A high-frequency radar high-speed target detection method and a computer readable medium

Abstract

The application provides a high-frequency radar high-speed target detection method and a computer readable medium. A radar system receiver receives a return signal, performs matched filtering on the return signal to obtain a pulse compression signal, and performs clutter preprocessing on the pulse compression signal to obtain a preprocessed pulse compression signal; the preprocessed pulse compression signal is subjected to coherent accumulation processing to obtain a coherent accumulation pulse compression signal, and a confidence motion trajectory is further calculated in combination with the coherent accumulation pulse compression signal; the preprocessed pulse compression signal is filtered along the confidence motion trajectory to filter out signals with signal-to-noise ratio flicker, thereby obtaining a filtered pulse compression signal; the filtered pulse compression signal is subjected to incoherent accumulation along the confidence motion trajectory, and target accurate motion parameters are obtained through peak value detection. The application utilizes a long-time accumulation algorithm to obtain an accurate target motion trajectory, thereby effectively improving the target detection probability.

Description

Technical Field

[0001] This invention belongs to the field of radar technology, and in particular relates to a high-frequency radar high-speed target detection method and a computer-readable medium. Background Technology

[0002] High-frequency (HF) radar utilizes the diffraction effect of electromagnetic waves to achieve over-the-horizon dynamic monitoring of the ocean surface and detection of low-flying targets or vessels. Due to the long wavelength of HF signals, HF radar has excellent anti-stealth capabilities and can provide early warning of various low-altitude targets.

[0003] Currently, compact HF radar technology is relatively mature in sea state inversion and ship target detection. To address the detection of high-speed targets, research is being conducted without altering the radar hardware and waveform parameters. Due to the long sweep period of high-frequency radar (a coherent accumulation period can last several minutes), high-speed targets exhibit range migration within a single coherent accumulation period. Similarly, due to the limited Doppler measurement range of HF radar for sea state detection, high-speed targets also exhibit Doppler frequency migration within a single coherent accumulation period. The existence of range migration and Doppler frequency migration prevents conventional moving target detection algorithms from fully accumulating the signal, leading to algorithm failure. Furthermore, the changing angle between the radar and the target during long coherent accumulation periods, combined with the scattering characteristics of HF electromagnetic waves, causes fluctuations in the signal-to-noise ratio (SNR) of the target echo received by the radar, a phenomenon known as SNR scintillation.

[0004] Long-term coherent accumulation algorithms utilize the amplitude and phase information of the target signal to correct range migration and Doppler frequency migration. Typical coherent accumulation algorithms include Keystone Transform (KT), Radon Fourier Transform (RFT), and Generalized Radon Fourier Transform (GRFT). The core of KT is to perform a scaling transformation on the slow time in the fast-slow time dimension, achieving decoupling between slow and fast time in coherent accumulation. KT is often used to correct the first-order range migration of uniformly moving targets. However, the KT transform needs to be performed before pulse compression, limiting its applicability. RFT achieves coherent accumulation through a joint search of the target and velocity. Its core idea is to extract the range migration trajectory of the target echo in the range-slow time plane according to preset search parameters, and then use Fourier transform to accumulate the extracted echo signal. The accumulation value of RFT reaches its maximum when the search parameters match the true target parameters. Like KT, RFT can only correct first-order range cell migration. GRFT is a higher-order generalization of RFT. It extracts the range migration trajectory of the target echo through a joint search of multi-dimensional motion parameters while simultaneously compensating for Doppler frequency migration, thereby achieving coherent accumulation. GRFT can correct for higher-order range migration and compensate for complex Doppler frequency migration by searching for higher-order motion parameters. However, in practical applications, the detection probability of GRFT needs further improvement under conditions of low signal-to-noise ratio and signal-to-noise ratio flicker.

[0005] In summary, high-speed target detection using HF radar still faces many challenges that need to be addressed. These mainly include: how to construct an accurate target motion model, how to cope with challenges such as target signal-to-noise ratio flicker, complex background noise, and low target signal-to-noise ratio. Existing high-speed target detection algorithms have encountered significant challenges when applied to HF radar, making it difficult to obtain accurate parameter estimation results. Summary of the Invention

[0006] This invention proposes a high-speed target detection method and computer-readable medium for high-frequency radar, which solves the problems of target motion model mismatch, signal-to-noise ratio flicker, complex background noise and low target signal-to-noise ratio in high-speed target detection by HF radar.

[0007] To solve the above-mentioned technical problems, the technical solution of the present invention is a high-frequency radar high-speed target detection method, the specific implementation steps of which are as follows:

[0008] Step 1: The radar system receiver receives the echo signal, performs matched filtering on the echo signal to obtain a pulse compression signal, and performs clutter preprocessing on the pulse compression signal to obtain a preprocessed pulse compression signal.

[0009] Step 2: Perform coherent accumulation processing on the preprocessed pulse compression signal to obtain a coherent accumulated pulse compression signal, and further calculate the confidence motion trajectory by combining the coherent accumulated pulse compression signal;

[0010] Step 3: Filter the preprocessed pulse compression signal along the confidence motion trajectory to remove the signal-to-noise ratio flickering signal and obtain the filtered pulse compression signal;

[0011] Step 4: Perform incoherent accumulation of the filtered pulse compression signal along the confident motion trajectory, and obtain the accurate motion parameters of the target through peak detection;

[0012] Preferably, the pulse compression signal in step 1 is:

[0013]

[0014] t m =mT r

[0015] m∈[1,M]

[0016] Among them, t m Let M represent the slow time of the m-th sweep cycle, and M represent the number of sweep cycles. This represents the fast time in the m-th sweep cycle. In the m-th sweep frequency cycle The pulse compression signal at time t, where B is the bandwidth of the carrier frequency signal, and R(t) m ) for in t m The instantaneous radial distance between the radar and the target at any given moment, where λ represents the wavelength of the signal, sinc is the sigma function, and T... r This indicates the duration of each sweep cycle;

[0017] Step 1 involves performing clutter preprocessing on the pulse compression signal, specifically as follows:

[0018] A high-pass filter bank is used to filter the pulse compression signal in the slow time dimension. The specific calculation is as follows:

[0019]

[0020] m∈[1,M]

[0021] in, In the m-th sweep frequency cycle The preprocessed pulse compression signal at time h HP (t m ) represents the high-pass filtering process during the slow time of the predetermined m-th sweep cycle, h HP (*) indicates a high-pass filter, and M indicates the number of sweep cycles;

[0022] Step 2 involves compressing the preprocessed pulse signal, as detailed below:

[0023] Step 2.1: Construct the motion model of the target;

[0024] For high-speed targets detected by HF radar, the motion model of the target during the coherent accumulation time is as follows:

[0025]

[0026] m∈[1,M]

[0027] Where R0 represents the initial radial distance between the radar and the target, V represents the target's true velocity, and t m Let represent the slow time of the m-th sweep cycle, M represent the number of sweep cycles, Θ represent the initial radial angle, and cos represent the cosine calculation.

[0028] Step 2.2: Determine the search range and interval for the motion parameters;

[0029] Based on the HF radar's range and waveform parameters, the initial radial range search range is determined as follows:

[0030] [r 0,min ,r 0,max ]

[0031] Where, r 0,min r represents the minimum initial radial distance to the target being searched. 0,max r represents the maximum initial radial distance of the target being searched. 0,min =ΔR,r 0,max =2T p / c, ΔR=c / 2B is the radar range resolution, T p Where is the pulse width, c is the speed of light, and B is the bandwidth of the carrier frequency signal;

[0032] Based on the type of target, the search range for the true speed is determined as follows:

[0033] [v min ,v max ]

[0034] Among them, v min v represents the minimum true speed of the search target. max v represents the maximum actual speed of the search target. min =0, v max =V max V max This represents the maximum actual speed of the target.

[0035] Based on the positional relationship between the HF radar and the target, the initial radial angle search range is determined as follows:

[0036] [θ min ,θ max ]

[0037] Where, θ min θ represents the minimum initial radial angle of the target being searched. max θ represents the maximum initial radial angle of the target being searched. min =0, θ max =180°;

[0038] The interval of the search parameters is determined based on the radar waveform parameters;

[0039] The intervals of the search parameters include: the search interval for the initial radial distance, the search interval for the true velocity, and the search interval for the initial radial angle.

[0040] The search interval for the initial radial distance is:

[0041] ΔR=c / 2B

[0042] Where c is the speed of light, and B represents the bandwidth of the carrier frequency signal.

[0043] The search interval for the actual speed is:

[0044] Δv=λ / 2MT r

[0045] Where λ represents the wavelength of the signal, M represents the number of sweep cycles, and T r Indicates the duration of each frequency sweep cycle.

[0046] The search interval for the initial radial angle is:

[0047] Δθ=Δvθ max / v max

[0048] Where Δv is the search interval for the true velocity, and θ max v is the maximum initial radial angle of the target being searched. max To achieve the maximum actual speed for the search target;

[0049] Based on the range and interval of the initial radial distance search, determine multiple sets of discretized initial radial distances.

[0050] r 0,i =r 0,min +iΔR

[0051] i∈[0,N r -1]

[0052] Where i represents the initial radial distance index of the search, and N r The initial radial range search count is represented by ΔR, where ΔR represents the radar's range resolution, and r... 0,minr represents the minimum initial radial distance to the target being searched. 0,i This represents the initial radial distance during the i-th search;

[0053] Based on the actual speed range and interval of the search, determine multiple sets of discretized actual speeds:

[0054] v p =v min +pΔv

[0055] p∈[0,N v -1]

[0056] Where p represents the index of the actual search speed, and N v The number of searches for the true speed is represented by Δv, where Δv represents the search interval for the true speed. min v represents the minimum true speed of the search target. p This represents the actual speed of the p-th search;

[0057] Based on the range and interval of the initial radial angles searched, multiple sets of discretized initial radial angles are determined.

[0058] θ q =θ min +qΔθ

[0059] q∈[0,N θ -1]

[0060] Where q represents the index of the initial radial angle of the search, and N θ Δθ represents the number of initial radial angle searches, and Δθ represents the search interval for the initial radial angle. min θ represents the minimum initial radial angle of the target being searched. q This represents the initial radial angle of the q-th search;

[0061] Step 2.3: Perform coherent accumulation processing to obtain the coherent accumulation pulse compression signal for each set of discretized search parameters;

[0062] The coherent accumulation pulse compression signal for each set of discretized search parameters is:

[0063]

[0064]

[0065] m∈[1,M]

[0066] Among them, G(r) 0,i ,v p ,θ q) represents the coherent accumulated pulse compression signal at the initial radial distance of the i-th search, the true velocity of the p-th search, and the initial radial angle of the q-th search, r(t) m ) indicates that at t m The instantaneous radial distance r between the radar and the target during constant search. 0,i This represents the initial radial distance in the i-th search, where i ∈ [0, N]. r -1],N r v represents the number of initial radial distance searches. p This represents the actual speed of the p-th search, where p∈[0,N]. v -1],N v θ represents the number of searches for the true speed. q Let q represent the initial radial angle of the q-th search, where q ∈ [0, N]. θ -1],N θ The initial radial angle search count is represented by λ, the signal wavelength is represented by sinc, and t is the sigma function. m R(t) represents the slow time of the m-th sweep cycle, M represents the number of sweep cycles, and R(t) represents the slow time of the m-th sweep cycle. m ) for in t m The instantaneous radial distance between the radar and the target at any given time, where j represents an imaginary number, ΔR represents the radar's range resolution, cos represents the cosine calculation, and T... r Indicates the duration of each frequency sweep cycle.

[0067] The calculation process for the confidence trajectory in step 2 is as follows:

[0068] At the initial radial distance of each search, the coherently accumulated pulse compression signal at the initial radial angle of the p-th search and the actual velocity of the q-th search is traversed, i.e., in N... v *N θ The maximum value is searched for in the coherent accumulated pulse compression signal, and the corresponding search parameters are:

[0069]

[0070] Indicates that in N r Given the initial radial distance of the i-th search, the true velocity of the p-th search when the absolute value of the coherent accumulation amplitude reaches its maximum value is given by the initial radial distance of the i-th search. and the initial radial angle of the q-th search where i∈[0,N] r -1],p∈[0,N v -1],q∈[0,N θ -1],N r N represents the number of initial radial distance searches. v N represents the number of searches performed at the actual speed.θ Indicates the number of initial radial angle searches;

[0071] The expression for the confidence trajectory is:

[0072]

[0073] m∈[1,M]

[0074] Where, r i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target, r 0,i This represents the initial radial distance for the i-th search. This represents the true velocity of the p-th search when the absolute value of the coherent accumulation amplitude reaches its maximum at the initial radial distance of the i-th search. Let represent the initial radial angle of the q-th search when the absolute value of the coherent accumulation amplitude reaches its maximum at the initial radial distance of the i-th search, where i∈[0,N]. r -1],p∈[0,N v -1],q∈[0,N θ -1],N r N represents the number of initial radial distance searches. v N represents the number of searches performed at the actual speed. θ t represents the number of initial radial angle searches. m Let represent the slow time of the m-th sweep cycle, cos represents the cosine calculation, and M represents the number of sweep cycles.

[0075] Step 3 involves filtering the preprocessed pulse compression signal along the confident motion trajectory to obtain a new pulse compression signal.

[0076] like:

[0077] and

[0078] m∈[1,M]

[0079] in, In the m-th sweep frequency cycle The preprocessed pulse compression signal at each moment. In the m-th sweep frequency cycle The preprocessed pulse compression signal at each moment. In the m-th sweep frequency cycle The preprocessed pulse compression signal at time r i (t m ) represents the trajectory t under the i-th confidence trajectory. mThe instantaneous radial distance between the radar and the target at any given time, t m Let ΔR represent the slow time of the m-th sweep cycle, ΔR represent the radar's range resolution, c represent the speed of light, and M represent the number of sweep cycles.

[0080] but:

[0081]

[0082] m∈[1,M]

[0083] in, In the m-th sweep frequency cycle A new pulse compression signal at every moment. In the m-th sweep frequency cycle The preprocessed pulse compression signal at time r i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target at any given time, t m Let ΔR represent the slow time of the m-th sweep cycle, ΔR represent the radar's range resolution, c represent the speed of light, and M represent the number of sweep cycles.

[0084] otherwise:

[0085]

[0086] m∈[1,M]

[0087] in, In the m-th sweep frequency cycle A new pulse compression signal at any given time, r i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target at any given time, t m Let ΔR represent the slow time of the m-th sweep cycle, ΔR represent the radar's range resolution, c represent the speed of light, and M represent the number of sweep cycles.

[0088] Step 4, which involves incoherently accumulating the new pulse compression signal along the confidence motion trajectory, is as follows:

[0089]

[0090] m∈[1,M]

[0091] in, This indicates that the search parameters are The amplitude value of the incoherent accumulated pulse compression signal, where Indicates that in N rGiven the initial radial distance of the i-th search, the true velocity of the p-th search when the absolute value of the coherent accumulation amplitude reaches its maximum value is given by the initial radial distance of the i-th search. and the initial radial angle of the q-th search where i∈[0,N] r -1],p∈[0,N v -1],q∈[0,N θ -1],N r N represents the number of initial radial distance searches. v N represents the number of searches performed at the actual speed. θ This indicates the number of initial radial angle searches. In the m-th sweep frequency cycle The new pulse compression signal at time T r r represents the duration of each sweep cycle. i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target at any given time, t m Let ΔR represent the slow time of the m-th sweep cycle, ΔR represent the radar's range resolution, and c represent the speed of light.

[0092] Furthermore, in step 4, the accurate motion parameters of the target are obtained through peak detection, and their expression is:

[0093]

[0094] Where argmax represents the value of the function's argument when the function reaches its maximum value. This indicates that the search parameters are The maximum value of the amplitude of the incoherent accumulated pulse compression signal, where Indicates that in N r Given the initial radial distance of the i-th search, the true velocity of the p-th search when the absolute value of the coherent accumulation amplitude reaches its maximum value is given by the initial radial distance of the i-th search. and the initial radial angle of the q-th search where i∈[0,N] r -1],p∈[0,N v -1],q∈[0,N θ -1],N r N represents the number of initial radial distance searches. v N represents the number of searches performed at the actual speed. θ This indicates the number of initial radial angle searches. This represents the estimated initial radial distance. This represents the estimated true speed. This represents the estimated initial radial angle.

[0095] The present invention also provides a computer-readable medium storing a computer program executed by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the high-frequency radar high-speed target detection method.

[0096] The beneficial effects of this invention are:

[0097] The motion model used in this invention is more in line with the actual situation and has higher cumulative gain without increasing computational complexity;

[0098] The pulse compression signal is first coherently accumulated to obtain the target's confident motion trajectory. Then, based on the characteristics of the pulse compression signal, signals with low signal-to-noise ratio on the confident motion trajectory are filtered out. Finally, incoherent accumulation is used to obtain the accurate target motion trajectory, which effectively improves the target detection probability. Attached Figure Description

[0099] Figure 1 : Flowchart of the method according to an embodiment of the present invention;

[0100] Figure 2 The spectrum of the pulse compression signal provided in Embodiment 1 of the present invention;

[0101] Figure 3 The spectrum of the preprocessed pulse compression signal provided in Embodiment 1 of the present invention;

[0102] Figure 4 : Profiles of the initial radial velocity and initial radial distance processed by the GRFT3 algorithm in this embodiment of the invention;

[0103] Figure 5 : Initial radial acceleration and initial radial distance profiles processed by the GRFT3 algorithm in this embodiment of the invention;

[0104] Figure 6 : A cross-sectional view of the actual velocity and initial radial distance in an embodiment of the present invention;

[0105] Figure 7 : A cross-sectional view of the initial radial angle and initial radial distance in an embodiment of the present invention;

[0106] Figure 8 A comparison chart of distance estimation error results from embodiments of the present invention;

[0107] Figure 9 Comparison chart of detection probability results in embodiments of the present invention. Detailed Implementation

[0108] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0109] In specific implementation, the method proposed in the technical solution of this invention can be automatically executed by those skilled in the art using computer software technology. System devices for implementing the method, such as computer-readable storage media storing the corresponding computer program of the technical solution of this invention and computer equipment including the computer program running the corresponding computer program, should also be within the protection scope of this invention.

[0110] The following is combined Figure 1-9 The technical solution of this invention is a high-speed radar target detection method, as described in the following details:

[0111] Step 1: The radar system receiver receives the echo signal, performs matched filtering on the echo signal to obtain a pulse compression signal, and performs clutter preprocessing on the pulse compression signal to obtain a preprocessed pulse compression signal.

[0112] The pulse compression signal mentioned in step 1 is:

[0113]

[0114] t m =mT r

[0115] m∈[1,M]

[0116] Among them, t m Let M represent the slow time of the m-th sweep cycle, and M = 512 represent the number of sweep cycles. This represents the fast time in the m-th sweep cycle. In the m-th sweep frequency cycle The pulse compression signal at time t, B = 300kHz is the bandwidth of the carrier frequency signal, R(t) m ) for in t m The instantaneous radial distance between the radar and the target at any given moment, λ = 23.08m represents the wavelength of the signal, sinc is the sigma function, and T... r =0.256s indicates the duration of each sweep cycle;

[0117] Step 1 involves performing clutter preprocessing on the pulse compression signal, specifically as follows:

[0118] A high-pass filter bank is used to filter the pulse compression signal in the slow time dimension. The specific calculation is as follows:

[0119]

[0120] m∈[1,M]

[0121] in, In the m-th sweep frequency cycle The preprocessed pulse compression signal at time h HP (t m ) represents the high-pass filtering process during the slow time of the predetermined m-th sweep cycle, h HP (*) indicates a high-pass filter, and M = 512 indicates the number of sweep cycles;

[0122] Step 2: Perform coherent accumulation processing on the preprocessed pulse compression signal to obtain a coherent accumulated pulse compression signal, and further calculate the confidence motion trajectory by combining the coherent accumulated pulse compression signal;

[0123] Step 2 involves compressing the preprocessed pulse signal, as detailed below:

[0124] Step 2.1: Construct the motion model of the target;

[0125] For high-speed targets detected by HF radar, the motion model of the target during the coherent accumulation time is as follows:

[0126]

[0127] m∈[1,M]

[0128] Where R0 = 15km represents the initial radial distance between the radar and the target, V = 200m / s represents the target's actual velocity, and t m The slow time of the m-th sweep cycle is represented by M = 512, the number of sweep cycles is represented by Θ = 2.5 rad, the initial radial angle is represented by cosine, and cos represents the cosine calculation.

[0129] Step 2.2: Determine the search range and interval for the motion parameters;

[0130] Based on the HF radar's range and waveform parameters, the initial radial range search range is determined as follows:

[0131] [r 0,min ,r 0,max ]

[0132] Where, r 0,min r represents the minimum initial radial distance to the target being searched. 0,max r represents the maximum initial radial distance of the target being searched. 0,min =ΔR = 0.5km, r 0,max =2T p / c = 30km, ΔR = c / 2B = 0.5km is the radar's range resolution, T p =0.2ms is the pulse width, c = 3 × 10 8 m / s is the speed of light, and B = 300kHz is the bandwidth of the carrier frequency signal;

[0133] Based on the type of target, the search range for the true speed is determined as follows:

[0134] [v min ,v max ]

[0135] Among them, v min v represents the minimum true speed of the search target. max v represents the maximum actual speed of the search target. min =0, v max =V max =300m / s, V max This represents the maximum actual speed of the target.

[0136] Based on the positional relationship between the HF radar and the target, the initial radial angle search range is determined as follows:

[0137] [θ min ,θ max ]

[0138] Where, θ min θ represents the minimum initial radial angle of the target being searched. max θ represents the maximum initial radial angle of the target being searched. min =0, θ max =180°;

[0139] The interval of the search parameters is determined based on the radar waveform parameters;

[0140] The intervals of the search parameters include: the search interval for the initial radial distance, the search interval for the true velocity, and the search interval for the initial radial angle.

[0141] The search interval for the initial radial distance is:

[0142] Δr=c / 2B

[0143] Where, c = 3 × 10 8 m / s is the speed of light, and B = 300kHz represents the bandwidth of the carrier frequency signal.

[0144] The search interval for the actual speed is:

[0145] Δv=λ2MT r

[0146] Where λ = 23.08m represents the wavelength of the signal, M = 512 represents the number of sweep cycles, and T r =0.256s indicates the duration of each sweep cycle.

[0147] The search interval for the initial radial angle is:

[0148] Δθ=Δvθ max / v max

[0149] Where Δv = 0.09 m / s is the search interval for the true velocity, and θ max =180° is the maximum initial radial angle of the target being searched, v max =300m / s is the maximum actual speed of the target being searched;

[0150] Based on the range and interval of the initial radial distance search, determine multiple sets of discretized initial radial distances.

[0151] r 0,i =r 0,min +iΔR

[0152] i∈[0,N r -1]

[0153] Where i represents the initial radial distance index of the search, and N r =60 represents the initial number of radial range searches, ΔR=0.5km represents the radar's range resolution, r 0,min =0.5km represents the minimum initial radial distance of the target being searched, r 0,i This represents the initial radial distance during the i-th search;

[0154] Based on the actual speed range and interval of the search, determine multiple sets of discretized actual speeds:

[0155] v p =v min +pΔv

[0156] p∈[0,N v -1]

[0157] Where p represents the index of the actual search speed, and N v =3333 represents the number of searches for the true speed, Δv = 0.09 m / s represents the search interval for the true speed, v min =0 indicates the minimum true velocity of the target being searched, v p This represents the actual speed of the p-th search;

[0158] Based on the range and interval of the initial radial angles searched, multiple sets of discretized initial radial angles are determined.

[0159] θ q =θ min +qΔθ

[0160] q∈[0,N θ -1]

[0161] Where q represents the index of the initial radial angle of the search, and N θ =3600 represents the number of initial radial angle searches, Δθ = 0.05° represents the search interval for the initial radial angle, θ min =0 indicates the minimum initial radial angle of the target being searched, θ q This represents the initial radial angle of the q-th search;

[0162] Step 2.3: Perform coherent accumulation processing to obtain the coherent accumulation pulse compression signal for each set of discretized search parameters;

[0163] The coherent accumulation pulse compression signal for each set of discretized search parameters is:

[0164]

[0165]

[0166] m∈[1,M]

[0167] Among them, G(r) 0,i ,v p ,θ q ) represents the coherent accumulated pulse compression signal at the initial radial distance of the i-th search, the true velocity of the p-th search, and the initial radial angle of the q-th search, r(t) m ) indicates that at t m The instantaneous radial distance r between the radar and the target during constant search. 0,i This represents the initial radial distance in the i-th search, where i ∈ [0, N]. r -1],N r =60 indicates the initial number of radial distance searches, v p This represents the actual speed of the p-th search, where p∈[0,N]. v -1],N v =3333 represents the number of searches for the actual speed, θ q Let q represent the initial radial angle of the q-th search, where q ∈ [0, N]. θ -1],N θ =3600 represents the initial number of radial angle searches, λ=23.08m represents the wavelength of the signal, sinc is the singer function, and t m R(t) represents the slow time of the m-th sweep cycle, M = 512 represents the number of sweep cycles, and R(t) represents the slow time of the m-th sweep cycle. m ) for in tm The instantaneous radial distance between the radar and the target at any given time, where j represents an imaginary number, ΔR = 0.5 km represents the radar's range resolution, cos represents the cosine calculation, and T... r =0.256s indicates the duration of each sweep cycle.

[0168] The calculation process for the confidence trajectory in step 2 is as follows:

[0169] At the initial radial distance of each search, the coherently accumulated pulse compression signal at the initial radial angle of the p-th search and the actual velocity of the q-th search is traversed, i.e., in N... v *N θ The maximum value is searched for in the coherent accumulated pulse compression signal, and the corresponding search parameters are:

[0170]

[0171] Indicates that in N r Given the initial radial distance of the i-th search, the true velocity of the p-th search when the absolute value of the coherent accumulation amplitude reaches its maximum value is given by the initial radial distance of the i-th search. and the initial radial angle of the q-th search where i∈[0,N] r -1],p∈[0,N v -1],q∈[0,N θ -1],N r =60 indicates the initial number of radial distance searches, N v =3333 represents the number of searches performed at the actual speed, N θ =3600 indicates the initial number of radial angle searches;

[0172] The expression for the confidence trajectory is:

[0173]

[0174] m∈[1,M]

[0175] Where, r i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target, r 0,i This represents the initial radial distance for the i-th search. This represents the true velocity of the p-th search when the absolute value of the coherent accumulation amplitude reaches its maximum at the initial radial distance of the i-th search. Let represent the initial radial angle of the q-th search when the absolute value of the coherent accumulation amplitude reaches its maximum at the initial radial distance of the i-th search, where i∈[0,N].r -1],p∈[0,N v -1],q∈[0,N θ -1],N r =60 indicates the initial number of radial distance searches, N v =3333 represents the number of searches performed at the actual speed, N θ =3600 represents the initial number of radial angle searches, t m Let m represent the slow time of the m-th sweep cycle, cos represents the cosine calculation, and M = 512 represents the number of sweep cycles.

[0176] Step 3: Filter the preprocessed pulse compression signal along the confidence motion trajectory to remove the signal-to-noise ratio flickering signal and obtain the filtered pulse compression signal;

[0177] Step 3 involves filtering the preprocessed pulse compression signal along the confident motion trajectory to obtain a new pulse compression signal.

[0178] like:

[0179] and

[0180] m∈[1,M]

[0181] in, In the m-th sweep frequency cycle The preprocessed pulse compression signal at each moment. In the m-th sweep frequency cycle The preprocessed pulse compression signal at each moment. In the m-th sweep frequency cycle The preprocessed pulse compression signal at time r i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target at any given time, t m Let ΔR = 0.5 km represent the slow time of the m-th sweep cycle, and let c = 3 × 10⁻⁶. 8 m / s is the speed of light, and M = 512 represents the number of sweep cycles.

[0182] but:

[0183]

[0184] m∈[1,M]

[0185] in, In the m-th sweep frequency cycle A new pulse compression signal at every moment. In the m-th sweep frequency cycle The preprocessed pulse compression signal at time r i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target at any given time, t m Let ΔR = 0.5 km represent the slow time of the m-th sweep cycle, and let c = 3 × 10⁻⁶. 8 m / s is the speed of light, and M = 512 represents the number of sweep cycles.

[0186] otherwise:

[0187]

[0188] m∈[1,M]

[0189] in, In the m-th sweep frequency cycle A new pulse compression signal at any given time, r i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target at any given time, t m Let ΔR = 0.5 km represent the slow time of the m-th sweep cycle, and let c = 3 × 10⁻⁶. 8 m / s is the speed of light, and M = 512 represents the number of sweep cycles;

[0190] Step 4: Perform incoherent accumulation of the filtered pulse compression signal along the confident motion trajectory, and obtain the accurate motion parameters of the target through peak detection;

[0191] Step 4, which involves incoherently accumulating the new pulse compression signal along the confidence motion trajectory, is as follows:

[0192]

[0193] m∈[1,M]

[0194] in, This indicates that the search parameters are The amplitude value of the incoherent accumulated pulse compression signal, where Indicates that in N r Given the initial radial distance of the i-th search, the true velocity of the p-th search when the absolute value of the coherent accumulation amplitude reaches its maximum value is given by the initial radial distance of the i-th search. and the initial radial angle of the q-th search where i∈[0,N] r -1],p∈[0,N v-1],q∈[0,N θ -1],N r =60 indicates the initial number of radial distance searches, N v =3333 represents the number of searches performed at the actual speed, N θ =3600 indicates the initial number of radial angle searches. In the m-th sweep frequency cycle The new pulse compression signal at time T r =0.256s represents the duration of each sweep cycle, r i (t m ) represents the trajectory t under the i-th confidence trajectory. m The instantaneous radial distance between the radar and the target at any given time, t m Let ΔR = 0.5 km represent the slow time of the m-th sweep cycle, and let c = 3 × 10⁻⁶. 8 m / s is the speed of light.

[0195] Furthermore, in step 4), the accurate motion parameters of the target are obtained through peak detection, and their expression is:

[0196]

[0197] Where argmax represents the value of the function's argument when the function reaches its maximum value. This indicates that the search parameters are The maximum value of the amplitude of the incoherent accumulated pulse compression signal, where Indicates that in N r Given the initial radial distance of the i-th search, the true velocity of the p-th search when the absolute value of the coherent accumulation amplitude reaches its maximum value is given by the initial radial distance of the i-th search. and the initial radial angle of the q-th search where i∈[0,N] r -1],p∈[0,N v -1],q∈[0,N θ -1],N r =60 indicates the initial number of radial distance searches, N v =3333 represents the number of searches performed at the actual speed, N θ =3600 indicates the initial number of radial angle searches. This represents the estimated initial radial distance. This represents the estimated true speed. This represents the estimated initial radial angle.

[0198] Figure 2The signal is the radar echo signal after pulse compression, where the background noise is the measured signal, the target echo is the simulated signal, the target's signal-to-noise ratio (SNR) is 0 dB, and the SNR scintillation rate is 39%. The SNR scintillation rate is defined as the number of SNR scintillation frames divided by the number of sweep cycles M. Figure 3 To Figure 2 The preprocessed pulse compressed signal is obtained by performing clutter preprocessing on the pulse compressed signal in the image. Experimental results are as follows: Figure 4-9 As shown, Figure 4-8 These are the experimental results provided in Embodiment 1 of the present invention. Figure 4 and Figure 5 The result of the GRFT3 algorithm, Figure 4 Given the initial radial velocity and initial radial distance, with an initial radial acceleration of 1.84 m / s², 2 Cross-sectional view, Figure 5 This is a cross-sectional view showing the initial radial acceleration and initial radial distance at an initial radial velocity of -202 m / s. Figure 6 and Figure 7 The result of the MGRFT algorithm of this invention, Figure 6 This is a cross-sectional view showing the actual velocity and initial radial distance at an initial radial angle of 2.5 rad. Figure 7 This is a cross-sectional view showing the initial radial angle and initial radial distance at a real velocity of 200 m / s. Figure 8 The error between the motion trajectory and the actual target motion trajectory is obtained using the GRFT3 and MGRFT algorithms. This is achieved through... Figures 4-7 It can be observed that the MGRFT algorithm has a greater accumulation gain than the GRFT3 algorithm, and the MGRFT algorithm has lower sidelobes. Through Figure 8 It can be observed that the MGRFT algorithm has more accurate parameter estimation performance compared to the GRFT3 algorithm. Figure 9 The experimental results provided in Embodiment 2 of this invention are shown in the figure. The figure illustrates the change curves of the detection probability of the GRFTa algorithm and the MFRFT algorithm with flicker rate under the condition of fixed target signal-to-noise ratios of -15dB, -20dB, and -25dB. Figure 9 It can be observed that under the conditions of target signal-to-noise ratio of -15dB, -20dB, and -25dB, the detection performance of the MGRFT algorithm is better than that of the GRFTa algorithm. Moreover, as the target signal-to-noise ratio decreases, the difference between the detection probability of the MGRFT algorithm and that of the GRFTa algorithm becomes larger and larger, proving that the MGRFT algorithm has a greater improvement in detection performance under low signal-to-noise ratio conditions.

[0199] A specific embodiment of the present invention also provides a computer-readable medium.

[0200] The computer-readable medium is a server workstation;

[0201] The server workstation stores the computer program executed by the electronic device. When the computer program runs on the electronic device, it causes the electronic device to perform the steps of the high-frequency radar high-speed target detection method according to the embodiments of the present invention.

[0202] It should be understood that any parts not described in detail in this specification belong to the prior art.

[0203] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A high-frequency radar high-speed target detection method, characterized in that, Includes the following steps: Step 1: The radar system receiver receives the echo signal, performs matched filtering on the echo signal to obtain a pulse compression signal, and performs clutter preprocessing on the pulse compression signal to obtain a preprocessed pulse compression signal. Step 2: Perform coherent accumulation processing on the preprocessed pulse compression signal to obtain a coherently accumulated pulse compression signal, including: Based on multiple initial radial distance conditions, the true velocity and initial radial angle are searched respectively to obtain multiple sets of discretized search parameters, and the corresponding coherent accumulation amplitude value is obtained by calculating each set of search parameters. Under each initial radial distance condition, the search parameter that maximizes the coherent accumulation amplitude value is selected as the trajectory parameter for the confidence motion trajectory; Step 3: Filter the preprocessed pulse compression signal along the confidence motion trajectory to remove the signal-to-noise ratio flickering signal and obtain the filtered pulse compression signal; Step 4: Perform incoherent accumulation of the filtered pulse compression signal along the confidence motion trajectory, and obtain the accurate motion parameters of the target through peak detection.

2. The high-frequency radar high-speed target detection method according to claim 1, characterized in that: The pulse compression signal mentioned in step 1 is: m ∈[1, M ] in, Indicates the first m The slow time of one sweep cycle, Indicates the number of sweep cycles. Indicates the first m The fast time of a frequency sweep cycle Indicates the first m Within each sweep frequency cycle The pulse compression signal at a given time. The bandwidth of the carrier frequency signal. In order to be in The instantaneous radial distance between the radar and the target at any given time. This represents the wavelength of the signal, and sinc is the sigma function. This indicates the duration of each sweep cycle; Step 1 involves performing clutter preprocessing on the pulse compression signal, specifically as follows: A high-pass filter bank is used to filter the pulse compression signal in the slow time dimension. The specific calculation is as follows: m ∈[1, M ] in, Indicates the first m Within each sweep frequency cycle The preprocessed pulse compression signal at each moment. For the predetermined first m High-pass filtering during the slow time of each sweep cycle. Indicates a high-pass filter. This indicates the number of sweep cycles.

3. The high-frequency radar high-speed target detection method according to claim 2, characterized in that: Step 2 involves compressing the preprocessed pulse signal, as detailed below: Step 2.1: Construct the motion model of the target; For high-speed targets detected by HF radar, the motion model of the target during the coherent accumulation time is as follows: m ∈[1, M ] in, This indicates the initial radial distance between the radar and the target. Indicates the target's true speed. This represents the slow time of the m-th sweep cycle. Indicates the number of sweep cycles. This represents the initial radial angle, and cos represents the cosine calculation. Step 2.2: Determine the search range and interval for the motion parameters; Based on the HF radar's range and waveform parameters, the initial radial range search range is determined as follows: in, This represents the minimum initial radial distance to the target being searched. This represents the maximum initial radial distance of the target being searched. , , For the range resolution of the radar, T p The pulse width. c At the speed of light, The bandwidth of the carrier frequency signal; Based on the type of target, the search range for the true speed is determined as follows: in, This represents the minimum actual speed required to search for the target. This indicates the maximum actual speed of the search target. , , This represents the maximum actual speed of the target. Based on the positional relationship between the HF radar and the target, the initial radial angle search range is determined as follows: in, This represents the minimum initial radial angle of the target being searched. Indicates the maximum initial radial angle of the target being searched. , ; The interval of the search parameters is determined based on the radar waveform parameters; The intervals of the search parameters include: the search interval for the initial radial distance, the search interval for the true velocity, and the search interval for the initial radial angle. The search interval for the initial radial distance is: in, c At the speed of light, B Indicates the bandwidth of the carrier frequency signal. The search interval for the actual speed is: in, λ Indicates the wavelength of the signal. Indicates the number of sweep cycles. Indicates the duration of each frequency sweep cycle. The search interval for the initial radial angle is: in, The search interval is the actual speed. The maximum initial radial angle of the target being searched. To achieve the maximum actual speed for the search target; Based on the range and interval of the initial radial distance search, determine multiple sets of discretized initial radial distances. in, Indicates the initial radial distance index for the search. This indicates the number of initial radial distance searches. This indicates the radar's range resolution. This represents the minimum initial radial distance to the target being searched. Indicates the first The initial radial distance for each search; Based on the actual speed range and interval of the search, determine multiple sets of discretized actual speeds: in, The index indicating the actual search speed. Indicates the number of searches performed at the actual speed. The search interval represents the actual speed. This represents the minimum actual speed required to search for the target. Indicates the first The actual speed of a search; Based on the range and interval of the initial radial angles searched, multiple sets of discretized initial radial angles are determined. in, The index representing the initial radial angle of the search. This indicates the number of initial radial angle searches. This represents the search interval for the initial radial angle. This represents the minimum initial radial angle of the target being searched. Indicates the first The initial radial angle for each search; Step 2.3: Perform coherent accumulation processing to obtain the coherent accumulation pulse compression signal for each set of discretized search parameters; The coherent accumulation pulse compression signal for each set of discretized search parameters is: m ∈[1, M ] in, Indicates the first The initial radial distance of the search is the first The actual search speed of the first search The coherent accumulation pulse compression signal at the initial radial angle of the search, Indicates in The instantaneous radial distance between the radar and the target being constantly searched. Indicates the first The initial radial distance for each search, , This indicates the number of initial radial distance searches. Indicates the first The actual speed of a search. , Indicates the number of searches performed at the actual speed. Indicates the first The initial radial angle of the search, , This indicates the number of initial radial angle searches. This represents the wavelength of the signal, and sinc is the sigma function. Indicates the first m The slow time of one sweep cycle, Indicates the number of sweep cycles. In order to be in The instantaneous radial distance between the radar and the target at any given time. j represents an imaginary number, This represents the radar's range resolution, and cos represents the cosine calculation. This indicates the duration of each sweep cycle.

4. The high-frequency radar high-speed target detection method according to claim 3, characterized in that: The calculation process for the confidence trajectory in step 2 is as follows: Given the initial radial distance of each search, traverse the... The actual search speed of the first search The coherently accumulated pulse compression signal at the initial radial angle of the search, i.e., in The maximum value is searched for in the coherent accumulated pulse compression signal, and the corresponding search parameters are: Indicates in In the initial radial distance of the search, the first Given the initial radial distance of the search, the absolute value of the coherent accumulation amplitude reaches its maximum value at the corresponding moment. The actual speed of a search and the The initial radial angle of the search ,in , , , This indicates the number of initial radial distance searches. Indicates the number of searches performed at the actual speed. Indicates the number of initial radial angle searches; The expression for the confidence trajectory is: m ∈[1, M ] in, Indicates the first Under the confidence trajectory The instantaneous radial distance between the radar and the target. Indicates the first The initial radial distance for each search, Indicates the first When the absolute value of the coherent accumulation amplitude reaches its maximum at the initial radial distance of the first search, the th search... The actual speed of a search. Indicates the first When the absolute value of the coherent accumulation amplitude reaches its maximum at the initial radial distance of the first search, the th search... The initial radial angle of the search, where , , , This indicates the number of initial radial distance searches. Indicates the number of searches performed at the actual speed. This indicates the number of initial radial angle searches. Indicates the first m The slow time of each sweep cycle, cos represents the cosine calculation. This indicates the number of sweep cycles.

5. The high-frequency radar high-speed target detection method according to claim 4, characterized in that: The new pulse compression signal obtained by filtering the preprocessed pulse compression signal along the confidence motion trajectory in step 3 is as follows: like: and m ∈[1, M ] in, Indicates the first m Within each sweep frequency cycle The preprocessed pulse compression signal at each moment. Indicates the first m Within each sweep frequency cycle The preprocessed pulse compression signal at each moment. Indicates the first m Within each sweep frequency cycle The preprocessed pulse compression signal at each moment. Indicates the first Under the confidence trajectory The instantaneous radial distance between the radar and the target. Indicates the first m The slow time of one sweep cycle, This indicates the radar's range resolution. c At the speed of light, Indicates the number of sweep cycles; but: m ∈[1, M ] in, Indicates the first m Within each sweep frequency cycle A new pulse compression signal at every moment. Indicates the first m Within each sweep frequency cycle The preprocessed pulse compression signal at each moment. Indicates the first Under the confidence trajectory The instantaneous radial distance between the radar and the target. Indicates the first m The slow time of one sweep cycle, This indicates the radar's range resolution. c At the speed of light, Indicates the number of sweep cycles; otherwise: m ∈[1, M ] in, Indicates the first m Within each sweep frequency cycle A new pulse compression signal at every moment. Indicates the first Under the confidence trajectory The instantaneous radial distance between the radar and the target. Indicates the first m The slow time of one sweep cycle, This indicates the radar's range resolution. c At the speed of light, This indicates the number of sweep cycles.

6. The high-frequency radar high-speed target detection method according to claim 5, characterized in that: Step 4, which involves incoherently accumulating the new pulse compression signal along the confidence motion trajectory, is as follows: m ∈[1, M ] in, This indicates that the search parameters are The amplitude value of the incoherent accumulated pulse compression signal, where Indicates in In the initial radial distance of the search, the first Given the initial radial distance of the search, the absolute value of the coherent accumulation amplitude reaches its maximum value at the corresponding moment. The actual speed of a search and the The initial radial angle of the search ,in , , , This indicates the number of initial radial distance searches. Indicates the number of searches performed at the actual speed. This indicates the number of initial radial angle searches. Indicates the first m Within each sweep frequency cycle The new pulse compression signal at that moment. This indicates the duration of each sweep cycle. Indicates the first Under the confidence trajectory The instantaneous radial distance between the radar and the target. Indicates the first m The slow time of one sweep cycle, This indicates the radar's range resolution. c It is the speed of light.

7. The high-frequency radar high-speed target detection method according to claim 6, characterized in that: The accurate motion parameters of the target obtained through peak detection in step 4 are expressed as follows: in, This represents the value of the function's independent variable when the function reaches its maximum value. This indicates that the search parameters are The maximum value of the amplitude of the incoherent accumulated pulse compression signal, where Indicates in In the initial radial distance of the search, the first Given the initial radial distance of the search, the absolute value of the coherent accumulation amplitude reaches its maximum value at the corresponding moment. The actual speed of a search and the The initial radial angle of the search ,in , , , This indicates the number of initial radial distance searches. Indicates the number of searches performed at the actual speed. This indicates the number of initial radial angle searches. This represents the estimated initial radial distance. This represents the estimated true speed. This represents the estimated initial radial angle.

8. A computer-readable medium, characterized in that, It stores a computer program executed by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method as described in any one of claims 1-7.

Images