Echo waveform compression and decompression method, system and product

Abstract

The embodiment of the invention provides an echo waveform compression and decompression method, system and product, and the method comprises the steps: carrying out the preprocessing of original echo waveform data, and obtaining a target time domain signal sequence; extracting a curvature maximum value point meeting a preset extraction rule in the sequence, and scoring the importance of the curvature maximum value point to obtain a score value of the curvature maximum value point; based on the score value of each curvature maximum value point, sorting the curvature maximum value points from large to small to obtain a target curvature maximum value point set; determining a target number of curvature maximum value points of which the score values are ranked in the front in the set as key points; determining an edge point bottom noise constant in the target time domain signal sequence; storing the bottom noise constants of the key points and the edge points as corresponding compressed data; and decompressing the compressed data to obtain reconstructed waveform data. The invention aims to obviously reduce the transmission bandwidth requirement of echo waveform data of a sensor, maintain the main characteristics of the waveform and improve the data transmission and processing efficiency.

Description

Technical Field

[0001] The present application relates to the field of data compression technology, and in particular to an echo waveform compression and decompression method, system and product. Background Art

[0002] With the rapid development of modern science and technology, sensor technology and digital signal processing technology have been significantly improved and widely used in radar, sonar, lidar, medical imaging and seismic exploration. These fields have increasing requirements for data acquisition accuracy, speed and capacity, and the amount of echo waveform data collected by sensors has therefore shown an exponential growth. Such a huge amount of data poses a severe challenge to communication bandwidth, storage capacity and processing power. Especially under limited resource conditions, how to compress data while ensuring real-time and processing efficiency has become a problem that needs to be solved urgently.

[0003] Currently, common echo waveform data compression methods include Fourier transform, wavelet transform and compressed sensing. However, these traditional methods have significant limitations when facing modern high-demand applications. Specifically, when applied to echo waveform data, there are generally problems of high computational complexity, poor real-time performance and insufficient feature retention. Especially when processing waveforms with key features such as mutation points and spikes, the compression process may cause these important information to be smoothed or lost, thereby affecting the effectiveness of subsequent tasks such as target detection, recognition and positioning. Summary of the invention

[0004] In view of this, the present application provides an echo waveform compression and decompression method, system and product, which are intended to significantly reduce the transmission bandwidth requirements of sensor echo waveform data while maintaining the main features of the waveform and improving the efficiency of data transmission and processing.

[0005] The first aspect of the present application provides an echo waveform compression and decompression method, the method comprising:

[0006] An echo waveform compression and decompression method, characterized in that the method comprises:

[0007] Preprocess the collected original echo waveform data to obtain the target time domain signal sequence;

[0008] Extracting the curvature maximum value points satisfying the preset extraction rule in the target time domain signal sequence to obtain a first curvature maximum value point set;

[0009] By scoring the importance of each curvature maximum point in the first curvature maximum point set, a score value of each curvature maximum point is obtained;

[0010] Based on the score value of each curvature maximum point, sort each curvature maximum point in the first curvature maximum point set from large to small to obtain a corresponding target curvature maximum point set;

[0011] Determine the target number of curvature maximum value points with the highest score in the target curvature maximum value point set as key points, wherein the key points include the time coordinates and amplitude values ​​of the key points;

[0012] Determine the noise floor constant of the edge points in the target time domain signal sequence through the noise estimation algorithm;

[0013] storing the key point and the edge point noise floor constants as compressed data corresponding to the original echo waveform data;

[0014] The compressed data is decompressed by a preset decompression algorithm to obtain reconstructed waveform data.

[0015] Optionally, extracting the curvature maximum value points satisfying a preset extraction rule in the target time domain signal sequence to obtain a first curvature maximum value point set includes:

[0016] Calculate the curvature of the target time domain signal sequence to obtain the corresponding curvature sequence;

[0017] Compare each curvature in the curvature sequence with the target threshold and its own front and back curvatures to obtain a comparison result;

[0018] When the comparison result satisfies the first condition, the corresponding curvature is determined as a candidate curvature maximum point;

[0019] According to the energy density and response time of the original echo waveform data, the candidate maximum value points are screened to obtain the curvature maximum value point;

[0020] All the curvature maximum points obtained by screening are determined as the first curvature maximum point set.

[0021] Optionally, by performing importance scoring on each curvature maximum point in the first curvature maximum point set, a score value of each curvature maximum point is obtained, including:

[0022] A scoring algorithm is used to score each curvature maximum point in the first curvature maximum point set to obtain a score value of each curvature maximum point. The scoring algorithm is:

[0023]

[0024] Among them, R i is the score value of the i-th curvature maximum point; κ′ max is the maximum value of curvature among all the maximum points of curvature; Ei is the pulse energy density at the i-th curvature maximum point; E max is the maximum value of the pulse energy density among all the maximum curvature points; T i is the response time of the i-th curvature maximum point; T max is the maximum value of the response time among all the maximum curvature points; F i is the frontier amplitude of the i-th curvature maximum point; F max is the maximum value of the frontier amplitude among all the maximum curvature points; α, β, γ, δ are weight coefficients, satisfying α+β+γ+δ=1.

[0025] Optionally, determine the target quantity, including:

[0026] The current available bandwidth and the storage space required for the key points are calculated by a key point number determination algorithm to determine the maximum allowed key point number. The key point number determination algorithm is:

[0027]

[0028] Among them, B is the current available bandwidth, S key The storage space required for each key point, T trans is the allowed time window for data transmission;

[0029] The maximum number of key points is determined as a target number.

[0030] Optionally, determine the target quantity, including:

[0031] Determining the energy difference value between wave peaks according to the pulse energy density of each curvature maximum point in the first curvature maximum point set;

[0032] Determining the number of dynamic key points according to the energy difference value between the peaks and the number of reference key points;

[0033] The number of dynamic key points is determined as a target number.

[0034] Optionally, determine the target quantity, including:

[0035] The current available bandwidth and the storage space required for key points are calculated by the key point quantity determination algorithm to determine the maximum allowed key point quantity;

[0036] Determining the energy difference value between wave peaks according to the pulse energy density of each curvature maximum point in the first curvature maximum point set;

[0037] Determining the number of dynamic key points according to the energy difference value between the peaks and the number of reference key points;

[0038] Based on a preset screening rule, one of the maximum number of key points and the number of dynamic key points is determined as a target number.

[0039] Optionally, the target number of curvature maximum value points with the highest score values ​​in the target curvature maximum value point set are determined as key points, including:

[0040] Sequentially selecting the curvature maximum value point with the largest score value from the target curvature maximum value point set;

[0041] Determine the time interval between two adjacent curvature maximum value points extracted, and compare the time interval with a preset time interval threshold;

[0042] In the case where the relationship between the time interval and the preset time interval threshold satisfies the second condition, determining the two adjacent curvature maximum value points as key points, and determining the number of current key points;

[0043] When the relationship between the time interval and the preset time interval threshold does not satisfy the second condition, the most recently extracted curvature maximum value point is eliminated, and the extraction of curvature maximum value points continues;

[0044] When the number of key points reaches the target number, the determination of the key points is terminated.

[0045] Optionally, the noise floor constant of the edge points in the target time domain signal sequence is determined by a noise estimation algorithm, including:

[0046] Extracting a preset number of first time domain signal sequences from a starting position of the target time domain signal sequence, and extracting a preset number of second time domain signal sequences from an ending position of the target time domain signal sequence;

[0047] Calculate the mean of the first time domain signal sequence and the second time domain signal sequence respectively to obtain the starting edge mean and the ending edge mean;

[0048] According to the starting edge mean and the ending edge mean, respectively calculating the variance of the first time domain signal sequence and the second time domain signal sequence to obtain the starting edge variance and the ending edge variance;

[0049] Determine a starting edge noise estimate and an ending edge noise estimate according to a starting edge mean, an ending edge mean, a starting edge variance, and an ending edge variance;

[0050] The edge point noise floor constant is determined based on the starting edge noise estimation and the ending edge noise estimation.

[0051] Optionally, decompressing the compressed data by a preset decompression algorithm to obtain reconstructed waveform data includes:

[0052] Constructing a cubic spline interpolation function through the time coordinates and amplitude values ​​of the key points in the compressed data;

[0053] Within the time range of the original echo waveform data, the time points are discretized based on a preset reconstruction sampling rate, and the amplitude value of the reconstructed waveform at each discretized position is determined based on the cubic spline interpolation function;

[0054] Based on the discretization result and the determined amplitude value of the reconstructed waveform, reconstructed waveform data corresponding to the original echo waveform data is obtained.

[0055] A second aspect of the present application provides an echo waveform compression and decompression system, the system comprising:

[0056] A preprocessing module is used to preprocess the collected original echo waveform data to obtain a target time domain signal sequence;

[0057] An extraction module, used to extract the curvature maximum value points satisfying the preset extraction rules in the target time domain signal sequence, and obtain a first curvature maximum value point set;

[0058] A scoring module, used for scoring the importance of each curvature maximum point in the first curvature maximum point set to obtain a score value of each curvature maximum point;

[0059] A sorting module, configured to sort each curvature maximum point in the first curvature maximum point set from large to small based on the score value of each curvature maximum point, to obtain a corresponding target curvature maximum point set;

[0060] A key point determination module, used to determine the target number of curvature maximum value points with the highest score in the target curvature maximum value point set as key points, wherein the key points include the time coordinates and amplitude values ​​of the key points;

[0061] An edge point noise floor constant determination module is used to determine the edge point noise floor constant in the target time domain signal sequence through a noise estimation algorithm;

[0062] A compression module, used for storing the key point and the edge point background noise constant as compressed data corresponding to the original echo waveform data;

[0063] The decompression module is used to decompress the compressed data using a preset decompression algorithm to obtain reconstructed waveform data.

[0064] The third aspect of the present application provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and running on the processor, wherein when the computer program is executed by the processor, the steps in the echo waveform compression and decompression method described in the first aspect of the present application are implemented.

[0065] The fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the echo waveform compression and decompression method described in the first aspect of the present application are implemented.

[0066] The echo waveform compression and decompression method provided by the present application has the following advantages:

[0067] An echo waveform compression and decompression method provided in an embodiment of the present application first pre-processes the collected original echo waveform data to obtain a target time domain signal sequence; extracts the curvature maximum points that meet the preset extraction rules in the target time domain signal sequence to obtain a first curvature maximum point set; scores each curvature maximum point in the first curvature maximum point set to obtain a score value for each curvature maximum point; based on the score value of each curvature maximum point, sorts each curvature maximum point in the first curvature maximum point set from large to small to obtain a corresponding target curvature maximum point set; determines a target number of curvature maximum points with top score values ​​in the target curvature maximum point set as key points, the key points including the time coordinates and amplitude values ​​of the key points; determines the edge point background noise constant in the target time domain signal sequence through a noise estimation algorithm; stores the key points and the edge point background noise constant as compressed data corresponding to the original echo waveform data; decompresses the compressed data through a preset decompression algorithm to obtain reconstructed waveform data. Therefore, compared with the high computational complexity of traditional Fourier transform, wavelet transform and compressed sensing methods in high sampling rate and large data volume application scenarios, which makes it difficult to achieve real-time processing, this application effectively reduces the number of data points to be processed through curvature maximum point selection and adaptive key point screening, significantly reduces the amount of calculation, and enables the algorithm to run efficiently in resource-constrained embedded systems to meet real-time processing requirements; secondly, the algorithm design is simple, and steps such as moving average smoothing and curvature maximum point extraction are used to ensure efficient data processing. The dynamic selection of key point mechanism further optimizes the compression process, improves the overall processing speed, and meets the real-time data compression and decompression requirements under high sampling rates; in addition, through curvature calculation and adaptive threshold combined with local extreme value detection technology, the key turning points and feature change points in the waveform are accurately extracted, and the edge point noise estimation is introduced to ensure that the compressed data highly restores the main features of the original waveform after decompression, and ensure the accuracy of subsequent target detection, recognition and positioning tasks. Finally, the transmission bandwidth requirements of sensor echo waveform data are significantly reduced, while maintaining the main features of the waveform and improving the efficiency of data transmission and processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor.

[0069] Figure 1 A flowchart of an echo waveform compression and decompression method shown in one embodiment of the present application;

[0070] Figure 2A schematic diagram of an echo waveform compression and decompression system is shown in accordance with an embodiment of the present application. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0072] refer to Figure 1 , Figure 1 The following is a flow chart of an echo waveform compression and decompression method according to an embodiment of the present application. Figure 1 As shown, the method includes:

[0073] Step S1: pre-process the collected original echo waveform data to obtain a target time domain signal sequence.

[0074] In this embodiment, the original echo waveform data is first collected by a sensor system (such as radar, sonar, etc.) to obtain a continuous waveform signal at a high sampling rate. The continuous waveform signal at a high sampling rate is the original echo waveform data. The original echo waveform data is a time domain signal sequence, where S={s 1 ,s 2 ,…,s N}, where N is the number of sampling points. In order to reduce the impact of noise on subsequent curvature calculation, this application uses a moving average filter method to smooth the original echo waveform data, thereby obtaining the corresponding target time series signal sequence. The specific method is to smooth the original echo waveform data through the following expression:

[0075]

[0076] Wherein, M is the moving average window size, and the moving average window size can select a suitable value according to the actual noise level and signal characteristics.

[0077] Step S2: extracting the curvature maximum points satisfying the preset extraction rules in the target time domain signal sequence to obtain a first curvature maximum point set.

[0078] In this embodiment, after the target time domain signal sequence is obtained in step S1, the curvature of each data point in the target time domain signal sequence is calculated. The specific calculation method is to calculate the curvature of each data point in the target time domain signal sequence through the following expression:

[0079]

[0080] Where κ[i] represents the curvature of the i-th data point in the target time domain signal sequence; s ′ [i] and s ′″ [i] are the first and second derivatives of the waveform, respectively. The first and second derivatives can be calculated by the following expressions corresponding to the finite difference method:

[0081]

[0082] s″[i]=S 平滑 [i+1]-2S 平滑 [i]+S 平滑 [i-1]

[0083] In this embodiment, based on the calculated curvature of each data point in the target time domain signal sequence, it is determined which data points have curvatures that satisfy preset extraction rules, and the data points corresponding to the curvatures that satisfy the preset extraction rules are determined as curvature maximum points. These data points determined as curvature maximum points are then stored in the same set to obtain a first curvature maximum point set.

[0084] In the present application, step S2 may include: performing curvature calculation on the target time domain signal sequence to obtain a corresponding curvature sequence; comparing each curvature in the curvature sequence with the target threshold and its own front and back curvatures to obtain a comparison result; when the comparison result satisfies the first condition, determining the corresponding curvature as a candidate curvature maximum point; screening the candidate maximum points according to the energy density and response time of the original echo waveform data to obtain the curvature maximum point; determining all the curvature maximum points obtained by screening as the first curvature maximum point set.

[0085] In this embodiment, for step S2, the present application provides an extraction method based on a combination of local maximum detection technology and an adaptive threshold method to determine the curvature maximum point, specifically: calculate the curvature of each data point in the target time domain signal sequence, and the calculation method is the same as the curvature calculation in the above-mentioned embodiment, which will not be repeated here. Then, the curvature of each data point in the target time domain signal sequence is compared with the target threshold and the curvature of a data point before and after itself to obtain the corresponding comparison result. When the comparison result satisfies the first condition, the corresponding curvature is determined as a candidate curvature maximum point. Wherein, when the curvature of a data point in the target time domain signal sequence is greater than the curvature of the data points before and after the data point in the target time domain signal sequence, and greater than the target threshold, it is determined that the curvature of the data point meets the first condition, and the data point is determined as a candidate curvature maximum point. For example, the curvature of the i-th data point in the target time domain signal sequence is κ[i], the target threshold is T, and when κ[i]>κ[i-1], κ[i]>κ[i+1], and κ[i]>T, the i-th data point corresponding to κ[i] is determined as a candidate curvature maximum point. Wherein, the target threshold can be dynamically adjusted according to the waveform energy density, and the target threshold is correspondingly taken as a larger value when the waveform energy density is larger.

[0086] In this embodiment, in order to ensure that the selected maximum point not only has significant curvature but also has significant characteristics in pulse intensity and response speed, after obtaining the candidate curvature maximum points, the candidate curvature maximum points are further screened based on the energy density and response time of the pulse, and the candidate curvature maximum points whose energy density exceeds a certain threshold and whose response time is lower than a certain threshold (that is, the response speed exceeds a certain threshold) are screened out as the final curvature maximum points, and all the screened out curvature maximum points constitute the first curvature maximum point set.

[0087] Step S3: Obtain a score value for each curvature maximum point by scoring the importance of each curvature maximum point in the first curvature maximum point set.

[0088] In this embodiment, in order to ensure that the significance of curvature and the energy and response characteristics of the pulse are taken into account when selecting key points from the curvature maximum points in the subsequent step, the present application will score the importance of each curvature maximum point in the first curvature maximum point set obtained in step S2 based on the energy density and response characteristics of the curvature maximum point, so as to obtain the score value of each curvature maximum point, which will serve as the basis for determining whether the curvature maximum point will be determined as a key point.

[0089] In the present application, step S3 may include: performing a scoring calculation on each curvature maximum point in the first curvature maximum point set by a scoring algorithm to obtain a scoring value of each curvature maximum point, and the scoring algorithm is:

[0090] Among them, R i is the score value of the i-th curvature maximum point; κ′ max is the maximum value of curvature among all the maximum points of curvature; E i is the pulse energy density at the i-th curvature maximum point; E max is the maximum value of the pulse energy density among all the maximum curvature points; T i is the response time of the i-th curvature maximum point; T max is the maximum value of the response time among all the maximum curvature points; F i is the frontier amplitude of the i-th curvature maximum point; F max is the maximum value of the frontier amplitude among all the maximum curvature points;

[0091] α, β, γ, and δ are weight coefficients, satisfying α+β+γ+δ=1.

[0092] In this embodiment, for step S3, the present application provides an optional calculation method to determine the importance score of each curvature maximum point. Specifically, the present application pre-constructs a scoring algorithm, and the expression of the algorithm is:

[0093] Among them, R i is the score value of the i-th curvature maximum point; κ′ max is the maximum value of curvature among all the maximum points of curvature; E i is the pulse energy density at the i-th curvature maximum point; E max is the maximum value of the pulse energy density among all the maximum curvature points; T i is the response time of the i-th curvature maximum point; T max is the maximum value of the response time among all the maximum curvature points; F i is the frontier amplitude of the i-th curvature maximum point; F max is the maximum value of the frontier amplitude among all the curvature maximum points; α, β, γ, δ are weight coefficients, satisfying α+β+γ+δ=1. Then, the scoring algorithm that comprehensively considers the pulse energy, response characteristics, and frontier characteristics directly related to the pulse emission characteristics, propagation characteristics, and interaction relationship with the target object is used to score the importance of each curvature maximum point, and the score value of each curvature maximum point is obtained.

[0094] Step S4: Based on the score value of each curvature maximum point, sort each curvature maximum point in the first curvature maximum point set from large to small to obtain the corresponding target curvature maximum point set.

[0095] In this embodiment, after calculating the score value of each curvature maximum point in the first curvature maximum point set through step S3, all curvature maximum points in the first curvature maximum point set are sorted in descending order based on the score value to obtain the sorted target curvature maximum point set.

[0096] Step S5: Determine the target number of curvature maximum value points with the highest score values ​​in the target curvature maximum value point set as key points, wherein the key points include the time coordinates and amplitude values ​​of the key points.

[0097] In this embodiment, after obtaining the sorted target curvature maximum point set in step S4, the target number of curvature maximum points with the highest score values ​​of the target curvature maximum point set are determined as key points, that is, the first target number of curvature maximum points with the largest score values ​​of the target curvature maximum point set are determined as key points, wherein the information of each key point includes time coordinates and amplitude values. The target number can be set according to the actual application scenario.

[0098] In the present application, determining the target number may include: calculating the current available bandwidth and the storage space required for the key points by a key point number determination algorithm to determine the maximum allowed key point number, wherein the key point number determination algorithm is: Among them, B is the current available bandwidth, S key The storage space required for each key point, T trans is the allowed time window for data transmission; and the maximum number of key points is determined as the target number.

[0099] In this embodiment, the present application provides an optional implementation method for determining the target number, specifically: firstly, a key point number determination algorithm is pre-constructed, and the corresponding expression is: Among them, B is the current available bandwidth, S key The storage space required for each key point, T trans is the allowed time window for data transmission. Then, the algorithm for determining the number of key points is used to calculate the current available bandwidth and the storage space required for the key points, determine the maximum number of key points allowed, and finally determine the maximum number of key points as the target number.

[0100] In the present application, determining the target number may include: determining the energy difference value between peaks based on the pulse energy density of each curvature maximum point in the first curvature maximum point set; determining the number of dynamic key points based on the energy difference value between peaks and the number of benchmark key points; and determining the number of dynamic key points as the target number.

[0101] In this embodiment, the present application provides another optional implementation for determining the target quantity, specifically: firstly, a peak-to-peak energy difference determination algorithm is pre-constructed, and the corresponding expression is: Among them, E i is the pulse energy density at the i-th curvature maximum point, K is the number of echo peaks, where E i Represents the pulse energy density of the i-th echo peak. Then, the pulse energy density of each curvature maximum point in the first curvature maximum point set is calculated by the inter-peak energy difference determination algorithm to obtain the corresponding inter-peak energy difference value. Then, the determined inter-peak energy difference value and the pre-set number of reference key points are calculated by the dynamic key point number determination algorithm to obtain the number of dynamic key points. Finally, the number of dynamic key points is directly determined as the target number. Among them, the dynamic key point number determination algorithm is K 0 +λ·ΔE, where K 0 is the number of benchmark key points, and λ is the adjustment coefficient, which is used to control the influence of ΔE on K.

[0102] In the present application, determining the target number may include: calculating the current available bandwidth and the storage space required for the key points through a key point number determination algorithm to determine the maximum allowed number of key points; determining the energy difference value between peaks based on the pulse energy density of each curvature maximum point in the first curvature maximum point set; determining the number of dynamic key points based on the energy difference value between peaks and the number of benchmark key points; and determining one of the maximum number of key points and the dynamic number of key points as the target number based on preset screening rules.

[0103] In this embodiment, the present application provides a third optional implementation method for determining the target number, specifically: in order to achieve the adaptability of the compression process, the number of key points is dynamically adjusted according to the current bandwidth limit and the peak difference. First, a key point number determination algorithm is pre-constructed, and the corresponding expression is: Among them, B is the current available bandwidth, S key The storage space required for each key point, T trans is the allowed time window for data transmission. Then, the algorithm for determining the number of key points is used to calculate the current available bandwidth and the storage space required for the key points to determine the maximum number of key points allowed.

[0104] At the same time, an algorithm for determining the energy difference between peaks is pre-built, and the corresponding expression is: Among them, E i is the pulse energy density of the i-th curvature maximum point. Then, the pulse energy density of each curvature maximum point in the first curvature maximum point set is calculated by the energy difference between peaks determination algorithm to obtain the corresponding energy difference between peaks. A larger energy difference between peaks indicates that there are significant differences between peaks, and more key points are needed to retain each significant feature; conversely, a smaller energy difference between peaks indicates that the waveform changes smoothly, and the number of key points can be appropriately reduced. Then, the determined energy difference between peaks and the pre-set number of benchmark key points are calculated by the dynamic key point number determination algorithm to obtain the number of dynamic key points, where the dynamic key point number determination algorithm is K 0 +λ·ΔE, where K 0 is the number of benchmark key points, and λ is the adjustment coefficient, which is used to control the influence of ΔE on K.

[0105] Finally, by using the preset screening rules, one of the calculated maximum number of key points and the number of dynamic key points is selected as the target number. Among them, an optional implementation of the preset screening rules is to select the smallest value. Through this implementation method of determining the target number, the target number of key points can be dynamically adjusted according to the bandwidth and peak difference, ensuring the optimal compression effect in different application scenarios.

[0106] Step S6: Determine the noise floor constant of the edge points in the target time domain signal sequence through a noise estimation algorithm.

[0107] In this embodiment, data points in the edge region are extracted from the target time domain signal sequence, and then the data points in the edge region are calculated using a noise estimation algorithm to obtain the edge point noise floor constant in the target time domain signal sequence.

[0108] In the present application, step S6 may include: extracting a preset number of first time domain signal sequences from the starting position of the target time domain signal sequence, and extracting a preset number of second time domain signal sequences from the ending position of the target time domain signal sequence; performing mean calculation on the first time domain signal sequence and the second time domain signal sequence, respectively, to obtain a starting edge mean and an ending edge mean; performing variance calculation on the first time domain signal sequence and the second time domain signal sequence, respectively, based on the starting edge mean and the ending edge mean, to obtain a starting edge variance and an ending edge variance; determining a starting edge noise estimate and an ending edge noise estimate based on the starting edge mean, the ending edge mean, the starting edge variance, and the ending edge variance; determining an edge point background noise constant based on the starting edge noise estimate and the ending edge noise estimate.

[0109] In this embodiment, a preset number of data points are extracted from the starting position of the target time domain signal sequence, and the preset number of data points are the first time domain signal sequence, and a preset number of data points are extracted from the end position of the target time domain signal sequence, and the preset number of data points are the second time domain signal sequence. The first time domain signal sequence and the second time domain signal sequence are both data points in the edge area of ​​the target time domain signal sequence. The first time domain signal sequence is extracted from the target time domain signal sequence, and is expressed as S start ={s 平滑 [1],s 平滑 [2],…,s 平滑 [M]}, extract the second time domain signal sequence from the target time domain signal sequence, denoted as S end ={s 平滑 [N-M+1],s 平滑 [N-M+2],…,s 平滑 [N]}, where M is a preset number, usually M = 0.05N or M = 0.1N, and N is the total number of data points in the target time domain signal sequence.

[0110] In this embodiment, in order to accurately estimate the background noise of the edge area, a noise estimation method based on statistical characteristics is adopted. In the edge area, it is assumed that the noise obeys a Gaussian distribution with a mean of zero. At this time, the noise floor is estimated by calculating the average value of the edge area. First, the means of the first time domain signal sequence and the second time domain signal sequence are calculated respectively by the following formula to obtain the starting edge mean N start and the end edge mean N end .

[0111]

[0112] In this embodiment, in order to improve the estimation accuracy, based on the calculated starting edge mean and ending edge mean, the variance of the first time domain signal sequence and the second time domain signal sequence is calculated by the following formula: and the end edge variance This is used to judge the noise level.

[0113]

[0114] In this embodiment, based on the calculated start edge variance and end edge variance, the threshold of the noise estimation is dynamically adjusted to adapt to waveform data in different noise environments. Specifically, the start edge mean, end edge mean, start edge variance and end edge variance are calculated by the following expression to obtain the start edge noise estimation: and end edge noise estimation

[0115]

[0116] Wherein, λ is the adjustment coefficient, which is usually set to λ=1 or adjusted according to the actual noise level.

[0117] In this embodiment, the calculated starting edge noise estimate and ending edge noise estimate are averaged to obtain the overall edge point noise constant N edge , the corresponding mean value is calculated by the following formula:

[0118]

[0119] Step S7: storing the key point and the edge point background noise constants as compressed data corresponding to the original echo waveform data.

[0120] In this embodiment, through the above steps S1 to S6, a target number of key points are obtained. i ,s i ) includes the time coordinate t of the key point i and amplitude s i , and then the key points of the target number and the calculated edge point noise floor constant are stored as compressed data corresponding to the collected original echo waveform data.

[0121] Step S8: decompressing the compressed data using a preset decompression algorithm to obtain reconstructed waveform data.

[0122] In this embodiment, after obtaining compressed data through step S7, the compressed data is packaged into a unified format for storage and transmission. When the compressed data is needed, the compressed data is decompressed by a preset decompression algorithm to obtain corresponding reconstructed waveform data. The reconstructed data will be similar to the original echo waveform data corresponding to the compressed data.

[0123] In combination with the above embodiments, in one implementation, the embodiment of the present application further provides an echo waveform compression and decompression method. In the echo waveform compression and decompression method, step S8 may include: constructing a cubic spline interpolation function through the time coordinates and amplitude values ​​of the key points in the compressed data; within the time range of the original echo waveform data, based on a preset reconstruction sampling rate, discretizing the time points, and determining the amplitude value of the reconstructed waveform at each discretized position based on the cubic spline interpolation function; based on the discretization result and the determined amplitude value of the reconstructed waveform, reconstructing and obtaining the reconstructed waveform data corresponding to the original echo waveform data.

[0124] In this embodiment, the compressed data packaged into a unified format obtained in step S7 is obtained from the storage medium, and the compressed data includes a target number of key points, and the key points include the time coordinates of the key points {t 1 ,t 2 ,…,t K} and amplitude value {s 1 ,s 2 ,…,s K} and edge point noise constant N edge Then the cubic spline interpolation method is used to approximate the waveform reconstruction. Specifically, the time coordinates of the key points {t 1 ,t 2 ,…,t K} and amplitude value {s 1 ,s 2 ,…,s K}Construct the cubic spline interpolation function S(t), the expression is as follows:

[0125] S(t)=CubicSpline({t i},{s i})

[0126] Then, within the time range of the original echo waveform data, the time point t is discretized according to the original sampling rate or the preset reconstruction sampling rate, and the corresponding amplitude value is calculated using the above cubic spline interpolation function for reconstruction. The calculation expression is as follows:

[0127] S 重构 [j] = S(t j ),j=1,2,…,N

[0128] Among them, t j is the jth time point of the reconstructed waveform, N is the total number of sampling points of the reconstructed waveform, S 重构 [j] is the amplitude value of the reconstructed waveform at the jth time point.

[0129] Finally, based on the discretization result and the determined amplitude value of the reconstructed waveform, reconstructed waveform data corresponding to the original echo waveform data is obtained.

[0130] In combination with the above embodiments, in one implementation, the embodiment of the present application further provides an echo waveform compression and decompression method. In the echo waveform compression and decompression method, step S5 may include: sequentially extracting the curvature maximum value point with the largest score value from the target curvature maximum value point set; determining the time interval between the two adjacent curvature maximum value points extracted, and comparing the time interval with the preset time interval threshold; when the relationship between the time interval and the preset time interval threshold satisfies the second condition, determining the two adjacent curvature maximum value points as key points, and determining the number of current key points; when the relationship between the time interval and the preset time interval threshold does not satisfy the second condition, eliminating the latest extracted curvature maximum value point, and continuing to extract the curvature maximum value points; when the number of key points reaches the target number, ending the determination of key points.

[0131] In this embodiment, the curvature maximum value point with the largest score value is sequentially taken from the target curvature maximum value point set; then the time interval between the two adjacent curvature maximum value points taken is determined, and the time interval is compared with the preset time interval threshold; when the time interval is greater than the preset time interval threshold, it is determined that the relationship between the two meets the second condition, at which time the two adjacent curvature maximum value points corresponding to the time interval are both determined as key points, and the number of current key points is determined. If the number of key points reaches the target number, the determination of the key points is terminated. If it does not reach the target number, the curvature maximum value point with the largest score value is continuously taken from the target curvature maximum value point set until the target number of key points is obtained. When the time interval is less than or equal to the preset time interval threshold, it is determined that the relationship between the two does not meet the second condition, at which time the curvature maximum value point that has been taken out is eliminated, and the curvature maximum value point is continuously taken out and the determination of whether it is a key point is continued until the target number of key points is obtained.

[0132] For example, there are 10 curvature maximum points a1 to a10 in the target curvature maximum point set, which are sorted from large to small according to the score value, and the target number is assumed to be 3. First, the curvature maximum point a1 with the largest value is taken out. At this time, there are no other curvature maximum points for comparison, so the remaining curvature maximum point a2 with the largest value in the target curvature maximum point set is taken out in turn, and it is determined that the time interval between the curvature maximum point a2 and the curvature maximum point a1 is greater than the preset time interval threshold, so the curvature maximum points a1 and a2 are both determined as key points; at this time, there are 2 key points, which does not reach the target number of 3, so continue to take the key points from the target curvature maximum point set in turn. The remaining maximum curvature point a3 with the largest value is taken out from the target curvature maximum point set, and the time interval between the curvature maximum point a3 and the curvature maximum point a2 is determined to be less than the preset time interval threshold, so the time interval between the curvature maximum point a3 and the curvature maximum point a2 does not meet the second condition, so the curvature maximum point a3 is eliminated, and the remaining maximum curvature point a4 with the largest value is taken out from the target curvature maximum point set in turn, and the time interval between the curvature maximum point a4 and the curvature maximum point a2 is determined to be greater than the preset time interval threshold, which meets the second condition, so the curvature maximum point a4 is determined as a key point. At this time, there are 3 key points, namely a1, a2 and a4, reaching the target number of 3, so the key point determination is no longer performed, and finally 3 key points a1, a2 and a4 are obtained.

[0133] An echo waveform compression and decompression method provided in an embodiment of the present application first pre-processes the collected original echo waveform data to obtain a target time domain signal sequence; extracts the curvature maximum points that meet the preset extraction rules in the target time domain signal sequence to obtain a first curvature maximum point set; scores each curvature maximum point in the first curvature maximum point set to obtain a score value for each curvature maximum point; based on the score value of each curvature maximum point, sorts each curvature maximum point in the first curvature maximum point set from large to small to obtain a corresponding target curvature maximum point set; determines a target number of curvature maximum points with top score values ​​in the target curvature maximum point set as key points, the key points including the time coordinates and amplitude values ​​of the key points; determines the edge point background noise constant in the target time domain signal sequence through a noise estimation algorithm; stores the key points and the edge point background noise constant as compressed data corresponding to the original echo waveform data; decompresses the compressed data through a preset decompression algorithm to obtain reconstructed waveform data. Therefore, compared with the high computational complexity of traditional Fourier transform, wavelet transform and compressed sensing methods in high sampling rate and large data volume application scenarios, which makes it difficult to achieve real-time processing, this application effectively reduces the number of data points to be processed through curvature maximum point selection and adaptive key point screening, significantly reduces the amount of calculation, and enables the algorithm to run efficiently in resource-constrained embedded systems to meet real-time processing requirements; secondly, the algorithm design is simple, and steps such as moving average smoothing and curvature maximum point extraction are used to ensure efficient data processing. The dynamic selection of key point mechanism further optimizes the compression process, improves the overall processing speed, and meets the real-time data compression and decompression requirements under high sampling rates; in addition, through curvature calculation and adaptive threshold combined with local extreme value detection technology, the key turning points and feature change points in the waveform are accurately extracted, and the edge point noise estimation is introduced to ensure that the compressed data highly restores the main features of the original waveform after decompression, and ensure the accuracy of subsequent target detection, recognition and positioning tasks. Finally, the transmission bandwidth requirements of sensor echo waveform data are significantly reduced, while maintaining the main features of the waveform and improving the efficiency of data transmission and processing. The key point selection of this method can dynamically adjust the number of key points according to the bandwidth restrictions in different application scenarios. It can not only retain more waveform details when bandwidth resources are sufficient, but also give priority to retaining core features when bandwidth is limited. It is highly adaptable and can flexibly respond to diverse application needs and improve data transmission and storage efficiency. At the same time, moving average smoothing and curvature maximum point extraction effectively filter out random noise, enhance the algorithm's anti-noise ability and robustness, enable it to work stably in complex environments, and ensure data reliability and accuracy.By storing only the key points of the selected number of targets, the data volume of the echo waveform is significantly compressed, the data transmission bandwidth requirement and storage space occupation are reduced, the burden on the communication and storage systems is reduced, and the overall data processing efficiency is improved. It is particularly suitable for large-scale data acquisition and long-term continuous monitoring application scenarios. Finally, due to the low computational complexity of the algorithm and the simple steps, it is easy to implement and integrate on various embedded systems and hardware platforms, which facilitates rapid deployment in practical engineering applications and improves overall system performance and response speed. In other words, this application overcomes the current deficiencies in computational complexity, real-time performance, and feature retention by optimizing the data compression process, dynamic key point selection, and efficient approximate reconstruction methods, and shows significant advantages in bandwidth adaptability, noise resistance, and system integration, greatly improving the efficiency and effect of sensor echo waveform data processing.

[0134] In this embodiment, the present application proposes an echo waveform compression and decompression method, which mainly optimizes the storage and transmission of waveform data through the following steps:

[0135] Smoothing: The original echo waveform usually contains noise, which will affect the subsequent curvature calculation and key point extraction. This application uses a moving average filtering method to smooth the original echo waveform to filter out random noise and ensure the cleanliness of the waveform data.

[0136] Maximum point extraction: Curvature is an important parameter to measure the rate of change of a waveform, which reflects the degree of curvature of the waveform at a specific position. This application accurately captures the key turning points and characteristic change points of the waveform by calculating the curvature of the smoothed waveform. In addition, factors such as pulse characteristics, energy density, and response time are also considered to further optimize the method of extracting maximum points. Specifically, the adaptive threshold method is combined with the local extreme value detection technology to ensure that the extracted maximum points not only have high curvature values, but also show significant characteristics in pulse intensity and response speed, thereby providing more reliable data support for subsequent compression algorithms.

[0137] Key point selection: The maximum points are sorted according to the curvature and the error introduced after removing the points. Then, in order to effectively select the number of key points, a function is designed for it. The specific design principle is as follows:

[0138] Bandwidth: Considering the bandwidth limitations in different application scenarios, the function will dynamically adjust the number of key points according to the available bandwidth. When bandwidth resources are relatively abundant, the number of key points will be appropriately increased to retain more waveform details; conversely, when bandwidth resources are tight, the number of key points will be reduced to prioritize the retention of core features, thereby achieving efficient waveform compression under limited bandwidth.

[0139] Peak Difference: The function also evaluates the relative importance of peaks based on the significant differences between different peaks in the waveform. When the differences between peaks are large, it indicates that there are more significant features in the waveform. In this case, the number of key points will be increased accordingly to ensure that these key features are fully preserved. On the contrary, if the differences between peaks are small, it means that the waveform changes more slowly. In this case, the number of key points can be appropriately reduced to reduce unnecessary data storage and transmission overhead.

[0140] Leading edge of the LiDAR waveform: The leading edge of a LiDAR waveform usually refers to the beginning of the pulse waveform, that is, the time period from the start of the pulse emission to the time when the pulse reaches a certain intensity (usually a preset threshold). This leading edge indicator is of great significance in radar technology because it is directly related to the pulse's emission characteristics, propagation characteristics, and interaction with the target object. Leading edge indicators may include:

[0141] Leading-edge time: The time from when a pulse starts to be emitted until its intensity reaches a preset threshold, which reflects the response speed and propagation speed of the pulse.

[0142] Leading-edge slope: describes the rate at which the pulse intensity changes over time and can help understand the energy distribution and shape of the pulse.

[0143] Leading edge amplitude: The intensity of the pulse leading edge when it reaches the preset threshold, reflecting the peak power or energy of the pulse.

[0144] In radar systems, leading edge indicators have an important impact on target detection, distance measurement, and signal processing. For example, in target detection, the accuracy of the leading edge time is crucial to determining the distance and position of the target. In distance measurement, even a slight change in the leading edge time may lead to significant errors in the measurement results. In signal processing, changes in the leading edge slope and leading edge amplitude can reflect information such as the reflection characteristics and surface structure of the target object. Therefore, when selecting key points, this application specifically considers the leading edge indicators to ensure that these important feature information is retained during the compression process.

[0145] Through the corresponding scoring function, the importance of each maximum point can be comprehensively evaluated to ensure that the key point selection takes into account both the significance of the curvature and the energy and response characteristics of the pulse, while retaining the frontier information of the radar waveform. This multi-dimensional comprehensive evaluation method makes the key point selection more scientific and reasonable, and can achieve a more flexible and efficient compression effect while ensuring the integrity of the main features of the waveform.

[0146] Edge point noise estimation: The edge area of ​​the waveform usually contains important signal information. By estimating the noise constant of the edge points of the compressed waveform, the accuracy of the edge points is retained in the compressed representation, avoiding the loss of edge information during the compression process.

[0147] Compressed representation: Only the coordinates of a selected number of key points of the target are stored as a compressed representation of the waveform, which significantly reduces the amount of data in the echo waveform and reduces the transmission bandwidth requirements while maintaining the main features of the waveform.

[0148] Echo waveform approximate decompression method: By using the key points selected in the compression stage, a nonlinear method is used to approximate the echo waveform and achieve rapid data recovery. Specifically, based on the key points of the target number selected in the compression stage, a cubic spline interpolation method is used to reconstruct an approximate waveform that is highly similar to the original echo waveform data. This method not only improves the accuracy of the reconstructed waveform, but also better preserves the main features of the waveform, ensuring that the decompressed data can meet the accuracy requirements of subsequent target detection and recognition.

[0149] Based on the same inventive concept, an embodiment of the present application provides an echo waveform compression and decompression system, such as Figure 2 As shown, the system 200 includes:

[0150] The preprocessing module 201 is used to preprocess the collected original echo waveform data to obtain a target time domain signal sequence;

[0151] An extraction module 202 is used to extract the curvature maximum value points that meet the preset extraction rules in the target time domain signal sequence to obtain a first curvature maximum value point set;

[0152] A scoring module 203 is used to score the importance of each curvature maximum point in the first curvature maximum point set to obtain a score value of each curvature maximum point;

[0153] A sorting module 204 is used to sort each curvature maximum value point in the first curvature maximum value point set from large to small based on the score value of each curvature maximum value point, so as to obtain a corresponding target curvature maximum value point set;

[0154] A key point determination module 205 is used to determine the target number of curvature maximum value points with the highest score values ​​in the target curvature maximum value point set as key points, wherein the key points include the time coordinates and amplitude values ​​of the key points;

[0155] The edge point noise floor constant determination module 206 is used to determine the edge point noise floor constant in the target time domain signal sequence by using a noise estimation algorithm;

[0156] A compression module 207, configured to store the key point and the edge point background noise constants as compressed data corresponding to the original echo waveform data;

[0157] The decompression module 208 is used to decompress the compressed data using a preset decompression algorithm to obtain reconstructed waveform data.

[0158] Optionally, the extraction module 202 includes:

[0159] The curvature calculation module is used to calculate the curvature of the target time domain signal sequence to obtain the corresponding curvature sequence;

[0160] A comparison module is used to compare each curvature in the curvature sequence with the target threshold and the curvatures before and after it to obtain a comparison result;

[0161] A maximum point determination module, configured to determine the corresponding curvature as a candidate curvature maximum point when the comparison result satisfies the first condition;

[0162] A screening module, used for screening the candidate maximum points according to the energy density and response time of the original echo waveform data to obtain the curvature maximum point;

[0163] The set determination module is used to determine all the curvature maximum value points obtained by screening as the first curvature maximum value point set.

[0164] Optionally, the scoring module 203 is used to score each curvature maximum point in the first curvature maximum point set by using a scoring algorithm to obtain a score value of each curvature maximum point, and the scoring algorithm is:

[0165]

[0166] Among them, R i is the score value of the i-th curvature maximum point; κ ′ max is the maximum value of curvature among all the maximum points of curvature; E i is the pulse energy density at the i-th curvature maximum point; E max is the maximum value of the pulse energy density among all the maximum curvature points; T i is the response time of the i-th curvature maximum point; T max is the maximum value of the response time among all the maximum curvature points; F i is the frontier amplitude of the i-th curvature maximum point; F max is the maximum value of the frontier amplitude among all the maximum curvature points; α, β, γ, δ are weight coefficients, satisfying α+β+γ+δ=1.

[0167] Optionally, the system 200 further includes a target quantity determination module, configured to determine the target quantity;

[0168] The target quantity determination module includes:

[0169] The maximum key point number determination module is used to calculate the current available bandwidth and the storage space required for the key points through a key point number determination algorithm to determine the maximum allowed key point number. The key point number determination algorithm is:

[0170]

[0171] Among them, B is the current available bandwidth, S key The storage space required for each key point, T trans is the allowed time window for data transmission;

[0172] The first target quantity determination module is used to determine the maximum key point quantity as the target quantity.

[0173] Optionally, the target quantity determination module includes:

[0174] a module for determining the energy difference value between wave peaks, used for determining the energy difference value between wave peaks according to the pulse energy density of each curvature maximum point in the first curvature maximum point set;

[0175] A dynamic key point quantity determination module, used to determine the number of dynamic key points according to the energy difference value between the peaks and the number of reference key points;

[0176] The second target quantity determination module is used to determine the quantity of the dynamic key points as the target quantity.

[0177] Optionally, the target quantity determination module includes:

[0178] A maximum key point number determination module is used to calculate the current available bandwidth and the storage space required for the key points through a key point number determination algorithm to determine the maximum allowed key point number;

[0179] a module for determining the energy difference value between wave peaks, used for determining the energy difference value between wave peaks according to the pulse energy density of each curvature maximum point in the first curvature maximum point set;

[0180] A dynamic key point quantity determination module, used to determine the number of dynamic key points according to the energy difference value between the peaks and the number of reference key points;

[0181] The third target quantity determination module is used to determine one of the maximum key point quantity and the dynamic key point quantity as the target quantity based on a preset screening rule.

[0182] Optionally, the key point determination module 205 includes:

[0183] A screening module, used to sequentially select the curvature maximum value point with the largest score value from the target curvature maximum value point set;

[0184] A comparison submodule, used to determine the time interval between two adjacent curvature maximum value points extracted, and compare the time interval with a preset time interval threshold;

[0185] A first key point determination module, configured to determine the two adjacent curvature maximum value points as key points and determine the number of current key points when the relationship between the time interval and the preset time interval threshold satisfies a second condition;

[0186] A second key point determination module is used to remove the most recently extracted curvature maximum point and continue to extract the curvature maximum point when the relationship between the time interval and the preset time interval threshold does not satisfy the second condition;

[0187] The key point end determination module is used to end the key point determination when the number of key points reaches the target number.

[0188] Optionally, the edge point noise constant determination module 206 includes:

[0189] A time domain signal sequence extraction module, configured to extract a preset number of first time domain signal sequences from a starting position of a target time domain signal sequence, and to extract a preset number of second time domain signal sequences from an ending position of the target time domain signal sequence;

[0190] A mean value calculation module, used to perform mean value calculation on the first time domain signal sequence and the second time domain signal sequence respectively to obtain a starting edge mean value and an ending edge mean value;

[0191] A variance calculation module, used to perform variance calculation on the first time domain signal sequence and the second time domain signal sequence according to the starting edge mean and the ending edge mean, to obtain a starting edge variance and an ending edge variance;

[0192] A noise estimation determination module, used to determine a starting edge noise estimation and an ending edge noise estimation according to a starting edge mean, an ending edge mean, a starting edge variance, and an ending edge variance;

[0193] The edge point noise floor constant determination submodule is used to determine the edge point noise floor constant according to the starting edge noise estimation and the ending edge noise estimation.

[0194] Optionally, the decompression module 208 includes:

[0195] An interpolation function determination module, used to construct a cubic spline interpolation function through the time coordinates and amplitude values ​​of key points in the compressed data;

[0196] A first reconstruction module is used to discretize the time points within the time range of the original echo waveform data based on a preset reconstruction sampling rate, and determine the amplitude value of the reconstructed waveform at each discretized position based on the cubic spline interpolation function;

[0197] The second reconstruction module is used to reconstruct and obtain the reconstructed waveform data corresponding to the original echo waveform data based on the discretization result and the determined amplitude value of the reconstructed waveform.

[0198] Based on the same inventive concept, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and running on the processor. When the computer program is executed by the processor, it implements the steps in an echo waveform compression and decompression method as described in the first aspect of the present application.

[0199] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in an echo waveform compression and decompression method as described in the first aspect of the present application are implemented.

[0200] As for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0201] It should be noted that, for the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

[0202] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0203] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the embodiments of the present application may adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0204] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0205] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0206] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable terminal device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0207] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0208] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0209] The above is a detailed introduction to the echo waveform compression and decompression method, system and product provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. An echo waveform compression and decompression method, characterized in that: The method comprises: Preprocess the collected original echo waveform data to obtain the target time domain signal sequence; Extracting the curvature maximum value points satisfying the preset extraction rule in the target time domain signal sequence to obtain a first curvature maximum value point set; By scoring the importance of each curvature maximum point in the first curvature maximum point set, a score value of each curvature maximum point is obtained; Based on the score value of each curvature maximum point, sort each curvature maximum point in the first curvature maximum point set from large to small to obtain a corresponding target curvature maximum point set; Determine the target number of curvature maximum value points with the highest score in the target curvature maximum value point set as key points, wherein the key points include the time coordinates and amplitude values ​​of the key points; Determine the noise floor constant of the edge points in the target time domain signal sequence through the noise estimation algorithm; storing the key point and the edge point noise floor constants as compressed data corresponding to the original echo waveform data; The compressed data is decompressed by a preset decompression algorithm to obtain reconstructed waveform data.

2. The echo waveform compression and decompression method according to claim 1, characterized in that: Extracting the curvature maximum value points satisfying the preset extraction rule in the target time domain signal sequence to obtain a first curvature maximum value point set includes: Calculate the curvature of the target time domain signal sequence to obtain the corresponding curvature sequence; Compare each curvature in the curvature sequence with the target threshold and its own front and back curvatures to obtain a comparison result; When the comparison result satisfies the first condition, the corresponding curvature is determined as a candidate curvature maximum point; According to the energy density and response time of the original echo waveform data, the candidate maximum value points are screened to obtain the curvature maximum value point; All the curvature maximum points obtained by screening are determined as the first curvature maximum point set.

3. The echo waveform compression and decompression method according to claim 1, characterized in that: By scoring the importance of each curvature maximum point in the first curvature maximum point set, a score value of each curvature maximum point is obtained, including: A scoring algorithm is used to score each curvature maximum point in the first curvature maximum point set to obtain a score value of each curvature maximum point. The scoring algorithm is: Among them, R i is the score value of the i-th curvature maximum point; κ′ max is the maximum value of curvature among all the maximum points of curvature; E i is the pulse energy density at the i-th curvature maximum point; E max is the maximum value of the pulse energy density among all the maximum curvature points; T i is the response time of the i-th curvature maximum point; T max is the maximum value of the response time among all the maximum curvature points; F i is the frontier amplitude of the i-th curvature maximum point; F max is the maximum value of the frontier amplitude among all the maximum curvature points; α, β, γ, δ are weight coefficients, satisfying α+β+γ+δ=1.

4. The echo waveform compression and decompression method according to claim 1, characterized in that: Determine target quantities, including: The current available bandwidth and the storage space required for the key points are calculated by a key point number determination algorithm to determine the maximum allowed key point number. The key point number determination algorithm is: Among them, B is the current available bandwidth, S key The storage space required for each key point, T trans is the allowed time window for data transmission; The maximum number of key points is determined as a target number.

5. The echo waveform compression and decompression method according to claim 1, characterized in that: Determine target quantities, including: Determining the energy difference value between wave peaks according to the pulse energy density of each curvature maximum point in the first curvature maximum point set; Determining the number of dynamic key points according to the energy difference value between the peaks and the number of reference key points; The number of dynamic key points is determined as a target number.

6. The echo waveform compression and decompression method according to claim 1, characterized in that: Determine target quantities, including: The current available bandwidth and the storage space required for key points are calculated by the key point quantity determination algorithm to determine the maximum allowed key point quantity; Determining the energy difference value between wave peaks according to the pulse energy density of each curvature maximum point in the first curvature maximum point set; Determining the number of dynamic key points according to the energy difference value between the peaks and the number of reference key points; Based on a preset screening rule, one of the maximum number of key points and the number of dynamic key points is determined as a target number.

7. The echo waveform compression and decompression method according to claim 1, characterized in that: The curvature maximum value points of the target number with the highest score in the target curvature maximum value point set are determined as key points, including: Sequentially selecting the curvature maximum value point with the largest score value from the target curvature maximum value point set; Determine the time interval between two adjacent curvature maximum value points extracted, and compare the time interval with a preset time interval threshold; In the case where the relationship between the time interval and the preset time interval threshold satisfies the second condition, determining the two adjacent curvature maximum value points as key points, and determining the number of current key points; When the relationship between the time interval and the preset time interval threshold does not satisfy the second condition, the most recently extracted curvature maximum value point is eliminated, and the extraction of curvature maximum value points continues; When the number of key points reaches the target number, the determination of the key points is terminated.

8. The echo waveform compression and decompression method according to claim 1, characterized in that: The noise estimation algorithm is used to determine the noise floor constant of the edge points in the target time domain signal sequence, including: Extracting a preset number of first time domain signal sequences from a starting position of the target time domain signal sequence, and extracting a preset number of second time domain signal sequences from an ending position of the target time domain signal sequence; Calculate the mean of the first time domain signal sequence and the second time domain signal sequence respectively to obtain the starting edge mean and the ending edge mean; According to the starting edge mean and the ending edge mean, respectively calculating the variance of the first time domain signal sequence and the second time domain signal sequence to obtain the starting edge variance and the ending edge variance; Determine a starting edge noise estimate and an ending edge noise estimate according to a starting edge mean, an ending edge mean, a starting edge variance, and an ending edge variance; The edge point noise floor constant is determined based on the starting edge noise estimation and the ending edge noise estimation.

9. The echo waveform compression and decompression method according to claim 1, characterized in that: Decompressing the compressed data by a preset decompression algorithm to obtain reconstructed waveform data includes: Constructing a cubic spline interpolation function through the time coordinates and amplitude values ​​of the key points in the compressed data; Within the time range of the original echo waveform data, the time points are discretized based on a preset reconstruction sampling rate, and the amplitude value of the reconstructed waveform at each discretized position is determined based on the cubic spline interpolation function; Based on the discretization result and the determined amplitude value of the reconstructed waveform, reconstructed waveform data corresponding to the original echo waveform data is obtained.

10. An echo waveform compression and decompression system, characterized in that: The system comprises: A preprocessing module is used to preprocess the collected original echo waveform data to obtain a target time domain signal sequence; An extraction module, used to extract the curvature maximum value points satisfying the preset extraction rules in the target time domain signal sequence, and obtain a first curvature maximum value point set; A scoring module, used for scoring the importance of each curvature maximum point in the first curvature maximum point set to obtain a score value of each curvature maximum point; A sorting module, configured to sort each curvature maximum point in the first curvature maximum point set from large to small based on the score value of each curvature maximum point, to obtain a corresponding target curvature maximum point set; A key point determination module, used to determine the target number of curvature maximum value points with the highest score in the target curvature maximum value point set as key points, wherein the key points include the time coordinates and amplitude values ​​of the key points; An edge point noise floor constant determination module is used to determine the edge point noise floor constant in the target time domain signal sequence through a noise estimation algorithm; A compression module, used for storing the key point and the edge point background noise constant as compressed data corresponding to the original echo waveform data; The decompression module is used to decompress the compressed data using a preset decompression algorithm to obtain reconstructed waveform data.

11. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and running on the processor, wherein the computer program, when executed by the processor, implements the steps in the echo waveform compression and decompression method as described in claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps in the echo waveform compression and decompression method according to claims 1 to 9 are implemented.

Images