Target detection processing method, terminal equipment and signal compensation method

Abstract

The invention discloses a target detection processing method and terminal equipment, and aims to ensure that energy of a TX dimension and energy of an RX dimension are correctly accumulated through coherent gain optimization in the embodiment of the invention, and finally improve the SNR of a target echo signal, thereby ensuring that a subsequent target echo signal is not missed in a coarse search stage. Furthermore, through TX Order solution, correct sorting of TX dimension data is ensured, so that the MIMO signal can be correctly decoded. Therefore, the calculation amount of subsequent data reordering is reduced, and the coherent gain of the TX dimension is improved.

Description

Technical Field

[0001] This application relates to, but is not limited to, radar signal processing technology, and particularly to a target detection and processing method and terminal equipment, and a signal compensation method. Background Technology

[0002] Target detection is one of the core functions of a radar system, referring to the detection and identification of targets from received echo signals. This typically includes extracting information such as range, velocity, and azimuth. In frequency-modulated continuous wave (FMCW) radar, target detection mainly relies on methods such as frequency domain analysis, coherent signal processing, and constant false alarm rate (CFAR) detection to distinguish targets from background noise.

[0003] Improving the accuracy of target detection is a technical problem that needs to be addressed. Summary of the Invention

[0004] This application provides a target detection processing method and terminal device, as well as a signal compensation method, which can improve the accuracy of target detection.

[0005] This invention provides a target detection processing method, which includes, before target detection: Perform one or any combination of the following gain optimization processes before target detection: coherent gain optimization, principal subarray summation, and incoherent gain weighted merging to improve the signal-to-noise ratio (SNR) of the target detection signal; The coherent gain optimization is used to align the signals of the transmit antenna (TX dimension) and the receive antenna (RX dimension) through phase compensation; the master subarray summation is used to coherently combine the signals of the master subarray; and the incoherent gain weighted combining is used to obtain the incoherent gain of the slave subarray by calculating the signal power of the slave subarray and weighting it with the power of the master subarray.

[0006] Optionally, before the target detection gain optimization process, the method further includes: The coherent gain of the RX dimension is extracted from the received target echo signal containing multiple TX dimension data, and the arrangement order of the transmit antenna dimension signals is determined by combining the encoded information after differential delay modulation (DDM) to ensure that the TX signals are correctly aligned.

[0007] Optionally, the coherent gain of the receiving antenna dimension can be extracted using the frequency domain transformation of the receiving end.

[0008] Optionally, determining the arrangement order of the transmitting antenna dimension signals includes: For the first input signal, a Fast Fourier Transform (FFT) is performed in the chirp dimension to obtain Doppler spectrum information. The first input signal is the target echo signal received at the RX end, which contains data in multiple TX dimensions. Perform RX FFT processing on all or part of the RX signals of the receiving antenna, calculate the coherence gain between RX signals, and arrange the antenna array at the RX end on the same horizontal line. The signal power of the third signal after RX FFT processing is calculated and the data is split according to the DDM Code information after DDM modulation to obtain the processed fourth signal; The fifth signal is obtained by accumulating and synthesizing the data from each DDM Code dimension; Perform a maximum value search on the fifth signal to obtain the maximum value index in the RX dimension. The maximum value index represents the position of the strongest signal in the RXFFT dimension. Extract the corresponding TX dimension power data from the fourth signal by using the maximum value index on the RX dimension; The extracted TX dimension power data and the DDM modulated coded information are convolved and the maximum value is calculated to obtain the arrangement order of the transmit antenna dimension signals.

[0009] Optional, also includes: The imbalance of each RX signal in the first input signal is compensated to obtain the first signal; the FFT in the Chirp dimension is to perform FFT on the first signal in the Chirp dimension.

[0010] Optionally, the step of compensating for the imbalance of each RX signal in the first input signal to obtain the first signal includes: Calculate the average signal amplitude for each RX channel to estimate the gain of each RX channel; Calculate the noise variance for each RX channel to estimate the noise level for each RX channel; The RX signal compensation factor is obtained based on the gain and noise level of each RX channel. And use the obtained RX signal compensation factor The first signal is obtained by compensating the target echo signal.

[0011] Optionally, the arrangement order of the transmitted antenna dimension signals is used for coherent gain D-CFAR calculation; or, for incoherent gain DA-CFAR calculation.

[0012] Optionally, for target detection using the Doppler-azimuth joint constant false alarm rate (DA-CFAR) algorithm, the incoherent gain of the subarray is obtained through processing including TX dimension compensation and any one or any combination of the following: data remapping and DOA FFT processing.

[0013] Optionally, obtaining the incoherent gain of the subarray includes: The signal Sig[dop, rx, tx] used for incoherent gain calculation is used as the second input signal. Phase compensation is performed along the TX dimension to obtain the seventh signal. The signal used for incoherent gain calculation is the TX dimension data in the Doppler spectrum information obtained after rearranging the first signal after imbalance compensation according to the arrangement order of the transmit antenna dimension signal and performing FFT in the Chirp dimension. For the seventh signal, the main subarray is remapped according to its array position based on the radar antenna array information; Perform an FFT on the first main subarray signal to extract the azimuth information and obtain the eighth signal; Calculate the power of the eighth signal to obtain the power information of the main subarray; The power of the first subarray signal is calculated and accumulated along the TX dimension to obtain the total power of the subarray. The incoherent gain of the slave array is obtained by weighted summation of the power information of the master subarray and the total power of the slave subarray.

[0014] Optionally, the weighting factors used in the weighted summation Used to control the power ratio of the first master subarray and the first slave subarray; The weighting factor It can be a fixed value; or it can be adjusted adaptively according to different application scenarios.

[0015] Optional, also includes: By constructing a Trade-Off objective function that maximizes the signal power / noise variance, the Trade-Off objective function is: ;in, The number of antennas represents the coherence gain. The number of antennas represents the incoherent gain. Indicates the energy of the target signal. Indicates noise power. This represents the weighting factor; By applying the Trade-Off objective function with respect to the weighting factors Find the derivative and calculate the optimal value. The value that makes the Trade-Off objective function reach its maximum value.

[0016] Optionally, for target detection using the Doppler dimension constant false alarm rate (CFAR) algorithm, the coherence gain of the slave array is obtained through processing including TX compensation and phase compensation, as well as any one or any combination of the following: master-slave array signal accumulation, TX dimension FFT, and RX-TX dimension optimization.

[0017] Optionally, obtaining the coherent gain of the subarray includes: Based on the arrangement order of the transmitted antenna dimension signals, the TX dimension signal is extracted from the fourth signal and used as the third input signal to compensate the TX dimension signal; For each TX signal that has been compensated for in the TX dimension, phase compensation is performed according to each grid point of the RX FFT to obtain the TX dimension phase compensation. The compensated signals include the second master subarray signal and the second slave subarray signal. The second master subarray signal is the TX dimension signal of the master subarray, which has been phase compensated according to each grid point of the RX FFT. The second slave subarray signal is the TX dimension signal of the slave subarray, which has been phase compensated according to each grid point of the RX FFT. The TX dimensions of the compensated second master array signal are summed to obtain the gain that maximizes the energy of the master array signal. The second subarray signal is combined with the main subarray gain signal and FFT is performed along the TX dimension to convert the TX dimension signal to the Doppler-antenna domain for signal enhancement in the TX dimension to obtain the thirteenth signal.

[0018] Optional, also includes: For the thirteenth signal, find the maximum value along the RX-FFT and TX-FFT dimensions to construct a three-dimensional detection cube signal.

[0019] This application also provides a target detection processing method applied in a MIMO FMCW radar, the method comprising: After performing FFT processing on the received radar echo data, the RDM (Range-Doppler-Channel Map) spectrum is obtained; and After calculating the emission order Tx Order based on the RDM spectrum, constant false alarm rate (CFAR) processing is performed.

[0020] Optionally, the antennas in the antenna array of the MIMO FMCW radar are arranged according to a preset rule; before solving the transmission order Tx Order based on the RDM spectrum, the method further includes: The receiving antenna Rx dimension is compensated based on the antenna arrangement rules.

[0021] Optionally, after the Tx Order calculation and before the CFAR processing, the following steps are also included: The transmitting antenna Tx dimension is compensated based on the antenna arrangement rules.

[0022] Optionally, in the antenna array of the MIMO FMCW radar, the receiving antennas Rx are arranged side by side along the azimuth direction, and in the method: After performing Rx amplitude and phase compensation on the RDM spectrum, the Tx order is calculated based on the Rx coherence gain, and then the Tx phase is supplemented based on the angle information before the CFAR processing is performed.

[0023] Optionally, the CFAR processing includes D-CFAR and DA-CFAR; the method further includes: Determine the task type processed by the CFAR; The task type is D-CFAR, which is processed using a fully coherent gain path. The task type is DA-CFAR, which is processed using a weighted incoherent merging path.

[0024] Optionally, the weighting coefficient k in the weighted incoherent merging path ranges from 0.7 to 0.9.

[0025] This application also provides a signal compensation method, which can be applied to the signal processing of MIMO FMCW radar. The method may include: performing FFT processing on the received radar echo data to obtain the RDM spectrum. Based on the RDM spectrum, compensation is made for non-ideal amplitude and phase errors between each receiving channel to ensure consistency between the receiving channels; and / or, After Tx Order calculation, compensation is performed to address non-ideal amplitude and phase errors between each transmission channel to ensure consistency among them; and / or, After Tx Order calculation and before coherent merging, angle-based ideal phase compensation is performed for the transmit channel.

[0026] Optionally, in the antenna array of the MIMO FMCW radar, the receiving antennas Rx are distributed along the same straight line in the azimuth dimension. The method may also include: taking advantage of the array feature that the receiving antennas Rx are distributed along the same straight line in the azimuth dimension, FFT or DBF can be performed on the Rx channel dimension first, and then Tx Order calculation can be continued.

[0027] In this embodiment of the application, the Tx Order calculation by fusing the coherent gain of the Rx channel is not only advanced but also changed in method. That is, by inputting all data (instead of just candidate points), and taking advantage of the array characteristics of the RX antennas on the same horizontal line, FFT (or DBF) is first performed on the RX channel dimension. The coherent gain of the RX dimension is obtained at the beginning of the Tx Order calculation, which greatly improves the success rate of the calculation and enables the output of global and highly reliable Tx Order calculation results.

[0028] This application embodiment further provides a terminal device, including a memory and a processor, wherein the memory stores the following instructions executable by the processor: for performing the steps of the target detection processing method described in any of the above claims.

[0029] The target detection processing method provided in this application embodiment ensures that the energy in the TX and RX dimensions is correctly accumulated through coherent gain optimization, thereby improving the SNR of the target echo signal and ensuring that subsequent target echo signals are not missed in the coarse search stage.

[0030] Furthermore, by calculating the TX order, the correct sorting of the TX dimension data is ensured, enabling the MIMO signal to be correctly decoded. This reduces the computational load of subsequent data reordering and improves the coherence gain in the TX dimension.

[0031] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0032] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0033] Figure 1 This is a schematic diagram of the target detection method combining coarse search and fine search in the embodiments of this application; Figure 2 This is a flowchart illustrating the target detection processing method in the embodiments of this application; Figure 3 This is a schematic diagram illustrating the process of solving the TX Order in an embodiment of this application; Figure 4 This is a schematic diagram illustrating the process of implementing TX Order calculation in the embodiments of this application; Figure 5This is a schematic diagram illustrating the process of obtaining the incoherent gain of the subarray in an embodiment of this application; Figure 6(a) is a schematic diagram of the Trade-Off function of 4T4R in the embodiment of this application; Figure 6(b) is a schematic diagram of the Trade-Off function of 8T8R in the embodiment of this application; Figure 7 This is a schematic diagram of the Doppler spectrum after weighted incoherent summation of the subarray and the principal subarray in an embodiment of this application; Figure 8 This is a schematic diagram illustrating the process of obtaining the coherent gain of the subarray in an embodiment of this application; Figure 9 This is a schematic diagram illustrating the implementation of TX phase compensation in an embodiment of this application; Figure 10 This is a comparative diagram showing the optimization of target detection performance using different compensation schemes in the embodiments of this application; Figure 11 This is a comparative diagram of different TX orders used in target detection in the embodiments of this application; Figure 12 This is a schematic diagram comparing the performance of different target detection gain schemes in the embodiments of this application; Figure 13 This is a schematic diagram of the target detection process in a traditional MIMO FMCW radar. Figure 14 This is a schematic diagram of the integrated target detection process provided in an embodiment of this application. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in detail below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be arbitrarily combined with each other.

[0035] To facilitate understanding of this application, a more complete description will be provided below with reference to the accompanying drawings, which illustrate embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of this application will be thorough and complete.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0037] It is understood that the terms "first" and "second" used in this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0038] It is understood that the term "connection" in the following embodiments should be understood as "electrical connection," "communication connection," etc., if the connected circuits, modules, units, etc., have electrical signal or data transmission with each other.

[0039] When used herein, the singular forms of “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising / including” or “having,” etc., specify the presence of the stated features, wholes, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0040] The steps illustrated in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases the steps shown or described may be performed in a different order than that presented here.

[0041] The digital signal processing flow of a frequency-modulated continuous wave (FMCW) radar can be divided into four main parts. First, in the digital front-end (DFE) stage, the system preprocesses and compensates the digital signal to optimize signal quality and provide a more stable input for subsequent processing. Second, in the target detection stage, the system analyzes different range bins and Doppler bins to determine the presence of target signals. Subsequently, in the direction of arrival (DOA) stage, the system estimates the direction of arrival for the detected targets to obtain their angle information. Finally, in the post-processing stage, further advanced processing such as target tracking and attitude estimation is performed to improve the system's intelligence level.

[0042] Currently, commonly used object detection methods combine coarse search and fine search, such as... Figure 1As shown, in the coarse search phase, single-input single-output (SISO) signal combining is first performed on each channel. Then, target detection is performed by setting a fixed threshold, thus filtering out potential target points. However, due to the limited gain in the coarse search phase, this process may lose some weaker target signals. In the fine search phase, the system further analyzes the target points detected in the coarse search phase. Specifically, this is done by coherently integrating the data from each channel corresponding to these target points. Subsequently, the system sets the threshold again for detection to eliminate false alarms that may have occurred in the coarse search phase. The coherent processing in the fine search phase can enhance the effective target signal while suppressing noise and interference signals, thereby improving detection accuracy.

[0043] While combining coarse and fine searches can improve target detection capabilities, it still has certain problems and limitations. First, the problem of missing small targets is a major drawback. Because the coarse search stage only performs SISO Combine, it fails to fully utilize signal gain, resulting in weaker target signals (small targets) possibly going undetected. Furthermore, since the fine search stage relies on the detection results of the coarse search stage, if a small target is missed in the coarse search, subsequent fine search processes cannot remedy this, leading to missed detections. Second, in this scheme, the detection threshold calculation is usually based on a fixed benchmark, such as the single-input single-output power (SISO Power) obtained after solving for the transmission order (TX Order). However, because the signal processing methods for coarse and fine searches differ, their statistical distributions are not consistent. Using the same threshold benchmark between the two may lead to false alarms due to differences in statistical distributions. Additionally, to address the false alarm problem caused by the different statistical distributions between the coarse and fine search stages, a larger bias can be set on the threshold benchmark to reduce the false alarm rate. However, increasing the bias also raises the detection threshold, causing some real target signals to be ignored, thus increasing the probability of missed detection, especially for weaker targets, making the optimization of the detection threshold a challenge.

[0044] In summary, the accuracy of target detection is affected by several factors, including but not limited to: Signal-to-Noise Ratio (SNR) – the stronger the target signal, the easier it is to detect; low SNR may lead to missed detections. Interference and noise – environmental noise, clutter, and other signal sources may affect the detection results. Coherence gain – obtaining the coherence gain between the transmit antenna (TX) and the receive antenna (RX) helps improve detection sensitivity. Detection threshold setting – too high a threshold will result in missed detections, while too low a threshold will produce false alarms.

[0045] To improve the accuracy of target detection, embodiments of this application provide a target detection processing method, such as... Figure 2 As shown, it includes: Step 200: Extract the coherent gain of the RX dimension from the received target echo signal containing multiple TX dimension data, and determine the arrangement order of the transmit antenna dimension signals by combining the DDM modulated coding information to ensure that the TX signals are correctly aligned.

[0046] It is important to note that Differential Delay Modulation (DDM) occurs at the TX end. DDM modulation is a special MIMO modulation scheme primarily used to control the time delay of the transmitted signal at the TX end, giving different TX transmitted signals unique phase characteristics through a specific time offset. If the TX end does not use DDM, the RX end does not need to perform TX Order Resolving because the arrangement of the TX dimension data is fixed and requires no additional parsing. If the TX end uses DDM, then the RX end needs to perform TX Order Resolving to correctly resolve the TX dimension signal. In other words, step 200, TX Order Resolving, can be omitted.

[0047] In one exemplary instance, the coherent gain in the receiving antenna (RX) dimension can be extracted using a frequency domain transformation at the receiver. In one embodiment, the coherent gain in the RX dimension can be extracted via a Fast Fourier Transform (FFT) of the RX dimension.

[0048] TX dimension refers to the signal dimension between different transmit antennas (TX). In a multiple transmit antenna (MIMO) system, each TX antenna transmits a different signal, and the RX end needs to correctly interpret the order of the TX signals. RX dimension refers to the signal dimension between different receive antennas (RX), used to enhance the coherence gain of the signal (e.g., RX FFT calculates the coherence gain between RXs).

[0049] The TX Order calculation in step 200 ensures the correct sorting of the TX dimension data, enabling correct decoding of the MIMO signal. This reduces the computational load of subsequent data reordering and improves the coherence gain in the TX dimension.

[0050] In one exemplary instance, such as Figure 9 As shown, step 200 may include: Step 2001: For each RX signal in the first input signal ( Figure 10 The imbalance of the target echo signal Sig[chirp,rx] is compensated to obtain the first signal. Figure 10 The first signal in the input signal is Sig'[chirp,rx]). Here, the first input signal is the target echo signal received at the RX end, which contains data in multiple TX dimensions.

[0051] Different receiving antennas (RX channels) at the RX end may have uneven gain, phase, and noise levels. By compensating for the signal at the RX end, the signal amplitude and phase on all RX channels can be made consistent, thereby improving detection performance and coherence gain.

[0052] According to the Generalized Likelihood Ratio Test (GLRT), the final RX signal compensation factor is... yes: ,in: This represents the average signal amplitude of the nth RX channel. This represents the noise power of the nth RX channel. RX signal compensation factor. It is a compensation coefficient that is normalized for each RX channel to balance the channel gain and noise level, so as to ensure that the signal amplitude and phase of each RX channel are consistent, thereby improving the coherence of the signal.

[0053] In one exemplary instance, step 2001 may include: Calculate the average signal amplitude for each RX channel , In this way, the gain of each RX channel can be estimated. Here, M represents the number of sampling points in the chirp dimension. The chirp dimension generally refers to the dimension along the chirp (frequency-modulated pulse) sequence, that is, the time dimension formed by a series of chirs acquired by the radar within the Coherent Processing Interval (CPI). In FMCW radar, the chirp dimension can also be understood as the fast time dimension, used for subsequent range-dimensional FFT processing.

[0054] Calculate the noise variance for each RX channel. , This allows us to estimate the noise level for each RX channel.

[0055] The RX signal compensation factor is obtained based on the gain and noise level of each RX channel. And use the obtained RX signal compensation factor Compensation is performed on the target echo signal Sig[chirp,rx], i.e., the first signal. This normalizes the gain and noise levels of all RX channels, improving signal consistency. The first signal... To process each RX signal in the first input signal ( Figure 10The result is obtained after compensating for the imbalance of the target echo signal Sig[chirp,rx]).

[0056] In one exemplary instance, in an object detection task, suppose This indicates that the target exists. This indicates that the target does not exist. Let the target be the complex amplitude value, then the GLRT formula is: ,in, This indicates that, under the assumption that the target exists, the signal The probability distribution; This indicates that the signal is valid under the assumption that the target does not exist. The probability distribution; This represents maximizing the likelihood function to estimate the optimal objective parameters. Under the Gaussian noise assumption, the GLRT formula can be transformed into: ,in, The steering vector (including channel amplitude compensation and phase correction factors). The noise floor level for each channel. Indicates the relationship between array geometry and target azimuth. The determined spatial phase factor reflects the physical phase difference of the signal between different receiving antennas. The gain is a complex number, which includes compensation for the amplitude and phase mismatch of the nth channel, thereby achieving channel uniformity and improving the signal synthesis gain. Indicates the antenna spacing. Indicates the signal wavelength. This represents the azimuth angle of the target. Under the Gaussian assumption, for After maximizing the solution, we can obtain... The numerator represents the signal power gain, which is mainly affected by the channel amplitude compensation factor. The denominator is a normalization factor, ensuring scale invariance of the detection quantity. In other words, the denominator is a constant, while the numerator represents the compensation coefficient for the nth channel. for Compensation coefficient It's not SNR, because the numerator is the amplitude, not the magnitude.

[0057] Step 2002: Perform FFT on the first signal after imbalance compensation in the Chirp dimension to obtain Doppler spectrum information.

[0058] In one exemplary instance, step 2002 may include: performing Doppler FFT processing on multiple Chirp signals of each RX channel to obtain a second signal Sig[dop, rx].

[0059] It should be noted that when the channel consistency at the RX end is good, the impact of compensation is small, and step 2001 can be omitted. In this case, step 2002 directly applies compensation to each RX signal in the first input signal. Figure 10 The target echo signal Sig[chirp,rx] is subjected to FFT in the Chirp dimension to obtain Doppler spectral information.

[0060] It should be noted that if the TX end also has multiple channels (such as MIMO, beamforming transmission, etc.), the power amplifiers, feeders, and antennas of each transmission channel will also have inconsistencies in amplitude and phase. During transmission, complex compensation factors can be added before each channel to adjust the amplitude and phase of the transmitted signal, so that the correct beam can be formed in the target direction, ensuring that the transmitted energy is focused.

[0061] Step 2003: Perform FFT on the RX signal and calculate the coherence gain between RX signals.

[0062] In this embodiment of the application, for the radar array, the antenna arrays at the RX end are arranged on the same horizontal line, while the antenna arrays at the TX end are not constrained in their arrangement. Therefore, performing RX FFT calculation is actually obtaining the coherent gain between the RX signals of the target echo signal to be detected, such as... Figure 10 In this context, the third signal Sig[dop, rx-fft] has a coherent gain between RX, which enables more stable resolution of the TX order.

[0063] In one embodiment, an FFT can be performed on the RX signals of all receiving antennas (receiving channels). Performing an FFT on all RX signals (i.e., full-array FFT processing) maximizes the coherence gain in the RX dimension and improves signal quality. In another embodiment, an FFT can be performed on the RX signals of only some receiving antennas (receiving channels), depending on the computational requirements of the system design and the detection target.

[0064] Step 2004: Calculate the signal power and split the data according to the coding information after DDM modulation to obtain the processed fourth signal Sig_pow[dop, rx-fft, code].

[0065] In one exemplary instance, step 2004 may include: Calculate the power spectrum signal of the third signal Sig[dop, rx-fft] after RX FFT processing; The obtained power spectrum signal is classified using the DDM code after DDM modulation at the TX end to obtain the fourth signal Sig_pow[dop, rx-fft,code].

[0066] In this step, after DDM Code parsing, signals of different TX dimensions can be accurately distinguished, ensuring the correctness of TX Order parsing. Thus, this embodiment of the application not only ensures the accuracy of the input data for detection gain calculation, but also overcomes the technical problem that insufficient DDM settlement accuracy may become a bottleneck for detection performance.

[0067] It should be noted that the third signal Sig[dop,rx-fft] after RX FFT processing still retains the phase information in the TX dimension, which is important in subsequent coherent gain calculations (such as coherent accumulation in the TX dimension). The power spectrum signal after power calculation is only a representation of signal intensity and has lost phase information, so it cannot be used for coherent accumulation, but it can be used for TXOrder parsing and DDM Code parsing.

[0068] Step 2005: Accumulate and synthesize the data from each DDM Code dimension to obtain the fifth signal Sig_pow[dop, rx-fft].

[0069] Step 2005 yields the combined power across all DDM Code dimensions, avoiding the problem of uneven power distribution across the RX FFT dimension that might be caused by different DDM Codes.

[0070] Step 2006: Perform a maximum value search on the fifth signal to obtain the maximum value index on the RX dimension.

[0071] In this step, the fifth signal is the power spectrum signal obtained by accumulating along the Code dimension after RX FFT processing. Here, all DDM Code dimension data are accumulated and synthesized.

[0072] In this step, the maximum value index Index_rx-fft[dop] represents the location of the strongest signal in the RX FFT dimension, that is, the point with the strongest signal in the RX dimension.

[0073] If Argmax is directly calculated on the fourth signal Sig_pow[dop, rx-fft, code], each Code needs to be calculated separately, which involves a large amount of computation. In this embodiment, the signals are accumulated and synthesized into the fifth signal Sig_pow[dop, rx-fft], and Argmax is directly calculated on it, which improves the parsing efficiency.

[0074] In one embodiment, step 2006 may include: Perform an Argmax operation on the fifth signal Sig_pow[dop, rx-fft] to find the index Index_rxfft[dop] with the highest signal power in the RX FFT dimension. This index Index_rxfft[dop] represents the point with the strongest signal power in the RX FFT dimension.

[0075] Step 2007: Extract the corresponding TX dimension power data from the fourth signal Sig_pow[dop, rx-fft,code] using the maximum value index on the RX dimension.

[0076] In this step, the TX dimension power data is... Figure 10 The sixth signal in the algorithm is Sig_pow[dop,code]. This step ensures that the input data for TX Order parsing comes from the strongest signal point in the RX FFT dimension, providing a technical guarantee to avoid noise interference and improve the accuracy of TX Order parsing.

[0077] Step 2008: Perform convolution operation on the extracted TX dimension power data and the DDM modulated coded information, and find the maximum value to obtain the arrangement order of the transmit antenna dimension signals.

[0078] In this step, the order of the transmitted antenna dimensional signals is as follows: Figure 10 The index in the table.

[0079] In one exemplary instance, the order of the resolved transmit antenna dimension signals can be used for coherent gain (D-CFAR) or non-coherent gain (DA-CFAR) calculations.

[0080] In one embodiment, when the coherent gain method of the subarray is used for the detection pre-gain method in the next stage, the TX dimension data in the third signal Sig[dop, rx-fft] obtained by performing FFT calculation on all RX signals in step 2003 is rearranged according to the arrangement order of the transmit antenna dimension signals, i.e., the index, to obtain the signal Sig[dop, rx-fft,tx] used for coherent gain calculation of the subarray.

[0081] In one embodiment, when the incoherent gain method of the subarray is used for the detection pre-gain method in the next stage, the TX dimension data of the Doppler spectrum information obtained after performing FFT in the Chirp dimension on the first signal after imbalance compensation in step 2002, i.e., the index, is rearranged according to the arrangement order of the transmitted antenna dimension signal, i.e., the index, to obtain the signal Sig[dop, rx, tx] for incoherent gain calculation.

[0082] The TX end transmits signals in a certain order (e.g., DDM modulation may introduce different TX time offsets). The RX end receives the mixed TX echo signals and needs to parse the TX order to correctly arrange the TX dimension data. In the embodiments of this application, the index obtained by TX Order Resolving in step 200 represents the correct arrangement of the TX dimension signals, ensuring that the data can be correctly input into subsequent signal processing procedures (such as DOA estimation, pre-detection gain calculation, etc.).

[0083] In one exemplary instance, the arrangement order of the TX-dimensional signals obtained by the TX Order parsing method provided in this application embodiment can be optimized by selecting an appropriate gain calculation method according to different detection tasks (such as Doppler dimension or DOA dimension). In one embodiment, after TX Order parsing, coherent accumulation is performed on the TX dimension to obtain a higher pre-detection gain. This method is applicable to Doppler dimension target detection, i.e., D-CFAR. In another embodiment, after TX Order parsing, incoherent accumulation is performed on the TX dimension to optimize the noise floor variance and improve the angle information extraction capability. This method is applicable to orientation angle dimension target detection, i.e., DA-CFAR.

[0084] Step 201: Before target detection, perform target detection gain optimization processing through one or any combination of the following: coherent gain optimization, master subarray summation, and incoherent gain weighted merging to improve the SNR of the target detection signal.

[0085] In one exemplary instance, the role of coherent gain optimization is to align the TX and RX dimension signals (including alignment of the TX dimension signals themselves, alignment of the RX dimension signals themselves, and alignment between the TX and RX dimension signals) through phase compensation, avoiding signal loss caused by phase mismatch. Coherent accumulation of the TX and RX dimensions can increase the target echo signal energy and maximize coherent gain. In one embodiment, coherent gain optimization may include: compensating for phase errors in the TX and RX dimensions to ensure that no phase mismatch occurs when all TX and RX channel signals are coherently superimposed. The phase compensation provided by this application avoids the problem of SNR decrease caused by coherent cancellation of signals in different TX and RX dimensions; and the coherent accumulation of the TX and RX dimensions, i.e., coherent accumulation in the TX and RX dimensions, enhances the target echo signal energy, improves the pre-detection gain, and avoids the problem of missed detection due to insufficient SNR of the target echo signal.

[0086] The coherent gain optimization in this embodiment ensures that the energy in the TX and RX dimensions is correctly accumulated, ultimately improving the SNR of the target echo signal, thereby ensuring that subsequent target echo signals are not missed in the coarse search stage.

[0087] In one exemplary instance, the role of principal subarray summation is to improve the target signal energy while reducing computational complexity through coherent merging of principal subarray signals. This is applicable to D-CFAR or DA-CFAR, improving detection efficiency by reducing the computational dimensionality. In one embodiment, principal subarray summation may include summing the principal subarray signals in both the TX and RX dimensions, merging the TX signals to reduce the computational dimensionality, and retaining all TX dimension gains. Compared to directly performing FFT in all TX dimensions, the principal subarray summation in this embodiment reduces the computational cost of principal subarray summation while optimizing the TX dimension gain.

[0088] In one exemplary instance, the role of incoherent gain weighted merging is to obtain the incoherent gain of the slave array by calculating the signal power of the slave array and weighting and merging it with the power of the master array (i.e., weighted incoherent accumulation), thereby reducing the noise variance and improving the stability of the detection threshold and detection performance. Weighted incoherent gain optimization is suitable for DA-CFAR target detection, especially for optimizing the noise floor level before detection to reduce the false alarm rate. In one embodiment, weighted incoherent gain optimization may include: incoherently accumulating the power of the master array and the slave array, and using a weighting factor to optimize the noise floor stability; by adjusting the weights of the master array and the slave array, the noise floor variance is reduced, thereby improving the stability of the detection threshold.

[0089] In this embodiment, before entering the target detection stage, the SNR is improved through one or any combination of the following pre-detection gain optimizations: coherent gain optimization, principal subarray summation, and incoherent gain weighted merging, to ensure that small targets are not missed in the coarse search stage. Specifically, phase compensation aligns the TX and RX dimensions, avoiding signal loss caused by phase mismatch. Coherent accumulation of the TX and RX dimensions increases the target signal energy. Weighted incoherent accumulation is used for noise reduction optimization to reduce noise variance, thereby improving the stability of the detection threshold.

[0090] In step 201, coherent gain optimization improves the target signal energy through phase compensation and coherent accumulation; principal subarray summation improves the TX dimension gain and reduces computational load by merging TX signals; and weighted incoherent accumulation optimizes the detection threshold stability by reducing noise variance. The ultimate goal is to ensure that small targets are not missed in the coarse search stage, thereby improving detection sensitivity. These three gain optimization methods before target detection can be used individually or in combination, depending on the detection requirements. For example, if the target detection task is DA-CFAR, weighted incoherent accumulation can be used to optimize the SNR and noise floor stability before detection. Similarly, if the target detection task is D-CFAR, coherent gain optimization combined with principal subarray summation can be used to improve the gain of the TX dimension signal.

[0091] In a preferred embodiment, phase mismatch causes partial cancellation of the TX and RX signals during coherent summation; therefore, phase compensation is a prerequisite for coherent gain optimization. Subarray summation involves the merging of TX-dimensional signals; therefore, it relies on phase compensation to better ensure that all TX signals do not lose gain due to phase errors during coherent synthesis. The signal after subarray summation can be further used for DA-CFAR target detection to improve the detection sensitivity of the target signal. Subarray power-weighted merging involves the accumulation of signal energy; therefore, weighted incoherent gain optimization relies on the subarray summation step to better ensure the correct merging of TX-dimensional signals. The signal after weighted incoherent accumulation enters DA-CFAR target detection, which can improve the detection robustness of the target signal, make the detection threshold more stable, and reduce false positives and false negatives.

[0092] The signal is optimized through the pre-detection gain acquisition processing, or TX Order calculation + pre-detection gain acquisition processing, provided in the embodiments of this application. This ensures that the input data of the radar is optimal before performing "coarse search + fine search." In other words, the data is optimized before target detection, thereby improving the accuracy and efficiency of target detection. Specifically, pre-detection gain optimization achieves better signal gain in the coarse search stage, reducing the problem of missed detections of small targets. TX Order parsing solves the signal matching problem under multi-TX dimension MIMO schemes, ensuring that multi-channel data can be correctly processed in both coarse and fine searches.

[0093] In one exemplary instance, for target detection using the Doppler-Angle Joint Constant False Alarm Rate (DA-CFAR) algorithm, the incoherent gain from the subarray can be obtained through processing including TX dimension compensation, and any or any combination of the following: data remapping, DOA FFT processing. Combined with... Figure 5 It can include: First, the signal Sig[dop, rx, tx] used for incoherent gain calculation is used as the second input signal, and phase compensation is performed along the TX dimension.

[0094] In one embodiment, TX dimension compensation may include: performing phase compensation on the TX dimension of the second input signal to correct the phase error at the TX end, ensuring that the TX dimension signals do not experience phase mismatch leading to gain loss during coherent superposition. Here, the second input signal is the signal Sig[dop, rx, tx] used for incoherent gain calculation. The signal after TX dimension compensation... (like Figure 5 The seventh signal The calculation is as follows: Through phase compensation, the correct alignment of signals in the TX dimension was achieved.

[0095] Next, regarding the seventh signal According to the radar antenna array information (the arrangement of RX / TX antennas), the main subarray is remapped according to its array position. This means classifying and indexing the TX dimension signals to ensure the correct arrangement of the main subarray data. This ensures that the signals are correctly organized according to the MIMO structure, avoids TX dimension data corruption, and thus correctly distinguishes the first main subarray signal Sig[dop,MIMO-TRX(main)] and the first slave subarray signal Sig[dop, MIMO-TRX(slave)].

[0096] Next, the azimuth information of the main subarray is calculated. In one embodiment, this may include: performing an FFT on the first main subarray signal Sig[dop, MIMO-TRX(main)] to transform the signal from the spatial domain (antenna dimension) to the angular domain (DOA dimension), and extracting the azimuth information, such as... Figure 5 As shown, the eighth signal Sig[dop, doa] is obtained.

[0097] It should be noted that digital beamforming (DBF) can also be used to obtain azimuth information. In this embodiment, DBF processing of the main subarray is implemented using Data Remap + FFT, which simplifies the computational complexity.

[0098] Next, the power of the eighth signal Sig[dop, doa] is calculated to obtain the power information of the main subarray (e.g., Figure 5 The ninth signal in the signal is Sig_pow[dop, doa].

[0099] Then, the power of the first slave subarray signal Sig[dop, MIMO-TRX(slave)] is calculated and accumulated along the TX dimension to obtain the total power of the slave subarray (e.g., ...). Figure 5The tenth signal, Sig_pow[dop], is shown in the middle.

[0100] Finally, a weighted sum is performed on the ninth signal Sig_pow[dop, doa] and the tenth signal Sig_pow[dop] to obtain the incoherent gain from the subarray, such as... Figure 5 The eleventh signal shown is cube[dop, doa]. In one embodiment, ,in, For weighting factors. In one embodiment, the weighting factor... It can be 0.5 or other fixed values. In one embodiment, the weighting factor... It can be adaptively adjusted according to different application scenarios to obtain the optimal weighting factor for the current application scenario. For the value and specific implementation details, please refer to the following text.

[0101] In this way, DA-CFAR target detection can be performed on the eleventh signal cube[dop, doa] to obtain the final target point.

[0102] The method for obtaining the incoherent gain of a subarray provided in this application embodiment is implemented as follows: Figure 5 As shown, by using TX dimension compensation, data remapping, and FFT transformation, the computational load is optimized and the detection efficiency is improved. Moreover, the method of non-coherent gain weighted merging is adopted to stabilize the noise level, improve the gain before detection, and ensure that small targets are not missed.

[0103] It should be noted that DA-CFAR performs target detection jointly in the Doppler and azimuth dimensions, meaning that azimuth information is directly obtained during detection. Therefore, users can choose whether to use the method provided in this application embodiment for obtaining the incoherent gain of the subarray to improve the pre-detection gain. In other words, the method provided in this application embodiment for obtaining the incoherent gain of the subarray is not a mandatory step for DA-CFAR; however, implementing this method can improve the pre-detection gain, enhance detection capability, reduce the risk of missed detections, and thus optimize target detection capabilities.

[0104] In one exemplary instance, weighting factors By controlling the power ratio of the master subarray and the slave subarray, a suitable weighting factor can be selected. To maximize the SNR of the merged power spectrum and prevent the target signal from being overwhelmed by noise, embodiments of this application may further include: First, a Trade-Off objective function is constructed to maximize the signal power / noise variance. The Trade-Off objective function is shown below: ; in, The number of antennas represents the coherence gain. The number of antennas represents the incoherent gain. Indicates the energy of the target signal. Indicates noise power. This represents the weighting factor.

[0105] Then, by adjusting the Trade-Off objective function with respect to the weight factors... Find the derivative and calculate the optimal value. The value that makes the Trade-Off objective function reach its maximum value.

[0106] In this way, the best approach can be adopted. Values ​​are weighted and combined using power: Optimized The value can improve the SNR, making the target signal clearer, thereby improving the stability and reliability of DA-CFAR target detection.

[0107] In one embodiment, taking the scenarios of FMCW radar 4T4R and 8T8R with half the number of main subarrays as examples, the Trade-Off objective functions shown in Figure 6(a) and Figure 6(b) can be obtained respectively. In Figure 6(a) and Figure 6(b), the horizontal axis represents the weighting factor. The value of , with the vertical axis representing the value of the Trade-Off objective function, as shown in Figure 6(a), under the 4T4R configuration, when At this time, the Trade-Off objective function reaches its peak, as shown in Figure 6(b). Under the 8T8R configuration, when At this point, the Trade-Off objective function reaches its peak. As can be seen from Figures 6(a) and 6(b), the weighting factors... A proper selection can maximize SNR. The optimal weighting factor is determined under different antenna configurations (4T4R / 8T8R). The values ​​of are different, meaning that in different systems, optimal detection performance is achieved through adaptive adjustment of the weights of the incoherent gain. Taking 4T4R as an example, when the Trade-Off objective function reaches its peak value, The Doppler spectrum at this time is as follows Figure 7 As shown, Figure 7 The horizontal axis represents Doppler bins, indicating the Doppler spectral index, corresponding to the target's velocity information. The vertical axis represents the power spectral density or signal strength, indicating the intensity of the target's echo signal. Curve 71 shows the original Doppler spectrum. (i.e., unweighted), curve 72 is the optimized Doppler spectrum of the embodiment of this application (using... (weighted), such as Figure 7 As shown, curve 72 has a lower noise floor and a more obvious target peak, indicating that the weighting factor has been optimized according to the embodiments of this application. After weighted optimization, the detection capability was improved. Figure 7 Shown in the best At this value, the noise floor of the Doppler spectrum is optimized, the target signal is more prominent, and the target detection capability is improved.

[0108] Optional, such as Figure 7 As shown, in the step of weighted incoherent summation of the subarray and the main subarray, the weighting coefficients can also be configured to be obtained based on other Tarde-Off functions, such as defining SNR as the ratio of the signal energy to the square of the mean of the noise floor plus the variance of the noise floor.

[0109] In one exemplary instance, for target detection using the Doppler Dimensional Constant False Alarm Rate (D-CFAR) algorithm, processing may include TX compensation and phase compensation, as well as any or any combination of the following: accumulation of master and subarray signals, TX-dimensional FFT, RX-TX-dimensional optimization, and obtaining the coherence gain of the slave array. Combined with... Figure 8 It can include: First, based on the index obtained from the TX Order parsing, the TX dimension signal Sig[dop, rx-fft, code] is extracted from the fourth signal Sig_pow[dop, rx-fft, code] obtained after classifying the power spectrum signal using the DDM code at the TX end in step 2004. This extracted TX dimension signal Sig[dop, rx-fft, tx] is then used as the third input signal to compensate for the TX dimension signal. Figure 8 As shown, the twelfth signal after compensation is obtained. This ensures that the TX signal does not exhibit amplitude or phase errors during subsequent coherent processing. In one embodiment, the compensation coefficient for compensating the TX dimension signal is a compensation factor calculated using the DDM Code at the TX end during TX Order parsing, primarily used to correct the initial phase error in the TX dimension.

[0110] Next, for each TX signal that has undergone TX dimension signal compensation, i.e., the twelfth signal... Phase compensation was performed on each grid point of the RX FFT to obtain the thirteenth signal after phase compensation in the TX dimension. This aligns the information in the RX FFT dimension with the TX dimension signal, ensuring that the TX dimension signal does not experience phase mismatch during coherent accumulation, thus preventing gain loss. In one embodiment, the phase compensation term is... , ,in, This indicates the relative position of the TX antenna in the lateral direction, which is known during array deployment; Indicates the signal wavelength; and These represent the target's azimuth and elevation angles, respectively. The phase term representing the pitch coupling azimuth is separated in the RX FFT dimension, therefore, compensation is performed separately for different RX FFT bins. Figure 9 This demonstrates the physical meaning of TX phase compensation, such as... Figure 9 As shown, the left figure displays the uncompensated (unaligned) TX-dimensional signal distribution, while the right figure shows the TX-dimensional signal after TX phase compensation according to the embodiments of this application. The TX-dimensional signal is "straightened" and the phase is consistent, thus enabling subsequent coherent accumulation. Figure 9 As shown, after compensation, all TX values ​​for the target echo signal have been "straightened" into a vertical direction, thus containing only a single dimension. After accumulating the echo signals from the master and slave arrays, the slave and master array signals can be merged and subjected to FFT, as described in step 4. At this point, the coherent gain of the TX values ​​of the master and slave arrays can be fully obtained.

[0111] like Figure 8 As shown, the thirteenth signal This includes the second master subarray signal Main_sig[dop,rx-fft,tx-shift] and the second slave subarray signal Slave_sig[dop,rx-fft,tx-shift]. The second master subarray signal Main_sig[dop,rx-fft,tx-shift] is the TX dimension signal of the master subarray, which has undergone phase compensation according to the RX FFT grid points and is mainly used for calculating the coherent gain of the master subarray. The second slave subarray signal Slave_sig[dop,rx-fft,tx-shift] is the TX dimension signal of the slave subarray, which has also undergone phase compensation according to the RX FFT grid points and is mainly used for calculating the coherent gain of the slave subarray.

[0112] Then, the second principal subarray signal Main_sig[dop, rx-fft, tx-shift] obtained after compensation is summed along each TX dimension, i.e., the principal subarray is summed along the TX dimension, to obtain the gain that maximizes the energy of the principal subarray signal, such as... Figure 8 The main subarray gain signal shown is Main_sig[dop, rx-fft].

[0113] Then, the second slave array signal Slave_sig[dop, rx-fft, tx-shift] is combined with the main subarray gain signal Main_sig[dop, rx-fft], and an FFT is performed along the TX dimension to transform the TX dimension signal to the Doppler-TX domain for TX dimension signal enhancement, such as... Figure 8 As shown, the thirteenth signal, Main_Slave_sig[dop, rx-fft, tx-fft], is obtained.

[0114] Finally, for the thirteenth signal Main_Slave_sig[dop, rx-fft, tx-fft], the maximum value is found along the RX-FFT and TX-FFT dimensions to construct the three-dimensional detection cube signal as follows: Figure 8 The fourteenth signal cube[dop] shown here has high resolution, which can optimize the D-CFAR target detection performance and thus improve the target detection sensitivity. It should be noted that, in one embodiment, the operation of finding the maximum value can be omitted, and detection can be performed directly in the 3D-cube in subsequent processes. This can also be applied to DAE-CFAR processes with lower azimuth resolution.

[0115] In this way, D-CFAR target detection can be performed on the fourteenth signal cube[dop] to obtain the final target point.

[0116] In this embodiment, the target detection gain optimization in step 201 is divided into two types for different application targets. One type is for DA-CFAR used for angle detection, which proposes an incoherent gain weighted merging method to obtain the incoherent gain from the subarray, reduce the influence of noise, improve the target detection gain, and make the target signal more prominent. The other type is for D-CFAR used for velocity detection, which enhances the target signal energy and improves the SNR by obtaining the coherent gain of all RX and TX dimensions, so that small targets will not be missed, reduce false detections, and thus improve the target detection sensitivity.

[0117] This application also provides a non-transitory computer-readable storage medium storing computer-readable instructions thereon, which, when executed by a processor, cause the processor to perform the target detection processing method as described in any of the preceding claims.

[0118] This application embodiment also provides a terminal device, including a memory and a processor, wherein the memory stores the following instructions executable by the processor for performing the target detection processing method described in any of the above embodiments.

[0119] The following simulation analysis uses 8T8R as an example. In this embodiment, it is assumed that the RX end adopts a horizontal array, while the TX end is not constrained, and the master sub-array and slave sub-array each account for half of the total number of channels.

[0120] Figure 10 This is a comparative diagram illustrating the optimization of target detection performance using different compensation schemes in the embodiments of this application, such as... Figure 10 As shown, the horizontal axis represents SNR, and the vertical axis represents the target detection probability. Curve 101 indicates that the complete target detection processing method provided in this application embodiment has been compensated, that is, gain and phase compensation has been performed on both the TX and RX signals. Curve 102 indicates that no gain and phase compensation has been performed on the RX end, and only compensation on the TX end is considered. Curve 103 indicates that no gain and phase compensation has been performed on the TX end, and only compensation on the RX end is considered. Curve 104 indicates the traditional SISO scheme that does not perform any compensation at all, that is, no adjustment is made to the gain and phase of the TX or RX end. Figure 10 As shown, curve 101 is the highest, indicating that simultaneously compensating for the gain and phase of both the TX and RX ends yields the best detection performance, effectively improving detection performance at low SNR. Curves 102 and 103 show that compensating only the TX or RX end still results in performance loss compared to complete compensation. Curve 104 is the lowest, indicating that not compensating leads to a significant decrease in detection capability, especially a substantial reduction in detection probability at low SNR. In other words, the target detection processing method provided in the embodiments of this application optimizes target detection capability and improves detection performance at low signal-to-noise ratios.

[0121] Figure 11 This is a comparative diagram of different TX orders used in target detection in the embodiments of this application, such as... Figure 11 As shown, the horizontal axis represents SNR, and the vertical axis represents the target detection probability. Curve 111 represents the TX Order parsing method in the target detection processing method provided in this application embodiment, and curve 112 represents the traditional SISO scheme. From Figure 12 The simulation results show that in the low SNR region (-10 dB ~ 0 dB), the TX Order parsing method in this application embodiment has a significant performance improvement and the detection probability is much higher than that of the SISO scheme. In other words, the TX Order parsing scheme provided in this application embodiment can effectively improve the correctness of signal arrangement, making the gain calculation before target detection more accurate, thereby optimizing the target detection capability.

[0122] Figure 12 This is a schematic diagram comparing the performance of different target detection gain schemes in the embodiments of this application, such as... Figure 12As shown, the horizontal axis represents SNR, and the vertical axis represents the target detection probability. Curve 121 represents the TX Order parsing scheme in this embodiment, using the method of obtaining the coherent gain from the subarray; curve 122 represents the TX Order parsing scheme in this embodiment, using the method of obtaining the incoherent gain from the subarray; curve 123 represents the TX Order parsing scheme in this embodiment, using only the coherent gain of the main subarray; curve 124 represents the TX Order parsing scheme provided in this embodiment, directly accumulating the coherent gain of the main subarray and the incoherent gain of the slave subarray; and curve 125 represents the use of the traditional SISO scheme. Figure 12 As can be seen, curve 121 exhibits the best performance, maintaining a high detection probability even in the low SNR region. Curve 122 also significantly improves detection performance, but is slightly lower than curve 121. Curves 123 and 124, compared to curves 121 and 122, show lower detection probabilities at low SNR, indicating that simple accumulation is less effective than a reasonable gain optimization scheme. Figure 12 The comparison shows that using the TX Order parsing scheme in the embodiments of this application, combined with pre-detection gain optimization (especially obtaining the coherent gain from the subarray), can significantly improve target detection performance.

[0123] This application also provides a target detection method, including: performing target detection based on a target echo signal provided by any one of the embodiments of this application.

[0124] This application also provides a target detection processing method, applied in a MIMO FMCW radar, which may include: After receiving the radar echo signal, the received radar echo data is processed by FFT to obtain the Range-Doppler-Channel Map (RDM) spectrum. In other words, by performing operations such as ADC (sampling), Range FFT, and Doppler FFT on the radar echo signal, a data cube containing parameters such as range, Doppler, and channel can be obtained. Subsequently, a Tx Order calculation based on fused Rx coherent gain can be performed, that is, Tx Order calculation is performed based on the RDM spectrum followed by CFAR processing.

[0125] Compared to the coarse search with incoherent merging performed directly after FFT, followed by a fine search including Tx Order calculation and coherent merging, this embodiment of the application advances the Tx Order calculation with fused Rx coherent gain to after FFT. This not only advances the order but also innovatively improves the method. Specifically, by using all data as input, rather than just candidate point data, and by leveraging pre-defined array features such as RX antennas being aligned in a straight line (e.g., a horizontal line), FFT (or DBF) can be performed on the RX channel dimension first. This allows the coherent gain of the RX dimension to be obtained at the beginning of Tx Order calculation, significantly improving the success rate and resulting in a global, highly reliable Tx Order calculation result.

[0126] In one exemplary instance, the antennas in the antenna array of the MIMO FMCW radar are arranged according to a preset rule; before solving the transmission order Tx Order based on the RDM spectrum, the following steps are also included: The receiving antenna Rx dimension is compensated based on the antenna arrangement rules.

[0127] In this embodiment, since the receiving antennas of a MIMO radar are typically arranged along the azimuth direction, amplitude imbalance, phase imbalance, or noise level differences may exist between different channels. Based on the antenna arrangement rules, Rx-dimensional amplitude and phase compensation is performed on the RDM to obtain the normalized signal; the compensation factor can be expressed as... This compensation ensures that the amplitude and phase of all Rx channels remain consistent, providing channel consistency for subsequent processing.

[0128] In one exemplary instance, after Tx Order calculation and before CFAR processing, the process may further include: compensating for the transmit antenna Tx dimension based on the antenna arrangement rules.

[0129] In this embodiment, after the Tx Order is calculated, since the amplitude and phase may still be inconsistent in the power amplifier, feeder and other components of the transmitting antenna, this embodiment further compensates for the Tx dimension.

[0130] In one exemplary instance, the receiving antennas Rx in the antenna array of a MIMO FMCW radar are arranged side-by-side along the azimuth direction, in the method: After Rx amplitude and phase compensation of the RDM spectrum, Tx order is calculated based on Rx coherent gain, and Tx phase is supplemented based on angle information before CFAR processing is performed.

[0131] In one exemplary instance, CFAR processing includes D-CFAR and DA-CFAR; it may also include: determining the task type of CFAR processing; if the task type is D-CFAR, a fully coherent gain path is used for processing; if the task type is DA-CFAR, a weighted incoherent merging path is used for processing. In this embodiment, different CFAR paths are selected according to the detection task type. If the detection task is D-CFAR, a fully coherent gain path is used, that is, coherent accumulation is performed in the Tx and Rx dimensions to maximize the target energy. If the detection task is DA-CFAR, a weighted incoherent merging path is used, that is, weighted incoherent accumulation is performed on the power of the master subarray and the slave subarray to optimize the noise floor variance.

[0132] Compared to the sequential operation of coarse search and fine search, in this embodiment, the corresponding processing path can be automatically selected according to the final detection task type. For example, by determining whether it is a D-CFAR detection task based on velocity information or a DA-CAFR detection task based on velocity-angle information, the fully coherent gain (as in D-CFAR) or weighted merging (as in DA-CFAR) method can be selected for the target detection task.

[0133] It should be noted that only two types of detection tasks are listed in the embodiments of this application. Those skilled in the art can also set a third type of detection task based on the needs of the scenario, and the corresponding judgment information also needs to be set accordingly. That is, as long as there are two or more types of detection tasks, adaptive selection can be achieved.

[0134] Optionally, the fully coherent gain for D-CFAR described above can be achieved by using the calculated Tx order and angle information to perform phase compensation on the TX channel, followed by operations such as merging and FFT, thereby obtaining the coherent gain of all TX and RX channels with lower computational cost. Meanwhile, the weighted merging scheme for DA-CFAR described above can obtain the azimuth spectrum by performing DBF on the principal subarray, then performing incoherent accumulation on the slave subarray, and finally merging the two using an optimized weighting coefficient k, ultimately achieving better detection gain while suppressing noise fluctuations. In one exemplary instance, the weighting coefficient k in the weighted incoherent merging path ranges from 0.7 to 0.9. In one embodiment, under a 4T4R configuration, the weighting coefficient k is 0.738. In another embodiment, under an 8T8R configuration, the weighting coefficient k is k=0.88. Therefore, for different antenna sizes, the optimal value for the weighting coefficient k can be within the range of 0.7–0.9.

[0135] In summary, the above embodiments solve the performance degradation problem caused by missed detections and statistical mismatches in the serial process by changing the timing and method of Tx Order calculation and by using adaptive path selection to replace the fixed serial process.

[0136] Figure 13 This is a schematic diagram of the target detection process of a traditional MIMO FMCW radar, such as... Figure 13 As shown, the process generally includes: performing Range FFT and Doppler FFT on the received ADC data to obtain a multi-channel RDM spectrum; judging candidate points on the RDM spectrum; if the conditions are met, proceeding to the fine search process, including Tx Order solution (SISO) and coherent combining (DBF), and finally completing the detection and angle measurement; if the conditions are not met, the data is discarded directly. Figure 13 In the traditional scheme shown, the coherent and incoherent merging processes are executed serially without channel consistency compensation, resulting in low gain and phase mismatch issues. Moreover, the Tx Order calculation relies on post-processing, and the front end cannot be optimized in advance, which limits the detection performance.

[0137] Figure 14 This is a schematic diagram of the integrated target detection process provided in the embodiments of this application, such as... Figure 14 As shown, the process generally includes: performing Range / Doppler FFT on the ADC data; performing Rx amplitude and phase compensation (global consistency) based on antenna arrangement rules; performing Tx order calculation on the compensated data (using Rx coherent gain to improve resolution accuracy); subsequently performing Tx dimension phase compensation (correction based on angle information) to ensure Tx channel consistency; selecting different detection paths according to the detection task type: if it is D-CFAR, entering the fully coherent gain path, maximizing signal energy through coherent accumulation of Tx and Rx dimensions; if it is DA-CFAR, entering the weighted incoherent merging path, optimizing noise floor stability through weighted accumulation of power between the master subarray and slave subarray. Figure 14 The proposed solution integrates Rx compensation, Tx order calculation, and Tx compensation before detection, forming a unified process that avoids the low gain and mismatch issues present in the candidate point stage of traditional solutions. Furthermore, it supports flexible selection of coherent / incoherent paths based on the detection task, improving the system's detection sensitivity and robustness. This embodiment overcomes the bottlenecks of traditional solutions and improves the accuracy and stability of MIMO FMCW radar target detection.

[0138] It should be noted that the target detection method in this application embodiment has an independent noise floor and can be extended to scenarios such as sidelobe suppression.

[0139] This application also provides a signal compensation method applicable to signal processing of MIMO FMCW radar. The method may include: performing FFT processing on received radar echo data to obtain an RDM spectrum; compensating for non-ideal amplitude and phase errors between each receiving channel based on the RDM spectrum to ensure consistency between each receiving channel; and / or, compensating for non-ideal amplitude and phase errors between each transmitting channel after Tx Order calculation to ensure consistency between each transmitting channel; and / or, performing angle-based ideal phase compensation for the transmitting channel after Tx Order calculation and before coherent combining.

[0140] Optionally, in the antenna array of the MIMO FMCW radar, the receiving antennas Rx are distributed along the same straight line in the azimuth dimension. The method may also include: taking advantage of the array feature that the receiving antennas Rx are distributed along the same straight line in the azimuth dimension, FFT or DBF can be performed on the Rx channel dimension first, and then Tx Order calculation can be continued.

[0141] Optionally, the aforementioned RX channel compensation can be initiated before TX order calculation, such as after completing Range FFT and Doppler FFT and before any channel-dimensional processing, to compensate for amplitude and / or phase deviations between the various receive channels (RX). This is because: for amplitude deviation, due to manufacturing tolerances and temperature drift of RF front-end components (such as amplifiers, mixers, and filters), the gain between different RX channels may be inconsistent; for example, some channels may receive stronger signals than others. For phase deviation, in addition to the ideal phase difference caused by the antenna position, the trace length, trace delay, and the chip itself can introduce non-ideal phase errors. In other words, through the above compensation operation, it can be ensured that all RX channels are on a level playing field, thus making subsequent coherent processing (such as RX-dimensional FFT and DBF) effective. Without compensation, these non-ideal errors may distort the beam shape, leading to inaccurate angle estimation and decreased gain. The compensation coefficients here can be obtained through initial calibration.

[0142] For TX channel compensation, it can be started after TX order calculation and before final coherent merging, such as before fine search in a cascaded process or before the fully coherent gain path (i.e., D-CFAR) in an integrated process, to compensate for amplitude and / or phase deviations between the various transmit channels (TX). The purpose and reason for this compensation are similar to those for RX channel compensation, to ensure the consistency of the TX channels.

[0143] For angle-based aroma compensation, it can be initiated after TX order calculation and before coherent combining to compensate for the phase difference caused by the different physical positions of the TX antennas. Typical DBF is "search-based": assuming an angle θ, it calculates the phase difference that all antennas should have at that assumed angle, then compensates it back, and checks if the output energy is maximized. In this embodiment, it is "computational," meaning that the approximate angular direction θ_est of a target has been estimated using the result of the previous RX-dimensional FFT. Then, based directly on this θ_est and the known specific position of each TX antenna, the phase delay that each TX channel should theoretically have is calculated and compensated for. This achieves dimensionality reduction; after compensation, the signals of all TX channels are "aligned" in phase, as if all TX antennas were located in the same virtual position. Simultaneously, it simplifies the calculation; after alignment, the data from these TX channels can be simply added (coherent combining) to obtain the TX-dimensional gain without the need for complex beamforming searches in the TX dimension. This operation reduces the computational complexity from O(N) to O(N). 2 The value was reduced to O(N).

[0144] In this embodiment, the signal compensation method first performs RX compensation to eliminate hardware errors, preparing for high-quality signal processing. Then, TX order calculation is performed (e.g., using RX coherent gain). After calculation, TX compensation is performed (which may include non-ideal error compensation and / or angle-based ideal phase compensation) to prepare for final coherent merging. Finally, a merging operation is performed to obtain the gain before detection. This staged compensation strategy provides a crucial guarantee for achieving high-performance detection with low computational complexity.

[0145] It should be noted that the signal compensation method in the embodiments of this application can be applied to the target detection method in the embodiments of this application, and the same or common parts can be borrowed or integrated from each other based on the understanding of those skilled in the art.

[0146] This application also provides an electromagnetic wave sensor, including: a signal transmitting module configured to transmit electromagnetic waves for target detection; a signal receiving module configured to receive echoes formed by reflection and / or scattering of the electromagnetic waves; and a processing module configured to perform signal and data processing on the echoes to achieve operations such as target judgment, location, and / or identification; wherein the processing module is further configured to implement any of the target detection methods provided in this application.

[0147] This application also provides an integrated circuit, which may include a radio frequency (RF) module, an analog signal processing module, and a digital signal processing module connected in sequence. The RF module is used to generate RF transmission signals and receive echo signals. The analog signal processing module is used to down-convert the echo signals to obtain intermediate frequency (IF) signals. The digital processing module is used to perform analog-to-digital conversion on the IF signals to obtain digital signals. The integrated circuit processes the digital signals based on the target detection method in this application to achieve target detection. For example, the integrated circuit may be a millimeter-wave radar chip (chip or die). The digital processing module may include sub-units such as a black box (BB) unit and an MCU unit, and each sub-unit can be configured to execute the corresponding steps in the target detection method described in the above embodiments.

[0148] In some optional embodiments, the integrated circuit may be an antenna-in-package (AiP) chip structure, an antenna-on-package (AoP) chip structure, an antenna-on-chip (AoC) chip structure, or a radiator-through-package (RoP) chip structure. Optionally, the RoP chip structure may involve setting a radiator on the chip package and surrounding the radiator with solder balls to form an air waveguide structure. That is, the radio frequency (RF) signal generated by the chip can be transmitted to an external antenna through the aforementioned radiator, the air cavity waveguide structure surrounded by the solder balls, and the air waveguide built into the PCB board, to radiate towards the target area.

[0149] According to some other embodiments of this application, an electromagnetic wave sensor is also proposed. This electromagnetic wave sensor may include an antenna and an integrated circuit as described above. The integrated circuit is electrically connected to the antenna and is used to transmit and receive electromagnetic wave signals. For example, the electromagnetic wave sensor may include: a carrier, an integrated circuit as described in any of the above embodiments, and an antenna, etc. The integrated circuit may be disposed on the carrier; the antenna may be disposed on the carrier, or integrated with the integrated circuit as a single device disposed on the carrier (i.e., the antenna may be an antenna disposed in an AiP, AoP, or AoC structure); wherein the integrated circuit is connected to the antenna (i.e., the sensing chip or integrated circuit does not integrate an antenna, such as a conventional SoC), and is used to transmit and receive electromagnetic wave signals. The carrier may be a printed circuit board (PCB), and the corresponding transmission line may be a PCB trace.

[0150] This application provides a device that may include: a device body; and an electromagnetic wave sensor as described above disposed on the device body; wherein the electromagnetic wave sensor is used for target detection and / or communication to provide reference information for the operation of the device body.

[0151] This application also provides a terminal device, which can be manifested in the form of a general computing device. The components of the terminal device may include, but are not limited to: at least one processing unit, at least one storage unit, a bus connecting different system components (including the storage unit and the processing unit), a display unit, etc. The storage unit stores program code, which can be executed by the processing unit to cause the processing unit to perform the methods described in this specification according to the various exemplary embodiments of this application. The storage unit may include a readable medium in the form of volatile storage units, such as random access memory (RAM) and / or cache memory units, and may further include read-only memory units (ROM).

[0152] The storage unit may also include a program / utility having a set (at least one) of program modules, including but not limited to: an operating system, one or more applications, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0153] A bus can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus that uses any of the various bus structures.

[0154] The terminal device can also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable users to interact with the terminal device, and / or any device that enables the terminal device to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication can be performed via input / output (I / O) interfaces. Furthermore, the terminal device can communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter. The network adapter can communicate with other modules of the terminal device via a bus. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with the terminal device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0155] For example, the terminal device in this application embodiment may further include: a device body; and an electromagnetic wave sensor disposed on the device body as described in any of the above embodiments; wherein the electromagnetic wave sensor can be used to realize functions such as target detection and / or wireless communication.

[0156] Specifically, based on the above embodiments, in one optional embodiment of this application, the electromagnetic wave sensor can be disposed outside the device body or inside the device body. In other optional embodiments of this application, the electromagnetic wave sensor can be partially disposed inside the device body and partially disposed outside the device body. This application does not limit the specific implementation; it can be determined according to the circumstances.

[0157] In an optional embodiment, the aforementioned device body can be a component or product applied in fields such as smart cities, smart homes, transportation, smart homes, consumer electronics, security monitoring, industrial automation, in-cabin detection (such as smart cockpits), medical devices, and healthcare. For example, the device body can be intelligent transportation equipment (such as automobiles, bicycles, motorcycles, ships, subways, trains, etc.), security equipment (such as cameras), liquid level / flow rate detection equipment, smart wearable devices (such as wristbands, glasses, etc.), smart home devices (such as robot vacuum cleaners, door locks, televisions, air conditioners, smart lights, etc.), various communication devices (such as mobile phones, tablets, etc.), as well as devices such as barriers, intelligent traffic lights, intelligent signs, traffic cameras, and various industrial robotic arms (or robots). It can also be various instruments for detecting vital signs parameters and various devices equipped with such instruments, such as in-cabin vital sign detection in automobiles, indoor personnel monitoring, smart medical devices, and consumer terminal devices.

[0158] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. The technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, or network device, etc.) to execute the methods described above according to the embodiments of this application.

[0159] Software products may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0160] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0161] Program code for performing the operations of this application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0162] Although the embodiments disclosed in this application are as described above, the content described is merely for the purpose of understanding this application and is not intended to limit this application. Any person skilled in the art to which this application pertains may make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in this application; however, the scope of patent protection of this application shall still be determined by the scope defined in the appended claims.

Claims

1. A target detection processing method, characterized in that, Before object detection, the following are included: Perform one or any combination of the following gain optimization processes before target detection: coherent gain optimization, principal subarray summation, and incoherent gain weighted merging to improve the signal-to-noise ratio (SNR) of the target detection signal; The coherent gain optimization is used to align the signals of the transmit antenna (TX dimension) and the receive antenna (RX dimension) through phase compensation; the master subarray summation is used to coherently combine the signals of the master subarray; and the incoherent gain weighted combining is used to obtain the incoherent gain of the slave subarray by calculating the signal power of the slave subarray and weighting it with the power of the master subarray.

2. The target detection processing method according to claim 1, further comprising, before the target detection gain optimization processing: The coherent gain of the RX dimension is extracted from the received target echo signal containing multiple TX dimension data, and the arrangement order of the transmit antenna dimension signals is determined by combining the encoded information after differential delay modulation (DDM) to ensure that the TX signals are correctly aligned.

3. The target detection processing method according to claim 2, wherein, The coherent gain of the receiving antenna dimension is extracted by frequency domain transformation at the receiving end.

4. The target detection processing method according to claim 2, wherein, The process of determining the arrangement order of the signals in the transmitting antenna dimension includes: For the first input signal, a Fast Fourier Transform (FFT) is performed in the chirp dimension to obtain Doppler spectrum information. The first input signal is the target echo signal received at the RX end, which contains data in multiple TX dimensions. Perform RX FFT processing on all or part of the RX signals of the receiving antenna, calculate the coherence gain between RX signals, and arrange the antenna array at the RX end on the same horizontal line. The signal power of the third signal after RX FFT processing is calculated and the data is split according to the DDM Code information after DDM modulation to obtain the processed fourth signal; The fifth signal is obtained by accumulating and synthesizing the data from each DDM Code dimension; Perform a maximum value search on the fifth signal to obtain the maximum value index in the RX dimension. The maximum value index represents the position of the strongest signal in the RX FFT dimension. Extract the corresponding TX dimension power data from the fourth signal by using the maximum value index on the RX dimension; The extracted TX dimension power data and the DDM modulated coded information are convolved and the maximum value is calculated to obtain the arrangement order of the transmit antenna dimension signals.

5. The target detection processing method according to claim 4 further includes: The imbalance of each RX signal in the first input signal is compensated to obtain the first signal; The FFT in the Chirp dimension refers to performing an FFT on the first signal in the Chirp dimension.

6. The target detection processing method according to claim 5, wherein, The step of compensating for the imbalance of each RX signal in the first input signal to obtain the first signal includes: Calculate the average signal amplitude for each RX channel to estimate the gain of each RX channel; Calculate the noise variance for each RX channel to estimate the noise level for each RX channel; The RX signal compensation factor is obtained based on the gain and noise level of each RX channel. And use the obtained RX signal compensation factor The first signal is obtained by compensating the target echo signal.

7. The target detection processing method according to claim 4 or 5, wherein, The arrangement order of the transmitted antenna dimension signals is used for coherent gain D-CFAR calculation; or, for incoherent gain DA-CFAR calculation.

8. The target detection processing method according to claim 5, wherein, For target detection using the Doppler-azimuth joint constant false alarm rate (DCFAR) algorithm, the incoherent gain of the subarray is obtained through processing including TX dimension compensation and any one or any combination of the following: data remapping and DOA FFT processing.

9. The target detection processing method according to claim 8, wherein, The step of obtaining the incoherent gain of the subarray includes: The signal Sig[dop, rx, tx] used for incoherent gain calculation is used as the second input signal. Phase compensation is performed along the TX dimension to obtain the seventh signal. The signal used for incoherent gain calculation is the TX dimension data in the Doppler spectrum information obtained after rearranging the first signal after imbalance compensation according to the arrangement order of the transmit antenna dimension signal and performing FFT in the Chirp dimension. For the seventh signal, the main subarray is remapped according to its array position based on the radar antenna array information; Perform an FFT on the first main subarray signal to extract the azimuth information and obtain the eighth signal; Calculate the power of the eighth signal to obtain the power information of the main subarray; The power of the first subarray signal is calculated and accumulated along the TX dimension to obtain the total power of the subarray. The incoherent gain of the slave array is obtained by weighted summation of the power information of the master subarray and the total power of the slave subarray.

10. The target detection processing method according to claim 9, wherein, The weighting factors used in the weighted summation Used to control the power ratio of the first master subarray and the first slave subarray; The weighting factor It can be a fixed value; or it can be adjusted adaptively according to different application scenarios.

11. The target detection processing method according to claim 10, further comprising: By constructing a Trade-Off objective function that maximizes the signal power / noise variance, the Trade-Off objective function is: ;in, The number of antennas represents the coherence gain. The number of antennas represents the incoherent gain. Indicates the energy of the target signal. Indicates noise power. This represents the weighting factor; By applying the Trade-Off objective function with respect to the weighting factors Find the derivative and calculate the optimal value. The value that makes the Trade-Off objective function reach its maximum value.

12. The target detection processing method according to claim 4 or 5, for target detection using the Doppler dimension constant false alarm rate (CFAR) algorithm, the coherence gain of the slave array is obtained by processing including TX compensation and phase compensation, and any or any combination of the following: master-slave array signal accumulation, TX dimension FFT, and RX-TX dimension optimization.

13. The target detection processing method according to claim 12, wherein, The step of obtaining the coherent gain of the subarray includes: Based on the arrangement order of the transmitted antenna dimension signals, the TX dimension signal is extracted from the fourth signal and used as the third input signal to compensate the TX dimension signal; For each TX signal that has been compensated for in the TX dimension, phase compensation is performed according to each grid point of the RX FFT to obtain the TX dimension phase compensation. The compensated signals include the second master subarray signal and the second slave subarray signal. The second master subarray signal is the TX dimension signal of the master subarray, which has been phase compensated according to each grid point of the RX FFT. The second slave subarray signal is the TX dimension signal of the slave subarray, which has been phase compensated according to each grid point of the RX FFT. The TX dimensions of the compensated second master array signal are summed to obtain the gain that maximizes the energy of the master array signal. The second subarray signal is combined with the main subarray gain signal and FFT is performed along the TX dimension to convert the TX dimension signal to the Doppler-antenna domain for signal enhancement in the TX dimension to obtain the thirteenth signal.

14. The target detection processing method according to claim 13, further comprising: For the thirteenth signal, find the maximum value along the RX-FFT and TX-FFT dimensions to construct a three-dimensional detection cube signal.

15. A target detection processing method, characterized in that, When applied to MIMO FMCW radar, the method includes: The received radar echo data is processed using FFT to obtain the RDM spectrum; and After calculating the emission order Tx Order based on the RDM spectrum, constant false alarm rate (CFAR) processing is performed.

16. The target detection processing method according to claim 15, wherein the antennas in the antenna array of the MIMO FMCW radar are arranged according to a preset rule; before solving the transmission order Tx Order based on the RDM spectrum, the method further includes: The receiving antenna Rx dimension is compensated based on the antenna arrangement rules.

17. The target detection processing method according to claim 16, further comprising, after the Tx Order calculation and before the CFAR processing: The transmitting antenna Tx dimension is compensated based on the antenna arrangement rules.

18. The target detection processing method according to claim 16, wherein, In the antenna array of the MIMO FMCW radar, the receiving antennas Rx are arranged side by side along the azimuth direction. In the method: After performing Rx amplitude and phase compensation on the RDM spectrum, the Tx order is calculated based on the Rx coherence gain, and then the Tx phase is supplemented based on the angle information before the CFAR processing is performed.

19. The target detection processing method according to any one of claims 15-18, wherein the CFAR processing includes D-CFAR and DA-CFAR; the method further includes: Determine the task type processed by the CFAR; The task type is D-CFAR, which is processed using a fully coherent gain path. The task type is DA-CFAR, which is processed using a weighted incoherent merging path.

20. The target detection processing method according to claim 19, wherein, The weighting coefficient k in the weighted incoherent merging path ranges from 0.7 to 0.

9.

21. A method for signal compensation, wherein, The method, applied to signal processing in MIMO FMCW radar, includes: The received radar echo data is processed by FFT to obtain the RDM spectrum; Based on the RDM spectrum, compensation is made for non-ideal amplitude and phase errors between each receiving channel to ensure consistency between the receiving channels; and / or, After Tx Order calculation, compensation is performed to address non-ideal amplitude and phase errors between each transmission channel to ensure consistency among them; and / or, After Tx Order calculation and before coherent merging, angle-based ideal phase compensation is performed for the transmit channel.

22. The target detection processing method according to claim 19, wherein, In the antenna array of the MIMO FMCW radar, the receiving antennas Rx are distributed along the same straight line in the azimuth dimension, and the method further includes: Taking advantage of the fact that the receiving antennas RX are distributed along the same straight line in the azimuth dimension, we can first perform FFT or DBF on the RX channel dimension and then continue to calculate the Tx Order.

23. A terminal device, comprising a memory and a processor, wherein, The memory stores the following instructions executable by a processor for performing the steps of the method according to any one of claims 1-16 or 17-23.

Images