A method and system for lidar signal processing

Abstract

The application relates to the technical field of microwave radars, and discloses a laser radar signal processing method and system. The method comprises the following steps: acquiring laser radar system parameters; the laser radar system parameters comprise target distance resolution, laser pulse time width and pulse repetition frequency; a linear frequency modulation radio frequency driving signal is generated according to the laser radar system parameters, and continuous laser is optically frequency-modulated through an electro-optical modulator to output a linear frequency modulation laser signal; receiving echo signals of the linear frequency modulation laser signal after atmospheric scattering, processing the echo signals through a multi-stage adaptive pulse compression processing link, and obtaining optimized wind field parameters. The application can realize synchronous improvement of detection resolution and long-distance detection capability.

Description

Technical Field

[0001] This invention relates to the field of microwave radar technology, and specifically to a laser radar signal processing method and system. Background Technology

[0002] With its advantages of high precision, non-contact operation, and high spatiotemporal resolution, pulsed laser wind radar has become a core remote sensing device for scenarios such as atmospheric boundary layer wind fields, upper-level wind profiles, atmospheric turbulence, and wind resource assessment. By emitting laser pulses to irradiate atmospheric aerosol particles, receiving backscattered echo signals, and calculating Doppler frequency shift, it can accurately detect atmospheric wind speed, wind direction, and layer distance information.

[0003] Currently available pulsed laser wind radars all employ traditional fixed pulse modulation mechanisms. This mechanism inherently possesses a strong coupling characteristic between time-width and bandwidth, leading to an irreconcilable technical contradiction between range resolution and long-range detection capability. Under the traditional fixed pulse mechanism, the range resolution of the laser radar is entirely determined by the transmitted pulse duration, and the signal bandwidth is passively bound to the pulse duration, lacking independent adjustability and constrained by an insurmountable mechanism.

[0004] Under traditional fixed-pulse modulation mechanisms, pulse duration and signal bandwidth are inherently strongly coupled, resulting in insurmountable performance constraints. Traditional mechanisms lack independent bandwidth adjustment capabilities; the signal bandwidth is strictly determined by the reciprocal of the pulse duration, thus range resolution is entirely dependent on the pulse duration. Narrow pulses correspond to wider natural signal bandwidth and higher range resolution, but their single-pulse emission energy is extremely low, making long-range, high-altitude weak wind field echoes easily submerged by system noise and turbulence noise, resulting in short effective detection range and insufficient weak signal detection capability. Wide pulses can increase single-pulse integration energy and extend detection range, but passively compress the effective signal bandwidth and degrade range resolution, making fine atmospheric stratification detection impossible. This duration-bandwidth coupling bottleneck is an inherent defect of the detection system and cannot be overcome through conventional parameter tuning, backend signal accumulation, or filtering optimization software methods, thus hindering the synchronous upgrading of the overall detection performance of pulsed laser wind radar in the long term.

[0005] From the perspective of the current state of industry technology, there are already mature solutions in the field of microwave radar to solve the problem of strong coupling between time and bandwidth. However, this technology cannot be directly applied to laser coherent detection scenarios. The two have fundamental differences in detection mechanism and signal physical properties: laser wind radar belongs to a narrow linewidth and high coherence detection mechanism, which has inherent characteristics such as short coherence duration, weak echo signal, and sensitivity to atmospheric turbulence. If the general fixed frequency modulation and open-loop filtering logic of microwave is directly applied, it will cause problems such as laser signal spectrum leakage, waveform distortion, and coherence attenuation, which will seriously degrade the detection accuracy and completely fail to meet the requirements of high-precision laser wind field detection. At present, there is no mature technical solution in the industry to solve the problem of strong coupling between time and bandwidth of laser radar. Summary of the Invention

[0006] To address the aforementioned shortcomings in existing technologies, this invention provides a lidar signal processing method and system. This invention abandons the traditional fixed pulse modulation system and the direct transplantation paradigm of general microwave linear frequency modulation. Through an integrated solution of "partial hardware supplementation and adaptation + laser-specific frequency modulation system innovation + adaptive pulse compression signal processing," it breaks through the temporal and bandwidth coupling constraints of the laser signal at the root of the detection system. This solves industry pain points such as the mutual exclusion of resolution and detection distance, low accuracy in weak wind fields, limited dynamic range, and poor turbulence adaptability in existing equipment, achieving a simultaneous leap in high-precision layered detection and long-distance weak wind field detection performance.

[0007] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows:

[0008] In a first aspect, the present invention proposes a laser radar signal processing method, comprising the following steps:

[0009] Acquire the parameters of the lidar system; the lidar system parameters include target range resolution, laser pulse duration, and pulse repetition frequency.

[0010] Based on the parameters of the lidar system, a linear frequency modulated radio frequency drive signal is generated, and the continuous laser is optically modulated by an electro-optic modulator to output a linear frequency modulated laser signal.

[0011] The echo signal of the linear frequency modulated laser signal after atmospheric scattering is received, and processed through a multi-stage adaptive pulse compression processing link to obtain optimized wind field parameters. The multi-stage adaptive pulse compression processing link includes: constructing a time-domain phase compensation function based on the refined Doppler frequency offset estimate to perform phase compensation on the echo signal; dynamically calculating the maximum effective coherent accumulation pulse number based on the laser linewidth and pulse repetition frequency, and using a 50% overlap sliding window strategy to perform coherent accumulation on the phase-compensated echo signal; constructing an optimized transmitted signal by Hamming window weighting of the linear frequency modulated laser signal, and real-time calling the optimal filter from a broadband filter bank covering the entire wind speed Doppler frequency offset range for pulse compression processing to suppress range sidelobes to below a preset threshold.

[0012] Furthermore, generating a linear frequency modulated radio frequency drive signal based on the lidar system parameters includes:

[0013] Calculate the required signal bandwidth based on the target distance resolution;

[0014] The frequency modulation slope is determined based on the laser pulse duration and the signal bandwidth.

[0015] The linear frequency modulated radio frequency drive signal is obtained by generating a direct digital synthesis control sequence based on the frequency modulation slope and the pulse repetition frequency.

[0016] Furthermore, after generating the direct digital synthesis control sequence, the process also includes: performing phase continuity correction between adjacent pulses.

[0017] Furthermore, optical frequency modulation of continuous laser light using an electro-optic modulator includes:

[0018] The linear frequency modulated laser signal is synchronized with the trigger timing of the electro-optic modulator.

[0019] Furthermore, before performing phase compensation on the echo signal, the method further includes:

[0020] A slow-time-dimensional fast Fourier transform is performed on the echo signal to roughly estimate the Doppler frequency shift caused by atmospheric wind field motion.

[0021] Furthermore, after roughly estimating the Doppler frequency shift, it also includes:

[0022] A Kalman filter is constructed based on a two-dimensional state-space model to refine the coarsely estimated Doppler frequency offset, resulting in a refined Doppler frequency offset estimate. The state vector of the two-dimensional state-space model includes the Doppler frequency offset value and its rate of change.

[0023] Furthermore, after obtaining the refined Doppler frequency offset estimate, the following is also included:

[0024] A time-domain phase compensation function is constructed based on the refined Doppler frequency offset estimate;

[0025] Phase compensation of the echo signal is performed using the aforementioned time-domain phase compensation function.

[0026] Furthermore, the coherent accumulation of the phase-compensated echo signal using a 50% overlap sliding window strategy includes:

[0027] Construct a sliding window based on the maximum effective coherent accumulation pulse number;

[0028] The sliding window is slid with a 50% overlap step size, and the echo signals within the window are coherently accumulated.

[0029] Furthermore, the pulse compression process involves real-time selection of the optimal filter from a broadband filter bank covering the entire wind speed Doppler frequency offset range, including:

[0030] Based on the preset Doppler frequency offset distribution step size, a filter bank covering the entire wind speed range is constructed;

[0031] Based on the real-time estimated Doppler frequency offset, the optimal filter with the center frequency matched is selected from the filter bank for pulse compression.

[0032] Secondly, this invention proposes a lidar signal processing system, which applies a lidar signal processing method as described above, including:

[0033] The parameter acquisition module is used to acquire the parameters of the lidar system; the lidar system parameters include target range resolution, laser pulse duration, and pulse repetition frequency.

[0034] The signal generation module is used to generate a linear frequency modulated radio frequency drive signal based on the parameters of the lidar system.

[0035] The electro-optic modulation control module is used to optically modulate the continuous laser through an electro-optic modulator and output a linear frequency modulated laser signal.

[0036] A multi-level processing module is used to receive the echo signal of the linear frequency modulated laser signal after atmospheric scattering, and process the echo signal through a multi-level adaptive pulse compression processing link to obtain optimized wind field parameters. The multi-level adaptive pulse compression processing link includes: constructing a time-domain phase compensation function based on the refined Doppler frequency offset estimate to perform phase compensation on the echo signal; dynamically calculating the maximum effective coherent accumulation pulse number according to the laser linewidth and pulse repetition frequency, and using a 50% overlap sliding window strategy to perform coherent accumulation on the phase-compensated echo signal; constructing an optimized transmitted signal by Hamming window weighting of the linear frequency modulated laser signal, and calling the optimal filter in real time from a broadband filter bank covering the entire wind speed Doppler frequency offset range for pulse compression processing to suppress range sidelobes to below a preset threshold.

[0037] The present invention has the following beneficial effects:

[0038] (1) This invention breaks through the inherent limitation of traditional fixed-pulse lidar, which cannot simultaneously achieve both resolution and detection range. By employing a novel laser-specific linear frequency modulation system, it achieves complete decoupling of pulse width and bandwidth, thus eliminating the rigid constraint of pulse width on resolution. While retaining the original long-range detection capability, it achieves at least a tenfold improvement in detection accuracy compared to traditional pulse lidar under the same pulse width. Under the same resolution conditions, the effective detection range of the lidar is increased by more than 37% compared to the traditional system, completely resolving the long-standing technical pain point of mutually exclusive dual performance in the industry, and achieving a leapfrog upgrade in comprehensive detection performance.

[0039] (2) This invention abandons the general microwave frequency modulation and signal processing paradigm and constructs a dedicated frequency modulation control and signal processing system adapted to the characteristics of narrow linewidth, high coherence and short coherence duration of laser. It avoids problems such as spectrum leakage, waveform distortion and coherence attenuation from the source of signal generation, and ensures the stability and accuracy of laser coherent detection. It solves the industry barrier that general microwave technology cannot be adapted to high-precision laser wind field detection.

[0040] (3) The present invention adopts an adaptive coherent accumulation strategy that matches the coherent characteristics of lasers, effectively suppresses atmospheric turbulent random noise, and significantly improves the signal-to-noise ratio of weak echo signals without zero resolution loss; combined with low sidelobe pulse compression filtering technology, the range sidelobe is suppressed to below -30dB, completely eliminating the problem of strong short-range scattering signals masking weak long-range signals, and greatly improving the radar detection dynamic range and the ability to identify weak wind fields in complex atmospheric environments.

[0041] (4) This invention relies entirely on the existing mature hardware architecture, optical circuits and core components of pulsed laser wind radar. No structural modifications or replacement of core components are required. Performance iteration can be completed simply by adding local functional modules and upgrading software algorithms and timing logic. The hardware modifications are minimal and the equipment compatibility is strong, which greatly reduces the hardware cost and engineering deployment difficulty of technology upgrades. It has strong engineering feasibility and large-scale promotion value.

[0042] (5) This invention pioneers a full-link technical solution adapted to laser wind field detection, consisting of "dedicated frequency modulation transmission + dynamic mismatch compensation + adaptive noise suppression + low sidelobe high-precision demodulation." It specifically addresses the unique technical challenges of lidar compared to microwave radar, establishing a novel detection system distinct from traditional fixed-pulse systems and microwave technology transplantation schemes. This provides a standardized and feasible technical path for high-precision, long-distance, and highly dynamic laser atmospheric wind field detection. This solution is not a simple patchwork of existing known technologies; rather, the various technical links are deeply coupled and synergistically adapted, breaking through conventional technical modification approaches in this field, filling a technological gap in the industry, and possessing outstanding originality and engineering promotion value. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of a laser radar signal processing method according to the present invention;

[0044] Figure 2 This is a schematic diagram of the transmitting section of a traditional pulse lidar.

[0045] Figure 3 This is a schematic diagram of the transmitting part of the linear frequency modulated pulse lidar of the present invention. Detailed Implementation

[0046] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0047] The core innovative principle of this invention lies in:

[0048] While fully retaining the existing radar's optical path, circuitry, mechanical architecture, and core detection devices, an EOM electro-optic frequency modulation module is added to supplement the laser optical frequency modulation capability, abandoning the traditional fixed pulse system and microwave universal linear frequency modulation adaptation scheme.

[0049] To address the unique detection characteristics of lasers—high coherence, short coherence duration, and turbulence sensitivity—a laser-specific linear frequency modulated (LFM) transmission and adaptive pulse compression reception system was constructed, overcoming the strong coupling defects of traditional pulse radar in terms of time-bandwidth ratio at the systemic level. The transmitter independently configures the pulse duration and signal bandwidth, breaking the traditional constraint that "pulse width determines resolution," achieving decoupling of time-bandwidth ratio and independent parameter control. The receiver uses multi-level signal optimization algorithms to correct signal distortion and suppress noise interference, ultimately achieving a simultaneous improvement in range resolution and long-range detection capability.

[0050] This invention is not a simple aggregation of known technologies such as microwave frequency modulation, Doppler compensation, pulse compression, and adaptive accumulation, and is completely different from the industry's conventional technology stacking solutions. Existing known modules are all general technologies for microwave radar, without any adaptation or modification for the specific characteristics of laser wind measurement radar. If they are directly applied individually or simply combined, fatal problems such as laser coherence attenuation, spectrum leakage, pulse compression mismatch, and noise redundancy will inevitably occur, making them unsuitable for practical application. The core innovation of this invention is not the modules themselves, but rather the customized design of adaptation parameters, linkage logic, timing matching, and closed-loop correction mechanisms for each module, specifically targeting the unique properties of lasers such as short coherence duration, high coherence, turbulence sensitivity, and weak echo detection. This reconstructs a laser-specific technical architecture that is completely different from microwave systems. The modules are deeply coupled and synergistically empowered, achieving technical effects that cannot be achieved by a single known module or a simple combination of modules.

[0051] like Figure 1 As shown, an embodiment of the present invention provides a laser radar signal processing method, including the following steps S1 to S3:

[0052] S1. Obtain the parameters of the lidar system; the lidar system parameters include target range resolution, laser pulse duration, and pulse repetition frequency.

[0053] In an optional embodiment of the present invention, step S1 obtains the set or automatically matched distance resolution. Laser pulse duration Parameters such as pulse repetition frequency (PRF) are included. Among them, the target range resolution... This determines the effective bandwidth of the required frequency modulation signal.

[0054] S2. Generate a linear frequency modulated radio frequency drive signal according to the parameters of the lidar system, and perform optical frequency modulation on the continuous laser through an electro-optic modulator to output a linear frequency modulated laser signal;

[0055] In an optional embodiment of the present invention, step S2, generating a linear frequency modulated radio frequency drive signal based on the lidar system parameters, includes:

[0056] Calculate the required signal bandwidth based on the target distance resolution;

[0057] The frequency modulation slope is determined based on the laser pulse duration and the signal bandwidth.

[0058] The linear frequency modulated radio frequency drive signal is obtained by generating a direct digital synthesis control sequence based on the frequency modulation slope and the pulse repetition frequency.

[0059] After generating the direct digital synthesis control sequence, the process also includes: performing phase continuity correction between adjacent pulses.

[0060] This embodiment first calculates the required frequency modulation bandwidth B based on the target resolution:

[0061] ;

[0062] Where ΔR is the range resolution of the lidar, referring to the minimum vertical distance between adjacent atmospheric sounding layers that the lidar can distinguish; c is the speed of light in a vacuum; and B is the bandwidth of the transmitted pulse signal. For example, when ΔR = 15 m, the calculated value is... .

[0063] Then, the frequency modulation slope is determined based on the laser pulse duration and the signal bandwidth. :

[0064] ;

[0065] For example, when When B = 1µs and B = 10MHz, the calculation yields... .

[0066] Next, the pulse width According to the DDS direct digital frequency synthesizer system clock cycle Discretize and segment to obtain The time step. time steps ( ), calculate the corresponding instantaneous frequency :

[0067] ;

[0068] in, Center frequency, Frequency modulation slope, Time step number, The clock cycle is the DDS clock cycle.

[0069] Instantaneous frequency Converted to DDS-recognizable frequency control word :

[0070] ;

[0071] in, The number of bits for the DDS phase accumulator (typically 32 bits). This is the DDS reference clock frequency. The calculated... Construct a length of An integer array, namely the DDS frequency control word sequence, serves as the digital waveform template for the linear frequency modulated laser signal.

[0072] Due to factors such as trigger delay and gating jitter in real-world systems, phase discontinuities may occur between adjacent pulses, leading to spectral spread. To prevent phase jumps between adjacent laser pulses that could cause spectral leakage and coherence degradation, this embodiment introduces a phase inheritance mechanism. The final phase value output by the DDS at the end of the previous pulse is recorded. and the starting phase of the current pulse Set to:

[0073] ;

[0074] in, This indicates modulo operation on the phase value. Operations to make it fall into Within the interval. Before triggering the current pulse, the initial phase register of the DDS is loaded. This achieves phase continuity across pulses.

[0075] In an optional embodiment of the present invention, step S2, which modulates the continuous laser frequency using an electro-optic modulator, includes:

[0076] The linear frequency modulated laser signal is synchronized with the trigger timing of the electro-optic modulator.

[0077] This embodiment also generates a pulse synchronization signal that is strictly synchronized with the linear frequency modulation signal. This signal is used to coordinate the timing of the EOM intensity modulator, the ADC sampling clock at the receiver, and the data acquisition system, ensuring that the operation error of each module in the system is controlled within ±5ns.

[0078] Through the above process, this embodiment realizes a linear frequency modulated signal generation method customized for pulsed laser wind radar application scenarios. It not only ensures the high accuracy and stability of the frequency modulated signal, but also effectively maintains the high coherence characteristics of the laser signal, providing a high-quality front-end signal guarantee for subsequent adaptive pulse compression processing.

[0079] This embodiment can flexibly adapt waveform parameters according to the detection scenario. The following are exemplary core parameters and signal expressions adapted to conventional atmospheric wind field detection, and are not considered as the only limiting parameters: linear frequency modulated signal bandwidth B, pulse width Tp, and radio frequency center frequency. It can be configured as needed, and the frequency modulation slope satisfies K=B / Tp. The time-domain expression of the linear frequency modulated transmission signal constructed by this invention is as follows:

[0080] ;

[0081] in, Let be the time-domain amplitude of the linearly frequency-modulated laser emission signal at time t; A is the amplitude of the laser emission signal, representing the laser emission intensity; exp is the natural exponential function; j is the imaginary unit, used to characterize the signal phase characteristics; is the radio frequency center frequency, and is the reference frequency of the laser frequency modulation signal; t is the time variable.

[0082] Traditional fixed-pulse modulation schemes inherently suffer from strong coupling between time and bandwidth. The effective bandwidth of the pulse signal is entirely determined by the pulse duration. The formula for calculating the binding relationship between the two is as follows:

[0083] ;

[0084] Substituting the above formula into the basic resolution formula From this, we can derive the limit calculation formula for the resolution being completely constrained by the pulse width under the traditional fixed pulse system:

[0085] ;

[0086] To intuitively verify the inherent defects of the traditional mechanism, and in conjunction with the industry-standard 1μs fixed pulse working condition, quantitative calculations were performed by substituting the values ​​into the above formula:

[0087] ;

[0088] Substituting the above calculations into the conventional 1μs pulse width operating condition, the quantitative solution yields a maximum range resolution of 150m for the traditional fixed pulse lidar, which directly demonstrates the inherent defects of the traditional system, such as time-width-bandwidth coupling and performance mutual exclusion.

[0089] This embodiment introduces a laser-specific linear frequency modulation (LFM) system, completely eliminating the strong coupling constraint of pulse duration and bandwidth in traditional fixed pulse modulation systems. This achieves complete decoupling of pulse duration and signal bandwidth, with independently controllable parameters. Under the LFM pulse compression system of this embodiment, distance resolution is no longer limited by pulse duration but is instead uniquely determined by the independently configured signal bandwidth. This completely overturns the mechanistic constraints of traditional fixed pulses, fundamentally solving the industry pain point of mutually exclusive resolution and detection range performance in traditional systems. It allows for refined distance detection by independently configuring a large bandwidth while retaining a large 1μs pulse duration and ensuring energy for long-distance detection.

[0090] To quantitatively verify the theoretical basis for the resolution improvement in this embodiment, a bandwidth calculation formula is derived based on a transformation of the basic distance resolution formula, serving as the core theoretical support for the high-precision detection design in this embodiment:

[0091] ;

[0092] Set high-precision target range resolution Substituting into the above formula, we can quantitatively derive the required signal bandwidth:

[0093] ;

[0094] Substituting the high-precision range resolution parameters of a 15m target into the above calculations, the minimum signal bandwidth required for this embodiment is 10MHz, obtained through quantization. The derivation results verify that this embodiment can achieve a fine-grained detection resolution of 15m with an independently configured 10MHz bandwidth while retaining the 1μs high-energy pulse duration, representing a tenfold improvement in accuracy compared to traditional systems.

[0095] S3. Receive the echo signal of the linear frequency modulated laser signal after atmospheric scattering, and process the echo signal through a multi-level adaptive pulse compression processing link to obtain optimized wind field parameters; wherein, the multi-level adaptive pulse compression processing link includes: constructing a time-domain phase compensation function based on the refined Doppler frequency offset estimate to perform phase compensation on the echo signal; dynamically calculating the maximum effective coherent accumulation pulse number according to the laser linewidth and pulse repetition frequency, and using a 50% overlap sliding window strategy to perform coherent accumulation on the phase-compensated echo signal; constructing an optimized transmitted signal by Hamming window weighting of the linear frequency modulated laser signal, and calling the optimal filter in real time from a broadband filter bank covering the entire wind speed Doppler frequency offset range for pulse compression processing to suppress the range sidelobes to below a preset threshold.

[0096] In an optional embodiment of the present invention, step S3, before performing phase compensation on the echo signal, further includes:

[0097] A slow-time-dimensional fast Fourier transform is performed on the echo signal to roughly estimate the Doppler frequency shift caused by atmospheric wind field motion.

[0098] Step S3, after roughly estimating the Doppler frequency offset, also includes:

[0099] A Kalman filter is constructed based on a two-dimensional state-space model to refine the coarsely estimated Doppler frequency offset, resulting in a refined Doppler frequency offset estimate. The state vector of the two-dimensional state-space model includes the Doppler frequency offset value and its rate of change.

[0100] Step S3, after obtaining the refined Doppler frequency offset estimate, also includes:

[0101] A time-domain phase compensation function is constructed based on the refined Doppler frequency offset estimate;

[0102] Phase compensation of the echo signal is performed using the aforementioned time-domain phase compensation function.

[0103] Step S3 employs a 50% overlap sliding window strategy to coherently accumulate the phase-compensated echo signal, including:

[0104] Construct a sliding window based on the maximum effective coherent accumulation pulse number;

[0105] The sliding window is slid with a 50% overlap step size, and the echo signals within the window are coherently accumulated.

[0106] Step S3 involves real-time selection of the optimal filter from a broadband filter bank covering the entire wind speed Doppler frequency offset range for pulse compression processing, including:

[0107] Based on the preset Doppler frequency offset distribution step size, a filter bank covering the entire wind speed range is constructed;

[0108] Based on the real-time estimated Doppler frequency offset, the optimal filter with the center frequency matched is selected from the filter bank for pulse compression.

[0109] First, this embodiment, based on the fundamental mechanism of pulse compression, theoretically breaks through the performance bottleneck of traditional systems, achieving an intrinsic gain improvement in signal-to-noise ratio and detection range, which is the core theoretical support for the entire receiving technology. The core principle of pulse compression is to utilize the energy accumulation characteristics of the large time-bandwidth product linear frequency modulated signal at the transmitting end to coherently accumulate the weak echo energy dispersed within the time domain of a long pulse. This completely breaks the inherent limitations of traditional fixed-pulse systems, where the time-bandwidth product is approximately 1 and there is no signal gain. The corresponding pulse compression signal-to-noise ratio gain quantization formula is as follows:

[0110] ;

[0111] in, For pulse compression signal-to-noise ratio gain; denoted as the laser pulse emission time width; B is the effective bandwidth of the linear frequency modulated laser signal; lg is a common logarithmic operation with base 10.

[0112] Traditional fixed-pulse systems are constrained by strong coupling between pulse duration and bandwidth, making signal gain amplification impossible. Detection performance is entirely limited by the single-pulse transmission energy, and weak signals at long distances are easily drowned out by noise. This embodiment innovates the transmitter system to achieve complete decoupling of pulse duration and signal bandwidth. It allows for independent configuration of a large duration to ensure transmission energy and a large bandwidth to ensure detection accuracy, constructing a duration-bandwidth product much greater than 1. Based on the pulse compression energy concentration mechanism described above, it can effectively concentrate weak echo energy, improve the signal-to-noise ratio, and enhance detection sensitivity.

[0113] In a lidar coherent detection system, the maximum effective detection range is determined by the echo signal-to-noise ratio threshold. According to classical radar detection theory, there is a fixed mathematical relationship between the detection range and the signal-to-noise ratio gain, which is the core theoretical basis for the long-range performance improvement in this embodiment.

[0114] ;

[0115] in, The maximum effective detection range of the radar is represented by ∝; ∝ indicates a proportional relationship. This represents the pulse compression signal-to-noise ratio gain. Under the premise that the hardware conditions such as radar transmit power, antenna parameters, target scattering characteristics, and system losses remain unchanged, this formula indicates that in a coherent detection system of lidar, the maximum detection range is directly proportional to the fourth root of the signal-to-noise ratio gain. The higher the signal-to-noise ratio gain, the higher the probability of detecting weak signals at long distances, and the farther the effective detection range.

[0116] In traditional fixed-pulse mode, the bandwidth product is approximately 1, resulting in a signal-to-noise ratio gain. This embodiment uses a matching condition with a 1μs pulse duration and a 10MHz signal bandwidth, and the time-width-bandwidth product... Substituting into the following formula, we can obtain the signal-to-noise ratio gain:

[0117] ;

[0118] ;

[0119] Substituting the above calculations into the operating parameters of the bandwidth product 10 in this embodiment, the quantization yields a pulse compression signal-to-noise ratio gain of up to 10 dB, verifying the significant signal-to-noise ratio improvement advantage of this embodiment compared to the traditional zero-gain system. Based on the signal-to-noise ratio gain quantization results, combined with the fourth root proportionality relationship of radar detection range, the detection range of this embodiment is found to be approximately 37% higher than that of the traditional system.

[0120] Based on the theoretical gain derivation above, it can be seen that pulse compression can achieve a significant improvement in signal-to-noise ratio and detection range. To avoid signal distortion, noise redundancy, and sidelobe interference offsetting the theoretical performance gain in complex atmospheric detection scenarios, and to ensure the stable implementation of the advantages of pulse compression technology, this embodiment constructs a multi-level collaborative signal optimization link. Through a three-level optimization mechanism of signal fidelity correction, adaptive noise reduction gain, and sidelobe interference suppression, the signal quality is improved step by step, the detection performance is strengthened, and the stable output of pulse compression performance is guaranteed in all aspects.

[0121] To address issues such as phase mismatch in linear frequency modulated echo signals, pulse compression main lobe broadening, and range resolution degradation caused by Doppler frequency shift due to dynamic atmospheric wind field movements, this embodiment provides a three-stage cascaded dynamic compensation mechanism: "FFT peak coarse estimation—Kalman filter fine optimization—time-domain phase compensation." This mechanism, through a progressive approach, first achieves rapid initial estimation of Doppler frequency shift, then performs time-series smoothing refinement, and finally completes accurate phase correction of the baseband echo signal, ensuring the matching and resolution stability of pulse compression across the entire wind speed range.

[0122] The specific implementation process of the three-level dynamic compensation method used in this embodiment is as follows:

[0123] First, baseband down-conversion preprocessing is performed to obtain a clean baseband echo signal.

[0124] The raw echo signal acquired by the radar receiver is a high-frequency carrier-modulated signal, expressed as:

[0125] ;

[0126] in, This is the echo amplitude attenuation factor; It is a delayed linear frequency modulated signal; The laser carrier frequency; This is due to Doppler frequency shift; For round-trip propagation delay; This refers to system noise and environmental clutter.

[0127] The signal is subjected to quadrature mixing and downconversion processing, and the local oscillator signal is... The complex baseband echo signal was obtained:

[0128] ;

[0129] in, This step removes residual noise after down-conversion. It eliminates the high-frequency carrier component, obtaining a low-frequency baseband signal containing only Doppler frequency shift information, which serves as input for subsequent Doppler estimation.

[0130] Then, FFT peak estimation is performed to achieve rapid initial judgment of Doppler frequency offset.

[0131] For a continuous The baseband echo data of each pulse are subjected to a slow-time FFT transform at the same distance gate to construct the slow-time echo matrix. ,in For distance gate index, This is the pulse sequence number.

[0132] Select target detection range gate Extract its slow time series to conduct Point FFT:

[0133] ;

[0134] Detection maximum value position The corresponding preliminary estimate of the Doppler frequency offset is:

[0135] ;

[0136] Where PRF is the pulse repetition frequency. To accumulate the number of pulses, this result is used as the initial observation input for the Kalman filter, exhibiting high noise but fast response.

[0137] Next, Kalman filtering is finely optimized to achieve time-series smoothing and trend prediction of Doppler frequency offset.

[0138] Construct a two-dimensional state-space model with the following state variables:

[0139] ;

[0140] in, For the current Doppler frequency shift, Its rate of change (acceleration term) reflects the trend of wind speed change.

[0141] The system state transition equation is:

[0142] ;

[0143] Wherein, the state transition matrix , The pulse repetition period (PRT) is the pulse repetition period. The noise is a process noise that follows a zero-mean Gaussian distribution.

[0144] The observation equation is:

[0145] ;

[0146] in, For the first Frame FFT coarse estimation, To observe noise.

[0147] The Kalman filter is executed according to the standard five-step recursive algorithm:

[0148] Predicted status: ;

[0149] Predicting covariance: ;

[0150] Calculate the Kalman gain: ;

[0151] Update status: ;

[0152] Update covariance: .

[0153] Output the optimal estimate:

[0154] ;

[0155] That is, the first-dimensional state quantity output by the Kalman filter is used as the final Doppler frequency bias estimate.

[0156] Finally, time-domain phase compensation is performed to complete the baseband echo signal repair.

[0157] Using precise Doppler frequency offset A phase compensation function is constructed to correct the original baseband echo signal point by point:

[0158]

[0159] in, This is the baseband echo signal after phase compensation; The original baseband echo signal after preprocessing; e is the natural constant; j is the imaginary unit; The value represents the estimated Doppler frequency offset caused by wind field motion; t is the time variable used to achieve accurate phase correction in the time dimension. This operation cancels out the linear phase rotation caused by wind speed, restores the time-frequency alignment characteristics of the signal, and ensures that it satisfies the conjugate matching condition with the local reference linear frequency modulated signal.

[0160] The compensated signal enters the subsequent adaptive coherent accumulation and pulse compression processing link, which significantly improves the sharpness of the main lobe of the pulse compression, reduces sidelobe interference, and ensures constant range resolution, laying a solid foundation for subsequent noise reduction optimization and high-precision demodulation.

[0161] Unlike microwave radar, which has a long coherence duration, lidar inherently has a short coherence duration. This makes traditional fixed-point accumulation modes highly susceptible to atmospheric turbulence noise, which can negate the theoretical gain of pulse compression. Addressing this specific challenge of lidar, this embodiment uses a bandwidth calculation formula combined with the laser linewidth and pulse repetition period to dynamically calculate the maximum effective number of coherent accumulation pulses. This precisely matches the physical boundary of the laser coherence duration, preventing noise redundancy caused by accumulation beyond the specified range.

[0162] ;

[0163] Where N_max is the maximum effective coherent accumulation pulse number of the lidar, which is the optimal accumulation number to adapt to the coherent characteristics of the laser; ⌊⌋ is the floor function operator, used to output the number of compliant integer pulses; π is a fixed constant of pi; Δν is the laser linewidth, which directly determines the coherent duration of the laser signal; PRT is the laser pulse repetition period, which refers to the time interval between the emission of adjacent laser pulses.

[0164] Simultaneously, a 50% overlap sliding window accumulation strategy is employed to maximize the purification of effective echo signals and suppress turbulent random noise without any resolution loss, further solidifying and amplifying the pulse compression signal-to-noise ratio gain. The specific steps of the 50% overlap sliding window accumulation strategy are as follows:

[0165] First of all, with Create a window with a length of [length]. Hamming window function:

[0166] ;

[0167] Then, the slow time series According to step size (i.e., 50% overlap) Sliding segmentation is performed to obtain several sub-sequence blocks:

[0168] ;

[0169] in, This represents the total number of sub-blocks.

[0170] Finally, for each sub-block Apply Hamming window and perform coherent accumulation:

[0171] ;

[0172] All This serves as input data for subsequent pulse compression processing. To generate a continuous wind field profile, adjacent... Perform interpolation or output directly as a time series.

[0173] This embodiment employs a 50% overlap sliding window accumulation strategy, maximizing the signal-to-noise ratio gain without sacrificing distance and temporal resolution. Traditional fixed-quantity accumulation modes require waiting for complete accumulation. The result is output only once every few pulses, resulting in a time resolution of 1 / 2. This makes it difficult to capture rapidly changing wind field dynamics; and non-overlapping processing also causes data discontinuity, introducing an equivalent loss of temporal resolution. This invention addresses this by setting the sliding step size to... (i.e., 50% overlap), so that each An accumulated result can be output in a short time, doubling the time resolution, and the output sequence is continuous, effectively preserving the transient characteristics of the wind field. More importantly, this strategy only changes the data output rhythm and does not introduce additional broadening or spectral leakage into the matched filtering process of pulse compression. The main lobe width after pulse compression is still determined by the signal bandwidth. The decision ensures that the distance resolution remains unchanged. At the same time, adjacent windows share some data, maintaining the original spectral concentration characteristics within the Fourier transform framework, without generating additional sidelobe interference or resolution degradation, thus essentially guaranteeing "zero resolution loss".

[0174] Simultaneously, signal-to-noise ratio gain was maximized without any resolution loss. Firstly, the maximum effective number of accumulated pulses... The laser linewidth and pulse repetition period are dynamically determined to ensure that all pulses involved in accumulation are within the same coherent time, fundamentally avoiding signal decoherence and signal-to-noise ratio degradation caused by time-out accumulation; based on this, [further details are needed]. Coherent averaging of each pulse can theoretically improve the signal-to-noise ratio. Secondly, the 50% overlap sliding window mechanism improves data utilization efficiency by 100%, outputting twice the effective samples compared to traditional non-overlapping accumulation within the same observation duration, significantly enhancing the statistical robustness of wind field parameter estimation. Simultaneously, the introduction of Hamming window weighting effectively suppresses spectral leakage at signal edges, further increasing the concentration of effective signal energy. The synergistic effect of these mechanisms not only fully leverages the theoretical gain potential of pulse compression but also minimizes the impact of random noise such as atmospheric turbulence by optimizing the data processing flow, thereby achieving a substantial and maximized improvement in signal-to-noise ratio gain.

[0175] After the first two stages of signal fidelity preservation and noise reduction gain optimization, signal accuracy and signal-to-noise ratio are effectively improved. However, the inherent problem of excessively high range sidelobes in conventional pulse compression still restricts the detection of weak signals at long distances, and the phenomenon of strong near-range scattering signals masking weak long-range echoes is prone to occur. To address this, this embodiment constructs a broadband filter bank covering the entire wind speed Doppler frequency offset range to adapt to the wind field detection needs of all scenarios. At the same time, Hamming window weighted optimization of the laser-specific linear frequency modulated reference signal is used to construct mismatch filtering to complete high-precision pulse compression processing, which can suppress the range sidelobes to below -30dB, completely solving the problem of near-far signal masking and significantly improving the weak wind field identification capability and detection dynamic range in complex scenarios. The specific process is as follows:

[0176] First, in this embodiment, a Hamming window weighting is applied to the linear frequency modulated signal at the transmitting end to construct an optimized transmitted signal:

[0177] ;

[0178] in, For Hamming window functions, .

[0179] In this embodiment, during the DDS waveform generation stage, the Hamming window coefficients are multiplied by the amplitude modulation portion of the linear frequency modulated signal to serve as the final output digital waveform template, ensuring that the transmitted signal itself is a weighted linear frequency modulated signal. Through Hamming window weighting, abrupt changes in the signal's time-domain envelope (Gibbs phenomenon) can be suppressed, reducing spectral leakage; the peak value of the range sidelobe after pulse compression can be weakened; and the filter's tolerance to Doppler frequency shift can be improved.

[0180] This embodiment then abandons traditional matched filtering and instead employs a mismatched filter. Its design goal is to suppress sidelobe levels to the maximum extent possible at the acceptable cost of main lobe widening. The desired compressed impulse response is defined as an ideal narrow impulse convolved with a low-sidelobe window function, for example:

[0181] ;

[0182] in, Taylor windows, Chebyshev windows, etc. can be selected, with the goal of making the side lobes ≤ -30 dB.

[0183] In the frequency domain, the frequency response of the mismatched filter Solve using the Wiener-Khinchin theorem or the least squares criterion:

[0184] ;

[0185] in, The spectrum of the weighted transmitted signal; To construct the spectrum of the desired response; This is a regularization factor used to suppress noise amplification.

[0186] Due to atmospheric wind field movement, Doppler frequency shift will occur. Directly using static filters can lead to mismatch. This invention constructs a broadband filter bank covering the entire wind speed range, based on the maximum detectable wind speed of the radar. (e.g., ±50 m / s), calculate the corresponding maximum Doppler frequency shift:

[0187] ;

[0188] exist Within the range, with step size (e.g., 100 Hz) generation Sub-filters Each corresponds to a center frequency offset. All filter coefficients are pre-calculated and stored in the signal processing module, forming a filter bank lookup table. λ is the laser wavelength.

[0189] The real-time filtering process in this embodiment is as follows: The receiver first obtains the Doppler frequency offset of the current range gate through "FFT coarse estimation + Kalman refinement". Then according to Find the closest filter index Then call the corresponding Perform frequency domain conjugate multiplication; finally, perform IFFT to obtain the low sidelobe compression result.

[0190] This invention achieves a sidelobe suppression effect below -30 dB through a collaborative design of "Hamming window weighted transmission + Doppler-adapted mismatched filter bank". The theoretical basis is that the Hamming window smooths the signal's temporal envelope, significantly reducing spectral leakage and lowering the first sidelobe of the traditional matched filter from -13.2 dB to approximately -42 dB. Building upon this, mismatched filtering optimization is further employed. Within an acceptable range of slight main lobe widening, residual sidelobes are suppressed through frequency domain weighting, ensuring the overall sidelobe level is stably controlled below -30 dB. Simultaneously, this scheme constructs a filter bank covering the entire wind speed Doppler frequency offset. Combined with real-time frequency offset estimation and filter lookup table retrieval mechanism, robustness under dynamic wind fields is guaranteed. Experimental verification shows that even when the echo from a nearby strong ground object is more than 40 dB higher than the signal from a distant weak wind field, the system can still clearly identify weak targets, improving the detection dynamic range by more than 30 dB, effectively solving the problem of signal masking between near and far distances.

[0191] Secondly, this invention proposes a lidar signal processing system, which applies a lidar signal processing method as described above, including:

[0192] The parameter acquisition module is used to acquire the parameters of the lidar system; the lidar system parameters include target range resolution, laser pulse duration, and pulse repetition frequency.

[0193] The signal generation module is used to generate a linear frequency modulated radio frequency drive signal based on the parameters of the lidar system.

[0194] The electro-optic modulation control module is used to optically modulate the continuous laser through an electro-optic modulator and output a linear frequency modulated laser signal.

[0195] A multi-level processing module is used to receive the echo signal of the linear frequency modulated laser signal after atmospheric scattering, and process the echo signal through a multi-level adaptive pulse compression processing link to obtain optimized wind field parameters. The multi-level adaptive pulse compression processing link includes: constructing a time-domain phase compensation function based on the refined Doppler frequency offset estimate to perform phase compensation on the echo signal; dynamically calculating the maximum effective coherent accumulation pulse number according to the laser linewidth and pulse repetition frequency, and using a 50% overlap sliding window strategy to perform coherent accumulation on the phase-compensated echo signal; constructing an optimized transmitted signal by Hamming window weighting of the linear frequency modulated laser signal, and calling the optimal filter in real time from a broadband filter bank covering the entire wind speed Doppler frequency offset range for pulse compression processing to suppress range sidelobes to below a preset threshold.

[0196] The overall technical framework of this invention is entirely based on the mature hardware topology of existing pulsed laser wind radar systems. It does not alter the original optical path arrangement, circuit topology, mechanical structure, or core detection devices. Instead, it establishes a dedicated linear frequency modulated (EOM) electro-optic modulation module and reconstructs the hardware driving timing and software algorithm system to create a proprietary integrated technology framework for laser coherent detection, combining linear frequency modulation transmission and adaptive pulse compression reception. This invention achieves a system upgrade while remaining fully compatible with existing equipment architectures, requiring minimal modifications, exhibiting strong engineering adaptability, and can be directly applied to the performance iteration and upgrade of existing equipment.

[0197] From a hardware perspective, traditional pulsed lidar only supports intensity switching modulation and lacks optical frequency modulation functionality, making it impossible to transmit linearly frequency-modulated signals. This invention, based on the existing mature lidar transmission optical path, adds an EOM electro-optic modulation module adapted to narrow-linewidth, high-coherence lasers. This is merely a functional hardware addition, without altering the original hardware topology, optical path structure, or device coupling relationships. It fills the gap in optical frequency modulation capabilities lacking in traditional hardware systems, providing the necessary hardware support for the implementation of this linear frequency modulation system.

[0198] This invention employs an in-situ serial embedded access method to complete module adaptation and installation, exhibiting strong compatibility with existing equipment and minimal modification scope. The modifications to the transmitting section are as follows: Figure 2 and Figure 3As shown.

[0199] The narrow-linewidth continuous laser and pulsed fiber amplifier in the original radar hardware link are retained. Only the AOM driver and AOM modulator in the original transmitting optical path are replaced with EOM driver and EOM modulator, and an intensity modulator is added. The receiving link remains the same as the original radar hardware and does not need to be changed. A complete laser linear frequency modulation (LFM) transmitting link can be built at low cost (the original narrow-linewidth LFM laser and pulsed fiber amplifier account for 85% of the cost of the entire link), realizing a smooth switch from the traditional fixed pulse system to the laser-specific LFM system.

[0200] The algorithm functional layer is the core innovation of this invention. It completely departs from the general microwave linear frequency modulation technology paradigm and traditional fixed pulse control logic, and reconstructs a completely new working mechanism based on the original hardware architecture. The algorithm functional layer specifically includes: a laser-specific linear frequency modulation module, a wind field Doppler dynamic compensation module, a laser coherent adaptive coherent accumulation module, a low sidelobe pulse compression and demodulation module, and a high-precision wind field parameter calculation module. These modules work together to form a complete closed-loop signal processing system, specifically addressing the pain points of lidar-specific technologies.

[0201] The overall implementation process of this invention is divided into three stages: hardware partial supplementation and upgrade and working condition reconstruction, echo signal acquisition and preprocessing, and laser-specific adaptive signal optimization. It relies on minimal hardware modifications combined with algorithm innovation to achieve system upgrade.

[0202] This invention adds a dedicated laser linear frequency modulated (EOM) electro-optic modulation module to the existing laser radar transmission optical path, while keeping the overall hardware architecture unchanged. Relying on the radar's inherent DDS direct digital frequency synthesizer, the invention uses a unique linear frequency modulation control algorithm to generate a linear scanning radio frequency drive signal with precise timing, controllable bandwidth, and stable frequency modulation slope, replacing the traditional fixed pulse trigger signal, which is then input to the newly added EOM electro-optic modulation module driver.

[0203] The "linear frequency modulation control algorithm" described in this invention is not a general frequency sweep command issuance, but a closed-loop parameter collaborative optimization algorithm constructed based on the working mode of pulsed laser wind radar, laser linewidth characteristics, EOM modulation response characteristics and atmospheric detection requirements. This algorithm ensures that the generated radio frequency drive signal has key performance characteristics such as accurate timing, controllable bandwidth, stable frequency modulation slope, and continuous phase without jumps, thus avoiding damage to laser coherence due to signal distortion.

[0204] The algorithm runs on the radar signal processing module and uses built-in control logic and lookup table mechanism to dynamically configure and output linear frequency modulated signal parameters in real time.

[0205] The core innovation of this invention's receiver is an integrated adaptive pulse compression technology system adapted to the characteristics of laser coherent detection, completely different from the fixed matched filtering scheme commonly used in the microwave field. Addressing the unique industry pain points of laser wind field detection—weak echoes, strong atmospheric turbulence interference, signal mismatch easily caused by dynamic wind fields, limited laser coherence duration, and mutual masking of near and far echo signals—this invention establishes a closed-loop signal processing link with multi-module, step-by-step collaborative optimization. The entire technology follows a progressive logic of "first achieving theoretical gain breakthroughs, then correcting signal fidelity, followed by noise reduction and gain amplification, and finally high-precision demodulation and adjustment." Each stage is closely integrated and mutually supportive, stably implementing the theoretical performance advantages of pulse compression, ultimately simultaneously improving radar range resolution, detection range, signal-to-noise ratio, and detection dynamic range.

[0206] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0207] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0208] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0209] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

[0210] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A laser radar signal processing method, characterized in that, Includes the following steps: Acquire the parameters of the lidar system; the lidar system parameters include target range resolution, laser pulse duration, and pulse repetition frequency. Based on the parameters of the lidar system, a linear frequency modulated radio frequency drive signal is generated, and the continuous laser is optically modulated by an electro-optic modulator to output a linear frequency modulated laser signal. The echo signal of the linear frequency modulated laser signal after atmospheric scattering is received, and processed through a multi-stage adaptive pulse compression processing link to obtain optimized wind field parameters. The multi-stage adaptive pulse compression processing link includes: constructing a time-domain phase compensation function based on the refined Doppler frequency offset estimate to perform phase compensation on the echo signal; dynamically calculating the maximum effective coherent accumulation pulse number based on the laser linewidth and pulse repetition frequency, and using a 50% overlap sliding window strategy to perform coherent accumulation on the phase-compensated echo signal; constructing an optimized transmitted signal by Hamming window weighting of the linear frequency modulated laser signal, and real-time calling the optimal filter from a broadband filter bank covering the entire wind speed Doppler frequency offset range for pulse compression processing to suppress range sidelobes to below a preset threshold.

2. The laser radar signal processing method according to claim 1, characterized in that, Generating a linear frequency modulated radio frequency drive signal based on the lidar system parameters includes: Calculate the required signal bandwidth based on the target distance resolution; The frequency modulation slope is determined based on the laser pulse duration and the signal bandwidth. The linear frequency modulated radio frequency drive signal is obtained by generating a direct digital synthesis control sequence based on the frequency modulation slope and the pulse repetition frequency.

3. The laser radar signal processing method according to claim 2, characterized in that, After generating the direct digital synthesis control sequence, the process also includes: performing phase continuity correction between adjacent pulses.

4. The laser radar signal processing method according to claim 1, characterized in that, Optical frequency modulation of continuous laser light using an electro-optic modulator includes: The linear frequency modulated laser signal is synchronized with the trigger timing of the electro-optic modulator.

5. The laser radar signal processing method according to claim 1, characterized in that, Before performing phase compensation on the echo signal, the method further includes: A slow-time-dimensional fast Fourier transform is performed on the echo signal to roughly estimate the Doppler frequency shift caused by atmospheric wind field motion.

6. The laser radar signal processing method according to claim 5, characterized in that, After roughly estimating the Doppler frequency shift, the following is also included: A Kalman filter is constructed based on a two-dimensional state-space model to refine the coarsely estimated Doppler frequency offset, resulting in a refined Doppler frequency offset estimate. The state vector of the two-dimensional state-space model includes the Doppler frequency offset value and its rate of change.

7. A laser radar signal processing method according to claim 6, characterized in that, After obtaining the refined Doppler frequency offset estimate, the following is also included: A time-domain phase compensation function is constructed based on the refined Doppler frequency offset estimate; Phase compensation of the echo signal is performed using the aforementioned time-domain phase compensation function.

8. The laser radar signal processing method according to claim 1, characterized in that, The coherent accumulation of the phase-compensated echo signal using a 50% overlap sliding window strategy includes: Construct a sliding window based on the maximum effective coherent accumulation pulse number; The sliding window is slid with a 50% overlap step size, and the echo signals within the window are coherently accumulated.

9. A laser radar signal processing method according to claim 1, characterized in that, Real-time selection of the optimal filter from a broadband filter bank covering the entire wind speed Doppler frequency offset range for pulse compression processing includes: Based on the preset Doppler frequency offset distribution step size, a filter bank covering the entire wind speed range is constructed; Based on the real-time estimated Doppler frequency offset, the optimal filter with the center frequency matched is selected from the filter bank for pulse compression.

10. A lidar signal processing system, employing a lidar signal processing method as described in any one of claims 1 to 9, characterized in that, include: The parameter acquisition module is used to acquire the parameters of the lidar system; the lidar system parameters include target range resolution, laser pulse duration, and pulse repetition frequency. The signal generation module is used to generate a linear frequency modulated radio frequency drive signal based on the parameters of the lidar system. The electro-optic modulation control module is used to optically modulate the continuous laser through an electro-optic modulator and output a linear frequency modulated laser signal. A multi-level processing module is used to receive the echo signal of the linear frequency modulated laser signal after atmospheric scattering, and process the echo signal through a multi-level adaptive pulse compression processing link to obtain optimized wind field parameters. The multi-level adaptive pulse compression processing link includes: constructing a time-domain phase compensation function based on the refined Doppler frequency offset estimate to perform phase compensation on the echo signal; dynamically calculating the maximum effective coherent accumulation pulse number according to the laser linewidth and pulse repetition frequency, and using a 50% overlap sliding window strategy to perform coherent accumulation on the phase-compensated echo signal; constructing an optimized transmitted signal by Hamming window weighting of the linear frequency modulated laser signal, and calling the optimal filter in real time from a broadband filter bank covering the entire wind speed Doppler frequency offset range for pulse compression processing to suppress range sidelobes to below a preset threshold.

Images