Signal processing method and system

Abstract

This application provides a signal processing method and system, relating to signal processing technology. The method includes: determining a first echo signal and a second echo signal based on the echo signal after a mixed signal is reflected from a target; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal; for each velocity assumption, determining the fine range estimation result and peak energy under each velocity assumption based on compensating the second echo signal and synthesizing a one-dimensional high-resolution range image; each velocity assumption is determined based on the system's velocity measurement requirements and the coarse estimation result obtained by processing the first echo signal; based on the deviation between the fine range estimation result and the coarse estimation result, and the peak energy, determining the target's range estimate and / or velocity estimate from each velocity assumption. This application enables long-range high-resolution detection while effectively improving target estimation accuracy.

Description

Technical Field

[0001] This application relates to signal processing technology, and more particularly to a signal processing method and system. Background Technology

[0002] Millimeter-wave radar has become a core sensor for intelligent driving perception due to its advantages such as all-weather operation, anti-interference, ease of implementation, and strong real-time performance. It is widely used in scenarios such as ranging and speed measurement and obstacle recognition. Among them, the linear frequency modulated continuous wave system has become the mainstream application system for vehicle-mounted millimeter-wave radar due to its ease of engineering implementation and stable performance.

[0003] Currently, in order to improve the range resolution capability of radar, known technologies often use frequency-stepped linear frequency modulated continuous wave signals as the transmitted waveform. High range resolution of the target is achieved by using the bandwidth synthesis characteristics of the stepped frequency signal. Then, Fourier transform, peak detection and parameter calculation are performed on the echo signal to complete the acquisition of range and velocity information of the detected target.

[0004] However, in practical applications, there is usually relative motion between the radar and the target, which can easily lead to a decrease in the accuracy of target estimation. Summary of the Invention

[0005] This application provides a signal processing method and system to reduce signal interference caused by the relative motion of the target, thereby improving the target estimation accuracy.

[0006] In a first aspect, this application provides a signal processing method, the method comprising:

[0007] The first echo signal and the second echo signal are determined based on the echo signal after the mixed signal is reflected by the target; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal.

[0008] For each velocity assumption, based on compensating the second echo signal and synthesizing a one-dimensional high-resolution range image, the precise range estimation result and peak energy under each velocity assumption are determined; each velocity assumption is determined based on the system velocity measurement requirements and the coarse estimation result obtained by processing the first echo signal.

[0009] Based on the deviation between the fine distance estimation result and the coarse distance estimation result, and the peak energy, the distance estimate and / or velocity estimate of the target are determined from each of the velocity assumptions.

[0010] In one possible implementation, determining the precise distance estimate and peak energy under each of the velocity assumptions includes:

[0011] Based on constant false alarm rate detection of the one-dimensional high-resolution range image under each of the aforementioned velocity assumptions, the accurate distance estimation result under each of the aforementioned velocity assumptions is determined;

[0012] The peak energy is extracted based on the one-dimensional high-resolution range image under each of the aforementioned velocity assumptions.

[0013] In one possible implementation, the compensation of the second echo signal and the synthesis of a one-dimensional high-resolution range image includes:

[0014] Based on the target coarse range estimate image corresponding to the second echo signal, each step linear frequency modulated continuous wave signal within the fine search range is extracted; the fine search range is determined based on the coarse range estimate result.

[0015] For any of the velocity assumptions, a two-dimensional high-resolution range image is obtained by performing phase compensation under the velocity assumptions on each of the step linear frequency modulated continuous wave signals.

[0016] The one-dimensional high-resolution range image is synthesized by extracting and stitching the two-dimensional high-resolution range image.

[0017] In one possible implementation, obtaining a two-dimensional high-resolution range image based on phase compensation under the velocity assumption for each of the said stepped linear frequency modulated continuous wave signals includes:

[0018] Based on the speed assumption, each of the step linear frequency modulated continuous wave signals is subjected to phase compensation processing to obtain the compensated step linear frequency modulated continuous wave signal.

[0019] Based on the windowing and inverse Fourier transform processing of the compensated step linear frequency modulated continuous wave signal, two-dimensional high-resolution range images corresponding to different velocity assumptions are generated.

[0020] And / or, extracting and stitching the two-dimensional high-resolution range image to synthesize the one-dimensional high-resolution range image, including:

[0021] Based on the same-distance discard method, the two-dimensional high-resolution range image under the velocity assumption is extracted to obtain the extracted range image;

[0022] The one-dimensional high-resolution range image is obtained by stitching the extracted range image along the frequency step dimension.

[0023] And / or, based on constant false alarm rate detection of the one-dimensional high-resolution range image under each of the aforementioned velocity assumptions, determine the precise distance estimation result under each of the aforementioned velocity assumptions, including:

[0024] Based on constant false alarm rate detection of the one-dimensional high-resolution range image under each of the aforementioned velocity assumptions, the detection peak value under each of the aforementioned velocity assumptions is obtained;

[0025] The distance corresponding to the detection peak under each speed assumption is determined as the distance estimation result under each speed assumption.

[0026] In one possible implementation, determining the distance estimate and / or velocity estimate of the target from each of the velocity assumptions based on the deviation between the fine distance estimate and the coarse distance estimate, and the peak energy, includes:

[0027] Based on the deviation and coarse range resolution under each of the velocity assumptions, candidate velocity assumptions are determined from each of the velocity assumptions; the candidate velocity assumptions are those whose deviation is less than the coarse range resolution.

[0028] Based on the peak energy, the distance estimate and velocity estimate of the target are determined from the candidate velocity assumptions.

[0029] In one possible implementation, determining the distance estimate and velocity estimate of the target from the candidate velocity assumptions based on the peak energy includes:

[0030] Based on the distance estimation result and velocity hypothesis corresponding to the velocity hypothesis with the largest peak energy among the candidate velocity hypotheses, the distance estimate and velocity estimate of the target are determined.

[0031] In one possible implementation, the speed assumptions are determined based on the system speed measurement requirements and the coarse estimation results obtained from processing the first echo signal, including:

[0032] Determine the corresponding speed expansion coefficient based on the system's speed measurement requirements;

[0033] Based on the speed expansion coefficient, the coarse speed estimation result in the coarse estimation result is extended over a range to generate multiple sets of different speed assumptions.

[0034] In one possible implementation, the method further includes:

[0035] Based on the sampling sequence or intermediate frequency sampling sequence corresponding to each linear frequency modulated continuous wave signal, a coarse estimated range profile corresponding to the target is obtained.

[0036] Based on the coarsely estimated range profile corresponding to the first echo signal, or the signal detection matrix constructed along the frequency modulation signal dimension from the coarsely estimated range profile, the coarse estimation result corresponding to the first echo signal is obtained.

[0037] In one possible implementation, the number of conventional linear frequency modulated continuous wave signals in the mixed signal is greater than the number of frequency-stepped linear frequency modulated continuous wave signals;

[0038] And / or, the method further includes:

[0039] The mixed signal is generated based on a preset method, which includes a time-division switching method and / or an amplitude mixing method.

[0040] The mixed signal is transmitted, and the echo signal reflected by the target is received.

[0041] Secondly, this application provides a signal processing apparatus, the apparatus comprising:

[0042] The determination module is used to determine the first echo signal and the second echo signal based on the echo signal after the mixed signal is reflected by the target; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal.

[0043] The compensation module is used to compensate the second echo signal based on the coarse estimation result of the first echo signal and the system speed measurement requirements, so as to obtain the fine estimation result of the target distance;

[0044] The determination module is also used to determine the distance estimate and / or velocity estimate of the target based on the deviation between the fine distance estimate and the corresponding coarse distance estimate.

[0045] Thirdly, this application provides an electronic device, including at least one processor and a memory communicatively connected to the processor;

[0046] The memory stores computer-executed instructions;

[0047] The processor executes computer execution instructions stored in the memory to implement the method as described in any of the first aspects.

[0048] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any of the first aspects.

[0049] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method as described in any of the first aspects.

[0050] Sixthly, this application provides a signal processing system, the system including a waveform generator, a radio frequency transmitter and receiver, and electronic equipment;

[0051] The waveform transmitter is used to generate a mixed signal; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal.

[0052] The radio frequency transmitter and receiver are used to transmit the mixed signal and receive the echo signal reflected by the target;

[0053] The electronic device is used to: determine a first echo signal and a second echo signal based on the echo signal after the mixed signal is reflected by the target; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal;

[0054] Based on the coarse estimation result of the first echo signal and the system speed measurement requirements, the second echo signal is compensated to obtain the fine estimation result of the target distance;

[0055] Based on the deviation between the fine distance estimation result and the corresponding coarse estimation result, the distance estimate and / or velocity estimate of the target are determined.

[0056] This application provides a signal processing method and system. The method proposes to first determine a first echo signal and a second echo signal based on the echo signal after the mixed signal is reflected by the target. Then, based on the compensation of the second echo signal and the synthesis of a one-dimensional high-resolution range image, the precise range estimation result and peak energy under each velocity assumption are determined. Furthermore, based on the deviation between the precise range estimation result and the coarse range estimation result, as well as the peak energy, the target's range estimate and / or velocity estimate are determined from each velocity assumption. The mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal. In this application, conventional linear frequency modulated continuous wave (LFM) signals and frequency-stepped LFM signals are mixed and transmitted. Based on the characteristics of the two types of signals, two echo signals are obtained. The coarse estimation result obtained from the first echo signal is used as the compensation benchmark, and motion compensation for the second echo signal is completed in combination with the system's velocity measurement requirements, thereby correcting the signal distortion caused by relative motion. At the same time, the deviation between the coarse and fine estimation results is combined for joint judgment to suppress the detection error caused by relative motion, reduce the negative impact of motion interference on the detection results, and ultimately improve the estimation accuracy of target distance and velocity, while meeting the requirements of long-distance high-resolution detection. Attached Figure Description

[0057] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0058] Figure 1 This is a schematic diagram illustrating an application scenario of a signal processing method provided in an embodiment of this application;

[0059] Figure 2 A flowchart illustrating a signal processing method provided in this application embodiment. Figure 1 ;

[0060] Figure 3A A mixed signal generation principle provided in this application embodiment Figure 1 ;

[0061] Figure 3B A mixed signal generation principle provided in this application embodiment Figure 2 ;

[0062] Figure 3C Schematic diagram three illustrating the principle of mixed signal generation provided in this application embodiment;

[0063] Figure 4A A flowchart illustrating a signal processing method provided in this application embodiment. Figure 2 ;

[0064] Figure 4B A schematic flowchart of a signal processing method provided in this application embodiment is shown in Figure 3.

[0065] Figure 5 This is a schematic diagram of the structure of a signal processing device provided in an embodiment of this application;

[0066] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application;

[0067] Figure 7 This is a schematic diagram of the structure of a signal processing system provided in an embodiment of this application.

[0068] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0069] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0070] With the rapid development of intelligent driving and vehicle perception technologies, millimeter-wave radar, with its all-weather, all-time operation, strong penetration, low power consumption, and excellent ranging and speed measurement performance, has become a core sensor for vehicle environmental perception and target detection and recognition, and is widely used in scenarios such as adaptive cruise control, collision warning, and blind spot monitoring. Linear frequency modulated continuous wave (LFM) radar, due to its simple structure and ease of hardware implementation, is currently the mainstream operating mode for vehicle-mounted millimeter-wave radar.

[0071] Currently, to improve radar range resolution, known technologies often employ frequency-stepped linear frequency modulated continuous wave (LFMC) signals as the transmitted waveform. These signals, by adjusting the carrier frequency stepwise, can synthesize an ultra-large effective bandwidth without increasing the hardware transmission bandwidth. Their excellent bandwidth synthesis characteristics enable high-range target resolution, effectively distinguishing nearby targets. In the conventional processing flow, the radar transmits a single-frequency stepped waveform and receives the target echo signal. The echo signal is then processed sequentially with deskewing sampling, Fourier transform, spectrum analysis, and peak detection. Combined with the waveform modulation rules, motion parameters such as target range and velocity are calculated, thereby acquiring target information. This method, due to its high-resolution characteristics, is widely used in high-precision radar detection scenarios.

[0072] However, in practical applications, relative motion between the radar and the target is common, which can easily lead to a decrease in the accuracy of range and velocity estimation. Specifically, single-frequency stepped linear frequency modulated continuous wave signals are highly sensitive to the target's motion state. The relative motion of the target can easily cause range drift and phase distortion in the echo signal, destroying the phase coherence between each stepped sub-pulse, resulting in defocusing and peak shift in the synthesized high-resolution range image. At the same time, the spectral aliasing caused by motion will further interfere with the parameter calculation of the velocity dimension, ultimately causing an increase in the deviation of the target range and velocity estimation results, making it difficult to meet the high-precision detection requirements of vehicle-mounted radar.

[0073] Therefore, embodiments of this application provide a signal processing method and system to solve the above-mentioned problems. Specifically, the method of this application proposes to transmit a mixed signal containing a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal within a single frame. Based on the characteristics of the mixed signal, two different types of echo signals are obtained. One echo signal is used to complete a coarse estimation of the target motion parameters. Based on the coarse estimation result and the system's velocity measurement requirements, motion compensation is performed on the other high-resolution echo signal to reduce phase distortion and signal distortion caused by relative motion. Then, by comparing the deviation between the coarse estimation result and the compensated fine estimation result, accurate and reliable target distance and velocity estimates are obtained through reverse screening.

[0074] It is understood that the signal processing method of this application is applicable to any scenario in which target detection is carried out using a linear frequency modulated continuous wave system and there is relative motion between the radar and the target. For example, as shown above, the method of this application can be used in vehicle-mounted radar target ranging and speed measurement and environmental perception scenarios. Figure 1 This is a schematic diagram illustrating an application scenario of a signal processing method provided in an embodiment of this application, such as... Figure 1 As shown, the method of this application is executed by the vehicle controller in this scenario.

[0075] In this scenario, when the vehicle-mounted radar uses frequency-stepped linear frequency modulated continuous wave signals to detect surrounding vehicles, pedestrians, and other targets, there is always relative motion between the radar and the detected targets during the vehicle's movement. This can easily cause phase distortion of the echo signal and distance movement issues, resulting in a decrease in the accuracy of target distance and speed estimation, which cannot meet the high-precision environmental perception requirements of intelligent driving.

[0076] Based on the method of this application, a mixed signal consisting of conventional linear frequency modulated continuous wave and frequency-stepped linear frequency modulated continuous wave is transmitted within a single frame detection period. The vehicle controller distinguishes two echo signals from the target reflected echo. The first echo signal is used to calculate the target coarse estimation result. The second high-resolution echo signal is used to perform motion compensation in combination with the system speed measurement requirements to obtain the distance fine estimation result. Finally, based on the deviation between the coarse distance estimation result and the fine distance estimation result, the accurate distance estimate and speed estimate of the target are determined.

[0077] In the above process, by integrating the advantages of the two types of frequency modulation signals, targeted motion compensation for high-resolution echo signals is achieved based on the coarse estimation results. This effectively suppresses signal distortion and detection errors caused by relative vehicle motion, overcomes the inherent defect of frequency step signals being sensitive to motion states, effectively improves the estimation accuracy of vehicle radar for the distance and speed of surrounding targets, and ensures the reliability of intelligent driving environment perception.

[0078] It should be understood that, in the above process, the executing entity of the method of this application can also be a control unit with computing and control capabilities, such as a vehicle-mounted radar main control unit, a domain controller, or a cloud server, and this embodiment does not limit this. Furthermore, the application scenarios of the method of this application can also be other scenarios involving radar target ranging, speed measurement, and high-resolution detection, such as security monitoring radar, roadside radar, and UAV airborne radar, and this embodiment does not limit this.

[0079] The following detailed description, with reference to the accompanying drawings and using any electronic device as the execution subject, outlines some embodiments of the signal processing method of this application. Where the embodiments do not conflict, the following embodiments and features thereof can be combined with each other.

[0080] This application provides a signal processing method. Figure 2 A flowchart illustrating a signal processing method provided in this application embodiment. Figure 1 ,like Figure 2 As shown, the method in this application embodiment includes:

[0081] S201. Based on the echo signal after the mixed signal is reflected by the target, determine the first echo signal and the second echo signal.

[0082] The mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal.

[0083] It should be understood that conventional linear frequency modulated continuous wave signals have a continuous linear frequency change within a single modulation period, a constant frequency increment, and a simple waveform modulation method, and have robust motion parameter detection capabilities; frequency-stepped linear frequency modulated continuous wave signals are composed of multiple sets of sub-frequency modulated signals, and the carrier frequencies between the sub-signals change stepwise according to certain rules, which can achieve ultra-high distance resolution through the synthesis bandwidth.

[0084] Accordingly, in this embodiment, the echo signal corresponding to the conventional linear frequency modulated continuous wave signal is defined as the first echo signal, which is mainly used to achieve coarse estimation of target motion parameters; the echo signal corresponding to the frequency step linear frequency modulated continuous wave signal is defined as the second echo signal, which is mainly used to achieve high-resolution fine estimation of the target.

[0085] In this embodiment, the electronic device splits the received mixed echoes according to the different modulation parameters inside the mixed signal, using bandpass filtering or signal separation algorithms, and directly separates the first echo signal and the second echo signal based on the differences in signal modulation slope and carrier frequency range.

[0086] In practical applications, a transmitter timing marking method can also be used to distinguish the timing of the two types of signals during the transmission phase. The receiver can directly classify and separate them based on the timestamp information to quickly determine the first echo signal and the second echo signal. This embodiment does not limit this method.

[0087] In this embodiment, the electronic device generates a mixed signal based on a preset method, transmits the mixed signal, and receives the echo signal reflected from the target. The preset method includes a time-division switching method and / or an amplitude mixing method.

[0088] Specifically, the time-division switching method involves sequentially transmitting a conventional linear frequency modulated (LFM) continuous wave (LFMC) signal and a frequency-stepped LFM LFMC signal at different times, with the two types of signals not overlapping in the time dimension. For example, in one frame of data, the first 15ms is a conventional LFM LFMC signal, and the last 5ms is a frequency-stepped LFM LFMC signal. The amplitude mixing method involves weighted superposition of the two types of signals at the same time, allowing them to coexist in the same transmission time slot. For example, a low-power frequency-stepped LFM LFMC signal can be superimposed on a high-power conventional LFM LFMC signal. The aforementioned two methods can be implemented individually or combined and multiplexed to form a hybrid waveform transmission format that combines time-division and amplitude mixing.

[0089] As a preferred example, a hybrid signal is generated through a time-division switching method and an amplitude mixing method. Specifically, the electronic device uses a time-division switching mode in part of the time slots of the same detection frame to generate and transmit a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal sequentially; in another part of the time slots, an amplitude mixing mode is used to generate and transmit the two types of signals simultaneously after amplitude weighting and superposition, ultimately forming a composite hybrid signal that is compatible with time-division and amplitude mixing within a whole frame detection period.

[0090] It is understood that the above-mentioned generated mixed signal is only an example of the present invention. In actual applications, there are many other mixed waveform combinations. As long as the mixed waveform contains both a traditional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal, this application does not limit it.

[0091] Meanwhile, the frequency-stepped linear frequency modulated continuous wave signal included in the above-mentioned mixed signal can be a uniform frequency stepping form with a fixed frequency increment, a random frequency stepping form with a non-fixed frequency increment, or a waveform form with any random combination and arrangement. This application does not limit this.

[0092] As an example, Figure 3A A mixed signal generation principle provided in this application embodiment Figure 1 , Figure 3B A mixed signal generation principle provided in this application embodiment Figure 2 , Figure 3C Schematic diagram three illustrates the principle of mixed signal generation in an embodiment of this application, wherein... Figure 3A The hybrid waveform structure generated by a single time-division switching method consists of a traditional linear frequency modulated continuous wave signal in the first half and a frequency-stepped linear frequency modulated continuous wave signal with a fixed frequency increment in the second half. Figure 3B It is also generated using a single time-division switching method. The first half is a traditional linear frequency modulated continuous wave signal, and the second half is a frequency-stepped linear frequency modulated continuous wave signal with random stepping and non-fixed frequency increment. Figure 3C To generate a hybrid waveform structure using a segmented time-division switching method, the signals are arranged in a time-division manner in the order of traditional linear frequency modulated continuous wave signal → frequency-stepped linear frequency modulated continuous wave signal → traditional linear frequency modulated continuous wave signal, forming a hybrid waveform that combines the three types of signals in segments.

[0093] Among them, B in the figure c The frequency modulation bandwidth is used to represent a single linear frequency modulated continuous wave signal, Tr is used to represent the transmission duration (repetition period / pulse width) of the single linear frequency modulated continuous wave signal, and f is used to represent the frequency modulation bandwidth. cThe carrier start frequency (center frequency) is used to represent a conventional linear frequency modulated continuous wave (LFM) signal, and Δf is used to represent the frequency step size of the LFM signal. N step Used to represent the number of steps (number of signals) in a frequency-stepped linear frequency modulated continuous wave signal.

[0094] It should be understood that the above processes in this embodiment are all executed autonomously by the electronic device. In actual applications, after determining the mixed signal generation logic, the electronic device can send control commands to the waveform generator so that the waveform generator generates a mixed signal that conforms to the modulation rules; at the same time, it can send transmission and acquisition commands to the radio frequency transmitter and receiver to control the radio frequency transmitter and receiver to complete the transmission of the mixed signal and the reception of the target echo signal. This application does not limit this.

[0095] In this embodiment, by flexibly employing time-division switching, amplitude mixing, and combinations thereof to generate and transmit hybrid signals, it is compatible with different waveform arrangements and transmission systems, and adapts to various frequency stepping rules such as uniform stepping and random stepping, thus broadening the range of waveform design and hardware adaptation. Simultaneously, it integrates the generation, transmission, and echo reception of hybrid signals, leveraging the dual advantages of stable motion parameter detection of conventional linear frequency modulated continuous wave signals and strong range resolution of frequency-stepped linear frequency modulated continuous wave signals. This provides reliable raw signal support for subsequent echo signal separation, motion compensation, and accurate target parameter calculation, effectively adapting to the practical application requirements of multi-scenario, multi-input multiple-output (MIMO) transmission systems for vehicle-mounted millimeter-wave radar, such as time division multiple access, Doppler frequency division multiple access, frequency division multiple access, and code division multiple access.

[0096] As a preferred example, in the aforementioned mixed signal, the number of conventional linear frequency modulated continuous wave (LFM) signals is greater than the number of frequency-stepped LFM signals. Specifically, when generating the mixed signal, the electronic device prioritizes allocating more time slots or power resources to conventional LFM signals to ensure the accuracy and stability of motion parameter estimation; simultaneously, frequency-stepped LFM signals are inserted into the remaining time slots or power resources to acquire high-resolution distance information.

[0097] Taking the time-division switching method as an example, and referring to the previous example, within a 20ms detection period of one frame, 15 conventional linear frequency modulated (LFM) continuous wave (RFMC) signals are transmitted sequentially in the first 15ms, and 5 frequency-stepped LFMMC signals are transmitted sequentially in the last 5ms. The two types of signals do not overlap in the time dimension, and the number of conventional signals (15) is greater than the number of stepped signals (5). Taking the amplitude mixing method as an example, and referring to the previous example, within the same transmission time slot, the 5 low-power frequency-stepped LFMMC signals are superimposed on the 15 high-power conventional LFMMC signals. The two types of signals completely overlap in the time domain, but are distinguished by power difference or code domain, and the number of conventional signals (15) is greater than the number of stepped signals (5). Taking the time-division and amplitude mixing method as an example, in a frame detection period, the first 10ms uses the time-division switching method to transmit conventional signals and step signals (quantity ratio 2:1), and the next 10ms uses the amplitude mixing method to transmit the two types of signals simultaneously (power ratio 3:1), ultimately forming a composite mixed signal, and the total number of conventional signals in the whole frame is still greater than the total number of step signals.

[0098] In practical applications, the ratio of conventional linear frequency modulated continuous wave (LFM) signals to frequency-stepped LFM signals is not limited to the preferred examples mentioned above, and can be adjusted according to the actual detection scenario and hardware capabilities. For example, in close-range high-resolution detection scenarios, an equal ratio (e.g., 1:1) can be used to balance the accuracy of motion parameter estimation and range resolution; in long-range detection scenarios, the proportion of conventional signals can be further increased (e.g., 5:1 or higher) to prioritize the stability of motion parameter calculation; in scenarios requiring rapid scanning, the number of stepped signals can be reduced (e.g., 10:1) to reduce data volume and processing latency. This application does not impose any limitations on this.

[0099] In this embodiment, by employing a preferred combination where the number of conventional signals exceeds the number of step signals, the robustness of motion parameter estimation can be improved while maintaining high-resolution range imaging. The larger number of conventional linear frequency modulated continuous wave signals provides sufficient slow-time dimension sampling points for velocity estimation, making Doppler-based velocity estimation more stable and reducing the impact of noise and clutter interference. While the number of frequency-stepped linear frequency modulated continuous wave signals is smaller, it is sufficient to achieve super-resolution in the range dimension through the synthesis bandwidth, avoiding a surge in data volume and processing latency caused by excessive step signals. At the same time, the reduction in the number of step signals means a reduction in the number of step channels to be processed, thereby reducing the computational load of the baseband processor and the power consumption of the RF front end. In addition, this combination can be flexibly adjusted, increasing the proportion of conventional signals for long-range detection and appropriately increasing the proportion of step signals for short-range high-resolution detection, achieving a dynamic balance between detection range and resolution capability, and effectively adapting to the resource allocation requirements under multiple-input multiple-output (MIMO) transmission systems such as time-division multiple access and frequency-division multiple access, avoiding inter-channel interference and improving the overall system performance.

[0100] S202. For each velocity assumption, based on compensating the second echo signal and synthesizing a one-dimensional high-resolution range image, determine the precise range estimation results and peak energy under each velocity assumption.

[0101] The speed assumptions are determined based on the system's speed measurement requirements and the coarse estimation results obtained from processing the first echo signal.

[0102] Among them, the system speed measurement requirement is the preset upper limit requirement for the maximum speed of the target in the radar detection scenario, which is used to determine the speed extension range during compensation processing; in this embodiment, the requirement is preset based on the radar's own performance indicators, application scenario safety specifications, or preset detection range parameters.

[0103] In practical applications, the aforementioned system speed measurement requirements can also be the target maximum speed input by the user, the speed threshold limited by the scenario, or the speed range obtained from historical detection data. This application does not limit these requirements.

[0104] In this embodiment, the precise distance estimation result is a high-precision estimate of the target distance obtained by calculating the compensated second echo signal after motion compensation processing. Compared with the coarse estimation result, it has higher distance resolution and parameter estimation accuracy. The one-dimensional high-resolution range image is a high-resolution range signal synthesized by extracting and stitching the compensated second echo signal. This signal has a fine resolution capability in the range dimension that exceeds the bandwidth limitation of a single signal. The peak energy is the signal amplitude or power value corresponding to the detected peak in the one-dimensional high-resolution range image for each velocity assumption, which is used to measure the confidence of the target response during subsequent coarse-fine deviation comparison.

[0105] Specifically, in this embodiment, the electronic device first extracts each step-linear frequency modulated continuous wave signal within the fine search range based on the target coarse estimation range image corresponding to the second echo signal; for any velocity assumption, a two-dimensional high-resolution range image is obtained by performing phase compensation under the velocity assumption on each step-linear frequency modulated continuous wave signal; finally, a one-dimensional high-resolution range image is synthesized by extracting and stitching based on the two-dimensional high-resolution range image. The fine search range is determined based on the coarse range estimation result.

[0106] More specifically, when determining the fine search range based on the coarse distance estimation result from the first echo signal, the electronic device first weighs the parameter configuration at the system level: under the premise of meeting the system distance resolution requirements, it takes into account suppressing the distance ambiguity caused by the grating lobe and reducing the influence of distance movement between step signals, and makes a compromise configuration between the maximum unambiguous distance corresponding to the inverse Fourier transform of a single sampling point and the number of step frequencies, so as to ensure that the maximum unambiguous distance is much greater than the coarse distance resolution and avoid the distance ambiguity problem.

[0107] Based on this, in order to balance high-resolution accuracy and computational complexity, relying on the coarse range estimation results of the target, and under the constraint that the maximum unambiguous range is much larger than the coarse range resolution, combined with the boundary limit of the minimum and maximum effective range of the radar system, a local range fine search range on the synthetic high-resolution range image is defined, and the corresponding fine search range index is obtained based on this fine search range. Only the intervals within the fine search range are subjected to subsequent high-resolution processing and motion phase compensation, which avoids range ambiguity and defocusing problems and reduces the overall computational load.

[0108] In practical applications, the size of the fine search range can be dynamically adjusted based on the bandwidth of the first echo signal, the composite bandwidth of the second echo signal, or the statistical results of historical detection data, or a preset fixed range can be used directly as the fine search range. This application does not limit this.

[0109] More specifically, regarding the method for determining multiple sets of velocity assumptions, Figure 4A A flowchart illustrating a signal processing method provided in this application embodiment. Figure 2 ,like Figure 4A As shown, the electronic device determines multiple sets of speed assumptions in the following manner:

[0110] S1. Determine the corresponding speed expansion coefficient based on the system speed measurement requirements.

[0111] S2. Based on the speed expansion coefficient, the rough speed estimate in the rough estimate is extended to generate multiple sets of different speed assumptions.

[0112] The electronic device obtains a coarse estimated range profile of the target based on the sampling sequence or intermediate frequency sampling sequence corresponding to each linear frequency modulated continuous wave signal; and obtains a coarse estimation result corresponding to the first echo signal based on the coarse estimated range profile corresponding to the first echo signal, or a signal detection matrix constructed along the frequency modulated signal dimension of the coarse estimated range profile.

[0113] In this embodiment, the electronic device performs down-conversion, amplification, and analog-to-digital conversion on the echo signal reflected from the target through the radio frequency receiving channel to directly obtain the original sampling sequence corresponding to the echo signal. This sampling sequence contains the amplitude, phase, and timing information of the echo signal.

[0114] As a further preferred design, the electronic device uses a mixer to mix the echo signal with the local oscillator signal multiplexed by the transmitter to obtain an intermediate frequency signal containing target distance information. This signal is then converted from analog to digital to obtain an intermediate frequency sampling sequence. This sequence filters out the carrier component, reducing the computational load of subsequent processing.

[0115] In this embodiment, for each linear frequency modulated continuous wave signal, the electronic device performs a one-dimensional fast Fourier transform on its corresponding intermediate frequency sampling sequence to obtain a coarse estimated range profile of the target. Further, based on the timing or modulation parameters of the first echo signal, the electronic device separates the corresponding portion from the overall coarse estimated range profile to obtain a coarse estimated range profile corresponding to the first echo signal; and for each sampling point on the coarse estimated range profile corresponding to the first echo signal, performs a two-dimensional fast Fourier transform along the frequency modulated signal dimension and calculates the amplitude to obtain a signal detection matrix.

[0116] It should be understood that the signal detection matrix is ​​a two-dimensional matrix constructed by associating distance and velocity information. The amplitude peaks in the matrix correspond to the combination of the target's distance and velocity, and are used to achieve two-dimensional detection of the target and coarse parameter estimation.

[0117] After obtaining the signal detection matrix, the electronic device performs constant false alarm rate (CFAR) detection and peak aggregation processing on the matrix to obtain a coarse estimate of the first echo signal. CFAR detection, based on a preset false alarm probability threshold, filters out valid target peaks exceeding the detection threshold from the signal detection matrix, suppressing background noise and clutter interference. Peak aggregation processing merges and categorizes adjacent valid peaks to obtain unique peak information corresponding to the target; the distance and velocity corresponding to this information are the coarse estimate.

[0118] In practical applications, a coarse range estimate can be obtained by performing a one-dimensional fast Fourier transform on the sampled sequence, and then the coarse estimate result can be obtained directly from the peak position in the coarse range estimate, without the need to construct a complete signal detection matrix. This application does not impose any restrictions on this.

[0119] In this embodiment, a coarse range profile is obtained by performing a one-dimensional fast Fourier transform on the sampling sequence or the intermediate frequency sampling sequence, which can quickly obtain the initial range information of the target. Based on this, the coarse estimation result can be directly calculated based on the range profile, or a two-dimensional joint estimation of range and velocity can be achieved by constructing a signal detection matrix. This balances processing efficiency and estimation reliability, provides a stable and accurate initial reference for subsequent motion compensation, and effectively reduces the computational complexity of subsequent fine processing.

[0120] After obtaining the system speed measurement requirements and coarse estimation results, the electronic equipment determines the speed expansion coefficient, i.e., the maximum speed expansion factor, based on the system speed measurement requirements. Specifically, the electronic equipment uses the ratio of the maximum target speed corresponding to the system speed measurement requirements to the coarse speed estimation result as the speed expansion coefficient, which is used to determine the expansion range of speed compensation. Subsequently, the electronic equipment expands the interval in both positive and negative directions based on the speed expansion coefficient, with the coarse speed estimation result as the center, to generate a set of speed assumptions containing multiple sets of discrete speed values. This set covers all possible movement speeds of the target under the system speed measurement requirements, providing a complete parameter space for phase compensation under different speed assumptions.

[0121] In practical applications, a fixed proportional coefficient can be directly preset as the velocity expansion coefficient, or the velocity expansion coefficient can be adaptively solved based on the radar velocity measurement accuracy and scene clutter intensity; multiple sets of velocity assumptions can also be generated by using an asymmetric interval expansion method or a segmented unequal interval discrete value method, which is not limited in this application.

[0122] In this embodiment, by expanding the range of the coarse velocity estimation results based on the system's velocity measurement requirements to generate multiple sets of velocity assumptions, it is possible to comprehensively cover the target's possible movement speeds and avoid compensation failures caused by velocity estimation deviations. At the same time, based on the coarse distance estimation results, the fine search range on the synthesized high-resolution range image is defined, limiting the high-resolution fine processing to the possible location range of the target. This avoids computational redundancy caused by full-range processing and ensures the targetedness and effectiveness of the compensation processing, providing a reasonable processing boundary for the subsequent generation of high-resolution range images and effectively balancing processing efficiency and estimation accuracy.

[0123] It should be understood that the above-mentioned processes of determining the search range and determining multiple sets of speed assumptions can be executed in parallel or sequentially in practical applications, and this application does not limit their specific execution order.

[0124] Furthermore, in this embodiment, referring to the aforementioned method for obtaining the coarse estimated range image, the electronic device also uses the sampling sequence or intermediate frequency sampling sequence corresponding to the second echo signal to process the target coarse estimated range image corresponding to the second echo signal through one-dimensional fast Fourier transform to obtain the target coarse estimated range image corresponding to the second echo signal; then, according to the distance index corresponding to the defined fine search range, the interval boundary is located from the target coarse estimated range image of the second echo signal, and the stepped linear frequency modulated continuous wave signal data contained in each segment within the fine search range are extracted.

[0125] For each step linear frequency modulated continuous wave signal within the search range, the electronic device traverses each set of preset velocity assumptions, introduces the corresponding velocity compensation phase factor one by one to perform phase correction on the signal, eliminates the range-velocity coupling phase shift caused by the moving target, restores the phase consistency between each step signal, and then processes and generates a two-dimensional high-resolution range image.

[0126] In one possible implementation, the electronic device performs phase compensation processing on each step-linear frequency modulated continuous wave signal based on a velocity assumption to obtain a compensated step-linear frequency modulated continuous wave signal; based on the windowing processing and inverse Fourier transform processing of the compensated step-linear frequency modulated continuous wave signal, a two-dimensional high-resolution range image corresponding to different velocity assumptions is generated.

[0127] Specifically, moving targets generate coupled phase terms between linear frequency modulated continuous wave signals with different phases, causing range cell crossings and phase disturbances. Phase compensation for each signal according to each velocity assumption can offset the phase distortion caused by Doppler motion. Windowing can suppress spectral leakage and sidelobe interference, improving the peak focusing effect of the range image. Then, the frequency domain is converted to the time domain by inverse Fourier transform, finally constructing a two-dimensional high-resolution range image with the range dimension and velocity dimension correlated.

[0128] More specifically, the electronic device calculates the corresponding phase compensation amount for each set of speed assumptions, and uses the obtained phase compensation amount to correct each extracted step linear frequency modulated continuous wave signal, eliminate the distance-velocity coupling phase term generated by the moving target, correct the disordered phase relationship between different step signals, thereby restoring the phase consistency of each step signal and completing the signal phase compensation.

[0129] Furthermore, the electronic device preloads a preset window function coefficient sequence, multiplies the compensated step linear frequency modulated continuous wave signal time-domain data with the window function coefficients point by point to complete time-domain windowing; then performs point-matched inverse fast Fourier transform on the windowed set of sampled data, arranges the output matrix data according to the distance dimension and velocity dimension, and directly generates a two-dimensional high-resolution range image under the corresponding velocity assumption.

[0130] In practical applications, the windowing process can be omitted according to system performance requirements, and the inverse Fourier transform can be directly performed on the compensated step linear frequency modulated continuous wave signal; alternatively, different window functions such as Hanning window, Hamming window, and Blackman window can be used for windowing smoothing, but this application does not limit this.

[0131] In this embodiment, by performing phase compensation on each step linear frequency modulated continuous wave signal according to different velocity assumptions, the actual motion state of the target can be matched, effectively correcting the phase defocusing problem introduced by motion. Combined with windowing processing, sidelobe clutter can be suppressed and spectral leakage can be reduced. Furthermore, by using inverse Fourier transform to achieve decoupling of two-dimensional information, the generated two-dimensional high-resolution range image has both high range resolution and velocity resolution capabilities, providing reliable basic data for subsequent one-dimensional synthesis processing and accurate calculation of target parameters.

[0132] Furthermore, due to the inherent characteristics of the step frequency system, a single two-dimensional high-resolution range image is difficult to achieve an ultra-high range resolution effect throughout the entire range. It is necessary to extract and stitch multiple sets of two-dimensional range images and integrate the information of each step channel to reconstruct a one-dimensional synthetic high-resolution range image.

[0133] In one possible implementation, the electronic device extracts two-dimensional high-resolution range images under different velocity assumptions based on the same-distance discard method to obtain the extracted range images; and synthesizes a one-dimensional high-resolution range image by stitching the extracted range images along the frequency step dimension.

[0134] Specifically, the same-distance discarding method is used to filter and remove redundant and repetitive same-distance units in two-dimensional high-resolution range images under the velocity assumption, retaining only the unique effective distance sampling information and avoiding information redundancy and peak confusion; segment-by-segment stitching along the frequency step dimension can integrate the bandwidth advantages of different step channels, break through the bandwidth limitation of a single signal, improve the overall range resolution, and achieve synthetic aperture-type high-resolution imaging effect.

[0135] More specifically, the electronic device uses a same-distance discarding method to discard repeated signals at the same distance index in the two-dimensional high-resolution range image, retaining only single-sampling data; then, based on the constraint relationship between the frequency modulation bandwidth of the single pulse signal and the frequency step, a fixed number of sampling points is set, and the signal is extracted point by point in the position sequence. If the number of remaining points at the end is less than the set value, the sampling continues cyclically from the starting position until the specified number of samplings is reached. Finally, all the extracted valid signals are spliced ​​together sequentially along the frequency step dimension.

[0136] In practical applications, the same-distance discarding method can be abandoned, and global sampling and stitching of the two-dimensional high-resolution range image can be performed directly; alternatively, stitching and reconstruction can be performed along the distance dimension or velocity dimension, and other equivalent filtering and stitching algorithms can be used to construct a one-dimensional synthetic high-resolution range image. This application does not limit this.

[0137] In this embodiment, the same-distance discarding method is used to extract the two-dimensional high-resolution range image, which can reduce the amount of data, eliminate redundant interference units, and avoid peak artifacts and range blur caused by duplicate information. The stitching along the frequency step dimension can fully integrate the frequency band resources of each step linear frequency modulated continuous wave signal, break through the single signal bandwidth constraint, effectively improve the range resolution and target focus of the one-dimensional high-resolution range image, and reduce the computational load of subsequent peak detection.

[0138] Finally, based on constant false alarm rate detection of one-dimensional high-resolution range images under each velocity assumption, the accurate distance estimation results under each velocity assumption are determined; and based on the one-dimensional high-resolution range images under each velocity assumption, peak energy is extracted.

[0139] In one possible implementation, the electronic device performs constant false alarm rate detection on a one-dimensional high-resolution distance image under each speed assumption to obtain the detection peak under each speed assumption; the distance corresponding to the detection peak under each speed assumption is determined as the distance estimation result under each speed assumption.

[0140] Specifically, in the one-dimensional synthetic high-resolution range image, the target is presented in the form of peaks. The electronic device first performs constant false alarm rate detection on the one-dimensional high-resolution range image under each velocity assumption, adaptively sets the detection threshold, removes false peaks caused by clutter and noise, and retains the effective peaks of the real target; then, the effective peaks that pass the detection threshold are extracted for position, and the distance corresponding to the detected peak under each velocity assumption is determined as the accurate distance estimation result under that velocity assumption, thereby suppressing false alarm interference and improving the accuracy of parameter estimation.

[0141] In practical applications, effective target screening can also be achieved by replacing constant false alarm rate (CFAR) detection with methods such as direct threshold comparison and peak clustering screening; different detection architectures such as unit average CFAR and ordered statistical CFAR can also be used to complete the screening judgment, and this application does not limit this.

[0142] It should be understood that the three processing steps of phase compensation and two-dimensional high-resolution range image generation, extraction and stitching to synthesize one-dimensional high-resolution range image, peak extraction and constant false alarm rate screening are preferred implementation methods. In practical applications, some unnecessary steps can be omitted or other equivalent signal processing procedures can be used instead. It is not necessary to strictly implement all of them. This application does not limit this.

[0143] Specifically, in this embodiment, the electronic device extracts the signal amplitude or power value corresponding to the detection peak in the one-dimensional high-resolution range image corresponding to the current velocity assumption, and uses it as the peak energy under that velocity assumption. This peak energy is used for subsequent target detection and joint parameter determination, reflecting the target's response intensity under that velocity assumption.

[0144] In practical applications, the energy weighted average of multiple distance cells around the peak can also be used as the peak energy to reduce the impact of noise. This application does not limit this.

[0145] In this embodiment, motion defocus is eliminated by multi-velocity hypothetical phase compensation, and a two-dimensional high-resolution range image is constructed by windowing and inverse Fourier transform. Then, the one-dimensional range image resolution is improved by extracting, stitching and fusing the frequency band advantages. Finally, false alarms are screened out by peak extraction and constant false alarm rate detection, and effective target range information is locked. The signal focusing ability, anti-clutter interference ability and parameter estimation accuracy are improved step by step, while taking into account the process substitutability and system adaptability flexibility.

[0146] As can be seen from the above, by extracting the stepped linear frequency modulated continuous wave signal within the fine search range and performing subsequent processing only on the local effective interval, the computational overhead can be effectively reduced. By traversing phase compensation through multiple velocity assumptions, the system can adapt to any target motion velocity state, effectively solving the problem of inconsistency between the distance movement and phase of the stepped frequency signal. By constructing a two-dimensional range image and extracting and stitching it together to synthesize a one-dimensional high-resolution range image, the system breaks through the bandwidth limitation to achieve super-resolution imaging. Finally, by solving the distance estimation results based on velocity assumptions, the system provides high-precision raw data for subsequent coarse and fine deviation comparison and joint determination of target distance and velocity. The overall process takes into account computational efficiency, distance resolution, motion adaptability, and parameter estimation reliability.

[0147] To further elaborate on the above-mentioned coarse estimation results, the determination process of the fine search range and multiple velocity assumptions, as well as the signal compensation process based on relevant parameters, the following explanation is based on specific signal processing principles.

[0148] Assuming the signal detection matrix obtained after a conventional linear frequency modulated continuous wave signal undergoes a one-dimensional Fourier fast transform and a two-dimensional Fourier fast transform is RD, and after constant false alarm rate detection and peak aggregation processing, the distance index and velocity index of a detected target are respectively... and The coarse estimation results (including the distance coarse estimation results) and the results of the rough speed estimate )for:

[0149]

[0150] in, The distance grid coordinates represent the process of multiplying the sampling point numbers of the one-dimensional Fourier transform with the single linear frequency modulated continuous wave distance resolution in sequence to obtain the complete distance scale axis. For velocity grid coordinates, it means that the indices of the velocity dimension of the two-dimensional Fourier transform are multiplied sequentially by the velocity resolution to obtain a complete velocity scale axis that is symmetrical with positive and negative values. Under complex sampling ADC (referring to an analog-to-digital converter that simultaneously acquires in-phase and quadrature components using IQ quadrature demodulation, outputting a complex baseband signal, which can completely retain the amplitude and phase information of the echo signal and can be directly used for two-dimensional Fourier transform, motion phase compensation, and target velocity calculation), they are as follows:

[0151]

[0152] in, and These represent the distance resolution under the frequency modulation bandwidth of a single linear frequency modulated continuous wave signal and the distance resolution of a conventional linear frequency modulated continuous wave signal group. Velocity resolution for a traditional single-carrier frequency linear frequency modulated continuous wave signal. , , This represents the speed of light, and its value is approximately 3 × 10⁻⁶. 8 meters per second This represents the frequency modulation bandwidth of a single linear frequency modulated continuous wave signal. This represents the transmission repetition period of a linear frequency modulated continuous wave signal. Indicates the radar operating wavelength. This refers to the number of ADC sampling points on a single linear frequency modulated continuous wave signal. The number of conventional linear frequency modulated continuous wave signals transmitted at a single carrier frequency.

[0153] If the system speed measurement requirement indicates a range of... ,fixed The maximum unambiguous speed is set to , Therefore, the required maximum speed expansion factor is determined as follows:

[0154]

[0155] Notice, To determine the required maximum speed expansion factor, which is also equivalent to the total number of different speed assumptions to be traversed.

[0156] Increase the maximum speed by a factor of 1 The velocity expansion factor within the range is array The length is equal to , This represents the minimum value of the negative velocity expansion coefficient. This represents the maximum value of the positive velocity expansion coefficient. The formulas for calculating both are as follows (where ceil() rounds up and floor() rounds down):

[0157]

[0158]

[0159] So, with different speed expansion coefficients The following is a rough estimate of the speed. The different speeds are assumed to be constructed as follows:

[0160]

[0161] in, Represents the velocity expansion factor The speed assumption corresponding to the time, The maximum unambiguous speed.

[0162] Assuming a one-dimensional Fast Fourier Transform is performed on the intermediate frequency sampling sequence of each linear frequency modulated continuous wave signal after deskewing of the echo signal, the coarse range profile of the target is obtained as follows: , .in, This represents the total number of linear frequency modulated continuous wave signals transmitted within one frame. , This refers to the number of linear frequency modulated continuous wave signals transmitted by a conventional linear frequency modulated continuous wave signal group. Let be the number of linear frequency modulated (LFM) continuous wave signals in the frequency-stepped LFM continuous wave signal group. Then, the coarsely estimated range profile corresponding to the frequency-stepped LFM continuous wave signal is: .

[0163] Assume the required distance resolution of the system is The required synthesized bandwidth of the system is Therefore, the number of stepped linear frequency modulated continuous wave signals required in the frequency stepped linear frequency modulated continuous wave signal group can be obtained as follows: ,in, It represents the fixed frequency step between adjacent stepped linear frequency modulated continuous wave signals.

[0164] It is worth noting that, in order to avoid range ambiguity caused by grating lobes in the synthesized high-resolution range image within the range search range, and to suppress the impact of large distance migration between stepped linear frequency modulated continuous wave signals, while meeting the system's range resolution requirements, the maximum unambiguous distance corresponding to the inverse fast Fourier transform of the single linear frequency modulated continuous wave signal in the stepped linear frequency modulated continuous wave signal must be maximized. Make it as large as possible, while also maximizing the number of steps. As small as possible, therefore it needs to be in and A comprehensive balance must be struck between them. , usually set , Far greater than Right now ,but .in, Used to indicate coarse distance resolution.

[0165] To improve the accuracy of high-resolution target range estimation and reduce computational complexity, based on the coarse range estimation results of the target... ,exist Far greater than In this case, the range for fine-tuning on the synthesized high-resolution range image is determined as follows:

[0166]

[0167] in, This is the minimum effective range of the radar system. This refers to the maximum effective range under the configured radar system parameters. In this embodiment, the 2 in the above formula is specifically calculated using system parameters. Specifically, it is assumed that the system's maximum velocity is -110 m / s, and the coarse range resolution... =0.8m, the repetition period of each linear frequency modulated continuous wave signal Tr=25μs, and the number of conventional linear frequency modulated continuous wave signals collected is 512, then Q=ceil(|(-110 512 25e-6)| / 0.8)=2, where ceil represents rounding up and 25e-6 represents 25μs. In practical applications, Q can also be other values, and this application does not limit this.

[0168] Based on this, the search range is determined by distance. The calculated fine search distance index is:

[0169]

[0170] For frequency-stepped linear frequency modulated continuous wave signals, moving targets generate phase terms that couple range and velocity, causing different step signals to fall into different range cells and disrupting the phase consistency between the step signals. Directly performing an inverse fast Fourier transform (IFT) cannot achieve effective coherent accumulation, resulting in reduced signal-to-noise ratio gain and severe defocusing of the synthesized high-resolution range image. Therefore, precise motion phase compensation must be performed before the IFT to restore the phase consistency of each step signal, thereby obtaining a well-focused synthesized high-resolution range image.

[0171] This embodiment employs a hybrid transmission waveform combining conventional linear frequency modulated (LFM) continuous wave (RFMC) and frequency-stepped LFM RFMC. The initial frequencies of all transmitted signals in the conventional LFM RFMC group are consistent, and the inter-frame phase difference remains constant. Effective coherent accumulation can be achieved through a two-dimensional fast Fourier transform (FFT), yielding a high-precision coarse velocity of the target. Using this coarse velocity as a benchmark, the velocity range is expanded to generate multiple sets of velocity hypotheses that include the target's true velocity. Phase compensation is then applied to the frequency-stepped LFM RFMC signal for each set of velocity hypotheses, followed by an inverse fast Fourier transform, resulting in a well-focused, composite high-resolution range image.

[0172] Before performing phase compensation and high-resolution imaging processing, it is necessary to extract all stepped linear frequency modulated continuous wave signal data within the fine search range from the coarsely estimated range image corresponding to the frequency stepped linear frequency modulated continuous wave, based on the aforementioned determined range fine search range and the calculated fine search range index.

[0173] Specifically, based on the calculated fine-search distance index Coarse estimation of range image from stepped linear frequency modulated continuous wave signal Extract all stepped linear frequency modulated continuous wave signals:

[0174]

[0175] Using different speed assumptions For all extracted step linear frequency modulated continuous wave signals Phase compensation is performed, and the velocity is assumed to be... The phase compensation amount is (where, For the nth step The starting frequency of each frequency-stepped linear frequency modulated continuous wave signal, f0 is the starting frequency of the first frequency-stepped linear frequency modulated continuous wave signal).

[0176]

[0177] Based on this, the velocity assumption is used The phase compensation amount calculated below For all extracted step linear frequency modulated continuous wave signals After phase compensation, the coarsely estimated range image after compensation is obtained as follows:

[0178]

[0179] In this embodiment, the compensated coarsely estimated distance image Add windows and perform The inverse fast Fourier transform of the points, where, , Representing 2 The power of the power yields the velocity assumption. Two-dimensional high-resolution range image of the target below:

[0180]

[0181] in, This indicates that the windowing function is the Hanning window, and the window length is equal to the number of steps in the linear frequency modulated continuous wave signal. .

[0182] Based on the frequency modulation bandwidth of a single linear frequency modulated continuous wave signal With frequency step The relationship between them: The same distance discard method is used to evaluate the assumed velocity. Two-dimensional high-resolution range image of the target Extraction and splicing:

[0183]

[0184]

[0185] in, Indicates each distance index The number of signal points after IFFT extraction. This indicates that the value within the parentheses will be rounded to the nearest integer. express Distance resolution corresponding to each frequency point of the inverse fast Fourier transform during point IFFT. , This indicates the starting position of the extracted signal after the inverse fast Fourier transform (IFT). If the last IFT-transformed signal is extracted, the number of extractions is less than [a certain value]. Then start drawing again from the beginning, until the desired result is drawn. until, This indicates the modulo operation.

[0186] Thus, we obtain the target's velocity assumption. The following one-dimensional composite high-range resolution image is denoted as .

[0187] Similarly, the phase compensation, windowed inverse fast Fourier transform, and decimation and stitching operations are repeated for other velocity assumptions of the target to obtain a one-dimensional synthetic high-resolution range image of the target under different assumed velocities. .

[0188] Finally, the targets under the velocity assumptions are discussed separately. One-dimensional high-range resolution image synthesized within a defined distance search range By performing constant false alarm rate (CFAR) testing, the distance estimation result is obtained as follows: ,in, This indicates the number of new targets obtained by performing high-resolution range analysis on the target.

[0189] S203. Based on the deviation between the fine distance estimation result and the coarse distance estimation result, and the peak energy, determine the distance estimate and / or velocity estimate of the target from each velocity assumption.

[0190] It should be understood that when performing phase compensation based on the erroneous velocity assumption, the step signal is severely defocused, resulting in peak distortion of the generated one-dimensional synthetic high-resolution range image. This leads to a significant deviation between the coarse and fine range estimates, and the peak energy of the synthetic high-resolution range image is relatively low. In contrast, the true velocity assumption achieves accurate phase compensation, resulting in a well-focused range image with smaller coarse and fine range deviations and higher peak energy. Therefore, by combining the coarse and fine range deviations with the peak energy as the final output parameter selection, accurate joint estimation of range and velocity can be achieved.

[0191] Specifically, in this embodiment, the electronic device determines candidate velocity assumptions from each velocity assumption based on the deviation and coarse range resolution under each velocity assumption; the candidate velocity assumption is a velocity assumption with a deviation less than the coarse range resolution; based on the peak energy, the target's range estimate and velocity estimate are determined from the candidate velocity assumptions.

[0192] More specifically, the electronic device determines candidate velocity assumptions and corresponding fine range estimates based on coarse range resolution. Based on peak energy and the range estimation deviations corresponding to the candidate velocity assumptions, it determines the target velocity assumption from the candidate velocity assumptions, and uses the target velocity assumption and its corresponding fine range estimate as the target's velocity estimate and range estimate, respectively. Specifically, a candidate velocity assumption is one whose deviation from the coarse-to-fine range estimation result is less than that corresponding to the coarse range resolution; the target velocity assumption is the candidate velocity assumption that satisfies the minimum deviation when the peak energy is at its maximum.

[0193] In this embodiment, the electronic device first filters candidate velocity assumptions according to deviation constraints. Specifically, the deviation constraints are: the maximum deviation between the coarse distance estimation result and the fine distance estimation result corresponding to different velocity assumptions is less than the distance resolution corresponding to the frequency modulation bandwidth of a single linear frequency modulated continuous wave signal; then, from the set of candidate velocity assumptions after filtering, the optimal velocity assumption that matches the actual motion state of the target is selected using the peak energy magnitude as the selection criterion, and then the corresponding fine distance estimation result is locked synchronously, thus completing the final determination of the target distance and velocity parameters.

[0194] In practical applications, other equivalent discrimination methods such as adaptive setting of deviation threshold, joint screening of multiple feature parameters, and ranking and selection can also be used to select candidate velocity hypotheses and match optimal parameters. This application does not limit this.

[0195] In this embodiment, by adopting a joint discrimination method that combines coarse and fine distance estimation deviation constraints with peak energy selection, the defocusing interference and parameter misjudgment problems caused by non-real velocity assumptions can be effectively eliminated. High-precision joint estimation of target distance and velocity can be achieved without adding an additional hardware detection module. At the same time, by relying on hierarchical screening logic to reduce computational complexity, improve the real-time performance and environmental adaptability of signal processing, and effectively improve the accuracy and stability of target parameter estimation in complex scenarios.

[0196] As a detailed explanation of the process of determining distance and velocity estimates based on deviation, assume there are N k Given a speed assumption, then Figure 4B A schematic flowchart of a signal processing method provided in this application embodiment is shown in Figure 3. Figure 4B As shown, the process includes: traversing all N... kFor each preset velocity assumption, constant false alarm rate (CFAR) detection is performed on the corresponding one-dimensional synthetic high-resolution range image to obtain the fine estimate of the target range and the maximum peak energy under that velocity assumption. Then, the deviation between each fine range estimate and the coarse target range estimate is calculated, and the maximum deviation is obtained. This process is repeated for all N values. k After iterating through the velocity hypotheses, candidate velocity hypotheses are obtained based on the deviation constraint condition, namely, the deviation of the coarse and fine distance estimation results is less than the coarse distance resolution of the single linear frequency modulated continuous wave signal. Then, the velocity hypothesis with the smallest deviation when the peak energy is the largest is selected as the optimal velocity hypothesis for the target. Finally, the fine distance estimation result corresponding to the optimal velocity hypothesis is used as the distance estimate of the target, and the optimal velocity hypothesis is used as the velocity estimate of the target.

[0197] As an illustration of a possible specific implementation process, as mentioned above, assume that the precise distance estimation result of the target is as follows: Then the electronic equipment further calculates the target's speed assumption. Lower distance precise estimation results Compared with the coarse distance estimation results The deviation between them, and the deviation and The maximum value in:

[0198] Next, based on the target velocity assumption One-dimensional high-range resolution image synthesized below Extracting the peak energy maximum value :

[0199] Similarly, the same steps are performed for other different velocity assumptions of the target to obtain the maximum coarse-to-fine distance estimation bias under different velocity assumptions. and extracted peak energy .

[0200] Because the synthesized high-resolution range image obtained after phase compensation for the erroneous velocity (i.e., non-true target velocity) is out of focus and produces waveform distortion, an incorrect fine range estimate is obtained. This results in a large deviation from the coarse target range estimate and a very small peak energy. The fine target range estimate is performed based on the coarse estimate, aiming to achieve a range resolution that is unattainable with coarse range resolution (i.e., under the frequency modulation bandwidth of a single linear frequency modulated continuous wave signal). Therefore, the deviation between the fine estimate and the coarse estimate of the true target range must be smaller than the coarse range resolution. Using the above principle, the output of the true high-resolution target range and velocity estimate is as follows:

[0201]

[0202]

[0203] in, The maximum coarse-to-fine distance estimation bias for different target velocities is less than the system's coarse distance resolution. Location, To satisfy the requirement that the coarse-to-fine distance estimation bias is less than The position of the peak energy maximum value among the candidate velocities. This represents the location of the minimum coarse-to-fine distance estimation bias corresponding to the location of the peak energy maximum value. The final output is the location of the target velocity assumption and the precise estimation result.

[0204] Finally, the target velocity assumption and the fine distance estimation result corresponding to the maximum deviation between the coarse and fine estimates of the target distance that is less than the coarse distance resolution under the frequency modulation bandwidth of the linear frequency modulated continuous wave signal and the minimum deviation when the peak energy is maximum are taken as the final velocity estimate and distance estimate.

[0205] In practical applications, electronic devices can flexibly configure the output content according to actual business needs: when only motion parameters need to be obtained, the filtered target velocity estimate can be output separately; when only position parameters need to be obtained, the matched target distance estimate can be output separately; or the distance estimate and velocity estimate can be output simultaneously. The parameter output format can be adaptively set as needed, and this application does not limit this.

[0206] The method provided in this application first divides the echo signal after the mixed signal reflection into a first echo signal and a second echo signal, realizing the separation and adaptation of echo signals of different systems; then, based on the coarse estimation result of the first echo signal output and combined with the system's velocity measurement requirements, the second echo signal is compensated and the target distance is precisely estimated; finally, the deviation correlation between the precise distance estimation result and the corresponding coarse estimation result is used to flexibly determine the target's distance estimate and / or velocity estimate. This hierarchical and progressive limiting logic has a clear division of labor and a complete link. It relies on the coarse estimation to provide benchmark support for precise compensation, and then achieves adaptive parameter output through the coarse-precise deviation correlation. This ensures both the regularity and feasibility of the signal processing flow, and lays a reliable logical foundation for the subsequent accurate selection of target distance and velocity parameters.

[0207] Furthermore, this application adopts a hybrid signal system combining conventional linear frequency modulated continuous wave (LFM) signals and frequency-stepped LFM signals. Coupled with a complete processing chain of coarse estimation, fine correction, and deviation optimization, it achieves both the stability of coarse estimation and the accuracy of high-resolution ranging without requiring additional hardware detection equipment or complex calculation modules. Simultaneously, it is compatible with multiple modes, including separate output of distance parameters, separate output of velocity parameters, or simultaneous output of both parameters, adapting to different detection scenarios and customized output requirements. This effectively improves the accuracy, anti-interference capability, and engineering applicability of target parameter estimation.

[0208] As an explanation of the effects of the method in this application, regarding the aforementioned Figure 3A The mixed signal shown is assumed to contain the number of conventional linear frequency modulated continuous wave signals. The number is 512, which represents the number of frequency-stepped linear frequency modulated continuous wave signals. The step count is 32, meaning the number of steps is 32. The maximum effective range is required to be 380m under a real-sampling ADC (which only acquires the amplitude information of the echo signal, without retaining phase information, suitable for scenarios where velocity calculation is not required). High distance resolution of 0.2m The required synthetic bandwidth The frequency step size between different step chirs in the SF-LFMCW group is 750MHz. The maximum unambiguous distance corresponding to the IFFT of a single linear frequency modulated continuous wave signal is 23.4375MHz. The calculated value is 6.4m, which ensures that no ambiguous range gratings are generated within twice the single Chirp range resolution, centered on the coarsely estimated range result, i.e., within the range of fine search determined on the true one-dimensional synthetic high range resolution image of the target.

[0209] This application, through the design of hybrid linear frequency modulated continuous wave transmission waveform, can solve the problems of inaccurate phase compensation of moving targets, poor real-time performance of distance-velocity estimation, difficulty in meeting the requirements of the maximum unambiguous velocity measurement range, and low frame data update rate in the existing technology when processing stepped frequency signals.

[0210] Furthermore, to verify the detection capability of the method of this application for high-speed moving targets, this application sets up two sets of high-speed moving target experiments with the same speed but different distances, as well as a set of speed estimation experiments under the system speed measurement range requirements.

[0211] Experiments 1 and 2 are high-speed moving target experiments with the same speed but different distances. Assuming the radar transmit carrier frequency is 77 GHz, the frequency modulation bandwidth of the single linear frequency modulated continuous wave signal is 200 MHz, the transmission repetition period is 31 μs, the corresponding range resolution is 0.75 m, the maximum unambiguous speed of the system is 31.42 m / s, and the required system velocity measurement range is -110 m / s to 55 m / s, requiring a 4-fold velocity extension. In Experiment 1, the two moving targets have a speed of -100 m / s, with distances of 80 m and 80.2 m respectively, and a signal-to-noise ratio of 10 dB for both. In Experiment 2, the two moving targets have a speed of 50 m / s, with distances of 200 m and 200.5 m respectively, and a signal-to-noise ratio of 10 dB for both. In the simulation, a 128-point windowed inverse Fourier transform is used to improve the frequency resolution; to reduce the loss of range and velocity accuracy due to the picket fence effect, a parabolic interpolation algorithm is used to compensate for the inherent bias in the Fourier transform frequency estimation.

[0212] In Experiment 1, four velocity hypotheses were obtained using the method described in this application: -99.9271 m / s, -37.2367 m / s, 25.7636 m / s, and 88.6040 m / s. The corresponding coarse and fine distance estimation biases were 0.4688 m, 0.5625 m, 1.3750 m, and 1.3125 m, respectively, and the corresponding peak energies were 1.0, 0.0893, 0.0120, and 0.3748, respectively. Among them, the coarse and fine distance estimation bias of the velocity hypothesis -99.9271 m / s was 0.4688 m, which is less than the coarse distance resolution of 0.75 m, and the peak energy was the largest (1.0). The corresponding fine distance estimation result was [80 m, 80.25 m], which is consistent with the actual target distance of 80 m and 80.2 m. The velocity estimate of -99.9271 m / s is basically consistent with the actual target velocity of -100 m / s. The coarse and fine distance estimates of the other velocity assumptions all had deviations greater than or close to the coarse distance resolution, and the peak energy was small, so they were correctly excluded.

[0213] In Experiment 2, four velocity assumptions were obtained using the method of this application: -138.4418 m / s, -75.6014 m / s, 12.7610 m / s, and 50.0793 m / s. The corresponding coarse and fine distance estimation biases were 1.3125 m, 1.3125 m, 0.0938 m, and 0.4688 m, respectively, and the corresponding peak energies were 0.5417, 0.0122, 0.5517, and 1.0, respectively. The coarse-to-fine distance estimation bias for the velocity assumption of 50.0793 m / s is 0.4688 m, which is less than the coarse range resolution of 0.75 m, and the peak energy is the largest (1.0). The corresponding fine distance estimation results are [200.0312 m, 200.5312 m], which are consistent with the actual target distances of 200 m and 200.5 m. The velocity estimate of 50.0793 m / s is basically consistent with the actual target velocity of 50 m / s. The coarse-to-fine distance estimation bias for the velocity assumption of 12.7610 m / s is 0.0938 m, which is less than the coarse range resolution, but the peak energy is only 0.5517, which is less than the peak energy of the velocity assumption of 50.0793 m / s, so it is correctly excluded.

[0214] The simulation results show that using only a traditional single linear frequency modulated continuous wave signal, the limited bandwidth of the single signal limits the range resolution to 0.75m, making it impossible to distinguish the two sets of targets. Furthermore, because the target speed exceeds the system's maximum unambiguous speed of 31.42 m / s, the velocity measurement results become blurred. The actual velocity of a moving target at -100 m / s is measured as 25.5289 m / s, and the actual velocity of a moving target at 50 m / s is measured as -12.7645 m / s, significantly different from the true target speeds. However, using the method described in this application, accurate high-resolution range and velocity estimation were achieved for both sets of targets in the experiments.

[0215] Experiment 3 is a velocity estimation experiment under the system's velocity measurement range requirements. Assuming the radar transmit carrier frequency is 77 GHz, the transmit repetition period is 31 μs, the maximum unambiguous velocity of the system is 31.42 m / s, and the system's velocity measurement range requirement is -110 m / s to 55 m / s, requiring a 4-fold velocity extension. The target distance is set to 50 m, and the moving speed starts from -110 m / s and increases in steps of 5 m / s to 55 m / s. The signal-to-noise ratio of the target echo is 10 dB. Simulation results show that the method proposed in this application can accurately estimate the target velocity under the velocity measurement range requirements of a vehicle-mounted millimeter-wave radar system, and achieve precise motion phase compensation for the asynchronous phase signals of the frequency-stepped linear frequency modulated continuous wave signal, thereby enabling accurate high-resolution range imaging of high-speed moving targets.

[0216] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0217] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0218] The above embodiments introduce a signal processing method from the perspective of process flow. The following embodiments introduce a signal processing device from the perspective of virtual module or virtual unit. For details, please refer to the following embodiments.

[0219] This application also provides a signal processing apparatus for implementing the methods described in the above method embodiments. Figure 5 This is a schematic diagram of the structure of a signal processing device provided in an embodiment of this application, as shown below. Figure 5 As shown, in this embodiment, the signal processing device may include:

[0220] The determination module 51 is used to determine the first echo signal and the second echo signal based on the echo signal after the mixed signal is reflected by the target; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal.

[0221] The compensation module 52 is used to determine the precise range estimation result and peak energy under each velocity assumption by compensating the second echo signal and synthesizing a one-dimensional high-resolution range image. Each velocity assumption is determined based on the system velocity measurement requirements and the coarse estimation result obtained by processing the first echo signal.

[0222] The determination module 51 is also used to determine the target's distance estimate and / or velocity estimate from each velocity assumption based on the deviation between the fine distance estimate and the coarse distance estimate, as well as the peak energy.

[0223] In one possible implementation of this application embodiment, the compensation module 52 is specifically used for:

[0224] Based on constant false alarm rate detection of one-dimensional high-resolution range images under various velocity assumptions, the accurate distance estimation results under each velocity assumption are determined.

[0225] Peak energy is extracted from one-dimensional high-resolution range images under various velocity assumptions.

[0226] In one possible implementation of this application embodiment, the compensation module 52 is specifically used for:

[0227] Based on the target coarse range estimate image corresponding to the second echo signal, extract each step linear frequency modulated continuous wave signal within the fine search range; the fine search range is determined based on the coarse range estimate results.

[0228] For any velocity assumption, a two-dimensional high-resolution range image is obtained by performing phase compensation under the velocity assumption on each step linear frequency modulated continuous wave signal.

[0229] A one-dimensional high-resolution range image is synthesized by extracting and stitching two-dimensional high-resolution range images.

[0230] In one possible implementation of this application embodiment, the compensation module 52 is specifically used for:

[0231] Based on the velocity assumption, phase compensation processing is performed on each step linear frequency modulated continuous wave signal to obtain the compensated step linear frequency modulated continuous wave signal.

[0232] Based on the windowing and inverse Fourier transform processing of the compensated step linear frequency modulated continuous wave signal, two-dimensional high-resolution range images corresponding to different velocity assumptions are generated.

[0233] And / or,

[0234] Based on the same-distance discard method, the two-dimensional high-resolution range image under the velocity assumption is extracted to obtain the extracted range image;

[0235] A one-dimensional high-resolution range image is obtained by stitching the extracted range image along the frequency step dimension.

[0236] And / or,

[0237] Based on constant false alarm rate detection of one-dimensional high-resolution range images under various velocity assumptions, the detection peak value under each velocity assumption is obtained;

[0238] The distance corresponding to the detection peak under each velocity assumption is determined as the accurate distance estimation result under each velocity assumption.

[0239] In one possible implementation of this application embodiment, the determining module 51 is specifically used for:

[0240] Based on the deviation and coarse range resolution under each velocity assumption, candidate velocity assumptions are determined from each velocity assumption; candidate velocity assumptions are those with deviations smaller than the coarse range resolution.

[0241] Based on peak energy, the target's range and velocity estimates are determined from candidate velocity hypotheses.

[0242] In one possible implementation of this application embodiment, the determining module 51 is specifically used for:

[0243] Based on the distance estimation results and velocity assumptions corresponding to the velocity assumption with the largest peak energy among the candidate velocity assumptions, the distance estimate and velocity estimate of the target are determined.

[0244] In one possible implementation of this application embodiment, the coarse estimation result includes a speed coarse estimation result and a distance coarse estimation result; the determining module 51 is specifically used for:

[0245] Determine the corresponding speed expansion coefficient based on the system speed measurement requirements;

[0246] Based on the velocity expansion coefficient, the coarse velocity estimation results in the coarse estimation results are extended in intervals to generate multiple sets of different velocity assumptions.

[0247] In one possible implementation of this application embodiment, the determining module 51 is further configured to:

[0248] Based on the sampling sequence or intermediate frequency sampling sequence corresponding to each linear frequency modulated continuous wave signal, a coarse range profile corresponding to the target is obtained.

[0249] Based on the coarsely estimated range profile corresponding to the first echo signal, or the signal detection matrix constructed along the frequency modulation signal dimension from the coarsely estimated range profile, the coarse estimation result corresponding to the first echo signal is obtained.

[0250] In one possible implementation of this application embodiment, the number of conventional linear frequency modulated continuous wave signals in the mixed signal is greater than the number of frequency-stepped linear frequency modulated continuous wave signals; and / or, the determining module 51 is further configured to:

[0251] The mixed signal is generated based on a preset method, which includes a time-division switching method and / or an amplitude mixing method.

[0252] It transmits a mixed signal and receives the echo signal reflected by the target.

[0253] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0254] This application provides an electronic device. Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 6 As shown, Figure 6 The illustrated electronic device includes at least one processor 61 and a memory 62. The processor 61 and the memory 62 are connected, for example, via a bus 63. Optionally, the electronic device may also include a transceiver 64. It should be noted that in practical applications, the transceiver 64 is not limited to one, and the structure of this electronic device does not constitute a limitation on the embodiments of this application.

[0255] Processor 61 may be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 61 may also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0256] Bus 63 may include a pathway for transmitting information between the aforementioned components. Bus 63 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Bus 63 may be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0257] The memory 62 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0258] The memory 62 stores computer execution instructions for implementing the scheme of this application, and the processor 61 controls the execution. The processor 61 executes the computer execution instructions stored in the memory 62 to implement the content shown in the foregoing method embodiments.

[0259] This application also provides a computer-readable storage medium, which may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. Specifically, the computer-readable storage medium stores computer-executable instructions, which are used to implement the methods in the above embodiments.

[0260] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the technical solution of the above method embodiments. Its implementation principle and technical effects are similar, and will not be repeated here.

[0261] This application also provides a signal processing system, which includes a waveform generator, a radio frequency transmitter and receiver, and electronic equipment. The waveform transmitter generates a mixed signal; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal; the radio frequency transmitter and receiver transmits the mixed signal and receives the echo signal reflected from the target.

[0262] The electronic device is used to: determine the first echo signal and the second echo signal based on the echo signal after the mixed signal is reflected by the target; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal; for each velocity assumption, based on compensating the second echo signal and synthesizing a one-dimensional high-resolution range image, determine the fine range estimate and peak energy under each velocity assumption; each velocity assumption is determined based on the system velocity measurement requirements and the coarse estimate obtained by processing the first echo signal; based on the deviation between the fine range estimate and the coarse range estimate, and the peak energy, determine the target's range estimate and / or velocity estimate from each velocity assumption.

[0263] Specifically, the system of this application relies on hardware-level division of labor to implement the aforementioned signal processing logic. A waveform generator generates a mixed waveform signal in a preset combination form, and the signal transmission and echo acquisition are completed through a radio frequency transmitter and receiver. Then, the signal processing unit inside the electronic device performs splitting, coarse parameter calculation, phase compensation, fine estimation and solution, and deviation screening and judgment on the echo signal, and finally outputs the target distance and velocity parameters. The target detection process under mixed waveform is fully realized by relying on the hardware collaborative architecture.

[0264] It should be understood that the system architecture disclosed in this application is only an illustrative example. In actual applications, the integration method, arrangement, and chip selection of each hardware module can be flexibly adjusted. For example, the waveform generator and the RF transceiver structure can be integrated into the same RF chip, or discrete devices can be used to build the hardware link. At the same time, this application does not limit the specific hardware form, processor model, or packaging form of the electronic device. Any equivalent hardware architecture that can implement the signal processing logic of this application is within the protection scope of this application, and this application does not limit it.

[0265] The system provided in this application relies on a hybrid waveform hardware transmission architecture combined with a hierarchical signal processing flow, which can achieve long-distance detection, high-speed speed measurement and high-resolution ranging capabilities with a simple hardware structure. It has a high hardware reuse rate and low sampling pressure, which effectively reduces the cost and power consumption of radar hardware. At the same time, the system has strong adaptability and can be compatible with various vehicle autonomous driving scenarios, which facilitates mass production of vehicle radar and universal adaptation to vehicle models.

[0266] As a preferred example, when the above system is applied to vehicle radar scenarios, Figure 7 This is a schematic diagram of the structure of a signal processing system provided in an embodiment of this application, as shown below. Figure 7 As shown, the system specifically includes a control and configuration module, a waveform generator, an RF transmitter and receiving antenna array, an RF receiver and descrambling unit, an ADC sampling module, a signal processing core DSP+MCU, a storage medium, and an on-board interface module.

[0267] Among them, the control and configuration module serves as the central control hub of the architecture. Its core functions are to realize dynamic configuration of waveform parameters, monitoring of module status, and fault diagnosis. Specifically, it receives vehicle scenario commands issued by the MCU and generates corresponding conventional linear frequency modulated continuous wave and frequency step linear frequency modulated continuous wave waveform parameters, including frequency range, bandwidth, period, and frequency modulation slope.

[0268] The waveform generator is used to generate the radio frequency signal of the mixed waveform and the local oscillator signal required for deskewing, so as to realize real-time detection with long distance, high speed and high distance resolution. The module supports time-division switching or amplitude mixing of two waveforms to adapt to different vehicle detection scenarios. At the same time, it reuses its own radio frequency signal as the local oscillator signal of the deskewing mixer, eliminating the need for additional local oscillator source design, reducing the number of hardware components and reducing the difficulty of hardware design.

[0269] The radio frequency transmitter is used to amplify the mixed waveform signal to meet the transmission power requirements of the vehicle radar and ensure the maximum detection range. The antenna array is used to realize the directional transmission of radio frequency signals and the reception of target echoes to ensure the accuracy of radar angle measurement.

[0270] The radio frequency receiver and de-chewing processing unit are used to receive the target echo signal. The de-chewing processing reduces the high-frequency radio frequency signal to a low-frequency intermediate frequency, which greatly reduces the ADC sampling rate. The de-chewing mixer mixes the echo signal output from the low-noise amplifier with the local oscillator signal from the waveform generator to output the intermediate frequency signal. The frequency difference caused by the target distance is only at the kilohertz to megahertz level, which can reduce the ADC sampling rate from the gigahertz level to the tens of megahertz level, significantly reducing the hardware sampling cost.

[0271] The ADC sampling module is used to convert the de-skewed analog intermediate frequency signal into a digital signal. It matches the corresponding resolution and sampling rate of the analog-to-digital converter chip according to the frequency of the intermediate frequency signal, thus balancing sampling accuracy and hardware cost.

[0272] The signal processing core consists of a DSP processing unit and an MCU processing unit. The high-speed DSP processing unit undertakes high-intensity digital signal processing tasks, performs fast Fourier transform on digital signals to achieve time-frequency conversion, completes target coarse estimation of conventional linear frequency modulated continuous wave signals through constant false alarm rate detection and peak aggregation processing, and achieves high-resolution imaging and speed solution of frequency step signals by relying on inverse fast Fourier transform. The MCU control processing unit is responsible for low-complexity tasks such as module collaborative control, vehicle data interaction, low power management, and fault handling.

[0273] The storage medium is used to store radar calibration parameters, program firmware, processing results and fault logs to meet the requirements of data retention and reuse; the vehicle interface module is used to realize bidirectional data interaction between the radar and the vehicle system, adapting to the autonomous driving vehicle ecosystem.

[0274] The aforementioned system architecture, through deskewing processing, can reduce the sampling rate from GHz to tens or even several MHz levels, significantly reducing ADC cost and power consumption. By employing integrated RF chips, low-to-medium speed ADCs, and MIMO antennas, the number of peripheral components is reduced, greatly simplifying wiring and thus lowering hardware design complexity. Furthermore, the collaborative work of DSP and MCU avoids the high power consumption of a single high-performance processor, and the hybrid waveform adapts to various advanced driver assistance systems and autonomous driving scenarios, such as adaptive cruise control, automatic emergency braking, and blind spot detection, without requiring hardware replacement; only software configuration of waveform parameters is needed, thereby reducing vehicle adaptation costs. Therefore, the vehicle millimeter-wave radar signal processing method and system architecture based on hybrid linear frequency modulated continuous wave provided by this invention are more suitable for high-performance, low-cost, and mass-producible automotive-grade millimeter-wave radar chips.

[0275] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0276] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0277] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A signal processing method, characterized in that, The method includes: The first echo signal and the second echo signal are determined based on the echo signal after the mixed signal is reflected by the target; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal. For each velocity assumption, based on compensating the second echo signal and synthesizing a one-dimensional high-resolution range image, the precise range estimation result and peak energy under each velocity assumption are determined; each velocity assumption is determined based on the system velocity measurement requirements and the coarse estimation result obtained by processing the first echo signal. Based on the deviation between the fine distance estimation result and the coarse distance estimation result, and the peak energy, the distance estimate and / or velocity estimate of the target are determined from each of the velocity assumptions.

2. The method according to claim 1, characterized in that, The determination of the distance estimation results and peak energy under each of the aforementioned velocity assumptions includes: Based on constant false alarm rate detection of the one-dimensional high-resolution range image under each of the aforementioned velocity assumptions, the accurate distance estimation result under each of the aforementioned velocity assumptions is determined; The peak energy is extracted based on the one-dimensional high-resolution range image under each of the aforementioned velocity assumptions.

3. The method according to claim 1 or 2, characterized in that, The compensation of the second echo signal and the synthesis of a one-dimensional high-resolution range image includes: Based on the target coarse range estimate image corresponding to the second echo signal, each step linear frequency modulated continuous wave signal within the fine search range is extracted; the fine search range is determined based on the coarse range estimate result. For any of the velocity assumptions, a two-dimensional high-resolution range image is obtained by performing phase compensation under the velocity assumptions on each of the step linear frequency modulated continuous wave signals. The one-dimensional high-resolution range image is synthesized by extracting and stitching the two-dimensional high-resolution range image.

4. The method according to claim 3, characterized in that, The process of obtaining a two-dimensional high-resolution range image based on phase compensation under the velocity assumption for each of the aforementioned stepped linear frequency modulated continuous wave signals includes: Based on the speed assumption, each of the step linear frequency modulated continuous wave signals is subjected to phase compensation processing to obtain the compensated step linear frequency modulated continuous wave signal. Based on the windowing and inverse Fourier transform processing of the compensated step linear frequency modulated continuous wave signal, two-dimensional high-resolution range images corresponding to different velocity assumptions are generated. And / or, extracting and stitching the two-dimensional high-resolution range image to synthesize the one-dimensional high-resolution range image, including: Based on the same-distance discard method, the two-dimensional high-resolution range image under the velocity assumption is extracted to obtain the extracted range image; The one-dimensional high-resolution range image is obtained by stitching the extracted range image along the frequency step dimension. And / or, based on constant false alarm rate detection of the one-dimensional high-resolution range image under each of the aforementioned velocity assumptions, determine the precise distance estimation result under each of the aforementioned velocity assumptions, including: Based on constant false alarm rate detection of the one-dimensional high-resolution range image under each of the aforementioned velocity assumptions, the detection peak value under each of the aforementioned velocity assumptions is obtained; The distance corresponding to the detection peak under each speed assumption is determined as the distance estimation result under each speed assumption.

5. The method according to claim 1 or 2, characterized in that, The step of determining the distance estimate and / or velocity estimate of the target from each of the velocity assumptions based on the deviation between the fine distance estimate and the coarse distance estimate, and the peak energy, includes: Based on the deviation and coarse range resolution under each of the velocity assumptions, candidate velocity assumptions are determined from each of the velocity assumptions; the candidate velocity assumptions are those whose deviation is less than the coarse range resolution. Based on the peak energy, the distance estimate and velocity estimate of the target are determined from the candidate velocity assumptions.

6. The method according to claim 5, characterized in that, The step of determining the distance estimate and velocity estimate of the target from the candidate velocity assumptions based on the peak energy includes: Based on the distance estimation result and velocity hypothesis corresponding to the velocity hypothesis with the largest peak energy among the candidate velocity hypotheses, the distance estimate and velocity estimate of the target are determined.

7. The method according to claim 1 or 2, characterized in that, Based on the system speed measurement requirements and the coarse estimation results obtained from processing the first echo signal, the speed assumptions are determined, including: Determine the corresponding speed expansion coefficient based on the system's speed measurement requirements; Based on the speed expansion coefficient, the coarse speed estimation result in the coarse estimation result is extended over a range to generate multiple sets of different speed assumptions.

8. The method according to claim 7, characterized in that, The method further includes: Based on the sampling sequence or intermediate frequency sampling sequence corresponding to each linear frequency modulated continuous wave signal, a coarse estimated range profile corresponding to the target is obtained. Based on the coarsely estimated range profile corresponding to the first echo signal, or the signal detection matrix constructed along the frequency modulation signal dimension from the coarsely estimated range profile, the coarse estimation result corresponding to the first echo signal is obtained.

9. The method according to claim 1 or 2, characterized in that, In the mixed signal, the number of conventional linear frequency modulated continuous wave signals is greater than the number of frequency step linear frequency modulated continuous wave signals; And / or, the method further includes: The mixed signal is generated based on a preset method, which includes a time-division switching method and / or an amplitude mixing method. The mixed signal is transmitted, and the echo signal reflected by the target is received.

10. A signal processing system, characterized in that, The system includes a waveform generator, a radio frequency transmitter and receiver, and electronic equipment; The waveform generator is used to generate a mixed signal; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal. The radio frequency transmitter and receiver are used to transmit the mixed signal and receive the echo signal reflected by the target; The electronic device is used to: determine a first echo signal and a second echo signal based on the echo signal after the mixed signal is reflected by the target; the mixed signal includes a conventional linear frequency modulated continuous wave signal and a frequency-stepped linear frequency modulated continuous wave signal; For each velocity assumption, based on compensating the second echo signal and synthesizing a one-dimensional high-resolution range image, the precise range estimation result and peak energy under each velocity assumption are determined; each velocity assumption is determined based on the system velocity measurement requirements and the coarse estimation result obtained by processing the first echo signal. Based on the deviation between the fine distance estimation result and the coarse distance estimation result, and the peak energy, the distance estimate and / or velocity estimate of the target are determined from each of the velocity assumptions.

Images