Millimeter wave satellite-oriented sensing integrated target positioning and identification method
By adopting synesthesia integrated target positioning and identification methods in satellite communications, and using morphological filtering and peak fusion algorithms, the problem of insufficient satellite navigation positioning and anti-interference capabilities is solved, and the capacity and security of satellite communications are improved.
Patent Information
- Application Number
- CN202410990619.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-22
AI Technical Summary
In satellite communications, especially in inter-star communication systems in millimeter wave E-band, it is difficult for the prior art to effectively realize the high-precision navigation and positioning of starlink satellites and anti-interference capabilities, especially in space environments with congested frequency band resources, resulting in Issues of insufficient satellite security and communication capacity.
A synesthesia integrated target positioning and identification method for millimeter wave satellites is adopted, and the identification and positioning of individual targets of Starlink are achieved through the direction angle adjustment, signal search, phase modulation, morphological filtering and peak fusion algorithm of prefabricated antennas. Dynamically track noise power changes in broadband channels to improve the accuracy of signal detection.
It improves the navigation and positioning accuracy and anti-interference ability of the satellite, enhances the system's communication capacity and signal detection performance, reduces false alarm and leak alarm rates, and ensures the safety and communication reliability of the satellite.
Smart Images

Figure CN119010987A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of satellite communication, and in particular to an inter-satellite communication system in the E frequency band of millimeter waves, and is suitable for large-capacity and high-speed communication between satellites and satellite target recognition and perception. Background Art
[0002] The millimeter wave band has abundant available frequency resources and can provide transmission rates from several GHz to hundreds of Gbps, which is comparable to the transmission rate of ground optical fiber. The main application scenarios of millimeter wave ultra-high-speed wireless communication are air-based backbone networks and air-to-air and air-to-ground ultra-high-speed communications. In the vast airspace from the ground to the adjacent space, the air-based backbone network is composed of adjacent space vehicles, and ultra-high-speed data transmission is achieved through millimeter wave links. In addition, millimeter wave electronic systems have a series of advantages such as high tracking and guidance accuracy, low susceptibility to electronic interference, good multi-target identification performance, high radar resolution, low multipath effect and ground clutter interference during low-angle tracking, etc., and are currently widely used in radar, guidance, tactical and strategic communications, electronic countermeasures, remote sensing, and radiation measurement instruments.
[0003] The United States' "Starlink" low-orbit Internet constellation plan plans to launch 42,000 satellites. Space resources will become increasingly crowded, and the safety of satellites will become increasingly severe. It is necessary to improve the accuracy and anti-interference ability of satellite navigation and positioning systems. The "Starlink" satellite frequency band has added the E band on the basis of the original Ku, Ka and V bands. The uplink feeder link is 81~86GHz, and the downlink feeder link is 71.0~76.0GHz. The downlink available bandwidth has increased by 3 times, greatly enhancing the system capacity. The communication capacity of a single satellite is about 20Gbps, and the data throughput of the whole system can reach 100Tbps. At present, the domestic satellite frequency band is still in the low frequency band and has not yet expanded to the E band. In the next few years, the domestic satellite frequency band will definitely move towards the E band. Summary of the invention
[0004] In response to the needs and background of space applications, the present invention proposes a millimeter-wave synaesthesia integrated system suitable for satellite communications, which can identify and locate individual Starlink targets, track and predict trajectories, and ensure its own safety.
[0005] The technical solution adopted by the present invention is:
[0006] A synaesthesia integrated target positioning and identification method for millimeter wave satellites, comprising the following steps:
[0007] Step 1: Before link establishment, the antenna pointing angle is pre-set, and the program guides the antenna to point to the target satellite to search for continuous signals and burst signals;
[0008] Step 2: After the antenna receives the continuous signal and the burst signal, the searched continuous signal is divided into two error signals, azimuth and elevation, and orthogonalized into a single-channel sum and difference signal, and then the corresponding phase modulation is performed to obtain the intermediate frequency signal;
[0009] Step 3, detect the intermediate frequency signal and demodulate the azimuth and elevation error voltage, and judge the in-band signal to locate and identify the target. Then adjust the pointing angle of the antenna according to the positioning and identification results of the target, and return to step 2 to continuously track the signal.
[0010] Furthermore, the specific process of detecting the intermediate frequency signal in step 3 is as follows:
[0011] Step 301: obtaining a contraction smooth envelope of the spectrum waveform by a closing and opening operation at a small scale, and obtaining an expansion smooth envelope of the spectrum waveform by an opening and closing operation at a small scale;
[0012] Step 302, after averaging the contraction smooth envelope and the expansion smooth envelope of the spectrum waveform, perform morphological opening operation at a large scale to estimate the noise waveform characteristics in the spectrum data.
[0013] Furthermore, the specific process of searching for continuous signals in step 1 is as follows:
[0014] Step 101, calculate the FFT of a frame of data in sequence, and obtain the instantaneous amplitude spectrum after taking the modulus of the FFT result; wherein the number of FFT points is a set value;
[0015] Step 102, performing L sliding averages on the instantaneous amplitude spectrum to obtain an average spectrum, wherein the number of averages L is a set value;
[0016] Step 103, when the average times are reached, the minimum value of the average spectrum is recorded to obtain the minimum maintained average spectrum;
[0017] Step 104, returning to step 101, until the number of statistical frames exceeds M, signal detection is performed based on the spectrum line fusion strategy, and the center frequency, bandwidth and signal-to-noise ratio of each searched signal are calculated.
[0018] Furthermore, the specific process of searching for the burst signal in step 1 is as follows:
[0019] Step 201, calculate the FFT of each frame of data, and obtain the instantaneous amplitude spectrum after taking the modulus of the FFT result; wherein the number of FFT points is a set value;
[0020] Step 202, calculating the instantaneous maximum hold spectrum according to the instantaneous amplitude spectrum, and performing L sliding averages on the instantaneous amplitude spectrum to obtain an average spectrum, and when the average times are reached, recording the minimum value of the average spectrum to obtain the minimum hold average spectrum; wherein the average times L is a set value;
[0021] Step 203, performing a difference operation on the instantaneous maximum hold spectrum and the minimum hold average spectrum to obtain a difference spectrum, and accumulating the difference spectrum to obtain a difference accumulation spectrum;
[0022] Step 204, return to step 201, and add 1 to the frame counter. When the number of statistical frames exceeds M, perform signal search based on the peak fusion strategy, and calculate the center frequency, bandwidth and signal-to-noise ratio of each searched signal.
[0023] Furthermore, the specific process of signal search based on the peak fusion strategy is as follows:
[0024] Based on the similarity of energy within the signal band and the difference of energy within and outside the band, the peak points with the same energy characteristics are attributed to one signal, and the signal boundary is analyzed and the center frequency is estimated. When all peak points are attributed, the in-band signal search process is completed.
[0025] The advantages of the present invention compared with the prior art are:
[0026] The adaptive threshold decision technology of the present invention will obtain better detection performance than the fixed threshold detection scheme. For millimeter wave broadband signals, a signal detection technology based on adaptive threshold is proposed. The morphological filtering method is applied to the determination of the adaptive threshold. The time-frequency changes of the noise power in the broadband channel can be dynamically tracked. On this basis, the peak fusion algorithm is adopted to complete the reliable detection of the signal. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 This is a flow chart of target identification and positioning processing of the present invention.
[0028] Figure 2 This is a broadband noise floor adaptive estimation diagram of the present invention.
[0029] Figure 3 This is a block diagram of the continuous signal search principle of the present invention.
[0030] Figure 4 This is a block diagram of the burst signal search principle of the present invention.
[0031] Figure 5 This is a block diagram of the peak fusion search principle of the present invention. DETAILED DESCRIPTION
[0032] The present invention will be further explained below in conjunction with the accompanying drawings.
[0033] A synaesthesia-integrated target positioning and recognition method for millimeter wave satellites, such as Figure 1 As shown, the specific steps include:
[0034] Step 1: Before link establishment, the antenna pointing angle is pre-set, and the program guides the antenna to point to the target satellite to search for continuous signals and burst signals;
[0035] Step 2: After the antenna receives the continuous signal and the burst signal, the searched continuous signal is divided into two error signals, azimuth and elevation, and orthogonalized into a single-channel sum and difference signal, and then the corresponding phase modulation is performed to obtain the intermediate frequency signal;
[0036] Step 3, detect the intermediate frequency signal and demodulate the azimuth and elevation error voltage, and judge the in-band signal to locate and identify the target. Then adjust the pointing angle of the antenna according to the positioning and identification results of the target, and return to step 2 to continuously track the signal.
[0037] The antenna tracking modes are mainly divided into program tracking, program guidance, automatic tracking and manual tracking, and there are other working modes such as standby, hand speed, etc. For satellite-borne antennas, due to the influence of space factors such as volume, weight, and power supply, program tracking or program guidance is mainly used, which uses the calculated space angle to guide the antenna to point to the target. At the same time, considering the accuracy of satellite attitude control, program tracking + autonomous tracking can also be used. Therefore, the working modes of the antenna include: program tracking, standby and autonomous tracking modes.
[0038] Before the link is established, the pointing angle of the antenna is preset in advance, and the program guides the pointing to the target satellite. During the link establishment process, the antenna angle is slightly adjusted. The establishment time is completed by the dual-satellite capture and tracking of the transmitted signal, which can generally be completed within 0.5s.
[0039] In step 3, the signal detection technology adopts the signal detection technology based on adaptive threshold, and applies the morphological filtering method to the determination of the adaptive threshold, which can dynamically track the time-frequency changes of the noise power in the broadband channel. On this basis, the peak fusion algorithm is adopted to complete the reliable detection of the signal.
[0040] The theoretical basis of morphological filtering is mathematical morphology, which utilizes shape-based nonlinear transformation methods and structural elements to modify the local morphology of the signal and approximate the essential characteristics of the signal.
[0041] If the signal's spectral waveform is processed by one-dimensional morphological filtering, the dilation operation will reduce the trough of the signal spectrum and expand the peak width; while the erosion operation will reduce the peak of the spectral waveform and expand the trough.
[0042] By combining the dilation and erosion operations, we can define the morphological opening and closing operations. Among them, the morphological opening operation is non-expansive because the erosion process is performed first, and the part smaller than the interval element will be weakened, so the opening operation can suppress signal glitches and mutations; the morphological closing operation is first expanded and has expansive properties. Through the expansion process, the troughs in the data waveform that are smaller than the interval element can be filled, thereby achieving the effect of suppressing the signal trough. The width of the peaks and troughs that can be filtered out by the morphological opening and closing operations depends on the width M of the interval element b used. By selecting interval elements that are larger than the width of the waveform to be processed, the opening and closing operations can be used to remove specific waveforms, thereby achieving signal smoothing and feature approximation.
[0043] The contraction smooth envelope of the spectrum waveform can be obtained by the closing and opening operation at a small scale, and the expansion smooth envelope of the spectrum waveform can be obtained by the opening and closing operation at a small scale. After averaging the results of the two morphological filtering processes, the morphological opening operation is further performed at a large scale to estimate the noise waveform characteristics in the spectrum data. The noise waveform estimation effect is shown in Figure 2 shown.
[0044] After multi-scale morphological filtering of one-dimensional broadband spectrum data, the in-band noise floor waveform curve can be estimated. The signal detection threshold determined on this basis has adaptive capabilities and can dynamically track the fluctuations of noise power in the broadband channel over time and frequency, thereby solving the false alarm and missed alarm problems introduced by a single fixed energy detection threshold and improving signal detection accuracy.
[0045] In step 1, different search principles are given for different signals:
[0046] 1) For continuous signal search, the corresponding module working principle block diagram is as follows Figure 3 As shown:
[0047] The detection of continuous signals is achieved through statistical analysis of the M-frame FFT amplitude spectrum, including the following steps:
[0048] Calculate the FFT of each frame of data in turn, and output the instantaneous amplitude spectrum after taking the modulus of the FFT result. The number of FFT points can be set.
[0049] The instantaneous amplitude spectrum is sent to the L-time sliding average module, and the average spectrum is output, where the average number L can be set;
[0050] When the cumulative number is reached, the minimum hold module records the minimum value of the average spectrum and outputs the minimum hold average spectrum;
[0051] When the number of statistical frames exceeds M, the minimum maintained average spectrum is sent to the spectrum line fusion search module;
[0052] The spectral line fusion search module completes the signal search based on the spectral line fusion strategy and calculates the center frequency, bandwidth and signal-to-noise ratio of each searched signal.
[0053] In order to ensure that the module is in good working condition, you need to pay attention to the setting of FFT points:
[0054] The spectrum resolution is appropriate, so that two adjacent signals can be clearly distinguished from the spectrum (observe the time-frequency diagram). The spectrum resolution can be adjusted by the number of FFT points;
[0055] The time resolution is appropriate, so that for burst signals, there must be a certain number of frames of data between two consecutive burst intervals to ensure the elimination of burst signals, which can be adjusted by the smoothing times L and the number of FFT points;
[0056] There is a contradiction between time resolution and frequency resolution, which requires a compromise in design.
[0057] 2) For burst signal search, the module principle block diagram is as follows Figure 4 As shown:
[0058] The detection of burst signals is achieved by statistical analysis of the M-frame FFT amplitude spectrum, including the following steps:
[0059] Calculate the FFT of each frame of data, and output the instantaneous amplitude spectrum after taking the modulus of the FFT result. The number of FFT points can be set;
[0060] One instantaneous amplitude spectrum is sent to the maximum hold module to calculate the instantaneous maximum hold spectrum;
[0061] One instantaneous amplitude spectrum is sent to the L-time sliding average module, and the average spectrum is output. The average number L can be set;
[0062] When the average times are reached, the minimum hold module records the minimum value of the average spectrum and outputs the minimum hold average spectrum;
[0063] After the instantaneous maximum hold spectrum and the minimum hold average spectrum are differentially calculated, the differential spectrum is output to the accumulation module. When the number of statistical frames exceeds M, the accumulation module sends the differential cumulative spectrum to the spectrum line fusion search module. At this time, only burst signals exist in the differential cumulative spectrum.
[0064] The spectrum line fusion search module completes the signal search based on the peak fusion strategy, and calculates the center frequency, bandwidth and signal-to-noise ratio of each searched signal. Since the signal search is based on the differential spectrum, the searched signal energy reflects the signal-to-noise ratio.
[0065] Each time a frame of data is calculated, the frame counter is incremented by 1. When the number of statistical frames exceeds M, the frame counter, the maximum holding module and the accumulation module are reset.
[0066] In order to ensure that the module is in good working condition, it is necessary to pay attention to the setting of the FFT points to ensure that the time resolution is appropriate, so that for a burst signal, there must be a frame of data within a certain burst signal duration, thereby reflecting the existence of the burst signal, that is, the shortest duration of the burst signal to be detected is associated with the FFT points through the sampling frequency.
[0067] If the shortest duration of the burst signal to be detected is Tw, the shortest burst interval is Td, and the system sampling frequency is Fs, then the number of FFT points K = min(Tw, Td) / 2*fs.
[0068] 3) For the spectrum line fusion search module, the module completes signal detection based on the adjacent peak fusion strategy. Its working principle is as follows: Figure 5 As shown:
[0069] Based on the basic property that the energy within the signal band is similar, while the energy within the band and outside the band are different, the peak points with the same energy characteristics are attributed to one signal, and their signal boundaries are further analyzed and the center frequency is estimated. When the attribution judgment of all peak points is completed, the in-band signal search process is completed.
Claims
1. A synaesthesia-integrated target positioning and recognition method for millimeter wave satellites, characterized in that: The following steps are involved: Step 1: Before establishing a link, pre-set the antenna pointing angle. The program guides the antenna to point to the target satellite and searches for continuous and burst signals. Step 2: After the antenna receives the continuous signal and the burst signal, it divides the searched continuous signal into two error signals, azimuth and elevation, and orthogonally converts them into a single sum and difference signal. It then performs corresponding phase modulation to obtain an intermediate frequency signal. Step 3: Detect the intermediate frequency signal and demodulate the azimuth and elevation error voltage. At the same time, judge the in-band signal to locate and identify the target. Then, adjust the pointing angle of the antenna according to the target positioning and identification results, and return to step 2 to continuously track the signal.
2. The method for synaesthesia-integrated target positioning and identification for millimeter wave satellites according to claim 1, characterized in that: The specific process of detecting the intermediate frequency signal in step 3 is as follows: Step 301: obtaining a contraction smooth envelope of the spectrum waveform by performing a closing and opening operation at a small scale, and obtaining an expansion smooth envelope of the spectrum waveform by performing an opening and closing operation at a small scale; Step 302 : After averaging the contraction smooth envelope and the expansion smooth envelope of the spectrum waveform, a morphological opening operation is performed on a large scale to estimate the noise waveform characteristics in the spectrum data.
3. The method for synaesthesia-integrated target positioning and identification for millimeter wave satellites according to claim 1, characterized in that: The specific process of searching for continuous signals in step 1 is: Step 101, sequentially calculating the FFT of a frame of data, taking the modulus of the FFT result to obtain the instantaneous amplitude spectrum; wherein the number of FFT points is a set value; Step 102, performing L sliding averages on the instantaneous amplitude spectrum to obtain an average spectrum; wherein the number of averages L is a set value; Step 103: When the number of averages is reached, the minimum value of the average spectrum is recorded to obtain the minimum maintained average spectrum; Step 104, returning to step 101, until the number of statistical frames exceeds M, performing signal detection based on the spectrum line fusion strategy, and calculating the center frequency, bandwidth and signal-to-noise ratio of each searched signal.
4. The method for synaesthesia-integrated target positioning and identification for millimeter wave satellites according to claim 1, characterized in that: The specific process of searching for the burst signal in step 1 is as follows: Step 201, calculate the FFT of each frame of data, and obtain the instantaneous amplitude spectrum after taking the modulus of the FFT result; wherein the number of FFT points is a set value; Step 202: Calculate the instantaneous maximum hold spectrum based on the instantaneous amplitude spectrum, and perform L sliding averages on the instantaneous amplitude spectrum to obtain an average spectrum. When the number of averages is reached, record the minimum value of the average spectrum to obtain the minimum hold average spectrum. The average number L is the set value; Step 203: performing a difference operation on the instantaneous maximum hold spectrum and the minimum hold average spectrum to obtain a difference spectrum, and performing accumulation to obtain a difference cumulative spectrum; Step 204, return to step 201, and add 1 to the frame counter. When the number of statistical frames exceeds M, perform signal search based on the peak fusion strategy, and calculate the center frequency, bandwidth and signal-to-noise ratio of each searched signal.
5. A millimeter wave satellite-oriented synaesthesia integrated target positioning and identification method according to claim 3 or 4, characterized in that: The specific process of signal search based on peak fusion strategy is as follows: Based on the similarity of signal in-band energy and the difference of in-band and out-band energy, peak points with the same energy characteristics are attributed to one signal, and the signal boundary is analyzed and the center frequency is estimated. When all peak points are attributed, the in-band signal search process is completed.
Citation Information
Cited By
E-band high-frequency tracking receiving system, method, equipment and medium
CN120320794A
Terminal equipment positioning method and device and electronic equipment
CN121078532A
