digital signal, it is output to the relevant processing module.

[0060] The digital acquisition module and the related processing module are connected via the FMC physical interface, and both use the JESD204B high-speed protocol for signal transmission.

[0061] Related processing modules

[0062] Connected to the digital acquisition module, it receives the time-domain digital signal transmitted by the digital acquisition module, performs filtering, complex correlation and other processing on the time-domain digital signal to obtain the complex correlation result, and then transmits the complex correlation result to the data transceiver module, while sending a status signal to the digital acquisition module.

[0063] The related processing module mainly consists of an FPGA board equipped with an XCKU115 chip. It takes the time-domain digital signal from the digital acquisition module as input and performs processing such as Hanning window filtering, FFT, cross-correlation, power and phase calculation, and accumulation on it. At the same time, it provides status signals to the digital acquisition module, including reference clock and timing signals, to control the data acquisition module and transmit the complex correlation results to the data transceiver module.

[0064] In actual solar observation, to reduce the length of the radio frequency cable, the digital acquisition module, the related processing module, and the analog front end are placed together under the antenna. The related processing module has an SFP+-supporting fiber optic interface and uses fiber optics to transmit the related processing results to the data transceiver module.

[0065] The complex correlation section is the core computational part of the digital correlator for processing digital signals, mainly including time-frequency conversion, conjugate multiplication, and other operations.

[0066] Figure 4 It is a flowchart of time-frequency conversion, such as Figure 4 As shown, after the 8 channels of data from each channel output by the digital acquisition module enter the FPGA, the 8 channels of data are first windowed using Hanning windows to achieve filtering and prevent spectral leakage. The window functions are generated by the MATLAB toolbox and amplified as necessary. Then, each windowed data channel is simultaneously processed by a 1k-point FFT. The sampling rate of the FFT processing is 1.25Gsps, and the number of FFT points is 8192, so the bandwidth of each channel is 153kHz, that is, the frequency resolution is 153kHz. Then, a three-level butterfly operation is performed to obtain the frequency domain data of that channel.

[0067] According to the Wiener-Khinchin theorem, the cross-correlation of signals in the time domain is equivalent to the conjugate multiplication of signals in the frequency domain. The conjugate multiplication of the two channel frequency domain data yields the observed visibility function. The correlation processing module then calculates the power and phase of the visibility function. To meet a certain time resolution and smooth out some noise, the calculated power and phase need to be accumulated over a period of time to obtain the final cross-correlation result, which is the cross-power spectrum and phase difference of the two channel radio signals.

[0068] Data transceiver module

[0069] It receives the complex correlation results output by the relevant processing module, performs packaging, framing and other processing, and then sends the results to the host computer for storage and plotting.

[0070] The data transceiver module frames the cross-correlation results. Each frame is 128KB in size and contains 1025×4 points. The first point of each 1025 points is the frame header data. The host computer reads this data, stores it, and plots it.

[0071] The data transceiver module connects to the indoor host computer via a PCIe×8 port. It receives the cross-correlation results from the relevant processing modules via fiber optic cable, and after processing such as packetization and framing, transmits them to the host computer via PCIe. The host computer then performs the next steps, such as drawing.

[0072] Example 2

[0073] Due to differences in the manufacturing precision and components of microwave devices (such as low-noise amplifiers and filters), the signal responses of two analog front-ends with the same parameters may differ. In particular, the amplitude and phase responses at different frequency points within the bandwidth will have different differences. This means that when an RF signal of the same power is split into two paths and passed through two analog front-ends with the same parameters, the amplitude characteristics of the output response within the bandwidth will differ. Both the phase difference and the amplitude difference will be non-zero. The amplitude and phase inconsistency problem can be reduced or even eliminated by adjusting the compensation coefficient.

[0074] This embodiment discloses a method for dynamic amplitude and phase calibration of a digital correlator applied to a zero-elimination interferometer;

[0075] Figure 5 This is a block diagram illustrating the implementation of the amplitude-phase dynamic calibration method, such as... Figure 5 As shown, a dynamic amplitude and phase calibration method for a digital correlator is proposed. The method uses a power sweep approach, which controls the power of the noise signal through the digital correlator to test the amplitude difference and phase difference under different power levels within the bandwidth, and performs dynamic compensation calibration for amplitude and phase consistency.

[0076] The aforementioned power sweep, within the acceptable power range of the analog front-end and digital correlator ADC, involves continuously varying the power from low to high (or from high to low). This continuously varying power is used as the preset power, and the following steps are iteratively performed:

[0077] (1) A noise signal at a preset power level is generated using a noise source. After passing through an amplifier and an adjustable attenuator, the noise signal is split into two paths by a power divider and input to two analog front-ends respectively.

[0078] There are two ways to set the signal power. One is to set it through a noise source, which can generate a noise signal with a certain power. The other is to change the actual signal power received by the ADC through an adjustable attenuator. However, the attenuation factor of the adjustable attenuator has a certain range, so the adjustable attenuator and the noise source need to work together to control the power and thus achieve the power sweep function.

[0079] In this embodiment, the digital correlator can adjust the attenuation factor of the adjustable attenuator through control commands, thereby changing the signal power.

[0080] (2) Noise signals of different power are amplified, mixed and filtered by analog front end and then output as intermediate frequency signals to digital correlator.

[0081] (3) The digital correlator calculates the amplitude difference and phase difference of the noise signal power after analog front-end processing at the power level by receiving the intermediate frequency signal, and packages and sends it to the host computer.

[0082] The digital correlator can also send control commands to the adjustable attenuator, that is, adjust the attenuation factor of the adjustable attenuator by controlling the output level.

[0083] (4) The host computer generates binary coefficients from the amplitude difference and phase difference information and feeds them back to the digital correlator for amplitude and phase consistency compensation.

[0084] (5) Set the next power to the new preset power, return to step (1), and perform a new round of compensation calibration.

Claims

1. A digital correlator applied to a zero-elimination interferometer, characterized in that, It includes a digital acquisition module, a related processing module, and a data transceiver module connected in sequence; The digital acquisition module performs analog-to-digital conversion on the two incoming intermediate frequency analog signals to obtain two time-domain digital signals. The correlation processing module performs complex correlation processing on the two-channel time-domain digital signals to obtain the complex correlation result; The data transceiver module uploads the correlation results to the host computer via the PCIe protocol; The complex correlation is obtained by filtering with a Hanning window, real-time parallel FFT processing and three-level butterfly operation to obtain frequency domain data. The frequency domain data is multiplied by conjugate to obtain the visibility function. Finally, the cross power spectrum and phase difference of the two radio signals are obtained based on the visibility function as the complex correlation result. The digital acquisition module is connected to the analog front end of the zero-canceling interferometer and receives two channels of intermediate frequency analog signals from the analog front end of the zero-canceling interferometer. Within the acceptable power range of the ADC in the analog front-end and digital correlator, a continuously varying power from low to high is used as the preset power. During the continuous variation, the following steps are iteratively executed: A noise signal at the preset power level is generated using a noise source. After passing through an amplifier and an adjustable attenuator, the noise signal is split into two paths by a power divider and input to the two-channel analog front-end. In the analog front-end, the noise signal is amplified, mixed, and filtered, and two-channel intermediate frequency analog signals are output to the digital correlator. In the digital correlator, the cross-power spectrum and phase difference of the two-channel signals at the preset power level are calculated and sent to the host computer. The host computer generates binary coefficients based on the cross-power spectrum and phase difference of the two-channel signals and feeds them back to the digital correlator for amplitude and phase consistency compensation.

2. The digital correlator applied to a zero-elimination interferometer as described in claim 1, characterized in that, The digital acquisition module, including an ADC chip, processes the intermediate frequency analog signal of each channel as follows: The ADC chip performs digital sampling at a sampling rate of 1.25 Gsps and a resolution of 14 bits. The signal from each sampled channel is channelized into 8 channels to obtain a time-domain digital signal.

3. A digital correlator for a zero-elimination interferometer as described in claim 1, characterized in that, The digital acquisition module and the related processing module are connected via an FMC physical interface and use the JESD204B high-speed protocol to transmit signals.

4. A digital correlator for a zero-elimination interferometer as described in claim 1, characterized in that, The related processing module, together with the digital acquisition module, is located below the antenna of the zero-elimination interferometer, and uses optical fiber to transmit the related processing results to the data transceiver module.

5. A digital correlator for a zero-elimination interferometer as described in claim 1, characterized in that, The relevant processing module consists of an FPGA board equipped with an XCKU115 chip, and the specific processing steps are as follows: Each of the eight data streams in each channel is windowed using a Hanning window. Real-time parallel FFT processing is performed on the windowed data of each channel, and two-channel frequency domain data is obtained through three-level butterfly operation; The visibility function is obtained by multiplying the two channel frequency domain data by their conjugates. The power and phase of the visibility function are calculated and accumulated over a preset time to obtain the cross power spectrum and phase difference of the two radio signals.

6. A digital correlator for a zero-elimination interferometer as described in claim 5, characterized in that, The sampling rate of the FFT processing is 1.25 Gsps, the number of FFT points is 8192, the channel bandwidth is 153 kHz, and the frequency resolution is 153 kHz.

7. A digital correlator for a zero-elimination interferometer as described in claim 1, characterized in that, The process of uploading the complex correlation results to the host computer via the PCIe protocol is as follows: the complex correlation results are packaged and framed, and then transmitted to the host computer via PCIe. The size of each frame is 128KB, and each frame contains 1025×4 points. The first point of each 1025 points is the frame header data.

8. A method for dynamic amplitude and phase calibration of a digital correlator as described in any one of claims 1-7, characterized in that, After calculating the cross power spectrum and phase difference, the digital correlator also sends control commands to the adjustable attenuator, adjusting the attenuation factor of the adjustable attenuator by controlling the output level.

Images