CN106154246A - Echo photon signal real-time processing method - Google Patents

Abstract

The present invention is relevant a kind of echo photon signal real-time processing method, and it comprises the following steps: energy accumulation detecting step, processes including echo photon signal compression, and motion compensation accumulates, and often walks slip, finds out useful photon and O-C value thereof；And time coherent detection step, including preliminary treatment, Robust filter process, iterative processing and data output.This processing method, it is possible to effectively suppress effect of noise, improves the probability of the most effectively echo detecting, can process in real time while improving the processing accuracy of low signal-to-noise ratio echo-signal.

Description

Echo photon signal real-time processing method

Technical field

The present invention relates to laser diffusion range finding field, real particularly to a kind of faint echo photon signal Time processing method.

Background technology

Satellite laser ranging (SLR) (Satellite Laser Ranging, SLR) is precision during extraterrestrial target is measured One of the highest technology, its certainty of measurement has reached grade.The object of conventional laser range finding is equipped with The cooperative extraterrestrial target of corner reflector.When detection target is noncooperative target, such as space junk, adopt When finding range with diffuse-reflectance return laser beam, owing to space junk reflective surface area is little, irreflexive echo Number of photons is few, and the exceptional value of detection is more, and signal to noise ratio is relatively low；Or when detection target is the conjunction of deep space When making target, because distance is remote, the exceptional value facing detection equally is more, the problem that signal to noise ratio is relatively low.

The most conventional returned photon numbers has following several according to processing method:

One, artificial screen processing method, the method precision is high, good to the adaptability of weak signal, but depends on Relying in interpretation personnel's experience, detection efficiency and automaticity are low.

Two, quick echo discrimination method, the method has feature fast and effectively, to low orbit satellite There is the highest detection probability (close to 100%), but for high rail satellite sounding success rate than relatively low (1% Below).

Three, related detecting method, the method principle is simple, but distance window size, threshold value, sky Vaporous condition is bigger on Effect on Detecting impact.

Four, for N/M (the M ＞ N) detection method of Low SNR signal, the method adaptability is good, but There is processing procedure amount of calculation bigger, it is impossible to meet the deficiency of requirement of real-time.

Five, Poisson (Poisson) statistical filtering method, the method is current Lunar LASER Ranging (Lunar Laser Ranging, LLR) and the main stream approach of satellite laser ranging (SLR) employing, Australia Si Telongluo Mountain observatory (Mt Stromlo Observatory) stands and applies Poisson statistics filtering method to achieve nothing The process of the full-automatic echo-signal of people's post, but the testing result of the method is filtered shadow by Poisson statistics Ring bigger.

At present, use above-mentioned existing echo photon processing method for the echo photon letter of low signal-to-noise ratio Number it is difficult to realize high accuracy simultaneously and real-time data process.

Summary of the invention

Because the defect existing for above-mentioned prior art, it is an object of the invention to, it is provided that a kind of time Glistening light of waves subsignal real-time processing method so that it is overcome that echo photon signal signal to noise ratio is low, abnormal observation Many, can process in real time while improving the processing accuracy of low signal-to-noise ratio echo photon signal.

To achieve these goals, a kind of echo photon signal side of process in real time proposed according to the present invention Method, including following steps: energy accumulation detecting step, processes including echo photon signal compression, Motion compensation accumulates, and often walks slip, finds out useful photon and O-C value (measured value-guiding value) thereof； And time coherent detection step, including preliminary treatment, Robust filter process, iterative processing and data are defeated Go out.

The present invention also can be applied to the following technical measures to achieve further.

Aforesaid echo photon signal real-time processing method, wherein said energy accumulation detecting step, root It is compressed processing to echo photon signal according to object reference orbital data, echo photon energy is concentrated To same Bo Mennei, the method for motion compensation accumulation is used to find out some at energy accumulation detection window Individual effective echo photon, detects one by one, it is thus achieved that the O-C value of each effective echo photon.

Aforesaid echo photon signal real-time processing method, wherein said preliminary treatment, is that utilization is described The effectively O-C value of echo photon fits to initial polynomial function g, and obtains current time predictive value, Compared with measured value by predictive value and determine whether current time measured value is virtual value.

Aforesaid echo photon signal real-time processing method, wherein said Robust filter processes, is with just Each effective echo photon that step processes the virtual value that determines of step and energy accumulation detecting step obtains Fit polynomial function g based on O-C value, it is thus achieved that estimating residual sequence, recycling Robust filter is again Estimate that initial polynomial function g ' carries out secondary detection, determine whether current time measured value is virtual value And export virtual value.

Aforesaid echo photon signal real-time processing method, wherein said iterative processing, is ought to determine to work as After front moment measured value is virtual value, ordered sequence increases the ordered sequence value of current time, and Remove the oldest ordered sequence value, use alternative manner to repeat initial processing step and Robust filter processes Step, completes whole processing procedure.

Data through iterative processing are wherein carried out by aforesaid echo photon signal real-time processing method Data export.

Aforesaid echo photon signal real-time processing method, the number of wherein said effective echo photon is 2/ signal to noise ratio.

Aforesaid echo photon signal real-time processing method, the number of wherein said effective echo photon can Signal to noise ratio self adaptation according to target is chosen.

The present invention compared with prior art has clear advantage and beneficial effect.By above-mentioned technical side Case, the echo photon signal real-time processing method of the present invention, at least have the advantage that

One, the echo photon signal real-time processing method of the present invention, by entering the echo photon of detection Row motion compensation and Robust filter process, it is possible to effectively suppress effect of noise, improve the most effectively The probability of echo detecting.

Two, the echo photon signal real-time processing method of the present invention, can improve low signal-to-noise ratio echo light Process in real time while the processing accuracy of subsignal.

Accompanying drawing explanation

Fig. 1 is echo-signal photon real-time processing method step schematic diagram of the present invention.

Fig. 2 is that echo-signal photon real-time processing method echo photon of the present invention accumulates schematic diagram.

Fig. 3 is that one embodiment of echo-signal photon real-time processing method of the present invention processes target 10517 Result schematic diagram.

Fig. 4 is that one embodiment of echo-signal photon real-time processing method of the present invention processes target 17590 Result schematic diagram.

Fig. 5 is that one embodiment of echo-signal photon real-time processing method of the present invention processes target 23769 Result schematic diagram.

Detailed description of the invention

By further illustrating the technological means and merit that the present invention taked by reaching predetermined goal of the invention Effect, below in conjunction with accompanying drawing and preferred embodiment, locates in real time to the echo photon signal proposed according to the present invention Its detailed description of the invention of reason method, step, structure, feature and effect thereof describe in detail.

Refer to shown in Fig. 1, be echo photon signal real-time processing method step schematic diagram of the present invention. Echo photon signal real-time processing method of the present invention includes energy accumulation detecting step S1 and time correlation inspection Survey step S2.Wherein energy accumulation detecting step S1 includes two processing procedures, i.e. echo photon signal Motion compensation accumulation S11, often step slip find out useful photon and O-C value S12 thereof；Time correlation detects Step S2 includes four processing procedures, i.e. preliminary treatment S21, Robust filter process at S22, iteration Reason S23 and data output S24, wherein O-C represents that measured value deducts guiding value, and wherein measured value refers to echo The moment delay, τ that photon produces_k, the time delay of guiding value feeling the pulse with the finger-tip mark reference orbit conversion

Preferably, in energy accumulation detecting step S1, according to object reference orbital data to echo-signal Carry out motion compensation accumulation, echo photon energy is focused on same echo Bo Mennei, at energy product In tired detection window M, find out p effectively echo photon, wherein p < M, by window forward slip, by Individual detection, it is thus achieved that the O-C value of m effective echo, for follow-up time correlation detection.

Preferably, in energy accumulation detecting step S1, motion compensation accumulation is preferably to go out from reference orbit Send and realize motion compensation.Refer to shown in Fig. 2, according to reference orbit data, obtain each pulse and send out Penetrate moment t_kMeasure the radial distance of target relative laserAnd it is converted into corresponding time delayFurther according to time delayProduce echo photon collection echo ripple door M_k, echo ripple door M_k Central instant correspondence time delay

In actual measurement, the moment delay, τ that echo photon produces_kThe time delay converted with object reference trackThere is time delay errorAs shown in Equation 1:

${\widehat{\mathrm{\&tau;}}}_k={\mathrm{\&tau;}}_k-{\hat{t}}_k\&Element;(-T_G/2T_G/2)$ ... ... ... formula 1

Wherein, T_GFor gatewidth.

Take interval certain the echo photon signal interior of accumulation to select as in reference signal, this preferred embodiment Signal on the basis of first echo, other echo photons are aligned by motion compensation.Method for optimizing As follows:

If the track in energy accumulation window meets state equation as shown in Equation 2:

X_k=Φ_k,1(X₁)+W_k... ... ... formula 2

Wherein, X₁And X_kFor t₁And t_kThe state vector in moment, Φ_k,1For state transition function, W_kFor State-noise vector, sets again t_kThe observation model in moment is as shown in Equation 3:

τ_k=h (X_k)+ε_k... ... ... formula 3

Wherein,

$h\left(X_k\right)=2\sqrt{{\left(\mathrm{\&Delta;}{X}_k\right)}^2+{\left(\mathrm{\&Delta;}{Y}_k\right)}^2+{\left(\mathrm{\&Delta;}{Z}_k\right)}^2}$

ΔX_k=X (t_k)-X₀

ΔY_k=Y (t_k)-Y₀

ΔZ_k=Z (t_k)-Z₀

Wherein, τ_kFor t_kThe observed quantity in moment, ε_kIt is respectively the random error of observed quantity, X (t_k),Y(t_k),Z(t_k) For t_kThe position of moment target；X₀,Y₀,Z₀Position for laser survey station.

Make x_k=X_k-X_k ^*, X_k ^*For t_kThe state vector of moment reference orbit,Time then It is poor to be delayedAs shown in Equation 4:

$\mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_k={\mathrm{\&tau;}}_k-{\widehat{\mathrm{\&tau;}}}_k=h\left(X_k\right)-h\left({X_k}^{*}\right)=\frac{\&PartialD;h}{\&PartialD;X}{|}_{x_k}{x}_k+{\mathrm{\&epsiv;}}_k$ $=H_k\mathrm{\&Phi;}(t_k,t_1)x_1+{\mathrm{\&epsiv;}}_k$ ... ... ... formula 4

Wherein, H_k=2 [a₁(k) a₂(k) a₃(k)]

$\left\{\begin{array}{c}{a}_1\left(k\right)=\mathrm{\&Delta;}{X}_k/R_k\\ a_2\left(k\right)=\mathrm{\&Delta;}{Y}_k/R_k\\ a_3\left(k\right)=\mathrm{\&Delta;}{Z}_k/R_k\end{array}\right.,$ $R_k=\sqrt{{\left(\mathrm{\&Delta;}{X}_k\right)}^2+{\left(\mathrm{\&Delta;}{Y}_k\right)}^2+{\left(\mathrm{\&Delta;}{Z}_k\right)}^2}$

Thus can obtain time delay errorAs shown in Equation 5:

$\mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_1={\mathrm{\&tau;}}_1-{\widehat{\mathrm{\&tau;}}}_1=H_1\mathrm{\&Phi;}(t_1,t_1)x_1+{\mathrm{\&epsiv;}}_1=H_1{x}_1+{\mathrm{\&epsiv;}}_1$ ... ... ... formula 5

Ignore the impact of random error, thenThen time delay errorAs shown in Equation 6:

$\mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_k=H_k\mathrm{\&Phi;}(t_k,t_1){H_1}^{-1}\mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_1+{\mathrm{\&epsiv;}}_k$ ... ... ... formula 6

Therefore, H_kΦ(t_k,t₁)H₁ ^-1It is exactly t₁The observed quantity in moment is to t_kThe motion compensation factor in moment. If t_kMoment and t₁Moment is same target measurement value, then t_kValue after moment compensation should be at t₁Time Carve in the neighborhood of measured value.

In motion compensation accumulation step S11, it is judged that the method for optimizing of error threshold δ is as described below:

The error of reference orbit can produce impact to the energy accumulation of photon, by reference orbit in the present invention The error of data is divided into radial velocity error, radial acceleration error and angular error, and wherein angle is by mistake Difference is azimuth angle error and angle of pitch error.

When angular error is bigger, target may be not in laser range of exposures, affects catching of target Obtain, but accumulated error is not affected.The most only need to analyze the radial velocity error of reference orbit, footpath To the acceleration error impact on accumulation.

Assume dbjective state transfer matrix Φ (t_k,t₁) meet Secondary movement model, and reference orbit data Radial distanceIt is approximately even acceleration to change, thenAs shown in Equation 7:

${\hat{R}}_k\&ap;{\hat{R}}_1+{\hat{v}}_1\&CenterDot;\mathrm{\&Delta;}{T}_k+{\hat{a}}_1\&CenterDot;{\left(\mathrm{\&Delta;}{T}_k\right)}^2/2$ ... ... ... formula 7

Wherein, Δ T_k=t_k-t₁,It is respectively t₁The radial velocity of moment reference orbit and radial direction Acceleration.

By t_kThe measurement in moment runs up to t₁Moment, then:

${\hat{R}}_1\&ap;{\hat{R}}_k-{\hat{v}}_1\&CenterDot;\mathrm{\&Delta;}{T}_k-{\hat{a}}_1\&CenterDot;{\left(\mathrm{\&Delta;}{T}_k\right)}^2/2$

R₁≈R_k-v₁·ΔT_k-a₁·(ΔT_k)²/2

${\hat{R}}_1-R_1\&ap;{\hat{R}}_k-R_k-({\hat{v}}_1-v_1)\&CenterDot;\mathrm{\&Delta;}{T}_k-({\hat{a}}_1-a_1)\&CenterDot;{\left(\mathrm{\&Delta;}{T}_k\right)}^2/2$ $={\hat{R}}_n-R_n-\mathrm{\&Delta;}{v}_1\&CenterDot;\mathrm{\&Delta;}{T}_n-\mathrm{\&Delta;}{a}_1\&CenterDot;\mathrm{\&Delta;}{T}_n^2/2$ ... ... ... formula 8

Wherein, v₁、a₁It is respectively t₁The radial velocity of moment actual track and radial acceleration.

Therefore, if t_kMoment and t₁Moment is same target measurement value, error threshold δ after its compensation Meet:

$\mathrm{\&delta;}<|\mathrm{\&Delta;}{v}_1\&CenterDot;\mathrm{\&Delta;}{T}_k+\mathrm{\&Delta;}{a}_1\&CenterDot;\mathrm{\&Delta;}{T}_k^2/2|$ ... ... ... formula 9

When accumulating interval and being shorter, single order amount is principal element, the impact of negligible acceleration, error Thresholding δ can be approximately as shown in Equation 10,

δ ＜ | Δ v | Δ T ... ... ... formula 10

By judging in energy accumulation detection window M that measured value compensation is to t the most in the same time₁The error door in moment Limit, finds out p effectively echo photon, wherein p < M.By window forward slip, detect one by one, it is thus achieved that The O-C value of m effective echo, for fitting of a polynomial.

In time correlation detecting step S2, it is as follows that initial processing step S21 is preferable to carry out method:

Seriality based on target travel, the measurement time of return laser beam has relevant in time series Property, analysis shows, in the short period of time, O-C value can be by a polynomial repressentation.Now t_kTime The time delay error carved can be expressed as equation 11:

$\mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_k={\widehat{\mathrm{\&tau;}}}_k-{\mathrm{\&tau;}}_k=g\left(t_k\right)=a+{\mathrm{bt}}_k=\left[\begin{array}{cc}1& t_k\end{array}\right]\left[\begin{array}{c}a\\ b\end{array}\right]$ ... ... ... formula 11

Then, m moment equations simultaneousness be expressed as shown in Equation 12 with matrix form:

$\left[\begin{array}{c}\mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_1\\ \mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_2\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ \mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_m\end{array}\right]=\left[\begin{array}{cc}1& t_1\\ 1& t_2\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ 1& t_m\end{array}\right]\left[\begin{array}{c}a\\ b\end{array}\right]$ ... ... ... formula 12

According to the principle of least square, the valuation of polynomial coefficient can be obtained as shown in Equation 13:

A=(T^TP^-1T)^-1T^TP^-1Y ... ... ... formula 13

Wherein, $a=\left[\begin{array}{c}a\\ b\end{array}\right],Y=\left[\begin{array}{c}\mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_1\\ \mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_2\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ \mathrm{\&Delta;}{\widehat{\mathrm{\&tau;}}}_m\end{array}\right],T=\left[\begin{array}{cc}1& t_1\\ 1& t_2\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ 1& t_m\end{array}\right],P_0=I_m$ (m rank unit matrix)

Use polynomial fitting to detect the new effectiveness measuring echo at this.

Preferably, in initial processing step S21, with t_kBefore moment as a example by m effective measurement, after matching Polynomial function be g, forecast t_kMoment measured value g (t_k), then prediction error δ v_kShown in formula 14:

$\mathrm{\&delta;}{v}_k={\widehat{\mathrm{\&tau;}}}_k-g\left(t_k\right)$ ... ... ... formula 14

When | δ v_k| ＜ δ_k, δ_kFor threshold value, typically takeThen think t_kMoment measured value is available point, It it is otherwise exceptional value.

In time correlation detecting step S2, it is as follows that Robust filter process step S22 is preferable to carry out method:

For improving the reliability of detection in real time, the method using Robust filter, real-time measurement values is carried out Secondary detection.

Owing to each observation is all separate, the robust being designed to adapt to have superseded district is weighed such as Shown in formula 15:

$\mathrm{\&psi;}\left(v_k\right)=\left\{\begin{array}{cc}{\stackrel{\&OverBar;}{v}}_k& \left|{\stackrel{\&OverBar;}{v}}_k\right|\&le;k_0\\ k_0\mathrm{sign}\left({\stackrel{\&OverBar;}{v}}_k\right)& k_0<\left|{\stackrel{\&OverBar;}{v}}_k\right|\&le;k_1\\ 0& \left|{\stackrel{\&OverBar;}{v}}_k\right|>k_1\end{array}\right.$ ... ... ... formula 15

$w\left(v_k\right)=\left\{\begin{array}{cc}1& \left|{\stackrel{\&OverBar;}{v}}_k\right|\&le;k_0\\ k_0/\left|{\stackrel{\&OverBar;}{v}}_k\right|& k_0<\left|{\stackrel{\&OverBar;}{v}}_k\right|\&le;k_1\\ 0& \left|{\stackrel{\&OverBar;}{v}}_k\right|>k_1\end{array}\right.$ ... ... ... formula 16

Wherein, w represents robust weight factor, wherein,Variancek₀Desirable 1.5～3.0, k₁Desirable 2.5～5.0.

Robust filter is used to obtain polynomial coefficient as shown in Equation 17:

$a={\left(T^T{\stackrel{\&OverBar;}{P}}^{-1}T\right)}^{-1}{T}^T{\stackrel{\&OverBar;}{P}}^{-1}Y$ ... ... ... formula 17

Wherein,As w (v_kDuring)=0, then reject the measured value in corresponding moment After, reappraise variance as shown in Equation 18:

${\mathrm{\&sigma;}}_k=\sqrt{\frac{1}{N-m}\underset{k=1}{\overset{N-m}{\mathrm{\&Sigma;}}}{v_k}^2}$ ... ... ... formula 18

Detection for subsequent window.

Preferably, Robust filter processes in step S22, if t_kMoment measured value is available point, by it Adding in the individual effectively measured value of above m, fit polynomial function g ' carries out secondary detection again, tries to achieve Estimate residual sequence δ v '_kAs shown in Equation 19:

$\mathrm{\&delta;}{v}_k^{\&prime;}={\widehat{\mathrm{\&tau;}}}_k-g^{\&prime;}\left(t_k\right)$ ... ... ... formula 19

When | δ v'_k| ＜ δ '_k, δ '_kFor threshold value, typically take 3 σ_k, then t is thought_kMoment measured value is available point, It it is otherwise exceptional value.

Preferably, in iterative processing steps S23, when determining that Robust filter processes the t of S22_kMoment measures After value is for effective echo photon, by this effective returned photon numbers according to output S24, simultaneously at ordered sequenceMiddle increase t_kThe ordered sequence in momentAnd remove the oldest valid dataAdopt Repeat preliminary treatment S21 with alternative manner and Robust filter processes S22, complete the detection of whole sequence.

For precision and the reliability of verification algorithm, as a example by Yunnan Observatory diffuse-reflectance LDMS, The rocket remains that have chosen some volumes bigger carry out actual measurement, and energy of lasers is 3.4J, pulse Frequency is 10hz.

Constraints in embodiment: using the TLE orbital tracking on the same day as reference orbit.Energy product Tired window M=20 (compensate the number of photons in thresholding > 2), fitting of a polynomial window n=9.

Embodiment one

On January 24th, 2011, having carried out test for the first time, target 10517 is a size of 6.4m × 2.0m Remains, within the observation period of 148s, observe 1004 measurement data, wherein available points altogether 73, signal to noise ratio is 1/14.Refer to shown in Fig. 3, be that echo-signal photon of the present invention processes in real time One embodiment of method processes the result schematic diagram of target 10517, and wherein a is measured value, and b is In real time result, c is result afterwards.

Embodiment two

On January 25th, 2011, having carried out second time test, target 17590 is a size of The remains of 10.4m × 3.9m, within the observation period of 148s, observe 1141 measurement data altogether, Wherein available point 75, signal to noise ratio is 1/15.Refer to shown in Fig. 4, be echo-signal light of the present invention One embodiment of sub-real-time processing method processes the result schematic diagram of target 17590, and wherein a is Measured value, b is real-time result, and c is result afterwards.

Embodiment three

On February 5th, 2011, having carried out third time test, target 23769 is a size of 5.9m × 2.4m Remains, within the observation period of 127s, observe 547 measurement data, wherein available point 60 altogether Individual, signal to noise ratio is approximately 1/10.Refer to shown in Fig. 5, be that echo-signal photon of the present invention is located in real time One embodiment of reason method processes the result schematic diagram of target 23769, and wherein a is measured value, b Being real-time result, c is result afterwards.

Table 1. the application method and control methods statistical result

The statistical result of above three embodiment is as shown in table 1, it can be seen that this application method utilizes mesh Target orbital characteristics, achieves the first detection of echo photon energy accumulation by motion compensation；Based on having Effect measurement error meets the characteristic of multinomial model, introduces the polynomial iteration Robust filter that slides, Improve the probability of the most effectively echo detecting, reduce the probability of error detection.Utilize Yunnan astronomical The repeatedly measured data of platform is tested, by contrasting real-time testing result and result afterwards: the party Method (< 2s) in the case of effective measured value continuous print, can realize detecting the most in real time substantially；When When effectively measured value is sparse, the probability of error detection then can increased, when measuring signal to noise ratio 1/15 Within, real-time detection probability is superior to 90%, and signal to noise ratio improves more than 30 times than original measurement value.By This, this method measurement data to diffuse-reflectance this low signal-to-noise ratio of range finding, it is possible to effectively suppress noise Impact, the biggest aspect solves the live signal test problems of current diffuse-reflectance laser ranging.

Although the present invention is disclosed above with preferred embodiment, so it is not limited to what the present invention implemented Scope, the simple equivalence made according to claims of the present invention and description changes and modifies, Still fall within the range of technical solution of the present invention.

Claims (8)

1. an echo photon signal real-time processing method, it is characterised in that comprise the following steps:

Energy accumulation detecting step, accumulates including echo photon signal motion compensation, often walks slip, looks for Go out useful photon and O-C value thereof；And

Time correlation detecting step, including preliminary treatment, Robust filter process, iterative processing and data Output.

2. echo photon signal real-time processing method as claimed in claim 1, it is characterised in that described Energy accumulation detecting step, is compressed processing to echo-signal according to object reference orbital data, will Echo photon energy focuses on same Bo Mennei, uses motion compensation to amass at energy accumulation detection window Tired method finds out several effective echo photons, detects one by one, it is thus achieved that each effective echo photon O-C value.

3. echo photon signal real-time processing method as claimed in claim 1, it is characterised in that described Preliminary treatment, is to utilize the O-C value of described effective echo photon to fit to initial polynomial function g, and Obtain current time predictive value, compare whether determine current time measured value by predictive value with measured value For virtual value.

4. echo photon signal real-time processing method as claimed in claim 1, it is characterised in that described Robust filter processes, and is the virtual value determined with initial processing step and the acquisition of energy accumulation detecting step Each effective echo photon O-C value based on fit polynomial function g, it is thus achieved that estimate residual error sequence Row, recycling Robust filter reappraises initial polynomial function g ' and carries out secondary detection, determines current Whether moment measured value is virtual value and exports virtual value.

5. echo photon signal real-time processing method as claimed in claim 1, it is characterised in that described During iterative processing, after determining that current time measured value is virtual value, increase current in ordered sequence The ordered sequence value in moment, and remove the oldest ordered sequence value, use alternative manner to repeat preliminary place Reason step and Robust filter process step, complete whole processing procedure.

6. echo photon signal real-time processing method as claimed in claim 1, it is characterised in that will be through The data crossing iterative processing carry out data output.

7. the echo-signal real-time processing method as described in any one of claim 1 to 6, its feature exists Number in described effective echo photon is [2/ signal to noise ratio].

8. the echo photon signal real-time processing method as described in any one of claim 1 to 6, it is special Levy and be that the number of described effective echo photon can be chosen according to the signal to noise ratio self adaptation of target.