Radar target detection method and system based on range gate noise normalization

Abstract

The invention relates to the technical field of thunder method target detection, in particular to a radar target detection method and system based on range gate noise normalization. Comprising the steps of setting a coherent accumulation strategy according to basic scanning data; performing coherent accumulation processing on the radar echo signal according to a coherent accumulation strategy, and generating an initial velocity spectrum according to a processing result; setting a normalization instruction of the initial velocity spectrum, and generating a velocity spectrum to be detected according to the normalization instruction; performing two-dimensional constant false alarm detection on the velocity spectrum to be detected, and outputting a real target trace point according to a detection result; through the noise normalization processing of the distance gate, the noise floor which is integrally raised due to the influence of strong clutters can be leveled, the shielded weak target signal is highlighted, the imbalance of background noise is eliminated, the signal-to-noise ratio of a real target is relatively improved, the false alarm is effectively suppressed, the real target is not excessively suppressed, and the false alarm rate of the real target is improved. Therefore, the detection performance of the radar on the tiny target is improved.

Description

Technical Field

[0001] This application relates to the field of radar target detection technology, and in particular to a radar target detection method and system based on range gate noise normalization. Background Technology

[0002] When radar detects low, small, and slow-moving targets, strong clutter (such as tall buildings) in urban environments poses a significant challenge. Due to the theoretical limits of radar system noise and the sidelobe suppression performance of window functions in signal processing, in traditional two-dimensional constant false alarm rate (CFAR) detection, the energy of strong clutter leaks into other Doppler channels, leading to two main problems: False alarm propagation: Strong clutter creates false targets on multiple velocity channels, significantly increasing the overall false alarm rate. Masking effect: Strong clutter raises the overall background noise (noise floor) of the range gate it's located in, causing real, small target signals near that range gate to be submerged and unable to be effectively detected. Traditional clutter maps or CFAR techniques mainly address clutter suppression between different range gates, but they struggle to effectively solve the masking effect on different velocity units caused by strong clutter sidelobe leakage. Summary of the Invention

[0003] The purpose of this application is to provide a radar target detection method and system based on range gate noise normalization in order to solve the above-mentioned technical problems, which aims to suppress strong clutter masking effects and improve the radar's detection performance for small targets.

[0004] In some embodiments of this application, noise normalization processing by range gate can "flatten" the overall noise floor that is raised due to strong clutter, making the weak target signal that has been blocked stand out, eliminating the unevenness of background noise, and relatively improving the signal-to-noise ratio of the real target. While effectively suppressing false alarms, it will not over-suppress the real target, thereby improving the radar's detection performance for small targets.

[0005] In some embodiments of this application, a noise reference is dynamically selected during the normalization process. The noise reference of each range gate can adapt to the local clutter and target distribution, avoiding the contamination of background estimation by strong targets and maintaining statistical efficiency in uniform regions, thereby improving the detection performance of radar in non-uniform environments.

[0006] In some embodiments of this application, a radar target detection method based on range gate noise normalization is provided, including: Set the coherent accumulation strategy based on the basic scan data; The radar echo signal is coherently accumulated according to the coherent accumulation strategy, and an initial velocity spectrum is generated based on the processing result. Set the normalization command for the initial velocity spectrum, and generate the velocity spectrum to be detected based on the normalization command; Two-dimensional constant false alarm rate (CFAR) detection is performed on the velocity spectrum to be detected, and the true target point is output based on the detection results. The initial velocity spectrum is a two-dimensional signal data.

[0007] In some embodiments of this application, the setting of the coherent accumulation strategy includes: Generate multiple coherent cumulative functions; A library of cumulative functions will be built based on all coherent cumulative functions. Establish multiple environmental signal characteristics; Select the target function sequentially from the cumulative function library; The target function is set according to the characteristics of all environmental signals to adapt to the scenario; Set the appropriate scenarios for each parameter accumulation function in sequence; Establish the target scenario based on the basic scan data; Generate the matching values ​​of the target expected scene and each adapted scene, and set the coherent accumulation function corresponding to the maximum value among all matching values ​​as the execution accumulation function; Generate a coherent accumulation strategy based on the execution of the accumulation function.

[0008] In some embodiments of this application, an initial velocity spectrum is generated based on the processing result, including: Acquire radar echo signals; An initial two-dimensional matrix is ​​generated based on the radar echo signal; Generate correction instructions and coherent accumulation instructions based on the coherent accumulation strategy; The initial two-dimensional matrix is ​​preprocessed according to the correction instructions; The initial velocity spectrum is output based on the coherent accumulation command and the preprocessing results.

[0009] In some embodiments of this application, the normalization command for setting the initial velocity spectrum includes: Establish a noise reference library; The noise reference library includes: multiple noise reference sub-components: Multiple range gates are generated based on the initial velocity spectrum; Select the target distance gate sequentially from all distance gates; Generate a background candidate set for the target range gate based on the initial velocity spectrum; Generate prior background features for the target distance gate; The normalization efficiency value of each noise reference component to the target distance gate is generated based on the background candidate set and prior background features. Set the noise reference sub-component corresponding to the maximum value among all normalized efficiency values ​​as the execution reference sub-component of the target distance gate; Set the execution reference sub-components for each distance gate in sequence; Generate a normalization instruction based on all executed reference sub-components.

[0010] In some embodiments of the present application, generating the background candidate set of the target range gate includes: Generate the first Doppler dimension data set according to the initial velocity spectrum; Set multiple sub-units for generating the target range gate according to the first Doppler dimension data set; Generate the power value of each sub-unit; Generate the rejection value of each sub-unit according to all the power values; Preset the rejection value threshold F1; If F1 < fi, i = 1, 2... n, generate a single rejection instruction according to the i-th sub-unit of the target range gate; Where, fi is the rejection value of the i-th sub-unit of the target range gate; n is the number of sub-units in the target range gate; Generate the background candidate set of the target range gate according to the output results of all the rejection instructions.

[0011] In some embodiments of the present application, generating the prior background features of the target range gate includes: Generate the second Doppler dimension data set of the target range gate according to the basic scan data; Generate the prior background features according to the second Doppler dimension data set; The prior background features include: mean, median, standard deviation, coefficient of variation, and quantile parameter.

[0012] In some embodiments of the present application, generating the normalization efficiency value of each noise reference sub-component for the target range gate includes: Sequentially select the target reference sub-component in the noise reference library; Generate the first-level processing value K1 of the target reference sub-component according to the prior background features of the target range gate; Generate the second-level processing value K2 of the target reference sub-component according to the background candidate set of the target range gate; Generate the normalization efficiency value d of the target reference sub-component, d = K1 + K2.

[0013] In some embodiments of the present application, a radar target detection system based on range gate noise normalization is provided, including: A signal processing unit, configured to set a coherent accumulation strategy according to the basic scan data; The signal processing unit is further configured to perform coherent accumulation processing on the radar echo signal according to the coherent accumulation strategy, and generate an initial velocity spectrum according to the processing result; A central control unit, configured to set a normalization instruction for the initial velocity spectrum, and generate a velocity spectrum to be detected according to the normalization instruction; The central control unit is also used to perform two-dimensional constant false alarm detection on the velocity spectrum to be detected, and output the real target point trace based on the detection result. The signal processing unit includes: The first processing module is used to generate various coherent accumulation functions; A library of cumulative functions will be built based on all coherent cumulative functions. Establish multiple environmental signal characteristics; Select the target function sequentially from the cumulative function library; The target function is set according to the characteristics of all environmental signals to adapt to the scenario; Set the appropriate scenarios for each parameter accumulation function in sequence; Establish the target scenario based on the basic scan data; Generate the matching values ​​of the target expected scene and each adapted scene, and set the coherent accumulation function corresponding to the maximum value among all matching values ​​as the execution accumulation function; Generate a coherent accumulation strategy based on the execution of the accumulation function.

[0014] In some embodiments of this application, the signal processing unit further includes: The second processing module is used to acquire radar echo signals; An initial two-dimensional matrix is ​​generated based on the radar echo signal; Generate correction instructions and coherent accumulation instructions based on the coherent accumulation strategy; The initial two-dimensional matrix is ​​preprocessed according to the correction instructions; The initial velocity spectrum is output based on the coherent accumulation command and the preprocessing results.

[0015] In some embodiments of this application, the central control unit includes: The first control module is used to establish a noise reference library; The noise reference library includes: multiple noise reference sub-components: Multiple range gates are generated based on the initial velocity spectrum; Select the target distance gate sequentially from all distance gates; Generate a background candidate set for the target range gate based on the initial velocity spectrum; Generate prior background features for the target distance gate; The normalization efficiency value of each noise reference component to the target distance gate is generated based on the background candidate set and prior background features. Set the noise reference sub-component corresponding to the maximum value among all normalized efficiency values ​​as the execution reference sub-component of the target distance gate; Set the execution reference sub-components for each distance gate in sequence; Generate normalized instructions based on all execution baseline subcomponents; Among them, generating a background candidate set for the target range gate includes: Generating a first Doppler dimension data set according to the initial velocity spectrum; Setting multiple subunits for generating the target range gate according to the first Doppler dimension data set; Generating the power value of each subunit; Generating the rejection value of each subunit according to all the power values; Presetting a rejection value threshold F1; If F1 < fi, i = 1, 2... n, generating a single rejection instruction according to the i-th subunit of the target range gate; Where, fi is the rejection value of the i-th subunit of the target range gate; n is the number of subunits in the target range gate; Generating a background candidate set for the target range gate according to the output result of all the rejection instructions.

[0016] Compared with the prior art, the beneficial effects of a radar target detection method and system based on range gate noise normalization in an embodiment of the present application are as follows: In some embodiments of the present application, through the noise normalization processing for each range gate, the noise floor that is overall elevated due to strong clutter can be "flattened", so that the weak target signals obscured are highlighted, the unevenness of the background noise is eliminated, the signal-to-noise ratio of the real target is relatively improved, while effectively suppressing false alarms, and it will not cause over-suppression of the real target, thereby improving the detection performance of the radar for small targets.

[0017] In some embodiments of the present application, by dynamically selecting the noise reference during the normalization process, the noise reference of each range gate can adapt to the local clutter and target distribution, avoiding the contamination of the background estimation by strong targets, and maintaining the statistical efficiency in the uniform region, improving the detection performance of the radar in a non-uniform environment. Description of the Drawings

[0018] Figure 1 is a flowchart of a radar target detection method based on range gate noise normalization in a preferred embodiment of an embodiment of the present application. Detailed Embodiments

[0019] The following combines the drawings and embodiments to further describe the detailed embodiments of the present application in detail. The following embodiments are used to illustrate the present application, but are not used to limit the scope of the present application.

[0020] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0021] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0022] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0023] like Figure 1 As shown, a preferred embodiment of this application provides a radar target detection method based on range gate noise normalization, comprising: S101: Set the coherent accumulation strategy based on the basic scan data; S102: Perform coherent accumulation processing on the radar echo signal according to the coherent accumulation strategy, and generate an initial velocity spectrum based on the processing result; S103: Set the normalization command for the initial velocity spectrum and generate the velocity spectrum to be detected according to the normalization command; S104: Perform two-dimensional constant false alarm detection on the velocity spectrum to be detected, and output the true target point traces based on the detection results; The initial velocity spectrum is a two-dimensional signal data.

[0024] Specifically, basic scan data refers to a rapid scan of the entire airspace using a lower resolution or fewer pulses, and the scan record data is set as basic scan data.

[0025] Specifically, multiple environmental signal features are selected based on historical parameters. These environmental signal features include, but are not limited to, parameters that affect the coherent cumulative processing effect, such as target density, strong signal-to-noise ratio, clutter background intensity, target velocity, and range resolution.

[0026] Specifically, the coherent accumulation strategy is set, including: Generate multiple coherent cumulative functions; A library of cumulative functions will be built based on all coherent cumulative functions. Establish multiple environmental signal characteristics; Select the target function sequentially from the cumulative function library; The target function is set according to the characteristics of all environmental signals to adapt to the scenario; Set the appropriate scenarios for each parameter accumulation function in sequence; Establish the target scenario based on the basic scan data; Generate the matching values ​​of the target expected scene and each adapted scene, and set the coherent accumulation function corresponding to the maximum value among all matching values ​​as the execution accumulation function; Generate a coherent accumulation strategy based on the execution of the accumulation function.

[0027] Specifically, various window functions and different levels of Keystone transform are selected based on historical processing parameters. The window function categories include, but are not limited to, rectangular windows, Hanning windows, Hamming windows, Blackman windows, 70dB Taylor windows, Kaiser windows, and other functions capable of weighting finite-length data. Different levels of Keystone transform include first-order Keystone transform, non-second-order Keystone transform, and Keystone transform disabled. Various coherent accumulation functions are generated based on random combinations of different window functions and different levels of Keystone transform.

[0028] Specifically, a single coherent cumulative function includes: a window function and a single level of Keystone transformation.

[0029] Specifically, all environmental signal features are quantized so that the reference values ​​of each environmental signal feature are within the same range.

[0030] Specifically, the appropriate scenarios for each coherent accumulator function are set based on historical processing parameters. For example, if the coherent accumulator function is a rectangular window and a first-order Keystone transform, the processing scenario for this coherent accumulator function is generated using Fenix ​​historical processing parameters: a clean background, target amplitude close to the noise floor, low target velocity, but not spanning multiple units. The value ranges corresponding to various environmental signal features are then generated based on the processing scenario, thus generating the appropriate scenario for the current coherent accumulator function. For example, when the coherent accumulator function is a Hamming window and a second-order Keystone transform, the corresponding processing scenario is: a large number of detected targets with similar velocities, and relatively high target velocity.

[0031] Specifically, by analyzing the basic scanning data, real-time reference values ​​of various environmental signal features are extracted. Based on all real-time reference values, a corresponding target expected scene is constructed. The matching value is set according to the degree of difference between the reference values ​​of various environmental signal features in the target expected scene and the current adapted scene. The smaller the degree of difference, the larger the matching value. The mapping relationship between the two can be set according to historical parameters.

[0032] Specifically, an initial velocity spectrum is generated based on the processing results, including: Acquire radar echo signals; An initial two-dimensional matrix is ​​generated based on the radar echo signal; Generate correction instructions and coherent accumulation instructions based on the coherent accumulation strategy; The initial two-dimensional matrix is ​​preprocessed according to the correction instructions; The initial velocity spectrum is output based on the coherent accumulation command and the preprocessing results.

[0033] Specifically, the radar echo signal includes multiple pulse echoes, and the echoes of multiple pulses are arranged into an initial two-dimensional matrix according to the range gate (fast time) and the pulse sequence number (slow time).

[0034] Specifically, correction instructions and coherent accumulation instructions are generated based on the execution accumulation function in the coherent accumulation strategy. The correction instructions can be set according to the Keystone transform level in the execution accumulation function. Based on the correction instructions, range travel correction is performed, aligning echoes of the same target to the same range gate. If the execution accumulation function does not initiate Keystone transform, the corresponding correction instructions do not perform range travel correction. Preprocessing results (i.e., echoes of the same target aligned to the same range gate) are generated based on the range travel correction results.

[0035] Specifically, the coherent accumulation instruction involves first applying a window function (i.e., executing the window function selected in the accumulation function), and then performing a Fourier transform (FFT) on the data of each distance gate along the slow time dimension. This achieves coherent accumulation and outputs the initial velocity spectrum.

[0036] Specifically, the initial velocity spectrum is two-dimensional signal data, in which one dimension is the range gate (target distance) and the other dimension is the Doppler channel (target radial velocity).

[0037] It is understood that in the above embodiments, by dynamically adjusting the coherent accumulation function, a dynamic balance is maintained between suppressing sidelobes (avoiding false alarms) and maintaining resolution (distinguishing neighboring targets), thereby improving the radar's detection performance in complex environments.

[0038] In a preferred embodiment of this application, the normalization command for setting the initial velocity spectrum includes: Establish a noise reference library; The noise reference library includes: multiple noise reference sub-components: Multiple range gates are generated based on the initial velocity spectrum; Select the target distance gate sequentially from all distance gates; Generate a background candidate set for the target range gate based on the initial velocity spectrum; Generate prior background features for the target distance gate; The normalization efficiency value of each noise reference component to the target distance gate is generated based on the background candidate set and prior background features. Set the noise reference sub-component corresponding to the maximum value among all normalized efficiency values ​​as the execution reference sub-component of the target distance gate; Set the execution reference sub-components for each distance gate in sequence; Normalized instructions are generated based on all execution baseline subcomponents.

[0039] Specifically, the noise reference components include, but are not limited to, the mean, truncated mean (including multiple truncation ratios, with different truncation ratios being different noise reference components), median, recursive smoothing value, and other noise references. Each noise reference component includes a normalization method that matches the noise reference. For example, if the noise reference is the mean, then the normalization method is mean normalization.

[0040] Specifically, all range gates are selected based on the lateral dimension (i.e., the range gate dimension) in the initial velocity spectrum.

[0041] Specifically, by dynamically selecting the noise reference during the normalization process, the noise reference of each range gate can adapt to the local clutter and target distribution, avoiding the contamination of background estimation by strong targets, and maintaining statistical efficiency in uniform regions, thereby improving the radar's detection performance in non-uniform environments.

[0042] Specifically, the background candidate set for generating the target distance gate includes: Generate the first Doppler dimensional dataset based on the initial velocity spectrum; Generate multiple subunits of a target range gate according to a first Doppler dimension dataset; Generate the power value of each subunit; Generate the rejection value of each subunit according to all power values; Preset a rejection value threshold F1; If F1 < fi, i = 1, 2... n, generate a single rejection instruction according to the ith subunit of the target range gate; Where, fi is the rejection value of the ith subunit of the target range gate; n is the number of subunits in the target range gate; Generate a background candidate set of the target range gate according to the output results of all rejection instructions.

[0043] Specifically, the first Doppler dimension dataset refers to all Doppler dimension data corresponding to the target range gate in the initial velocity spectrum. And it is divided into multiple subunits (i.e., Doppler units), and the power values of each subunit are generated in turn along the Doppler dimension. Specifically, set the corresponding rejection value according to the difference between the power value of the current subunit and the power values of the two adjacent subunits on the left and right (the current subunit minus the adjacent subunit). The larger the difference, the larger the local peak of the current subunit, and the larger the corresponding rejection value. The mapping relationship between the two can be set according to historical parameters.

[0044] Specifically, the rejection value threshold can be set according to historical parameters. If the rejection value of the current subunit is greater than the preset rejection value threshold, it means that the current subunit is a suspected target unit. At this time, the current subunit and several protection units in the attachment need to be removed from the background according to the rejection instruction. In this application, the protection units are preferably two subunits on the left and right respectively.

[0045] Specifically, generate a background candidate set of the target range gate according to the rejection result. At the same time, generate statistical parameters of the background candidate set, and the statistical parameters include: mean, median, coefficient of variation (the ratio of standard deviation to mean), and number of samples, so as to judge the clutter degree corresponding to the target range gate. The above statistical parameters are all corresponding parameters generated from the data in the background candidate set. Specifically, generate the prior background features of the target range gate, including: Generate a second Doppler dimension dataset of the target range gate according to the basic scan data; Generate prior background features according to the second Doppler dimension dataset; The prior background features include: mean, median, standard deviation, coefficient of variation, and quantile parameters.

[0046] Specifically, the second Doppler dimension dataset is all Doppler dimension data corresponding to the target range gate in the basic scan data, and judge whether the target range gate is a uniform clutter area according to the generated prior background features.

[0047] Specifically, the normalized efficiency values ​​of each noise reference sub-component with respect to the target range gate are generated, including: Select the target reference sub-components sequentially from the noise reference library; The first-level processing value K1 of the target reference sub-component is generated based on the prior background features of the target distance gate; The secondary processing value K2 of the target reference sub-component is generated based on the background candidate set of the target distance gate; Generate the normalized efficiency value d of the target reference sub-component, d = K1 + K2.

[0048] Specifically, the clutter uniformity of the target range gate is generated based on prior background features, and the processing effect of the target reference sub-component on the current clutter uniformity is generated. The better the processing effect, the larger the corresponding first processing value (for example, the mean and truncated mean have a good processing effect on high clutter uniformity, while the median has a good processing effect on poor clutter uniformity. The processing effect of each noise reference sub-component on different clutter uniformity can be obtained by analyzing historical parameters). The mapping relationship between the processing effect and the first processing value can be set according to historical parameters.

[0049] Specifically, the target density of the target distance gate is generated based on the background candidate set, and the processing effect of the target reference sub-component on the current target density is generated. The better the processing effect, the larger the corresponding second processing value. (For example, the mean is better for no targets, a lower proportion of the truncated mean is better for a small number of targets, and the median and a higher proportion of the truncated mean are better for a large number of targets. The processing effect of each noise reference sub-component on different numbers of targets can be obtained by analyzing historical parameters.) The mapping relationship between the processing effect and the second processing value can be set according to historical parameters.

[0050] Specifically, the first and second processed values ​​have the same range.

[0051] Specifically, the larger the normalization efficiency value, the better the current noise reference component normalizes the initial velocity spectrum.

[0052] It is understandable that, in the above embodiments, the noise normalization processing by the distance gate can "flatten" the overall noise floor that is raised by strong clutter, making the weak target signal that is blocked stand out, eliminating the unevenness of background noise, and relatively improving the signal-to-noise ratio of the real target. While effectively suppressing false alarms, it will not over-suppress the real target, thereby improving the radar's detection performance for small targets.

[0053] In another preferred embodiment of the radar target detection method based on range gate noise normalization according to any of the above preferred embodiments, this preferred embodiment provides a radar target detection system based on range gate noise normalization, including: The signal processing unit is used to set the coherent accumulation strategy based on the basic scan data; The signal processing unit is also used to perform coherent accumulation processing on the radar echo signal according to the coherent accumulation strategy, and generate an initial velocity spectrum based on the processing result; The central control unit is used to set the normalization command for the initial velocity spectrum and generate the velocity spectrum to be detected according to the normalization command. The central control unit is also used to perform two-dimensional constant false alarm detection on the velocity spectrum to be detected, and output the true target point trace based on the detection results; The signal processing unit includes: The first processing module is used to generate various coherent accumulation functions; A library of cumulative functions will be built based on all coherent cumulative functions. Establish multiple environmental signal characteristics; Select the target function sequentially from the cumulative function library; The target function is set according to the characteristics of all environmental signals to adapt to the scenario; Set the appropriate scenarios for each parameter accumulation function in sequence; Establish the target scenario based on the basic scan data; Generate the matching values ​​of the target expected scene and each adapted scene, and set the coherent accumulation function corresponding to the maximum value among all matching values ​​as the execution accumulation function; Generate a coherent accumulation strategy based on the execution of the accumulation function.

[0054] In a preferred embodiment of this application, the signal processing unit further includes: The second processing module is used to acquire radar echo signals; An initial two-dimensional matrix is ​​generated based on the radar echo signal; Generate correction instructions and coherent accumulation instructions based on the coherent accumulation strategy; The initial two-dimensional matrix is ​​preprocessed according to the correction instructions; The initial velocity spectrum is output based on the coherent accumulation command and the preprocessing results.

[0055] In a preferred embodiment of this application, the central control unit includes: The first control module is used to establish a noise reference library; The noise reference library includes: multiple noise reference sub-components: Multiple range gates are generated based on the initial velocity spectrum; Select the target distance gate sequentially from all distance gates; Generate a background candidate set for the target range gate based on the initial velocity spectrum; Generate the prior background features of the target range gate; Generate the normalization efficiency values of each noise reference sub-component for the target range gate based on the background candidate set and the prior background features; Set the noise reference sub-component corresponding to the maximum value among all the normalization efficiency values as the execution reference sub-component of the target range gate; Set the execution reference sub-components of each range gate in sequence; Generate a normalization instruction based on all the execution reference sub-components; Among them, generating the background candidate set of the target range gate includes: Generate the first Doppler dimension data set based on the initial velocity spectrum; Set multiple sub-units for generating the target range gate according to the first Doppler dimension data set; Generate the power values of each sub-unit; Generate the rejection value of each sub-unit based on all the power values; Preset the rejection value threshold F1; If F1 < fi, i = 1, 2... n, generate a single rejection instruction according to the i-th sub-unit of the target range gate; Among them, fi is the rejection value of the i-th sub-unit of the target range gate; n is the number of sub-units in the target range gate; Generate the background candidate set of the target range gate according to the output results of all the rejection instructions.

[0056] According to the first concept of the present application, through the noise normalization processing for each range gate, the noise floor that is overall elevated due to strong clutter can be "flattened", making the weak target signals that are masked stand out, eliminating the imbalance of background noise, relatively increasing the signal-to-noise ratio of real targets, effectively suppressing false alarms without over-suppressing real targets, thereby improving the radar's detection performance for small targets.

[0057] According to the second concept of the present application, by dynamically selecting the noise reference during the normalization process, the noise reference of each range gate can adapt to the local clutter and target distribution, avoiding the contamination of background estimation by strong targets, and maintaining statistical efficiency in uniform regions, improving the radar's detection performance in non-uniform environments.

[0058] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present application, several improvements and substitutions can still be made, and these improvements and substitutions should also be regarded as the protection scope of the present application.

Claims

1. A radar target detection method based on range gate noise normalization, characterized in that, Comprising: Setting a coherent integration strategy according to the basic scan data; Performing coherent integration processing on the radar echo signal according to the coherent integration strategy, and generating an initial velocity spectrum according to the processing result; Setting a normalization instruction for the initial velocity spectrum, and generating a to-be-detected velocity spectrum according to the normalization instruction; Performing two-dimensional constant false alarm detection on the to-be-detected velocity spectrum, and outputting real target traces according to the detection result; Wherein, the initial velocity spectrum is two-dimensional signal data; The setting of the coherent integration strategy includes: Generating multiple coherent integration functions; Establishing an integration function library according to all the coherent integration functions; Establishing multiple environmental signal characteristics; Sequentially selecting target functions in the integration function library; Setting an adaptation scenario for the target function according to all the environmental signal characteristics; Sequentially setting the adaptation scenarios for each coherent integration function; Establishing a target expected scenario according to the basic scan data; Generating a matching value between the target expected scenario and each adaptation scenario, and setting the coherent integration function corresponding to the maximum value among all the matching values as the execution integration function; Generating a coherent integration strategy according to the execution integration function; The setting of the normalization instruction for the initial velocity spectrum includes: Establishing a noise reference library; The noise reference library includes: multiple noise reference sub-components: Generating multiple range gates according to the initial velocity spectrum; Sequentially selecting a target range gate among all the range gates; Generating a background candidate set for the target range gate according to the initial velocity spectrum; Generating a prior background feature for the target range gate; Generating a normalization efficiency value of each noise reference sub-component for the target range gate according to the background candidate set and the prior background feature; Setting the noise reference sub-component corresponding to the maximum value among all the normalization efficiency values as the execution reference sub-component for the target range gate; Sequentially setting the execution reference sub-components for each range gate; Generating a normalization instruction according to all the execution reference sub-components.

2. The radar target detection method based on range gate noise normalization as described in claim 1, characterized in that, Generating an initial velocity spectrum according to the processing result, including: Obtaining the radar echo signal; Generating an initial two-dimensional matrix according to the radar echo signal; Generating a correction instruction and a coherent integration instruction according to the coherent integration strategy; Performing preprocessing on the initial two-dimensional matrix according to the correction instruction; Outputting an initial velocity spectrum according to the coherent integration instruction and the preprocessing result.

3. The radar target detection method based on range gate noise normalization as described in claim 2, characterized in that, The generation of the background candidate set for the target range gate includes: Generating a first Doppler dimension data set according to the initial velocity spectrum; Setting multiple sub-units for generating the target range gate according to the first Doppler dimension data set; Generating the power value of each sub-unit; Generating an exclusion value for each sub-unit according to all the power values; Presetting an exclusion value threshold F1; If F1 < fi, i = 1, 2... n, generating a single exclusion instruction according to the i-th sub-unit of the target range gate; Wherein, fi is the exclusion value of the i-th sub-unit of the target range gate; n is the number of sub-units in the target range gate; Generating a background candidate set for the target range gate according to the output result of all the exclusion instructions.

4. The radar target detection method based on range gate noise normalization as described in claim 3, characterized in that, The generation of the prior background feature for the target range gate includes: Generating a second Doppler dimension data set for the target range gate according to the basic scan data; Generating a prior background feature according to the second Doppler dimension data set; The prior background feature includes: mean, median, standard deviation, coefficient of variation and quantile parameter.

5. The radar target detection method based on range gate noise normalization as described in claim 4, characterized in that, The generation of the normalization efficiency value of each noise reference sub-component for the target range gate includes: Select target reference sub-components from the noise reference library in sequence; Generate the first-level processing value K1 of the target reference sub-component according to the prior background characteristics of the target range gate; Generate the second-level processing value K2 of the target reference sub-component according to the background candidate set of the target range gate; Generate the normalized efficiency value d of the target reference sub-component, where d = K1 + K2.

6. A radar target detection system based on range gate noise normalization, employing the radar target detection method based on range gate noise normalization as described in any one of claims 1-5, characterized in that, It includes: A signal processing unit for setting a coherent accumulation strategy according to the basic scan data; The signal processing unit is also used to perform coherent accumulation processing on the radar echo signal according to the coherent accumulation strategy, and generate an initial velocity spectrum according to the processing result; A central control unit for setting a normalization instruction for the initial velocity spectrum and generating a velocity spectrum to be detected according to the normalization instruction; The central control unit is also used to perform two-dimensional constant false alarm detection on the velocity spectrum to be detected and output real target traces according to the detection result; Among them, the signal processing unit includes: A first processing module for generating multiple coherent accumulation functions; Establish a cumulative function library according to all the coherent accumulation functions; Establish multiple environmental signal characteristics; Select target functions from the cumulative function library in sequence; Set the adaptation scenario of the target function according to all the environmental signal characteristics; Set the adaptation scenarios of each coherent accumulation function in sequence; Establish a target expected scenario according to the basic scan data; Generate the fit value between the target expected scenario and each adaptation scenario, and set the coherent accumulation function corresponding to the maximum value among all the fit values as the execution accumulation function; Generate a coherent accumulation strategy according to the execution accumulation function.

7. The radar target detection system based on range gate noise normalization as described in claim 6, characterized in that, The signal processing unit also includes: A second processing module for acquiring the radar echo signal; Generate an initial two-dimensional matrix according to the radar echo signal; Generate a correction instruction and a coherent accumulation instruction according to the coherent accumulation strategy; Preprocess the initial two-dimensional matrix according to the correction instruction; Output an initial velocity spectrum according to the coherent accumulation instruction and the preprocessing result.

8. The radar target detection system based on range gate noise normalization as described in claim 7, characterized in that, The central control unit includes: A first control module for establishing a noise reference library; The noise reference library includes: multiple noise reference sub-components: Generate multiple range gates according to the initial velocity spectrum; Select a target range gate from all the range gates in sequence; Generate a background candidate set of the target range gate according to the initial velocity spectrum; Generate the prior background characteristics of the target range gate; Generate the normalized efficiency value of each noise reference sub-component for the target range gate according to the background candidate set and the prior background characteristics; Set the noise reference sub-component corresponding to the maximum value among all the normalized efficiency values as the execution reference sub-component of the target range gate; Set the execution reference sub-components of each range gate in sequence; Generate a normalization instruction according to all the execution reference sub-components; Among them, generating the background candidate set of the target range gate includes: Generate a first Doppler dimension data set according to the initial velocity spectrum; Set multiple sub-units for generating the target range gate according to the first Doppler dimension data set; Generate the power value of each sub-unit; Generate the rejection value of each sub-unit according to all the power values; Preset a rejection value threshold F1; If F < fi, i = 1, 2... n, generate a single rejection instruction according to the i-th sub-unit of the target range gate; Among them, fi is the rejection value of the i-th sub-unit of the target range gate; n is the number of sub-units in the target range gate; Generate the background candidate set of the target range gate according to the output result of all the rejection instructions.

Images