Motion scene frequency agility signal time difference and speed difference estimation method, device and system

Abstract

The invention discloses a motion scene frequency agility signal time difference and speed difference estimation method, device and system, and belongs to the field of radio monitoring, and the method comprises the steps: carrying out the least square linear fitting of an intra-second pulse string for time difference estimation, and achieving the time-varying compensation of a first-order difference observed quantity and multi-pulse non-coherent accumulation for improving the estimation precision; for speed difference estimation, correlation processing is carried out on intra-second pulse string frequency domain spectrum phase data based on a cross-ambiguity function, and meanwhile, compensation on second-order difference observed quantity time-varying characteristics and multi-pulse coherent accumulation are realized for improving estimation precision. According to the method, the problem of high-precision estimation of the time difference and the speed difference of the radar frequency agility signal under the motion scene time-varying condition is solved.

Description

Technical Field

[0001] The present invention relates to the field of radio monitoring, and more specifically, to a method, device and system for estimating time difference and speed difference of agile frequency signals in a motion scene. Background Art

[0002] In the existing time-frequency difference estimation scheme, the time-varying influence analysis and compensation method are as follows: the literature "Time-varying influence of time-frequency difference accuracy and compensation estimation method", (Li Zhenqiang, Huang Zhen, Chen Xi, etc., System Engineering and Electronic Technology, 2016, 38 (3): 481-486.). The basic idea of ​​this method is: in view of the problem that the performance of time-varying parameter estimation in motion observation scenarios deteriorates with the increase of accumulation time, the expansion effect of relative frequency expansion on the peak of the mutual ambiguity function in the frequency difference direction is analyzed, the influence of uncompensated relative frequency expansion on the frequency difference estimation accuracy is derived, the frequency difference estimation accuracy attenuation factor and the optimal accumulation time are given, and on the basis of theoretical analysis, a time-varying compensated time-frequency difference estimation method is proposed to overcome the time and frequency expansion effects caused by time variability.

[0003] The contribution of the above existing methods is that they are the first to model and quantitatively analyze the effect of time variability on the expansion of the mutual ambiguity function peak in a motion scene. Since the time variability of electromagnetic signals in different areas within the coverage of the moving observation station is different, the proposed time-frequency difference estimation time variability compensation method only compensates for the time variability of the velocity difference (first-order difference observation), which can meet the requirements for time difference estimation accuracy; for frequency difference estimation, as the signal accumulation time increases, some areas (where the velocity difference change rate / acceleration difference is large) may still have serious deformation or even errors in the mutual ambiguity function peak. At the same time, considering that the frequency of some radiation source signals changes rapidly in actual situations, and the signal frequency jump distribution characteristics are shown within a certain bandwidth range, the mutual ambiguity function method cannot achieve high-precision frequency difference estimation through long-term accumulation, and cannot meet the growing demand for high-precision frequency difference estimation for positioning accuracy improvement. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method, device and system for estimating the time difference and speed difference of agile frequency signals in a motion scene, which solves the problem of high-precision estimation of the time difference and speed difference of radar agile frequency signals under time-varying conditions of motion scenes.

[0005] The object of the present invention is achieved through the following solutions:

[0006] A method for estimating time difference and speed difference of frequency agile signals in a motion scene, comprising:

[0007] The frequency difference measurement is replaced by the radial velocity difference measurement between multiple observation stations, so that the high-precision positioning of the radiation source can be carried out based on the time difference and velocity difference measurement results between multiple observation stations.

[0008] Furthermore, the radial velocity difference measurement between multiple observation stations is used to replace the frequency difference measurement, so as to carry out high-precision positioning of the radiation source based on the time difference and velocity difference measurement results between multiple observation stations, specifically including:

[0009] For time difference estimation, least squares linear fitting is performed on the pulse train within seconds to achieve compensation for the time variation of the first-order differential observations and non-coherent accumulation of multiple pulses to improve the estimation accuracy. For speed difference estimation, correlation processing is performed on the frequency domain spectral phase data of the pulse train within seconds based on the mutual ambiguity function, and compensation for the time variation of the second-order differential observations and coherent accumulation of multiple pulses are achieved to improve the estimation accuracy.

[0010] Furthermore, for the time difference estimation, the least square linear fitting is performed on the pulse train within a second to achieve time-varying compensation for the first-order difference observation quantity and non-coherent accumulation of multiple pulses to improve the estimation accuracy, which specifically includes the following sub-steps:

[0011] Step 1: Stations A and B and stations A and C perform intra-second pulse train pairing and extract the time difference based on the TOA frontier. The k-th pulse time difference τ 1k and τ 2k , assuming that the number of pulses is N, the repetition period is T, and the pulse train TOA vector is T = [t0,...,t0+(k-1)T,...,t0+(M-1)T] T , the pulse train sequence time difference vectors are:

[0012]

[0013] Among them, the arrival time difference between stations A and B and the ground radiation source is τ1, and the speed difference is Δv1; the arrival time difference between stations A and C and the ground radiation source is τ2, and the speed difference is Δv2, and M is the number of pulse width quantization points;

[0014] Step 2: τ 1k =(ΔR 01 +Δv1(t0+(k-1)T)) / c,ΔR 01 is the distance difference between stations A and B and the ground radiation source, let α1=ΔR 01 / c, α2=Δv1 / c, then let α=[α1,α2] T ; Let β1 = ΔR 02 / c, β2=Δv2 / c, then let β=[β1,β2] T , ΔR 02 is the distance difference between stations A and C and the ground radiation source; construct the time-varying compensation matrix H, then τ1 and τ2 are:

[0015]

[0016]

[0017] Step 3: Based on the least squares fast compensation method, the minimum root mean square fitting is realized. The likelihood function of stations A and B is constructed as follows:

[0018]

[0019] Among them, S(α) is the likelihood function constructed by stations A and B, is the estimation result of the time difference change rate vector of stations A and B;

[0020] Step 4: Minimize S(α) and find the partial derivative of the formula, where The first term is the time difference measurement result after time-varying compensation, and the second term is the estimated result of the time difference change rate:

[0021]

[0022] The time difference measurement results after time variation compensation of stations A and C are as follows:

[0023] β ls =(H T ×H) -1 H T ×τ k ;

[0024] Among them, β ls is the time difference change rate vector of stations A and C.

[0025] Furthermore, for the velocity difference estimation, the frequency domain spectrum phase data of the pulse train within a second is correlated based on the mutual ambiguity function, and the time-varying property of the second-order difference observation is compensated and multi-pulse coherent accumulation is performed to improve the estimation accuracy, which specifically includes the following sub-steps:

[0026] Step 1: Consider the time-varying effect to establish the phase difference model of frequency-agile radar signal, where (M k -1)T is the central time corresponding to the kth pulse when time-varying property is not considered, t k To consider the time-varying property, the leading edge moment corresponding to the kth pulse is: is the measured value of the first pulse carrier frequency, τ 11 and τ 21 are the first pulse time difference measurement values, and the time-varying model is as follows:

[0027]

[0028] in, is the phase difference of the Kth pulse between stations A and B, τ 1k is the time difference between stations A and B, T s is the sampling time interval, Δv 1kis the speed difference of the Kth pulse between stations A and B, Δa 01 is the acceleration difference between stations A and B;

[0029] Step 2: To realize the accumulation of the phase difference of the pulse train within the pulse, assuming that the number of paired pulses in one second is N, the first term of stations A and B is compensated by the phase term generated by the time difference, that is, the compensation phase factor is constructed pulse by pulse Perform phase compensation on the established frequency-agile radar signal phase difference model, and we get:

[0030]

[0031] in, is the phase difference of the Kth pulse between stations A and B after phase compensation;

[0032] Step 3: Use the improved mutual fuzzy function method based on high-order differential observations to construct the inclusion variable Δv of stations A and B pulse by pulse 01 and Δa 01 Search vector and construct A and C stations containing variable Δv 02 and Δa 02 The search vector is given based on the time difference measurement and the initial positioning result of the target, combined with the spatial distribution range of the velocity difference and the acceleration difference, and the variable Δv is given. 01 , Δv 02 and Δa 01 , Δa 02 The value range of is used, and the two-dimensional mutual fuzzy operation of velocity difference and acceleration difference is performed within this range to form the velocity difference estimation value corresponding to the maximum value of the search peak value, which is the high-precision measurement result of the velocity difference:

[0033]

[0034]

[0035] in, The Kth pulse, the pth velocity difference search value, the qth acceleration difference search value for the phase difference between stations A and B, Δv 01p The p-th speed search value corresponds to the speed difference between stations A and B, Δa 01q The qth acceleration difference search value corresponds to the acceleration difference between stations A and B; The Kth pulse, the pth velocity difference search value, the qth acceleration difference search value for the phase difference between stations A and C, Δv 02p is the speed difference between stations A and C corresponding to the p-th speed search value, Δa 02q The qth acceleration difference search value corresponds to the acceleration difference between stations A and C. is the estimated speed difference between stations A and B, is the estimated value of the acceleration difference between stations A and B, is the estimated value of the speed difference between stations A and C, is the estimated value of the acceleration difference between stations A and C.

[0036] A device for estimating time difference and speed difference of agile frequency signals in a motion scene, wherein the computer device comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is loaded by the processor, the method described in any one of the above items is executed.

[0037] A system for estimating time difference and speed difference of agile frequency signals in a motion scene comprises the device as described above.

[0038] The beneficial effects of the present invention include:

[0039] The present invention solves the problem of high-precision estimation of velocity difference (frequency difference) of radar frequency-agile signals under time-varying conditions of motion scenes, and proposes a method for estimating time difference and velocity difference of frequency-agile signals in motion scenes based on compensation for time-varying properties of high-order differential observations. The generation mechanism of time difference and velocity difference in motion scenes is derived, the time difference and velocity difference of frequency-agile radar signals are modeled, and a time-varying compensation measurement model is given based on the spatial distribution of time-varying properties of time difference and velocity difference.

[0040] Through simulation analysis, it can be known that when the signal accumulation time reaches 1s, the time difference measurement accuracy of the method of the present invention can reach 12ns (rms); the speed difference measurement accuracy can reach 0.00135m / s (rms); taking the agile frequency signal (center frequency 3105MHz, agile bandwidth 100MHz) as an example, the frequency difference measurement accuracy can reach 0.014Hz (rms). It can be seen that the parameter measurement accuracy requirements of multi-station time-frequency (speed) difference high-precision positioning can be met, and the effectiveness and advancement of the method proposed in the present invention are also verified.

[0041] The method of the present invention is applicable to conventional pulse signals, intra-pulse modulation signals such as frequency modulation and phase modulation, and inter-pulse modulation signals such as repetition frequency stagger, jitter, and slip, and has a wide range of adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0043] Figure 1 This is a schematic diagram of the observation model of ground radiation sources by a three-station formation;

[0044] Figure 2 is a flowchart of the steps of the method according to the embodiment of the present invention;

[0045] Figure 3 The relationship between the time difference measurement accuracy and the signal accumulation time length of the method according to the embodiment of the present invention;

[0046] Figure 4 is the measurement result of the velocity difference of the mutual ambiguity function (without compensating the time-varying property of the higher-order observables);

[0047] Figure 5 is the measurement result of the velocity difference of the mutual ambiguity function (after compensating for the time-varying property of the higher-order observations);

[0048] Figure 6 Comparison of the accuracy of the velocity difference estimation results of the mutual ambiguity function before and after compensating for the time-varying properties of high-order observations. DETAILED DESCRIPTION

[0049] All features disclosed in all embodiments in this specification, or steps in all methods or processes implicitly disclosed, except for mutually exclusive features and / or steps, can be combined and / or expanded or replaced in any manner.

[0050] In view of the current situation in the background, the inventor of the present invention believes after creative thinking that:

[0051] Since there is a positive proportional relationship between the Doppler frequency difference and the radial velocity difference, that is, Δf d =-Δv×f c / c, and the radial velocity difference is only positively correlated with the relative motion between the observation station and the radiation source, but has nothing to do with the signal carrier frequency. Therefore, the present invention proposes to replace the frequency difference measurement with the radial velocity difference (Velocity Difference of Arrival, abbreviated as VDOA, hereinafter referred to as velocity difference) measurement between multiple observation stations, so as to carry out high-precision positioning of the radiation source based on the time difference and velocity difference measurement results between multiple observation stations. In view of the time-varying compensation requirements for frequency difference (velocity difference) in motion scenes, the present invention proposes a motion scene agile frequency signal velocity difference estimation method based on time-varying compensation of high-order differential observations, and applies this method to high-precision estimation of velocity difference of agile frequency radar pulse signals.

[0052] Furthermore, the present invention aims to solve the problem of high-precision estimation of time difference and speed difference (frequency difference) of radar frequency-agile signals under time-varying conditions of motion scenes. Considering that the frequency of some radiation source signals is agile in actual situations, and the signal frequency jump distribution characteristics are presented within a certain bandwidth range, the mutual ambiguity function method cannot achieve high-precision measurement of frequency difference through long-term accumulation. At the same time, due to the rapid movement of the observation station, the time difference and frequency difference have time-varying characteristics within a certain time difference of signal accumulation. Directly performing long-term signal coherent processing based on the classical mutual ambiguity function may result in unclear or even erroneous accumulation of frequency difference peaks. In an embodiment of the present invention, a method for estimating time difference and speed difference of frequency-agile signals in motion scenes based on time-varying compensation of high-order differential observations is proposed. For time difference estimation, least squares linear fitting is performed on the pulse train within seconds to achieve compensation for the time variation of the first-order differential observation (velocity difference) and non-coherent accumulation of multiple pulses to improve the estimation accuracy. For velocity difference estimation, correlation processing is performed on the frequency domain spectral phase data (sparse data) of the pulse train within seconds based on the mutual ambiguity function, and at the same time, compensation for the time variation of the second-order differential observation (acceleration difference) and coherent accumulation of multiple pulses are achieved to improve the estimation accuracy.

[0053] The present invention includes deriving the generation mechanism of time difference and speed difference in motion scenes, modeling the time difference and speed difference of frequency-agile radar signals, and providing a time-varying compensation measurement model based on the spatial distribution of the time-varying nature of the time difference and speed difference, which are described in detail as follows.

[0054] 1. Mechanism of time difference and speed difference

[0055] Assume that at a certain moment, three observation stations ABC are detecting and processing ground radar signals. The geometric relationship of their spatial positions is shown as follows: Figure 1 As shown, the position vectors of stations A, B, and C are p0 = [x0, y0, z0] T , p1=[x1,y1,z1] T and p2 = [x2, y2, z2] T , the position of the ground radiation source is p = [x, y, z] T , the position velocity vectors of the three stations A, B, and C are v0=[v x0 ,v y0 ,v z0 ] T ,v1=[v x1 ,v y1 ,v z1 ] T and v2=[v x2 ,v y2 ,v z2 ] T , the velocity vector of the ground radiation source is v = [v x ,v y ,vz ] T , the position acceleration vectors of the three stations A, B, and C are A0=[a x0 ,a y0 ,a z0 ] T , A1=[a x1 ,a y1 ,a z1 ] T and A2 = [a x2 ,a y2 ,a z2 ] T , the acceleration vector of the ground radiation source is A = [a x ,a y ,a z ] T , the distances between the ground radiation source and stations A, B, and C are r0, r1, and r2.

[0056]

[0057] The time difference between stations A and B and the ground radiation source is τ1, and the time difference between stations A and C and the ground radiation source is τ2. The speed of light is c, then

[0058]

[0059] Since the space-based observation stations move at a high speed, especially the low-orbit platform, the speed can reach several thousand meters / order of magnitude, and the Doppler frequency f generated between each observation station and the ground radiation source d0 、f d1 and f d2 It can be expressed as formula (3):

[0060]

[0061] Considering that the Doppler frequency and radial velocity from the observation station to the ground radiation source satisfy the positive proportional relationship, the radial velocity between stations A, B, and C and the ground radiation source can be expressed as:

[0062]

[0063] The radial velocity difference between stations A and B and the ground radiation source is Δv1, and the radial velocity difference between stations A and B and the ground radiation source is Δv2, which can be expressed as:

[0064]

[0065] The radial acceleration between stations A, B, and C and the ground radiation source can be expressed as:

[0066]

[0067] The radial acceleration difference between stations A and B and the ground radiation source is Δa1, and the radial velocity difference between stations A and B and the ground radiation source is Δa2, which can be expressed as:

[0068]

[0069] (II) Modeling of time difference and speed difference of frequency-agile radar signals

[0070] The signal receiving models of the three observation stations A, B, and C are developed. The difference between the agile frequency signal and the conventional fixed frequency signal is that its signal carrier frequency presents pulse-to-pulse hopping characteristics. The RF signal receiving models of the three stations A, B, and C are:

[0071]

[0072] Among them, f c is the signal carrier frequency, f dA 、f dB and f dC is the Doppler frequency of stations A, B and C, τ A0 , τ B0 and τ C0 is the initial delay, is the initial phase of the received signal. The RF signal is down-converted to the intermediate frequency signal. Assuming that the number of pulse width quantization points is M, k The arrival time of the pulses is N kA T s , the estimated value of the signal intermediate frequency is Then for the kth pulse After DFT is performed at , the intermediate frequency signal of station A in equation (8a) can be expressed as:

[0073]

[0074] Then the phase of the kth pulse at station A is:

[0075]

[0076] Similarly, the phase of the kth pulse at station B is:

[0077]

[0078] Defines the measurement value of the signal carrier frequency Then for the kth pulse It is expressed as shown in formula (11). Similarly, the phase difference between stations A and C is It is expressed as (12)

[0079]

[0080]

[0081] The high-precision measurement of frequency difference depends on the long-term (usually hundreds of milliseconds to one second) coherent accumulation of the signal. The carrier frequency f of each pulse of the frequency-agile signal ck Different, then each pulse Doppler frequency difference Δf d1k and Δf d2k Also different, does not meet the phase difference accumulation conditions, the speed difference Δv1 and Δf d1k satisfy Equivalent relationship, c is the speed of light, since the speed difference has nothing to do with the signal carrier frequency agility, the present invention introduces the speed difference instead of the frequency difference to accumulate the phase difference between pulses, then equations (13) and (14) can be transformed into:

[0082]

[0083]

[0084] (III) Time-varying compensation of time difference and velocity difference and high-precision measurement method

[0085] According to equations (2) and (5), the time difference between stations A and B and the ground radiation source is τ1, and the speed difference is Δv1; the time difference between stations A and C and the ground radiation source is τ2, and the speed difference is Δv2.

[0086] Define the initial sampling time t0, at which the distance difference between stations A and B and the ground radiation source is ΔR 01 , the speed difference is Δv 01 , the acceleration difference (rate of change of velocity) is Δa 01 , the acceleration difference (acceleration change rate) is Δb 01 Similarly, the distance difference between stations A and C and the ground radiation source is ΔR 02 , the radial velocity difference is Δv 02 , the acceleration difference (rate of change of velocity) is Δa 02 , the acceleration difference (acceleration change rate) is Δb 02 Taking stations A and B as examples, the relationships between the time difference, radial velocity difference and each order variable are derived.

[0087] Assume Δa1(t) represents the acceleration difference (rate of change) between stations A and B and the ground radiation source at time t, then:

[0088]

[0089] Assume Δv1(t) represents the velocity difference between stations A and B and the ground radiation source at time t, then

[0090]

[0091] Assume ΔR1(t) represents the distance difference between stations A and B and the ground radiation source at time t, then

[0092]

[0093] Similarly, let ΔR2(t) represent the distance difference between stations A and C and the ground radiation source at time t, then:

[0094]

[0095] In the above equations (17) and (18), the time-varying property of the distance difference directly reflects the time-varying property of the time difference, where the velocity difference, acceleration difference and jerk difference are first-, second- and third-order differential observations, respectively.

[0096] Taking the observation scenario of three stations A, B, and C running at an altitude of 900 km from the ground as an example, the station spacing is about one-tenth of the height. The center positions of stations A and B at time t0 are projected on the ground as point O, covering a diameter of 5000 km in the longitude and latitude directions. Assuming that the ground radiation source is stationary, the distribution of time difference, velocity difference, acceleration difference, and acceleration difference change rate are simulated and analyzed respectively. The results are shown in Table 1 below.

[0097] Table 1 Time difference change rate, speed difference / acceleration difference / jerk difference distribution range

[0098]

[0099] For time difference measurement: the maximum value of the first-order variable speed difference A and B is 727.79m / s, the distance difference changes by 727.79m and the time difference changes by 2.426us in 1 second, and its time variability needs to be compensated; the maximum value of the second-order variable acceleration difference A and B is 5.1463m / s2, the distance difference changes by 2.5732m and the time difference changes by 8.5772ns in 1 second, which has little effect on the time difference measurement, and its time variability can be ignored; similarly, the third-order acceleration is smaller, and its time variability can be ignored.

[0100] For velocity difference measurement: the maximum value of the second-order variable acceleration difference between stations A and B is 5.1463m / s2, and the velocity difference changes by 5.1463m / s within 1 second, which is equivalent to a frequency difference change of 17.1543Hz / GHz. Its time variability needs to be compensated; the maximum value of the third-order acceleration measurement between stations A and C is -0.0145m / s3, and the velocity difference changes by -0.0073m / s within 1 second, which is equivalent to a maximum frequency difference change of -0.0242Hz / GHz. It has little impact on multi-station positioning and its time variability can be ignored.

[0101] The proposed compensation model for time difference and velocity difference time variation can be expressed as:

[0102]

[0103]

[0104] Taking the compensation and high-precision estimation of the time difference and speed difference of the pulse signal of the frequency-agile radar by three moving stations as an example, the process of the implementation method of the present invention is further explained as follows: Figure 1 The specific implementation steps are as follows:

[0105] (1) The implementation process of the time difference measurement method based on the least squares (first-order difference observation) fast compensation is as follows:

[0106] Step 1: Stations A and B and stations A and C perform intra-second pulse train pairing and extract the time difference based on the TOA frontier. The k-th pulse time difference τ 1k and τ 2k , assuming that the number of pulses is N, the repetition period is T, and the pulse train TOA vector is T = [t0,...,t0+(k-1)T,...,t0+(M-1)T] T , the pulse train sequence time difference vectors are:

[0107]

[0108] Step 2: From formula (19), we know that τ 1k =(ΔR 01 +Δv1(t0+(k-1)T)) / c, let α1=ΔR 01 / c, α2=Δv1 / c, then α=[α1,α2] T , similarly let β1=ΔR 02 / c, β2=Δv2 / c, then β=[β1,β2] T , construct the time-varying compensation matrix H as shown in formula (22), then τ1 and τ2 are:

[0109]

[0110]

[0111] Step 3: Based on the least squares fast compensation method, the minimum root mean square fitting is realized, and the likelihood function of stations A and B is constructed as shown in formula (24);

[0112]

[0113] Step 4: To minimize S(α), find the partial derivative of the equation, as shown in equation (25), where The first item is the time difference measurement result after time-varying compensation, and the second item is the estimated result of the time difference changing rate.

[0114]

[0115] Similarly, the time difference measurement results of stations A and C after time-varying compensation are shown in formula (26):

[0116] β ls =(H T ×H) -1 H T ×τ k (26)

[0117] (2) The implementation process of the velocity difference measurement method based on high-order differential observation compensation is as follows:

[0118] Step 1: Take stations A and B as examples, and consider the time-varying effect of the phase difference model of the frequency-agile radar signal (27), where (M k -1)T is the central moment corresponding to the kth pulse without considering the time-varying property, t k To consider the time-varying property, the leading edge moment corresponding to the kth pulse is: is the measured value of the first pulse carrier frequency, τ 11 and τ 21 are the first pulse time difference measurement values, and the time-varying model is as follows:

[0119]

[0120] Step 2: To realize the accumulation of the phase difference of the pulse train within the pulse, assuming that the number of paired pulses in one second is N, the first term of stations A and B is compensated by the phase term generated by the time difference, that is, the compensation phase factor is constructed pulse by pulse By performing phase compensation on equation (27), we can obtain:

[0121]

[0122] Step 3: Use the improved mutual fuzzy function method based on high-order differential observations to construct the inclusion variable Δv of stations A and B pulse by pulse 01 and Δa 01 Search vector, and similarly construct A and C stations containing variables Δv 02 and, Δa 02 The search vector is shown in formula (29). Based on the time difference measurement and the initial positioning result of the target, combined with the spatial distribution range of the velocity difference and the acceleration difference, the variable Δv is given. 01 , Δv 02 and Δa 01 , Δa 02 The value range of the velocity difference and acceleration difference is determined, and two-dimensional mutual fuzzy operation is performed within this range to form a velocity difference estimation value corresponding to the maximum value of the search peak value, which is the high-precision measurement result of the velocity difference.

[0123]

[0124]

[0125] The method proposed in the present invention is used to simulate and analyze the effectiveness and measurement accuracy of time-varying compensation for time difference and speed difference measurements of frequency-agile signals in motion observation scenarios.

[0126] Simulation conditions: Take the observation scenario of three stations A, B, and C running at an altitude of 900km from the ground as an example. The station spacing is about one-tenth of the height. At a certain moment, the center positions of stations A and B are projected on the ground as point O. The signal center frequency is 3105MHz, the agility range is 100MHz, the pulse width is 6.4us, the repetition period is 1000us, and the signal longitude and latitude positions are (124.890537°, 5.99771°).

[0127] Depend on Figure 3 It can be seen that when the accumulation time is 50 ms, the time difference accuracy of the method proposed in the present invention is 18 ns (rms); when the accumulation time reaches 1 s, the time difference accuracy of the method proposed in the present invention is 12 ns (rms), which gradually increases with the accumulation time.

[0128] Depend on Figure 4 , Figure 6 It can be seen that when the time-varying compensation of the second-order difference component (acceleration difference) is not performed, the velocity difference measurement accuracy first increases and then gradually decreases with the increase of the accumulation time. When the accumulation time reaches more than 0.4s, the peak shape of the mutual ambiguity function gradually deforms or even becomes erroneous.

[0129] Depend on Figure 5 and Figure 6 It can be seen that after the time-varying compensation of the second-order differential component (acceleration difference) is carried out by the method proposed in the present invention, the speed difference measurement accuracy improves with the increase of the accumulation time. When the accumulation time is 50ms, the speed difference measurement accuracy is 0.044m / s (rms), and the frequency difference measurement accuracy is 0.455Hz (rms) calculated at the center frequency of the agile frequency. When the accumulation time reaches 1s, the speed difference measurement accuracy is 0.00135m / s (rms), and the frequency difference measurement accuracy is 0.014Hz (rms) calculated at the center frequency of the agile frequency.

[0130] From the above quantitative simulation analysis results, it can be seen that the measurement accuracy of time difference and velocity difference (frequency difference) based on the method proposed in the present invention meets the requirements of high-precision positioning of multi-station time-frequency (velocity) differences.

[0131] The units involved in the embodiments of the present invention may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, limit the units themselves.

[0132] According to one aspect of an embodiment of the present invention, a computer program product or a computer program is provided, the computer program product or the computer program includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in the above various optional implementations.

[0133] As another aspect, an embodiment of the present invention further provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiment.

Claims

1. A method for estimating time difference and speed difference of frequency-agile signals in motion scenes, characterized in that: include: The frequency difference measurement is replaced by the radial velocity difference measurement between multiple observation stations, so that the high-precision positioning of the radiation source can be carried out based on the time difference and velocity difference measurement results between multiple observation stations.

2. The method for estimating time difference and speed difference of frequency agile signals in motion scenes according to claim 1, characterized in that: The method of replacing the frequency difference measurement with the radial velocity difference measurement between the multiple observation stations, thereby performing high-precision positioning of the radiation source based on the time difference and velocity difference measurement results between the multiple observation stations, specifically includes: For time difference estimation, least squares linear fitting is performed on the pulse train within seconds to achieve compensation for the time variation of the first-order differential observations and non-coherent accumulation of multiple pulses to improve the estimation accuracy. For speed difference estimation, correlation processing is performed on the frequency domain spectral phase data of the pulse train within seconds based on the mutual ambiguity function, and compensation for the time variation of the second-order differential observations and coherent accumulation of multiple pulses are achieved to improve the estimation accuracy.

3. The method for estimating time difference and speed difference of frequency agile signals in motion scenes according to claim 2, characterized in that: For the time difference estimation, the least square linear fitting is performed on the pulse train within a second to achieve time-varying compensation for the first-order difference observation quantity and non-coherent accumulation of multiple pulses to improve the estimation accuracy, which specifically includes the following sub-steps: Step 1: Stations A and B and stations A and C perform intra-second pulse train pairing and extract the time difference based on the TOA frontier. The k-th pulse time difference τ 1k and τ 2k , assuming that the number of pulses is N, the repetition period is T, and the pulse train TOA vector is T = [t0,...,t0+(k-1)T,...,t0+(M-1)T] T , the pulse train sequence time difference vectors are: Among them, the arrival time difference between stations A and B and the ground radiation source is τ1, and the speed difference is Δv1; the arrival time difference between stations A and C and the ground radiation source is τ2, and the speed difference is Δv2, and M is the number of pulse width quantization points; Step 2: τ 1k =(ΔR 01 +Δv1(t0+(k-1)T)) / c,ΔR 01 is the distance difference between stations A and B and the ground radiation source, let α1=ΔR 01 / c, α2=Δv1 / c, then let α=[α1,α2] T ; Let β1 = ΔR 02 / c, β2=Δv2 / c, then let β=[β1,β2] T , ΔR 02 is the distance difference between stations A and C and the ground radiation source; construct the time-varying compensation matrix H, then τ1 and τ2 are: Step 3: Based on the least squares fast compensation method, the minimum root mean square fitting is realized. The likelihood function of stations A and B is constructed as follows: Among them, S(α) is the likelihood function constructed by stations A and B, is the estimation result of the time difference change rate vector of stations A and B; Step 4: Minimize S(α) and find the partial derivative of the formula, where The first term is the time difference measurement result after time-varying compensation, and the second term is the estimated result of the time difference change rate: The time difference measurement results after time variation compensation of stations A and C are as follows: β ls =(H T ×H) -1 H T ×τ k ; Among them, β ls is the time difference change rate vector of stations A and C.

4. The method for estimating time difference and speed difference of frequency agile signals in motion scenes according to claim 2, characterized in that: For the velocity difference estimation, the frequency domain spectrum phase data of the pulse train within a second is correlated based on the mutual ambiguity function, and the time-varying property of the second-order difference observation is compensated and multi-pulse coherent accumulation is performed to improve the estimation accuracy, which specifically includes the following sub-steps: Step 1: Consider the time-varying effect to establish the phase difference model of frequency-agile radar signal, where (M k -1)T is the central time corresponding to the kth pulse when time-varying property is not considered, t k To consider the time-varying property, the leading edge moment corresponding to the kth pulse is: is the measured value of the first pulse carrier frequency, τ 11 and τ 21 are the first pulse time difference measurement values, and the time-varying model is as follows: in, is the phase difference of the Kth pulse between stations A and B, τ 1k is the time difference between stations A and B, T s is the sampling time interval, Δv 1k is the speed difference of the Kth pulse between stations A and B, Δa 01 is the acceleration difference between stations A and B; Step 2: To realize the accumulation of the phase difference of the pulse train within the pulse, assuming that the number of paired pulses in one second is N, the first term of stations A and B is compensated by the phase term generated by the time difference, that is, the compensation phase factor is constructed pulse by pulse Perform phase compensation on the established frequency-agile radar signal phase difference model, and we get: in, is the phase difference of the Kth pulse between stations A and B after phase compensation; Step 3: Use the improved mutual fuzzy function method based on high-order differential observations to construct the inclusion variable Δv of stations A and B pulse by pulse 01 and Δa 01 Search vector and construct A and C stations containing variable Δv 02 and Δa 02 The search vector is given based on the time difference measurement and the initial positioning result of the target, combined with the spatial distribution range of the velocity difference and the acceleration difference, and the variable Δv is given. 01 , Δv 02 and Δa 01 , Δa 02 The value range of is used, and the two-dimensional mutual fuzzy operation of velocity difference and acceleration difference is performed within this range to form the velocity difference estimation value corresponding to the maximum value of the search peak value, which is the high-precision measurement result of the velocity difference: in, The Kth pulse, the pth velocity difference search value, the qth acceleration difference search value for the phase difference between stations A and B, Δv 01p The p-th speed search value corresponds to the speed difference between stations A and B, Δa 01q The qth acceleration difference search value corresponds to the acceleration difference between stations A and B; The Kth pulse, the pth velocity difference search value, the qth acceleration difference search value for the phase difference between stations A and C, Δv 02p is the speed difference between stations A and C corresponding to the p-th speed search value, Δa 02q The qth acceleration difference search value corresponds to the acceleration difference between stations A and C. is the estimated speed difference between stations A and B, is the estimated value of the acceleration difference between stations A and B, is the estimated value of the speed difference between stations A and C, is the estimated value of the acceleration difference between stations A and C.

5. A device for estimating time difference and speed difference of frequency-agile signals in a motion scene, characterized in that: The computer device comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is loaded by the processor, the method according to any one of claims 1 to 4 is executed.

6. A system for estimating time difference and speed difference of frequency-agile signals in motion scenes, characterized in that: Comprising the device as claimed in claim 5.

Images