CN109164454A - A kind of fuzzy method for solving of the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation - Google Patents
A kind of fuzzy method for solving of the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation Download PDFInfo
- Publication number
- CN109164454A CN109164454A CN201810965428.8A CN201810965428A CN109164454A CN 109164454 A CN109164454 A CN 109164454A CN 201810965428 A CN201810965428 A CN 201810965428A CN 109164454 A CN109164454 A CN 109164454A
- Authority
- CN
- China
- Prior art keywords
- echo
- mta
- measurement
- section
- high frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/06—Systems determining position data of a target
- G01S17/08—Systems determining position data of a target for measuring distance only
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Electromagnetism (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Radar, Positioning & Navigation (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Remote Sensing (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation that the invention proposes a kind of obscures method for solving, uses BP neural network primary Calculation detection target potentially to measure apart from section and each section distribution probability and inputs as initial calculation;The surveying range of echo is further calculated using the continuity Characteristics of incidence relation and atural object between the complexity and inside of echo-signal in Full wave shape signal;The laser pulse emission time is modulated using pseudo-random binary sequence, according to the criterion that the noise energy of sequence target reflection echo minimizes, accurate sounding section is calculated, determines echo and seed light corresponding relationship, calculates accurately measurement distance accordingly.The present invention effectively can accurately solve range ambiguity in high-frequency laser radar, can obtain higher measurement accuracy and accuracy simultaneously with closeer measurement pin point in farther away measurement distance in high-frequency lidar measurement.
Description
Technical field
The present invention relates to lidar measurement field, specially a kind of medium-long range high frequency lasers based on pscudo-random codc modulation
Radar range finding obscures method for solving.
Technical background
Precision, density and efficiency are to measure three important indicators of airborne laser radar ability.With Full wave shape laser thunder
Up to the application of technology, more stringent requirements are proposed for measurement distance and laser footpoint density of the people to airborne laser radar.However,
When the laser emitter of high repetition frequency is applied to airborne laser radar, laser firing pulses and echo are inevitably brought
The problem of ambiguity is associated between signal, i.e., in high frequency lasers radar, when the echo reception moment is away from seed light emission time
When the more than one time interval even more than impulse ejection of time interval, echo can not be directly determined from timing signal sequence
, thus there is range ambiguity in the correct corresponding relationship of signal and its seed photo-signal.
Summary of the invention
It is sharp to provide a kind of medium-long range high frequency based on pscudo-random codc modulation aiming at the deficiencies in the prior art by the present invention
Optical radar range ambiguity method for solving reduces remote high-frequency laser radar range fuzzy problem, improves measurement accuracy.
To achieve the goals above, the medium-long range high frequency lasers radar based on pscudo-random codc modulation designed by the present invention is surveyed
Away from fuzzy method for solving, which comprises the following steps:
Echo waveform characteristic information and echo range information model are established using BP neural network, calculates single measurement echo
All potential measurements are apart from section and each section distribution probability, and wherein echo waveform morphological feature is BP neural network input
?;
All-wave graphic data is decomposed into multiple Gauss echoes using waveform decomposition algorithm, it is special according to the continuity of area target
Optimization algorithm is put and used, the optimization measurement distance range of measurement sequence can be further calculated, calculate the measurement of echo
Apart from section;
The laser pulse emission time is modulated using pseudo-random binary sequence, according to the noise energy of sequence target reflection echo
The criterion minimized is measured, accurate sounding section is calculated.
Further, the echo waveform morphological feature includes echo complex energy parameter, echo signal-to-noise ratio parameter, echo
Mean breadth parameter, height parameter, mean energy parameter, position of centre of gravity parameter.
Further, detailed process is as follows for the foundation of the BP neural network: being acquired using low-frequency laser single
Laser Full wave shape echo-signal, using the wave configuration feature information of extraction as BP neural network input item, according to low-frequency
Laser collected time scale calculate the waveform section MTA that the place of identical echo is collected in HF laser
[i], time when MTA (Multiple Time Around) interval value i is defined as echo reception are located between i-1 pulse repetition
Every time and i pulse recurrence interval between, i.e. interval between two transmitted waves.By the waveform section MTA at the place of echo
[i] and the probability in different surveying ranges establish 4 layers of BP neural network, intermediate state transfer as the output item of neural network
Function selects linear function and piecewise function respectively.
Further, the waveform decomposition algorithm is Levenberg-Marquardt decomposition algorithm.
Further, described to be advanced optimized using surveying range input item of the waveform decomposition algorithm to BP neural network
Optimization process it is specific as follows:
According to the continuity features of area target and optimization algorithm is used, the optimal of measurement sequence can be further calculated
Change measurement distance range MTA [X];Assuming that one shares N number of area MTA, laser is t at the time of emitting i-th of pulseT[i], detection
Device is t at the time of detecting j-th of echoE[j], emitted laser pulse number is G [j] at this time, by prosthomere neural network
The distribution probability for exporting result available Qi MTA area k (k=1,2 ... N) is pj,k, measure distance Lj,kWith such as ShiShimonoseki
System:
Ru×(k-1)≤Lj,k≤Ru×k
So for the continuous N echo since the 1st, this N number of section MTA probability distribution matrix and measurement away from
It is respectively as follows: from matrix
If the Mean Vector in the section MTA of this continuous N echo are as follows:
The Mean Vector of measurement distance is respectively as follows:
It since ground object target geometric shape can be considered as continuously in subrange, and be not in size is more than one
The case where a area MTA;Using Least Square Method, distance is measured with the expectation that straight line f is fitted this M echo, even if formula
(14) value is minimum, can obtain the optimal estimation of each echometric measurement distance, the i.e. section MTA optimal estimation value:
Further, detailed process is as follows for the criterion that the noise energy of the sequence target reflection echo minimizes:
Since the constant low frequency component of the distance signal in each section MTA is that can ignore not to the contribution of noise energy
Meter, therefore the difference vector Δ R of calculating continuous measurements can be passed throughMTA,j, it is specific as follows:
ΔRMTA,j=(r2-r1,r3-r2,...,rM-rM-1)T
From each distance vector RMTA,jMiddle elimination high frequency noise components, then, all potential sections MTA are associated each
Difference vector Δ RMTA,jNoise energyIt can be given by:
The smallest noise energyIt is not influenced by PPM, j value at this time is associated with the correct section MTA;Institute
Noise signal energy of each measurement in all potential sections MTA apart from group (M), selection tool can be distributed to by comparing
There is the section MTA at the place of minimal noise energy as a result.
Further, the pseudo-random binary sequence is modulated depending on different topography and landform characters, including mountain area,
City, rural area are pacifically regional, according to the particularity and complexity of detection target, use the distance measurement value and different measurement zones of statistics
Between noise energy determine the measurement confidence level in different sections, adjust the length and delay length τ of pseudo-random binary code.
Further, the noise energy of the sequence target reflection echo minimizes, and calculates obtained waveform measurement area
Between and correctness, while extracting priori training samples information of the information from objective pattern as the BP neural network of echo, join
With model training.
The present invention has the advantages that
It proposes the medium-long range high frequency lasers radar range finding based on the modulation of pseudo-random binary code and obscures method for solving, be applicable in
Range ambiguity in solution high frequency Full wave shape airborne laser radar.Using BP neural network model, according to waveform morphology
The characteristic information and echo range information model that information is established in single echo waveform are to calculate all potentially possible measurement zones
Between and and its distribution probability, fully considered the waveform morphology information of each waveform component of Full wave shape echo;Pass through multiple waves
The internal correlation information of shape echo-signal further determines that surveying range and as initial surveying range information, improves operation
Accuracy and confidence level;According to different terrain geomorphologic characteristics, the distance measurement value of statistics and the noise energy of different surveying ranges are used
The measurement confidence level for determining different sections determines therefrom that the length and delay size of pseudo-random binary code sequence, enhances this
The universality of method, it is simple and effective;Pulse launch time is adjusted using pseudo-random binary code, passes through the difference of continuous measurements
Vector characterizes noise information, can preferably reduce the influence of high frequency noise components, improves and calculates while reducing calculation amount
Confidence level.
Detailed description of the invention
Fig. 1 is the schematic diagram of range ambiguity during the high frequency lasers that the present invention solves measure.
Fig. 2 is general design idea schematic diagram of the invention.
Fig. 3 is the BP neural network model schematic established according to echo character and echo section.
Fig. 4 is that pseudo-random binary code modulates afterpulse transmitting schematic diagram.
Fig. 5 is that cloud distribution schematic diagram is put under different surveying ranges.
Specific embodiment
The present invention is described in further detail in the following with reference to the drawings and specific embodiments:
The present invention proposes that a kind of medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving, uses
The range ambiguity in solution high-frequency Full wave shape laser radar.In lidar measurement, with to airborne laser radar
The requirement of measurement distance and laser footpoint density be continuously improved, inevitably there is range ambiguity in high-frequency laser radar
Problem, such as Fig. 1, i.e., before the exomonental echo reception of the first beam, because tranmitting frequency is excessively high, interlude may be sent out again
A branch of or even multiple laser pulse is penetrated, therefore, it is impossible to which correctly correspond to seed light and echo directly from time series is associated with pass
Thus system leads to the problem of measurement fuzziness.When laser emission frequency (PRR, Pulse Repetition Rate) is 400kHz
When, when realistic objective distance is 1650m, directly according to all possible distance of time series calculating are as follows: 150m, 525m,
900m, 1275m and 1650m, here it is range ambiguities.
It is modulated using BP neural network and pseudo-random binary code to solve the method overall process stream of range ambiguity
Journey is as shown in Figure 2.In detail the following steps are included:
S1, single laser Full wave shape echo-signal, extraction echo range information and echo waveform characteristic information are used.Assuming that
Transmitting laser pulse signal is H (t), and atmospheric attenuation rate is β, and observation point and target spacing are L, and target has N layers, gross area A,
I-th layer of corresponding reflectivity, level product and layer height are respectively αi, A and li, then the echo in detector surface can be write as:
Therefore, received echo-signal on the detector are as follows:
Full wave shape echo complex energy are as follows:
Full wave shape echo average energy are as follows:
According to the range equation of laser radar, the signal-to-noise ratio of signal can be received are as follows:
Wherein, I is beam intensity;Ω is the solid angle (sr) of target;NEI is receiver equivalent noise irradiation level (W/
m2), it is determined by receiver performance;F is the target retroreflectance (1/sr) of unit solid angle;ηTReliability equivalence factor is broadened for pulse;
α is atmospheric extinction coefficient (1/m), including atmospheric molecule absorbs and scattering, aerosol absorb and scattering, other molecular scatterings or suction
It receives.
Analysis mode (4)-(6) it is found that influence laser echo signal complex energy relevant to ground object target, average energy and
The principal element of signal-to-noise ratio is the reflectivity of observation point and target spacing, ground object target form and ground object target.Therefore using sharp
The principal element and waveform morphology information of optical echo signal synthesis energy, average energy and signal-to-noise ratio include waveform position of centre of gravity,
Waveform widths, waveform height information etc. can establish and rise detection target and contacts away from measurement range between observation point.
S2, echo waveform characteristic information and echo range information model are established using BP neural network, determine n-th echo
The distance range and distribution probability of corresponding target.Single laser Full wave shape echo-signal, root are acquired using low-frequency laser
Extracting wave configuration feature according to S1, (echo complex energy parameter, echo signal-to-noise ratio parameter, echo mean breadth parameter, height are joined
The echoes shape informations such as number, mean energy parameter, position of centre of gravity parameter) information and as BP neural network input item, according to low
The laser of frequency collected time scale calculate the waveform area that the place of identical echo is collected in HF laser
Between MTA [i], time when MTA (Multiple Time Around) interval value i is defined as echo reception is located at i-1 pulse
Between the time of recurrence interval and i pulse recurrence interval, by the waveform section MTA [i] at the place of echo and in different measurements
Probability in section establishes 4 layers of BP neural network as the output item of neural network, and intermediate state transfer function distinguishes selection line
Property function and piecewise function, such as Fig. 3.Using multiple measured data as train samples.
S3, all-wave graphic data medium wave shape signal decomposition (waveform signal of laser radar decomposes).Due to answering for ground object target
Polygamy, incident laser beam may be divided into the different back scattering source of several intensity by ground object target, so that laser radar connects
The digitized wave forms signal received will no longer be a class Gaussian waveform, but several amplitudes and the different class high bass wave of width
The superposition of shape, therefore need to carry out waveform for airborne Full wave shape echo-signal and decompose preferably to analyze ground object target, it uses
Waveform decomposition algorithm be Levenberg-Marquardt decomposition algorithm.
Assuming that Full wave shape signal is made of N number of echo, then the waveform can be expressed as the superposition of N number of Gaussian component,
That is:
Wherein, Ak、μk、ωkRespectively represent the waveform maximum amplitude, waveform center and waveform half-breadth of k-th of waveform component.
The purpose that waveform decomposes is exactly to seek a series of most suitable (Ak、μk、ωk), so that the error of fitting is minimum, i.e., so that fitting
Gap between waveform and original waveform is small as far as possible or is less than error tolerance ∈.
All Gaussian components to be asked are expressed as P (p1, p2, p3... pm), m=3*N, then analog waveform can be with table
It is shown as: f (xi, P), given function initial value p first0, by f (xi, P) and in p0Place carries out Taylor expansion, and omits quadratic term and more
High-order term.It is possible thereby to extrapolate objective function:
Obtain objective function:
The ultimate aim of LM algorithm is so that objective function Q is minimum.To objective function ask first derivative make its 0, pass through
Matrix operation, the classical iterative formula of available LM algorithm:
P=p0+ [H (x, p0)+λE]-1JT(x, p0) [y-f (x, p0)] (11)
Wherein, J is that have function f (xi, P) first-order partial derivative composition Jacobian matrix, H be its second-order partial differential coefficient form
Hesse matrices, consider for practical calculation amount and feasibility, replaced using quasi- Hesse matrices, it may be assumed that
H=JTJ (12)
LM algorithm is the process of a continuous iteration, when formula to calculating result and actual amount gap very little or reach most
When big the number of iterations, that is, think that iteration terminates, wherein initial parameter information p in iteration0Second order peak detection method can be used
It obtains.
S4, the available measurement distance of S2 and probability distribution information are used according to the measurement sequence target that S3 is decomposed,
According to the continuity features of area target and use optimization algorithm, can further calculate measurement sequence optimization measurement away from
From range MTA [X].Assuming that one shares N number of area MTA, laser is t at the time of emitting i-th of pulseT[i], detector detects
It is t at the time of j-th of echoE[j], emitted laser pulse number is G [j] at this time, exports result by prosthomere neural network
The distribution probability of available Qi MTA area k (k=1,2 ... N) is pJ, k, measure distance LJ, kWith following relationship:
Ru×(k-1)≤LJ, k≤Ru×k (13)
So for the continuous N echo since the 1st, this N number of section MTA probability distribution matrix and measurement away from
It is respectively as follows: from matrix
If the Mean Vector in the section MTA of this continuous N echo are as follows:
The Mean Vector of measurement distance is respectively as follows:
It since ground object target geometric shape can be considered as continuously in subrange, and be not in size is more than one
The case where a area MTA.Using Least Square Method, distance is measured with the expectation that straight line f is fitted this M echo, even if formula
(14) value is minimum, can obtain the optimal estimation of each echometric measurement distance, the i.e. section MTA optimal estimation value.
S5, the length for optimizing pseudo-random binary code is determined.When the sequence echo quantity for participating in optimal MTA interval estimation
When M value is excessive, it can be runed counter to comprising the multiple echoes individually emitted with the supposed premise of optimal estimation;When M value is too small,
It may be unable to get preferable fitting result, therefore according to different terrain geomorphic feature, in mountain area, city, rural area and flat land
Area is determined according to the particularity and complexity of detection target using the distance measurement value of statistics and the noise energy of different surveying ranges
The measurement confidence level in different sections adjusts the length and delay length τ of pseudo-random binary code.Delay length τ is generally and laser
Pulsewidth is related, such as delay length τ is the 1/3 of laser pulse width.
S6, pulse launch time is adjusted according to the pseudo-random binary code sequence of generation, 0-pulse launch time is not more
Change, 1-pulse launch time postpones 1 time scale τ, such as Fig. 4;
S7, according to S4, M continuous echo-signals can be calculated in the distance vector R in each section MTAMTA,j=(r1,
r2,...,rM)T, it represents the M measurements in the jth time section MTA of the one of pseudo-random binary sequence of scan line
Distance.Since the constant low frequency component of the distance signal in each section MTA is that can be ignored to the contribution of noise energy
, therefore the difference vector Δ R of calculating continuous measurements can be passed throughMTA,j, such as formula 21, from each distance vector RMTA,jIn
Eliminate high frequency noise components.So, the associated each difference vector Δ R in all potential sections MTAMTA,jNoise energyIt can be provided by formula 22, such as Fig. 5.
ΔRMTA,j=(r2-r1,r3-r2,...,rM-rM-1)T (21)
S8, according to the noise figure calculated in S7, it is clear that the smallest noise energyIt is not influenced by PPM, j at this time
It is worth associated with the correct section MTA.So can distribute to by comparing each measurement in all potential sections MTA away from
Noise signal energy from group (M), selects the section MTA at the place with minimal noise energy as a result.
S9, seed light and echo corresponding relationship are determined according to S8, the corresponding measurement distance of each echo in calculating Full wave shape.
The present invention is based on the medium-long range high frequency lasers radar range findings of pseudo-random binary code modulation to obscure method for solving, is applicable in
Range ambiguity in solution high frequency Full wave shape airborne laser radar.Using BP neural network model, according to waveform morphology
The characteristic information and echo range information model that information is established in single echo waveform are to calculate all potentially possible measurement zones
Between and and its distribution probability, fully considered the waveform morphology information of each waveform component of Full wave shape echo;Pass through multiple waves
The internal correlation information of shape echo-signal further determines that surveying range and as initial surveying range information, improves operation
Accuracy and confidence level;According to different terrain geomorphologic characteristics, the distance measurement value of statistics and the noise energy of different surveying ranges are used
The measurement confidence level for determining different sections determines therefrom that the length and delay size of pseudo-random binary code sequence, enhances this
The universality of method, it is simple and effective;Pulse launch time is adjusted using pseudo-random binary code, passes through the difference of continuous measurements
Vector characterizes noise information, can preferably reduce the influence of high frequency noise components, improves and calculates while reducing calculation amount
Confidence level.
Above embodiments are merely to illustrate design philosophy and feature of the invention, and its object is to make technology in the art
Personnel can understand the content of the present invention and implement it accordingly, and protection scope of the present invention is not limited to the above embodiments.So it is all according to
It is within the scope of the present invention according to equivalent variations made by disclosed principle, mentality of designing or modification.
Claims (8)
1. a kind of medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving, which is characterized in that packet
Include following steps:
Echo waveform characteristic information and echo range information model are established using BP neural network, it is all to calculate single measurement echo
Potential measurement is apart from section and each section distribution probability, and wherein echo waveform morphological feature is BP neural network input item;
Advanced optimized using surveying range input item of the waveform decomposition algorithm to BP neural network, calculate the measurement of echo away from
From section;
The laser pulse emission time is modulated using pseudo-random binary sequence, most according to the noise energy of sequence target reflection echo
The criterion of smallization calculates accurate sounding section.
2. the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving according to claim 1,
It is characterized by:
The echo waveform morphological feature include echo complex energy parameter, echo signal-to-noise ratio parameter, echo mean breadth parameter,
Height parameter, mean energy parameter, position of centre of gravity parameter.
3. the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving according to claim 1,
It is characterized by:
Detailed process is as follows for the foundation of the BP neural network: acquiring single laser Full wave shape echo using low-frequency laser
Signal is collected using the wave configuration feature information of extraction as BP neural network input item according to low-frequency laser
Time scale calculate collected in HF laser identical echo place waveform section MTA [i], MTA (Multiple
Time Around) time of interval value i when being defined as echo reception be located at i-1 pulse recurrence interval time and i pulse
Between recurrence interval, using the waveform section MTA [i] at the place of echo and the probability in different surveying ranges as neural network
Output item establish 4 layers of BP neural network, intermediate state transfer function selects linear function and piecewise function respectively.
4. the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving according to claim 1,
It is characterized by: the waveform decomposition algorithm is Levenberg-Marquardt decomposition algorithm.
5. the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving according to claim 1,
It is characterized by:
It is described that the optimization process tool advanced optimized is done using surveying range input item of the waveform decomposition algorithm to BP neural network
Body is as follows:
According to the continuity features of area target and optimization algorithm is used, the optimization that can further calculate measurement sequence is surveyed
It measures distance range MTA [X];Assuming that one shares N number of area MTA, laser is t at the time of emitting i-th of pulseT[i], detector inspection
It is t at the time of measuring j-th of echoE[j], emitted laser pulse number is G [j] at this time, is exported by prosthomere neural network
As a result the distribution probability of available Qi MTA area k (k=1,2 ... N) is pj,k, measure distance Lj,kWith following relationship:
Ru×(k-1)≤Lj,k≤Ru×k
So for the continuous N echo since the 1st, in the probability distribution matrix in this N number of section MTA and measurement apart from square
Battle array is respectively as follows:
If the Mean Vector in the section MTA of this continuous N echo are as follows:
The Mean Vector of measurement distance is respectively as follows:
It since ground object target geometric shape can be considered as continuously in subrange, and be not in size is more than one
The case where area MTA;Using Least Square Method, distance is measured with the expectation that straight line f is fitted this M echo, even if formula
(14) value is minimum, can obtain the optimal estimation of each echometric measurement distance, the i.e. section MTA optimal estimation value:
6. the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving according to claim 1,
It is characterized by:
The pseudo-random binary sequence modulation is depending on different topography and landform characters, including mountain area, city, rural area peace
Ground area uses the distance measurement value of statistics and the noise energy of different surveying ranges according to the particularity and complexity of detection target
It determines the measurement confidence level in different sections, adjusts the length and delay length τ of pseudo-random binary code.
7. the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving according to claim 5,
It is characterized by:
Detailed process is as follows for the criterion that the noise energy of the sequence target reflection echo minimizes:
Since the constant low frequency component of the distance signal in each section MTA is that can be ignored to the contribution of noise energy
, therefore the difference vector Δ R of calculating continuous measurements can be passed throughMTA,j, it is specific as follows:
ΔRMTA,j=(r2-r1,r3-r2,...,rM-rM-1)T
From each distance vector RMTA,jMiddle elimination high frequency noise components, then, all potential associated each differences in the section MTA
Vector Δ RMTA,jNoise energyIt can be given by:
The smallest noise energyIt is not influenced by PPM, j value at this time is associated with the correct section MTA;So can
To distribute to noise signal energy of each measurement in all potential sections MTA apart from group (M) by comparing, selection has most
The section MTA at the place of small noise energy is as a result.
8. the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation obscures method for solving according to claim 7,
It is characterized by:
The noise energy of the sequence target reflection echo, which minimizes, calculates obtained waveform measurement section and correctness, can be with
The information from objective pattern for extracting echo simultaneously is further used as the priori training samples information of BP neural network, participates in model instruction
Practice.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810965428.8A CN109164454B (en) | 2018-08-23 | 2018-08-23 | Pseudo-random code modulation-based medium-long range high-frequency laser radar ranging ambiguity solving method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810965428.8A CN109164454B (en) | 2018-08-23 | 2018-08-23 | Pseudo-random code modulation-based medium-long range high-frequency laser radar ranging ambiguity solving method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109164454A true CN109164454A (en) | 2019-01-08 |
CN109164454B CN109164454B (en) | 2020-10-30 |
Family
ID=64896424
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810965428.8A Active CN109164454B (en) | 2018-08-23 | 2018-08-23 | Pseudo-random code modulation-based medium-long range high-frequency laser radar ranging ambiguity solving method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109164454B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110221241A (en) * | 2019-04-29 | 2019-09-10 | 西安电子科技大学 | A kind of low elevation angle DOA estimation method based on RBF neural |
CN111308489A (en) * | 2020-03-16 | 2020-06-19 | 杭州爱莱达科技有限公司 | Double-channel pseudo-random code phase modulation optical domain analysis focusing coherent laser method and device |
WO2020233415A1 (en) * | 2019-05-17 | 2020-11-26 | 深圳市速腾聚创科技有限公司 | Laser radar, and anti-jamming method therefor |
CN113204027A (en) * | 2021-05-06 | 2021-08-03 | 武汉海达数云技术有限公司 | Pulse type laser radar cross-period ranging method for accurately selecting ranging period |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5227784A (en) * | 1990-12-10 | 1993-07-13 | Mazda Motor Corporation | System for detecting and determining range of target vehicle |
CN104931973A (en) * | 2015-06-17 | 2015-09-23 | 浙江理工大学 | Asymmetric phase encoding ranging method applied to laser radar system |
CN105738890A (en) * | 2016-02-25 | 2016-07-06 | 沈阳承泰科技有限公司 | Method for improving radar measurement scope and measurement precision and radar |
CN106054204A (en) * | 2016-07-26 | 2016-10-26 | 北京邮电大学 | Long distance and high accuracy oriented compound laser range finding method and system |
CN107402386A (en) * | 2017-08-02 | 2017-11-28 | 中南大学 | A kind of method of estimation of underground metalliferous pipe radius and buried depth based on BP neural network |
-
2018
- 2018-08-23 CN CN201810965428.8A patent/CN109164454B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5227784A (en) * | 1990-12-10 | 1993-07-13 | Mazda Motor Corporation | System for detecting and determining range of target vehicle |
CN104931973A (en) * | 2015-06-17 | 2015-09-23 | 浙江理工大学 | Asymmetric phase encoding ranging method applied to laser radar system |
CN105738890A (en) * | 2016-02-25 | 2016-07-06 | 沈阳承泰科技有限公司 | Method for improving radar measurement scope and measurement precision and radar |
CN106054204A (en) * | 2016-07-26 | 2016-10-26 | 北京邮电大学 | Long distance and high accuracy oriented compound laser range finding method and system |
CN107402386A (en) * | 2017-08-02 | 2017-11-28 | 中南大学 | A kind of method of estimation of underground metalliferous pipe radius and buried depth based on BP neural network |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110221241A (en) * | 2019-04-29 | 2019-09-10 | 西安电子科技大学 | A kind of low elevation angle DOA estimation method based on RBF neural |
WO2020233415A1 (en) * | 2019-05-17 | 2020-11-26 | 深圳市速腾聚创科技有限公司 | Laser radar, and anti-jamming method therefor |
CN111308489A (en) * | 2020-03-16 | 2020-06-19 | 杭州爱莱达科技有限公司 | Double-channel pseudo-random code phase modulation optical domain analysis focusing coherent laser method and device |
CN111308489B (en) * | 2020-03-16 | 2022-02-11 | 杭州爱莱达科技有限公司 | Optical domain analysis focusing method and device of pseudo-random code phase modulation laser radar |
CN113204027A (en) * | 2021-05-06 | 2021-08-03 | 武汉海达数云技术有限公司 | Pulse type laser radar cross-period ranging method for accurately selecting ranging period |
CN113204027B (en) * | 2021-05-06 | 2024-06-11 | 武汉海达数云技术有限公司 | Pulse type laser radar cross-period ranging method capable of precisely selecting ranging period |
Also Published As
Publication number | Publication date |
---|---|
CN109164454B (en) | 2020-10-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Smith et al. | Land ice height-retrieval algorithm for NASA's ICESat-2 photon-counting laser altimeter | |
CN109164454A (en) | A kind of fuzzy method for solving of the medium-long range high frequency lasers radar range finding based on pscudo-random codc modulation | |
Wagner et al. | Gaussian decomposition and calibration of a novel small-footprint full-waveform digitising airborne laser scanner | |
CN108426856B (en) | Assess laser radar Satellite observation Atmospheric CO2The overall analysis system of concentration performance | |
CN104833979A (en) | Laser ranging method and laser ranging data signal processing method | |
Banakh et al. | Estimation of the turbulence energy dissipation rate from the pulsed Doppler lidar data | |
CN108445471A (en) | A kind of range accuracy appraisal procedure under the conditions of single-photon laser radar multi-detector | |
CN110927735B (en) | Multi-target distance measuring method based on multi-channel full-waveform laser radar data | |
CN106872969B (en) | Radar target angle estimation method based on MTD pulse accumulation and sliding processing | |
CN109633601B (en) | Satellite-borne laser radar foot point accurate positioning method based on earth surface model | |
CN101839981A (en) | Method and device for acquiring laser imaging echo waveform and level characteristics | |
CN105353355B (en) | A kind of multistatic radar multi-target orientation method based on sparse reconstruct and projection imaging | |
CN110031856A (en) | A kind of unrestrained attenuation coefficient extracting method of airborne LiDAR depth measurement data | |
CN105068063A (en) | Ground object characteristic obtaining method based on echo full-waveform decomposition | |
CN108845313A (en) | Moving target detection method based on Orthogonal Subspaces projection under limited training sample | |
CN114509409B (en) | Chlorophyll concentration vertical section inversion method of satellite-borne single photon laser radar | |
CN107255814A (en) | A kind of radar target detection method based on LFMSK waveforms | |
CN114637021B (en) | Sub-centimeter-level full-waveform laser radar ranging method and device | |
Banakh et al. | Determination of optical turbulence intensity by atmospheric backscattering of laser radiation | |
CN114235173A (en) | Photon counting satellite-borne marine laser radar detection simulation method | |
Guenther et al. | Laser applications for near-shore nautical charting | |
CN111780726A (en) | Ocean laser sounding method and system | |
CN111007490A (en) | Sky wave over-the-horizon radar coordinate registration method based on buoy geographic information | |
CN116136590A (en) | Method, device and storage medium for calibrating Doppler laser radar focal length | |
Rieger et al. | Resolving range ambiguities in high-repetition rate airborne lidar applications |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |