A long synthetic positioning method based on distributed coherent moving platform

Abstract

The present invention discloses a long synthesis positioning method based on a distributed coherent moving platform, and relates to the technical field of target positioning. The present invention is oriented to the distributed moving platform collaborative positioning scenario, and adopts a beam-forming multi-platform collaborative scanning method to ensure that all platforms can receive target area signals; and proposes an intermittent signal transmission method to reduce the demand for data communication and computing costs, and make full use of the position information of the platform's long-term movement, while reducing the number of data transmissions and signal phase alignment times; in view of the nonlinearity of the near-field signal source position information and the time-varying inconsistency of the long-term space vector, a nonlinear approximate motion compensation factor is established, and the distributed array signal phase is compensated to form a spatially equivalent virtual aperture; in view of the problem that the motion compensation factor contains unknown signal source parameters, an improved MUSIC-DPD algorithm is proposed to achieve synchronous estimation of the motion compensation factor and the signal source position.

Description

Technical Field

[0001] The present invention relates to the technical field of target positioning, and in particular to a long synthesis positioning method based on a distributed coherent moving platform. Background Art

[0002] The precise positioning of signal sources can enable us to gain intelligence advantages in electronic reconnaissance and accurately strike targets, so it is of great significance. Direct Position Determination (DPD) technology directly uses the signal source position information contained in the received signal for positioning, and has higher positioning accuracy than the two-step positioning method. Among them, the direct positioning technology based on distributed platform signal coherence directly uses the signal source position information contained in the phase relationship of each platform signal. Through coherent processing and spatial spectrum estimation, the positioning accuracy and resolution can be greatly improved. The algorithm principle of distributed platform coherent positioning is super-resolution spatial spectrum estimation technology, among which the multiple signal classification (MUSIC) algorithm is the most representative. With the help of spatial spectrum estimation technology, the distributed array is first used to realize the direct positioning of the signal source based on the maximum likelihood estimation (MLE), but this method does not use the properties of the signal subspace, and the positioning error is large. Therefore, the MUSIC idea can be introduced into direct positioning and the MUSIC-DPD algorithm is proposed. The cost function is constructed by using a set of bases of the signal subspace and the correlation of the steering vector to realize the direct positioning of multiple signal sources, but this method does not fully combine the position information of each platform from the signal level. The synthetic aperture theory was proposed, and its principle is to use the motion characteristics of the platform to synthesize a virtual aperture. However, although the corresponding existing methods have improved the positioning performance through synthetic aperture technology, they only use a single moving station for positioning and do not take advantage of the ultra-long baseline advantage of the distributed platform. The positioning accuracy and resolution still need to be improved. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a long synthetic positioning method based on a distributed coherent moving platform, which solves the problem that the prior art fails to utilize the ultra-long baseline advantage of the distributed platform, resulting in low positioning accuracy and resolution.

[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0005] A long synthetic positioning method based on a distributed coherent moving platform is provided, which comprises the following steps:

[0006] S1. Set up M motion platforms;

[0007] S2. Sampling the signal source through each motion platform to obtain corresponding sampling signal data;

[0008] S3. Based on each sampled signal data, a distributed multi-platform direct positioning signal model is constructed;

[0009] S4, using the improved MUSIC-DPD algorithm to process the distributed multi-platform direct positioning signal model to obtain the corresponding spatial spectrum;

[0010] S5. Determine the position of the signal source based on the spatial spectrum to complete the positioning of the signal source.

[0011] Furthermore, the sampling in step S2 is performed in an intermittent manner, and the corresponding process is:

[0012] Set the sampling interval T; according to the sampling interval T, perform K intermittent sampling of the signal source through each motion platform; the single sampling duration is T r ; Among them, the time period is T r Contains l signal samples;

[0013] The sampled signal data includes K×l signal samples.

[0014] Furthermore, the formula of the distributed multi-platform direct positioning signal model in step S3 is:

[0015]

[0016] Among them, t k,l represents the sampling time corresponding to the kth sampling, Y(t k,l ) represents the sampling time t k,l The distributed multi-platform direct positioning signal model output, φ s , They respectively represent the direction angle and the downward angle of the reference point to the signal source. represents the sampling time t k,l The steering vector, s(t k,l ) represents the sampling time t k,l The amplitude of the corresponding sampled signal data, exp{·} represents the exponential function with the natural constant e as the base, j represents the imaginary number, π represents the circumference of a circle, and λ s represents the carrier wavelength, Represents the unit vector from the reference point to the signal source, Respectively represent the first motion platform, the second motion platform and the Mth motion platform at the sampling time t k,l location information; Indicates that all moving platforms are at t k,l The sampled signal data.

[0017] Further, step S4 includes the following steps:

[0018] S4-1. Set the parameters of the distributed multi-platform direct positioning signal model and determine the spatial spectrum range and accuracy;

[0019] S4-2, select as the current search value and fix it;

[0020] S4-3, using a motion-compensated synthetic aperture method to process the output of the distributed multi-platform direct positioning signal model at all sampling moments to generate a sample matrix of a virtual array;

[0021] S4-4, calculating the spatial spectrum amplitude of the current search value based on the sample matrix of the virtual array;

[0022] S4-5. Repeat steps S4-2 to S4-5 until all search values ​​are traversed to obtain corresponding spatial spectra.

[0023] Further, step S4-3 includes the following steps:

[0024] S4-3-1. Calculate the displacement of the motion platform at all sampling moments according to the position information of the motion platform at all sampling moments;

[0025] S4-3-2, constructing corresponding motion compensation factors based on the displacement of the motion platform at all sampling moments;

[0026] S4-3-3. According to the motion compensation factor, the sampling signal data of each motion platform is phase-shifted to obtain the sample matrix of the corresponding virtual array.

[0027] Furthermore, the calculation formula of the motion compensation factor in step S4-3-2 is:

[0028]

[0029] Among them, V m represents the moving speed of the mth moving platform, Φ m represents the motion compensation factor of the mth moving platform, and T represents the slow time domain time interval between two samplings of each platform.

[0030] Furthermore, the calculation formula of the sample matrix in step S4-3-3 is:

[0031] Y m =[y m (t1)Φ m K-1 y m (t2)Φ m K-2 … y m (t K )Φ m K-K ]

[0032] Y=[Y1 Y2… Y M ] T

[0033] Among them, Y represents the sample matrix, Y1, Y2, Y m , Y M Respectively represent the sample data after phase shift of the first motion platform, the second motion platform, the mth motion platform and the Mth motion platform, t1, x2 ,t K Respectively represent the first, second and Kth sampling times, Φ m K-1 , Φ m K-2 , Φ m K-K The motion compensation factors of the mth moving platform corresponding to the first, second and Kth phase shifts, y m (t1),y m (t2),y m (t K ) represent the sampling times t1, t2, t K The corresponding signal sample of the mth moving platform, (·) T Represents the transposed matrix.

[0034] Furthermore, the calculation formula of the spatial spectrum amplitude in step S4-4 is:

[0035]

[0036] in, represents the spatial spectrum amplitude, E n represents the noise subspace of the sample matrix, represents the steering vector corresponding to the sampling time K, (·) H represents the conjugate transpose.

[0037] The beneficial effects of the present invention are:

[0038] 1. This method is aimed at the scenario of distributed moving platform collaborative positioning. It adopts a beam-forming multi-platform collaborative scanning method to ensure that all platforms can receive signals in the target area. It also proposes an intermittent signal transmission method to reduce the demand for data communication and computing costs, and make full use of the location information of the platform's long-term movement, while reducing the number of data transmissions and signal phase alignment times.

[0039] 2. This method aims at the nonlinearity of the position information of the near-field signal source and the time-varying inconsistency of the long-term space vector, and establishes a nonlinear approximate motion compensation factor to form a spatially equivalent virtual aperture by compensating the distributed array signal phase;

[0040] 3. To address the problem that the motion compensation factor contains unknown signal source parameters, this method proposes an improved MUSIC-DPD algorithm to achieve synchronous estimation of the motion compensation factor and the signal source position. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of the method of the present invention;

[0042] Figure 2 It is a schematic diagram of the collaborative scanning method of the distributed platform;

[0043] Figure 3 Schematic diagram of the sampling process;

[0044] Figure 4 A three-dimensional motion model diagram of the motion platform;

[0045] Figure 5 Schematic diagram of the synthetic aperture method based on motion compensation. DETAILED DESCRIPTION

[0046] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0047] like Figure 1 As shown, a long synthetic positioning method based on a distributed coherent moving platform comprises the following steps:

[0048] S1. Set up M motion platforms;

[0049] The radar on the motion platform does not transmit signals, but only receives signals from external signal sources. Each motion platform is distributed at different fixed heights and flies in a straight line at a fixed speed. The signal source is located somewhere on the ground, with a Z coordinate of 0 and unknown X and Y coordinates.

[0050] The collaborative scanning method of the distributed motion platform is as follows: Figure 2 As shown. In the figure, the moving platform m and the moving platform n are taken as examples. The two fly in straight lines along their respective azimuths and continuously scan the target area with a spotlight. On the one hand, the beam vectors of each platform at the current moment point to the same target area to ensure that each platform can receive the signal of the target area at the same time for coherent synthesis; on the other hand, the beam vectors of each platform at the historical moment also point to the same target area to ensure that the moving platform can receive the signal of the target area at different times to synthesize the virtual aperture.

[0051] S2. Sampling the signal source through each motion platform to obtain corresponding sampling signal data;

[0052] The sampling of step S2 is performed in an intermittent manner, and the corresponding process is as follows:

[0053] like Figure 3 As shown, a sampling interval T is set; according to the sampling interval T, the signal source is intermittently sampled K times through each motion platform; the single sampling duration is T r ; Among them, the time period is T r Contains l signal samples;

[0054] The sampled signal data includes K×l signal samples.

[0055] During the entire signal acquisition process, only a short period of time T is transmitted and used every time cos(·) r The signal, T r The time period contains l signal samples, and a total of K T r For radar signals, K×l is the pulse repetition period, T r is the pulse width; for communication signals, K×l duration is in milliseconds, T r The duration is in the microsecond level, and the continuous signal can also be processed as a pulse signal. This intermittent transmission method of signal samples, on the one hand, leaves sufficient time for the transmission and data calculation of two groups of signal samples through the sampling interval T, which can eliminate the time domain error caused by signal transmission and data calculation. On the other hand, it can reduce the number of data transmissions and the amount of calculation. This intermittent transmission method only needs K data transmissions, each time transmitting l samples, while the continuous transmission method needs K×l data transmissions; in addition, when performing passive synthetic aperture, the continuous transmission method needs to phase shift all K×l signal samples, while the intermittent transmission method only needs to phase shift the signals of K sampling periods, and the amount of data calculation is significantly reduced.

[0056] S3. Based on each sampled signal data, a distributed multi-platform direct positioning signal model is constructed;

[0057] Taking the motion platform m as an example, its three-dimensional motion model is as follows Figure 4 As shown. Take a point O in space as the reference point and establish a spatial rectangular coordinate system. The motion platform m moves at a speed V m Flying at a constant speed parallel to the ground in a known direction at t k-1 to k+1 During the period, it moves from point A to point C. The moving platform m at t k At the moment, it is at point B, and its coordinates are The displacement from point B to point C is recorded as The signal source S is placed somewhere on the ground, and its coordinates are r(s)=[X s Y s Z s ], where X s and Y s Unknown, Z s is 0. The vector from the moving platform m to the signal source S is recorded as

[0058] For convenience of representation, let the unit vector from the reference point O to the target be φ is the direction angle, is the pitch angle. Represents the unit vector pointing to the signal source. Since the Z of the signal source S s The coordinate is 0, so it is still X s , Y s The function is:

[0059]

[0060] Among them, cos(·) and sin(·) represent the cosine function and sine function respectively, (·) T Represents the transposed matrix.

[0061] Because the reference point O is arbitrarily selected, when it is selected as the center of the distributed array, the distance r from the reference point to the signal source is (s) is much larger than the distance to the platform, that is Therefore, the distance between the motion platform m and the signal source S is It can be written as:

[0062]

[0063] This formula converts the unknown Separate the complex nonlinear relationship to facilitate subsequent phase shifting of the signal to achieve virtual aperture synthesis.

[0064] The modulation signal bandwidth emitted by the signal source is narrow. When the platform sampling rate is high, the received signal can be considered to be a single-frequency signal. Taking a sampling moment as an example, due to the propagation delay The influence of the signal received by the moving platform m relative to the reference point O for:

[0065]

[0066] Among them, s(t) represents the amplitude, f s represents the carrier frequency, and t represents the sampling time.

[0067] For f = fs The signal is multiplied by the known quantity exp{-j2πf s t k}, get the output signal y of the motion platform m m (t k,l )for:

[0068]

[0069] Among them, t k Indicates the sampling time.

[0070] The output signals of M platforms at the same sampling time are combined to obtain a distributed moving platform direct positioning signal model, so the formula of the distributed multi-platform direct positioning signal model in step S3 is:

[0071]

[0072] Among them, t k,l Indicates the sampling time corresponding to the kth sampling, that is, the starting time of the kth sampling, Y(t k,l ) represents the sampling time t k,l The distributed multi-platform direct positioning signal model output, φ s , They respectively represent the direction angle and the downward angle of the reference point to the signal source. represents the sampling time t k,l The steering vector, s(t k,l ) represents the sampling time t k,l The amplitude of the corresponding sampled signal data, exp{·} represents the exponential function with the natural constant e as the base, j represents the imaginary number, π represents the circumference of a circle, and λ s represents the carrier wavelength, Represents the unit vector from the reference point to the signal source, Respectively represent the first motion platform, the second motion platform and the Mth motion platform at the sampling time t k,l location information; Indicates that all moving platforms are at t k,l The sampled signal data.

[0073] S4, using the improved MUSIC-DPD algorithm to process the distributed multi-platform direct positioning signal model to obtain the corresponding spatial spectrum;

[0074] The actual received signal contains not only the signal from the signal source, but also interference noise. Noise m (t k,l ) is a normal distribution with a mean of 0 and a variance equal to the noise power.

[0075] The basic principle of spatial spectrum estimation is to estimate the autocorrelation matrix using sample signals, and use the orthogonality of the noise subspace and the signal subspace to generate the spatial spectrum, and then use the exhaustive search method to find the spectrum peak. Since the steering vector can also be spanned into the signal subspace, the orthogonality of the noise subspace and the steering vector can be used to generate the spatial spectrum.

[0076] Using the actual sample matrix Estimated signal autocorrelation function for:

[0077]

[0078] right Perform eigenvalue decomposition to obtain the noise subspace E n The phase of all time period signals is shifted by the synthetic aperture method based on motion compensation, and compared with t K The phase of the time period signal is aligned, so t K The guiding vector at the moment Span the signal subspace:

[0079]

[0080] Then, using the noise subspace E n and the steering vector The orthogonality of the spatial spectrum is used to construct the motion compensation factor Φ. This method does not calculate the motion compensation factor Φ through prior information, but when using the exhaustive search method to find the spectrum peak, the When the cost function (spatial spectrum amplitude) reaches the maximum value, the motion compensation factor Φ is the actual value, so the motion compensation factor and the signal source position are estimated synchronously.

[0081] Step S4 includes the following steps:

[0082] S4-1. Set the parameters of the distributed multi-platform direct positioning signal model and determine the spatial spectrum range and accuracy;

[0083] S4-2, select as the current search value and fix it;

[0084] S4-3, using a motion-compensated synthetic aperture method to process the output of the distributed multi-platform direct positioning signal model at all sampling moments to generate a sample matrix of a virtual array;

[0085] The synthetic aperture principle of the present invention is: by constructing a suitable motion compensation factor, the phase of the received sampling signal of each sampling period is shifted, and the phase of the received sampling signal of each sampling period is shifted. kThe phase of the signal is aligned, and the shifted signal is equivalent to a virtual synthetic array in the sampling period t k Received signal.

[0086] Consider t k+1 The received signal of the time period is Y(t k+1 ),in, is the motion platform m at t k to k+1 The displacement of the time period, when the motion platform moves in a uniform linear motion, the received signal Y(t k+1 )for:

[0087]

[0088] The left side of each row is t k The received signal of the time period, the right formula is the phase change caused by the platform movement between adjacent time periods, which contains the signal source position information Thus, a motion compensation factor Φ can be constructed to compensate for the phase change of the signal in adjacent sampling periods to achieve phase shift of the signal.

[0089] Synthetic aperture methods based on motion compensation such as Figure 5 As shown in Figure 2. Taking the moving platform m as an example, it has experienced K sampling periods in total, and each sampling period is sampled L times uniformly. k For example, all the sampling signals in this period are combined into y1(t k )for:

[0090] y1(t k )=[y1(t k,1 ) y1(t k,2 ) … y1(t k,L )]

[0091] Among them, y1(t k,1 ), y1(t k,2 ), y1(t k,L ) represent the sampling period t k The first, second and last sampled signals.

[0092] Next, the virtual aperture is synthesized. The signal is phase shifted from t k Move the time period to t k+1 The time period needs to be multiplied by the motion compensation factor Φ1, and so on, and moved to t K The time period needs to be multiplied by Φ1 K-K Similarly, move the signals of the moving platform m in all sampling periods to t K During this period, the received signal of the moving platform m after passing through the synthetic aperture is obtained.

[0093] The virtual array signals of all M motion platforms are combined to obtain a sample matrix Y. That is, step S4-3 includes the following steps:

[0094] S4-3-1. Calculate the displacement of the motion platform at all sampling moments according to the position information of the motion platform at all sampling moments;

[0095] S4-3-2, constructing corresponding motion compensation factors based on the displacement of the motion platform at all sampling moments;

[0096] The calculation formula of the motion compensation factor in step S4-3-2 is:

[0097]

[0098] Among them, V m represents the moving speed of the mth moving platform, Φ m represents the motion compensation factor of the mth moving platform, and T represents the slow time domain time interval between two samplings of each platform.

[0099] S4-3-3. According to the motion compensation factor, the sampling signal data of each motion platform is phase-shifted to obtain the sample matrix of the corresponding virtual array.

[0100] The calculation formula of the sample matrix in step S4-3-3 is:

[0101] Y m =[y m (t1)Φ m K-1 y m (t2)Φ m K-2 … y m (t K )Φ m K-K ]

[0102] Y=[Y1 Y2 … Y M ] T

[0103] Among them, Y represents the sample matrix, Y1, Y2, Y m , Y M Respectively represent the sample data after phase shift of the first motion platform, the second motion platform, the mth motion platform and the Mth motion platform, t1, t2, t K Respectively represent the first, second and Kth sampling times, Φ m K-1 , Φ m K-2 , Φm K-K The motion compensation factors of the mth moving platform corresponding to the first, second and Kth phase shifts, y m (t1),y m (t2),y m (t K ) represent the sampling times t1, t2, t K The sample matrix Y is equivalent to the virtual array formed by the entire distributed moving platform at the sampling time t K The received signal can then be estimated based on the sample matrix Y. The size of the sample matrix Y is M×KL.

[0104] S4-4, calculating the spatial spectrum amplitude of the current search value based on the sample matrix of the virtual array;

[0105] The calculation formula of the spatial spectrum amplitude in step S4-4 is:

[0106]

[0107] in, represents the spatial spectrum amplitude, E n represents the noise subspace of the sample matrix, represents the steering vector composed of each platform corresponding to the sampling time K, (·) H represents the conjugate transpose.

[0108] S4-5. Repeat steps S4-2 to S4-5 until all search values ​​are traversed to obtain corresponding spatial spectra.

[0109] S5. Determine the position of the signal source based on the spatial spectrum to complete the positioning of the signal source.

[0110] In summary, the present invention is aimed at the distributed moving platform collaborative positioning scenario, and adopts a beam-forming multi-platform collaborative scanning method to ensure that all platforms can receive the target area signal; and proposes an intermittent signal transmission method to reduce the demand for data communication and computing costs, and make full use of the position information of the platform's long-term movement, while reducing the number of data transmissions and signal phase alignment times; in view of the nonlinearity of the near-field signal source position information and the time-varying inconsistency of the long-term spatial vector, a nonlinear approximate motion compensation factor is established, and a spatially equivalent virtual aperture is formed by compensating the distributed array signal phase; in view of the problem that the motion compensation factor contains unknown signal source parameters, an improved MUSIC-DPD algorithm is proposed to achieve synchronous estimation of the motion compensation factor and the signal source position.

Claims

1. A long synthetic positioning method based on a distributed coherent moving platform, characterized in that: The following steps are involved: S1. Settings A sports platform; S2. Sampling the signal source through each motion platform to obtain corresponding sampling signal data; S3. Based on each sampled signal data, a distributed multi-platform direct positioning signal model is constructed; S4, using the improved MUSIC-DPD algorithm to process the distributed multi-platform direct positioning signal model to obtain the corresponding spatial spectrum; S5. Determine the position of the signal source based on the spatial spectrum to complete the positioning of the signal source; The step S4 comprises the following steps: S4-1. Set the parameters of the distributed multi-platform direct positioning signal model and determine the spatial spectrum range and accuracy; S4-2. Select unit vector as the current search value and fix it; , They respectively represent the direction angle and the downward angle of the reference point to the signal source; S4-3, using a motion-compensated synthetic aperture method to process the output of the distributed multi-platform direct positioning signal model at all sampling moments to generate a sample matrix of a virtual array; S4-4, calculating the spatial spectrum amplitude of the current search value based on the sample matrix of the virtual array; S4-5. Repeat steps S4-2 to S4-5 until all search values ​​are traversed to obtain corresponding spatial spectra.

2. The long synthetic positioning method based on a distributed coherent moving platform according to claim 1 is characterized in that: The sampling in step S2 is performed in an intermittent manner, and the corresponding process is as follows: Set the sampling interval ; According to the sampling interval , through each motion platform to the signal source Intermittent sampling; The single sampling duration is ; The time period is Included signal samples; The sampled signal data includes signal samples.

3. The long synthetic positioning method based on a distributed coherent moving platform according to claim 2 is characterized in that: The formula of the distributed multi-platform direct positioning signal model in step S3 is: in, Indicated in The sampling time corresponding to the sampling time is Indicates the sampling time The distributed multi-platform direct positioning signal model output, Indicates the sampling time The guiding vector, Indicates the sampling time The corresponding amplitude of the sampled signal data, represents an exponential function with the natural constant e as the base, represents an imaginary number, represents pi, represents the carrier wavelength, Represents the unit vector from the reference point to the signal source, , , Represent the first motion platform, the second motion platform and the The motion platform at the sampling time location information; Indicates that all motion platforms are The sampled signal data.

4. The long synthetic positioning method based on a distributed coherent moving platform according to claim 3 is characterized in that: The step S4-3 comprises the following steps: S4-3-1. Calculate the displacement of the motion platform at all sampling moments according to the position information of the motion platform at all sampling moments; S4-3-2, constructing corresponding motion compensation factors based on the displacement of the motion platform at all sampling moments; S4-3-3. According to the motion compensation factor, the sampling signal data of each motion platform is phase-shifted to obtain the sample matrix of the corresponding virtual array.

5. The long synthetic positioning method based on a distributed coherent moving platform according to claim 4 is characterized in that: The calculation formula of the motion compensation factor in step S4-3-2 is: in, Indicates The moving speed of the motion platform, Indicates Motion compensation factor for a motion platform, Indicates the slow time domain time interval between two samplings on each platform.

6. The long synthetic positioning method based on a distributed coherent moving platform according to claim 4 is characterized in that: The calculation formula of the sample matrix in step S4-3-3 is: in, represents the sample matrix, , , , Respectively represent the first motion platform, the second motion platform, The first motion platform and the The sample data after phase shift of a moving platform, , , Represents the first, second and The corresponding sampling time is , , Represents the first, second and The corresponding phase shift Motion compensation factor for a motion platform, , , Respectively represent the sampling time , , The corresponding Signal samples of a moving platform, Represents the transposed matrix.

7. The long synthetic positioning method based on a distributed coherent moving platform according to claim 4 is characterized in that: The calculation formula of the spatial spectrum amplitude in step S4-4 is: in, represents the spatial spectrum amplitude, represents the noise subspace of the sample matrix, Indicates the sampling time Corresponding to the guidance vector composed of each motion platform, represents the conjugate transpose.

Images