A laser doppler spectrum correction method based on power spectrum analysis

Abstract

The application discloses a dynamic laser Doppler spectrum correction method, and the traditional laser Doppler spectrum correction method only suppresses fixed position noise of a spectrum, the position of Doppler spectrum noise corresponding to different laser velocimeter systems is not necessarily the same, therefore, the method is not universal, the suppression effect of the method is poor, and the spectrum noise is still unstable, thereby causing interference to spectrum resolution of the velocimeter system, and therefore, the application provides a laser Doppler spectrum correction method based on power spectrum analysis.The application performs signal suppression based on power spectrum analysis, has good universality, the spectrum noise suppression effect is obvious, the spectrum noise is stable, and the application lays a foundation for accurately extracting a Doppler signal in subsequent spectrum resolution.

Description

Technical Field

[0001] This invention relates to a laser Doppler spectrum correction method, and more particularly to a laser Doppler spectrum correction method based on power spectrum analysis. Background Technology

[0002] Compared to odometers, laser Doppler velocimeters are non-contact measurement devices that utilize the Doppler effect of lasers to achieve high-precision measurement of vehicle speed. They can isolate the adverse effects of tire pressure changes, turning maneuvers, and skidding, making them an effective way to achieve high-precision autonomous positioning and adaptability for vehicles. They offer significant advantages such as fast response speed, high spatial resolution, and a large measurement range. Accurate extraction of the laser Doppler frequency signal is crucial in achieving high-precision laser velocimetry in engineering practice, directly affecting the accuracy of speed measurement.

[0003] However, due to interference from circuit noise, the correct interpretation of the signal is usually affected. In severe cases, incorrect speeds are calculated and output to the positioning and orientation equipment, resulting in increased positioning errors. Summary of the Invention

[0004] The purpose of this invention is to provide a laser Doppler spectrum correction method based on power spectrum analysis to solve the problem of unstable laser Doppler spectrum noise.

[0005] The specific steps of a laser Doppler spectral noise suppression method based on power spectrum analysis are as follows:

[0006] The first step is to build a laser Doppler spectrum processing system based on power spectrum analysis.

[0007] A laser Doppler spectrum processing system based on power spectrum analysis includes: a spectrum acquisition module and a spectrum processing module.

[0008] The function of the spectrum acquisition module is to acquire high-precision Doppler frequency domain signals, and to store the digital Doppler signals obtained by A / D conversion through the FPGA chip in the system's internal first-in-first-out FIFO unit for use by the spectrum processing module.

[0009] The function of the spectrum processing module is as follows: with the ARM chip STM32H743 as the core, the first-in-first-out FIFO unit transmits the spectrum signal to the ARM chip through timing control, and then performs spectrum noise correction processing on the power spectrum of the Doppler signal, extracts the background noise of the power spectrum and performs suppression processing.

[0010] The second step involves acquiring the power spectrum of the Doppler signal using the spectrum acquisition module.

[0011] The spectrum acquisition module converts the laser Doppler signal acquired by the photodetector into a photocurrent signal, then amplifies and filters it before converting the analog signal into a digital time-domain Doppler signal. The FPGA chip then uses a Fast Fourier Transform (FFT) to convert the detected time-domain Doppler signal into a spectrum, which is the power spectrum of the Doppler signal.

[0012] The core of laser velocimetry is to find the frequency value corresponding to the highest frequency point in the power spectrum, and then calculate the velocity value based on that frequency value.

[0013] The third step involves extracting the static power spectrum background noise using the spectrum processing module.

[0014] The Doppler signal power spectrum contains not only the effective Doppler signal detected by the laser velocimeter, but also noise signals from the circuit, which are randomly distributed across various frequency points in the power spectrum. Without suppression of this circuit noise, if the amplitude of the noise signal exceeds the effective Doppler signal amplitude, the noise signal will be mistakenly identified as a valid signal, leading to incorrect velocity calculations. Therefore, it is necessary to suppress the circuit noise signal.

[0015] Although the noise signal is randomly distributed across the frequency points of the power spectrum, the shape of the noise signal's envelope in the power spectrum remains unchanged. Based on the distribution characteristics of the noise signal, the spectrum processing module extracts the envelope of the noise signal. During the static placement of the velocimeter, there is no Doppler signal output, and the power spectrum only contains circuit noise information. Therefore, 400 frames of power spectrum are collected as samples. Each frame of power spectrum consists of 8192 frequency points, and the average amplitude of each frequency point of the power spectrum is calculated, expressed by formula (1):

[0016]

[0017] Where F i Let represent the power spectrum of the i-th frame. The obtained mean power spectrum can reflect the envelope distribution of circuit noise, and this envelope distribution is used as the background noise.

[0018] The fourth step involves the spectrum processing module performing static power spectrum floor noise analysis.

[0019] While the spectrum processing module collects 400 frames of static power spectrum background noise, it also analyzes the basis noise of the power spectrum, which is expressed by formula (2):

[0020]

[0021] This base noise is used in subsequent signal-to-noise ratio calculations.

[0022] The fifth step involves the spectrum processing module performing power spectrum signal-to-noise ratio analysis.

[0023] The spectrum processing module determines the power spectrum signal-to-noise ratio of the velocimeter, expressed by formula (3):

[0024]

[0025] In formula (3), F is the current power spectrum, and the signal-to-noise ratio Q represents the strength of the highest peak frequency in the current power spectrum.

[0026] When the laser velocimeter is stationary, it will not be sensitive to the effective Doppler signal, the signal-to-noise ratio Q will remain stable below 100, and it will not be affected by circuit noise.

[0027] When the laser velocimeter moves with the carrier, it will be sensitive to the effective Doppler signal, which will appear as a distinct spike signal in the entire power spectrum. The signal-to-noise ratio Q of this spike signal is usually greater than 1000, and the velocity can be accurately calculated based on the position of this spike in the power spectrum.

[0028] The sixth step, the spectrum processing module, dynamically updates the background noise and floor noise.

[0029] Although the 400 frames of background noise and floor noise collected can reflect the envelope shape of the power spectrum, the temperature of the circuit hardware will change over time, which will have a certain impact on the envelope shape of the power spectrum. The spectrum processing module dynamically updates the background noise and floor noise in real time.

[0030] The background noise update is expressed by formula (4):

[0031] F ave =k·F ave1 +(1-k)·F ave2 (4)

[0032] In formula (4), F ave For the updated background noise, F ave1 For historical background noise, F ave2 The background noise of the current frame is denoted as k, and the weighting coefficient is k = 0.9995. Since the change of the power spectrum envelope in each frame is very small, the weighting coefficient is k = 0.9995.

[0033] The update of the basis noise is expressed by formula (5):

[0034]

[0035] In formula (5), F sqr For the updated background noise, F sqr1 For historical basis noise, F ave1 The background noise is the historical noise, F is the Doppler spectrum of the current frame, and k is the weighting coefficient, which is set to k = 0.9995.

[0036] Thus, a laser Doppler spectrum correction method based on power spectrum analysis has been completed.

[0037] This invention uses power spectrum analysis for signal suppression, which has good universality, significant effect on spectral noise suppression, and stable spectral noise, laying the foundation for accurate extraction of Doppler signals in subsequent spectrum calculations. Detailed Implementation

[0038] The specific steps of a laser Doppler spectral noise suppression method based on power spectrum analysis are as follows:

[0039] The first step is to build a laser Doppler spectrum processing system based on power spectrum analysis.

[0040] A laser Doppler spectrum processing system based on power spectrum analysis includes: a spectrum acquisition module and a spectrum processing module.

[0041] The function of the spectrum acquisition module is to acquire high-precision Doppler frequency domain signals, and to store the digital Doppler signals obtained by A / D conversion through the FPGA chip in the system's internal first-in-first-out FIFO unit for use by the spectrum processing module.

[0042] The function of the spectrum processing module is as follows: with the ARM chip STM32H743 as the core, the first-in-first-out FIFO unit transmits the spectrum signal to the ARM chip through timing control, and then performs spectrum noise correction processing on the power spectrum of the Doppler signal, extracts the background noise of the power spectrum and performs suppression processing.

[0043] The second step involves acquiring the power spectrum of the Doppler signal using the spectrum acquisition module.

[0044] The spectrum acquisition module converts the laser Doppler signal acquired by the photodetector into a photocurrent signal, then amplifies and filters it before converting the analog signal into a digital time-domain Doppler signal. The FPGA chip then uses a Fast Fourier Transform (FFT) to convert the detected time-domain Doppler signal into a spectrum, which is the power spectrum of the Doppler signal.

[0045] The core of laser velocimetry is to find the frequency value corresponding to the highest frequency point in the power spectrum, and then calculate the velocity value based on that frequency value.

[0046] The third step involves extracting the static power spectrum background noise using the spectrum processing module.

[0047] The Doppler signal power spectrum contains not only the effective Doppler signal detected by the laser velocimeter, but also noise signals from the circuit, which are randomly distributed across various frequency points in the power spectrum. Without suppression of this circuit noise, if the amplitude of the noise signal exceeds the effective Doppler signal amplitude, the noise signal will be mistakenly identified as a valid signal, leading to incorrect velocity calculations. Therefore, it is necessary to suppress the circuit noise signal.

[0048] Although the noise signal is randomly distributed across the frequency points of the power spectrum, the shape of the noise signal's envelope in the power spectrum remains unchanged. Based on the distribution characteristics of the noise signal, the spectrum processing module extracts the envelope of the noise signal. During the static placement of the velocimeter, there is no Doppler signal output, and the power spectrum only contains circuit noise information. Therefore, 400 frames of power spectrum are collected as samples. Each frame of power spectrum consists of 8192 frequency points, and the average amplitude of each frequency point of the power spectrum is calculated, expressed by formula (1):

[0049]

[0050] Where F i Let represent the power spectrum of the i-th frame. The obtained mean power spectrum can reflect the envelope distribution of circuit noise, and this envelope distribution is used as the background noise.

[0051] The fourth step involves the spectrum processing module performing static power spectrum floor noise analysis.

[0052] While the spectrum processing module collects 400 frames of static power spectrum background noise, it also analyzes the basis noise of the power spectrum, which is expressed by formula (2):

[0053]

[0054] This base noise is used in subsequent signal-to-noise ratio calculations.

[0055] The fifth step involves the spectrum processing module performing power spectrum signal-to-noise ratio analysis.

[0056] The spectrum processing module determines the power spectrum signal-to-noise ratio of the velocimeter, expressed by formula (3):

[0057]

[0058] In formula (3), F is the current power spectrum, and the signal-to-noise ratio Q represents the strength of the highest peak frequency in the current power spectrum.

[0059] When the laser velocimeter is stationary, it will not be sensitive to the effective Doppler signal, the signal-to-noise ratio Q will remain stable below 100, and it will not be affected by circuit noise.

[0060] When the laser velocimeter moves with the carrier, it will be sensitive to the effective Doppler signal, which will appear as a distinct spike signal in the entire power spectrum. The signal-to-noise ratio Q of this spike signal is usually greater than 1000, and the velocity can be accurately calculated based on the position of this spike in the power spectrum.

[0061] The sixth step, the spectrum processing module, dynamically updates the background noise and floor noise.

[0062] Although the 400 frames of background noise and floor noise collected can reflect the envelope shape of the power spectrum, the temperature of the circuit hardware will change over time, which will have a certain impact on the envelope shape of the power spectrum. The spectrum processing module dynamically updates the background noise and floor noise in real time.

[0063] The background noise update is expressed by formula (4):

[0064] F ave =k·F ave1 +(1-k)·F ave2 (4)

[0065] In formula (4), F ave For the updated background noise, F ave1 For historical background noise, F ave2 The background noise of the current frame is denoted as k, and the weighting coefficient is k = 0.9995. Since the change of the power spectrum envelope in each frame is very small, the weighting coefficient is k = 0.9995.

[0066] The update of the basis noise is expressed by formula (5):

[0067]

[0068] In formula (5), F sqr For the updated background noise, F sqr1 For historical basis noise, F ave1 The background noise is the historical noise, F is the Doppler spectrum of the current frame, and k is the weighting coefficient, which is set to k = 0.9995.

[0069] Thus, a laser Doppler spectrum correction method based on power spectrum analysis has been completed.

Claims

1. A laser Doppler spectral noise suppression method based on power spectrum analysis, characterized in that... The specific steps are as follows: The first step is to build a laser Doppler spectrum processing system based on power spectrum analysis. A laser Doppler spectrum processing system based on power spectrum analysis includes: a spectrum acquisition module and a spectrum processing module; The second step involves acquiring the power spectrum of the Doppler signal using the spectrum acquisition module. The spectrum acquisition module converts the laser Doppler signal acquired by the photodetector into a photocurrent signal, then passes it through a preamplifier and filter, and finally converts the analog signal into a digital time-domain Doppler signal; the FPGA chip converts the detected time-domain Doppler signal into a spectrum using a Fast Fourier Transform (FFT), which is the power spectrum of the Doppler signal. The core of laser velocimetry is to find the frequency value corresponding to the highest frequency point in the power spectrum, and then calculate the velocity value based on that frequency value. The third step involves extracting the static power spectrum background noise using the spectrum processing module. The Doppler signal power spectrum contains not only the effective Doppler signal detected by the laser velocimeter, but also noise signals from the circuit. These noise signals are randomly distributed across various frequency points in the power spectrum. Without suppression of the circuit noise, if the amplitude of the circuit noise signal exceeds the effective Doppler signal amplitude, the noise signal will be mistakenly identified as an effective signal, leading to incorrect velocity calculations. Therefore, it is necessary to suppress the circuit noise signal. Although the noise signal is randomly distributed at various frequency points in the power spectrum, the shape of the noise signal envelope in the power spectrum will not change. Based on the distribution characteristics of the noise signal, the spectrum processing module extracts the envelope of the noise signal. During the static placement of the speedometer, there is no Doppler signal output, and the power spectrum only contains circuit noise information. Therefore, 400 frames of power spectrum are collected as samples. Each frame of power spectrum consists of 8192 frequency points, and the average amplitude of each frequency point of the power spectrum is calculated and expressed by formula (1): Where F i Let represent the power spectrum of the i-th frame. The mean power spectrum obtained can reflect the envelope distribution of circuit noise, and this envelope distribution is used as the background noise. The fourth step involves the spectrum processing module performing static power spectrum floor noise analysis. While the spectrum processing module collects 400 frames of static power spectrum background noise, it also analyzes the basis noise of the power spectrum, which is expressed by formula (2): This base noise is used in subsequent signal-to-noise ratio calculations; The fifth step involves the spectrum processing module performing power spectrum signal-to-noise ratio analysis. The spectrum processing module determines the power spectrum signal-to-noise ratio of the velocimeter, expressed by formula (3): In formula (3), F is the current power spectrum, and the signal-to-noise ratio Q represents the strength of the highest peak frequency in the current power spectrum. When the laser velocimeter is stationary, it will not be sensitive to the effective Doppler signal, the signal-to-noise ratio Q will remain stable below 100, and it will not be affected by circuit noise. When the laser velocimeter moves with the carrier, the velocimeter will be sensitive to the effective Doppler signal, which will appear as a distinct peak signal in the entire power spectrum. The signal-to-noise ratio Q of this peak signal is usually greater than 1000, and the velocity can be accurately calculated based on the position of this peak in the power spectrum. The sixth step, the spectrum processing module, dynamically updates the background noise and floor noise. Although the 400 frames of background noise and floor noise collected can reflect the envelope shape of the power spectrum, the temperature of the circuit hardware will change with the accumulation of usage time, which will have a certain impact on the envelope shape of the power spectrum. The spectrum processing module dynamically updates the background noise and floor noise in real time. The background noise update is expressed by formula (4): F ave =k·F ave1 +(1-k)·F ave2 (4) In formula (4), F ave For the updated background noise, F ave1 For historical background noise, F ave2 The background noise of the current frame is k, which is the weighting coefficient. Since the change of the power spectrum envelope in each frame is very small, the weighting coefficient is taken as k = 0.9995. The update of the basis noise is expressed by formula (5): In formula (5), F sqr For the updated background noise, F sqr1 For historical basis noise, F ave1 The background noise is the historical noise, F is the Doppler spectrum of the current frame, and k is the weighting coefficient, which is set to k = 0.9995. Thus, a laser Doppler spectrum correction method based on power spectrum analysis has been completed.

2. The laser Doppler spectral noise suppression method based on power spectrum analysis according to claim 1, characterized in that... The function of the spectrum acquisition module is to acquire high-precision Doppler frequency domain signals, and to store the digital Doppler signals obtained by A / D conversion in the system's internal first-in-first-out FIFO unit through the FPGA chip for use by the spectrum processing module.

3. The laser Doppler spectral noise suppression method based on power spectrum analysis according to claim 1, characterized in that... The function of the spectrum processing module is as follows: with the ARM chip STM32H743 as the core, the first-in-first-out FIFO unit transmits the spectrum signal to the ARM chip through timing control, and then performs spectrum noise correction processing on the power spectrum of the Doppler signal, extracts the background noise of the power spectrum and performs suppression processing.

Images