CN105158817B - Deep space probe Doppler frequency passive measurement method - Google Patents
Deep space probe Doppler frequency passive measurement method Download PDFInfo
- Publication number
- CN105158817B CN105158817B CN201510471208.6A CN201510471208A CN105158817B CN 105158817 B CN105158817 B CN 105158817B CN 201510471208 A CN201510471208 A CN 201510471208A CN 105158817 B CN105158817 B CN 105158817B
- Authority
- CN
- China
- Prior art keywords
- signal
- fitting
- polynomial
- filtering
- doppler
- 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.)
- Active
Links
Abstract
The present invention provides a kind of deep space probe Doppler frequency passive measurement method, including：1st, primary signal is filtered, fitting of a polynomial, Doppler effect correction and down-sampled processing, to obtain the first polynomial fitting and reconstruction signal；2nd, filtering is tracked to reconstruction signal using the first phaselocked loop, to obtain its phase and filtering signal；3rd, fitting of a polynomial is carried out to the phase after reconstruct signal filtering, to obtain the second polynomial fitting；4th, the second model signals are constructed according to the second polynomial fitting；5th, by the filtering signal conjugate multiplication of the second model signals and reconstruction signal, the second thermal compensation signal is obtained；6th, filtering is tracked to the second thermal compensation signal using the second phaselocked loop, obtains its frequency；7th, according to the frequency acquisition Doppler frequency of the first polynomial fitting, the second polynomial fitting and the second thermal compensation signal.Present invention design simple and stable, good wave filtering effect, measurement accuracy is high.
Description
Technical field
The present invention relates to doppler frequency measurement field, more particularly to deep space probe Doppler frequency passive measurement side
Method.
Background technology
High-precision Doppler is for deep space probe micrometric measurement rail and radio science using significant.It is more general
Strangling measurement has unidirectionally, the Three models such as two-way and three-dimensional.Their main distinction is transmitting terminal and receiving terminal frequency reference mark
Accurate difference.The frequency source that the frequency standard of One Way Doppler is carried by detector is provided；Two-way Doppler receives and sent equal
In the same survey station in ground, same atomic frequency source is used；Three-dimensional Doppler is using two ground survey stations, and one of them is to detection
Device sends upward signal, is received after being forwarded through detector by another survey station, and two stations use respective frequency reference standard.
Existing Doppler measurement method mainly has following several：
(1) several weeks counting method, it is adaptable to which the two-way and three-dimensional of deep space observation station (such as Jiamusi, Keshen, Sanya) is surveyed
Amount.Deep space observation station has the ability of transmitting upward signal, on the premise of the details of known connecting station transmission signal, energy
By the method for several weeks number, the lane number change for counting beginning and end is obtained, so as to obtain two-way or three-dimensional distance, then is asked
Its time derivative, you can obtain two-way/three-dimensional doppler data.This method is in document 1【Three-dimensional e measurement technology is in survey of deep space
Application study Huangs it is of heap of stone, Wang Hong, Fan Min Spacecraft TT＆C journals, the phase of volume 31 the 3rd in 2012:6-10 pages】In it is on the books.
The measuring method is disadvantageous in that：It should be understood that the details of connecting station, higher to equipment requirement, it is only suitable
For Deep Space Station.
(2) instantaneous Doppler measuring method.
Document 2【Deep space probe passive type high accuracy Doppler measurement method is the people, Ma Maoli, Wang Wen with application Zheng
It is refined.Aerospace journal, o. 11th of volume 34 in 2013:1462-1467】Transient measurement method is described, is comprised the following steps that：
(1) FFT spectrum analysis is carried out to the signal of reception, obtains the Doppler frequency of rough estimate；
(2) using the frequency construction reference signal of rough estimate, then the signal conjugate multiplication with reality, reference signal and reality are obtained
The phase difference of border signal；
(3) fitting of a polynomial is carried out to phase difference, obtains phase difference and change with time relational expression；
(4) derivation is carried out to phase place change relational expression, obtains difference on the frequency and change with time relational expression, by interpolation, obtain
Obtain residual frequency of the actual signal relative to any time of reference signal；
(5) reference frequency and residual frequency are combined, actual measurement Doppler frequency is obtained.
The measuring method is disadvantageous in that：Because the change of measured signal frequency is very fast, the reference signal of construction is only fitted
For the shorter time (depending on the movement velocity of detector, generally 1s), also pole is restricted for time of integration Δ T selection.
Δ T is smaller, and the measurement noise of phase is bigger, but Δ T is bigger, easily produces the fuzziness that can not be eliminated.Δ T typically takes 1~
10ms, equivalent to 1000~100Hz of filtering bandwidth.The time span of the reference signal of construction and time of integration Δ T selection
Experience is typically from, this causes system unstable.
(3) phaselocked loop measuring method.
Document 3【Meng Lingpeng, Zheng is that people extracts the high-precision Doppler frequency of deep space probe using software phlase locking loop technique
Shanghai Observatory, CAS's annual, 2012 the 0th phase P83-91】Phaselocked loop (PLL) measuring method is described, step is such as
Under：
(1) by FFT, the Doppler frequency of rough estimate is obtained；
(2) frequency by the use of rough estimate, using phase discriminator, measurement is tracked to signal as the centre frequency of phaselocked loop,
Obtain the residual frequency and phase relative to centre frequency；
(3) residual frequency and centre frequency are combined, actual measurement Doppler frequency is obtained.
The measuring method is disadvantageous in that：PLL tracking accuracy and loop bandwidth (Loop bandwidth, abbreviation
BL) relevant, BL is smaller, and tracking accuracy is higher, but the time that locking needs is longer, and system is more unstable.Due to measured signal frequency
The quick change of rate, the BL that this method is used needs to be maintained at hundred more than Hz, and the motion of detector line of vision is faster, and BL values are bigger,
So as to limit measurement accuracy.
(4) document 5【The software open loop Doppler measurement technique detected for YH_1 radio sciences --- preliminary research and development
With application Zhang Su armies, river, still Kun are read in letter, etc..Mapping circular:1-14】Describe a kind of open loop Doppler measurement method, step
For：
(1) FFT is carried out to the signal that survey station is received, obtains preliminary Doppler frequency；
(2) Doppler signal received to survey station carries out bandpass filtering in time domain；
(3) suitable reference frequency is set, down coversion is carried out to the signal after bandpass filtering；
(4) LPF is carried out to the signal after down coversion；
(5) Doppler count is carried out to the signal after LPF, obtains phase information.
(phase average in 6 pairs of a period of times, that is, obtain Doppler frequency；Or, phase is intended using multinomial
Close, the phase to fitting is averaged, and also obtains Doppler frequency.
The measuring method is disadvantageous in that：The signal that detector is received includes main carrier, ranging modulated signal etc..For
Main carrier separated with ranging modulated signal, the wave filter that this method is used must high-order.50 ranks as described in document,
100 ranks, 1000 ranks, exponent number are higher, and amount of calculation is bigger；When Doppler's change is very fast, such as 1s change 50Hz, 5s change 250Hz,
It is required that the bandwidth of wave filter must not be less than 250Hz, filter effect is limited；The time of integration is longer, how general due to dopplerbroadening
The frequency spectrum for strangling signal is wider, is not easy to measurement, i.e., can not export the frequency measurement of any time of integration；And have passed through band logical,
Low pass is filtered repeatedly, is designed relatively complicated.
The content of the invention
For above-mentioned the deficiencies in the prior art, the present invention provides a kind of deep space probe Doppler frequency passive measurement side
Method, with effectively, accurately measure Doppler frequency.
To achieve these goals, the present invention uses following technical scheme：
A kind of deep space probe Doppler frequency passive measurement method, comprises the following steps：
Step 1, the primary signal that survey station is received is filtered, fitting of a polynomial, Doppler effect correction and down-sampled processing,
To obtain the first polynomial fitting and a reconstruction signal；
Step 2, filtering is tracked to the reconstruction signal using the first phaselocked loop, with obtain its filtered phase and
Filtering signal；
Step 3, the filtered phase to the reconstruction signal carries out fitting of a polynomial, multinomial to obtain the second fitting
Formula；
Step 4, the second model signals are constructed according to second polynomial fitting；
Step 5, by the filtering signal conjugate multiplication of second model signals and the reconstruction signal, to obtain the
Two thermal compensation signals；
Step 6, filtering is tracked to second thermal compensation signal using the second phaselocked loop, to obtain its filtered frequency
Rate；And
Step 7, according to the filter of first polynomial fitting, second polynomial fitting and second thermal compensation signal
Frequency acquisition Doppler frequency after ripple.
In one embodiment, the step 1 comprises the following steps：
Step A11, using instantaneous Doppler measuring method, preliminary Doppler frequency is sought according to the primary signal；
Step A12, carries out fitting of a polynomial, to obtain first polynomial fitting to the preliminary Doppler frequency；
Step A13, one first model signals are constructed according to first polynomial fitting；
First model signals and the primary signal are carried out conjugate multiplication by step A14, to obtain the first compensation letter
Number；And
Step A15, subsection integral is carried out to first thermal compensation signal, to obtain its segment phase, and according to described the
The segment phase of one thermal compensation signal sets up the reconstruction signal.
In another embodiment, the step 1 comprises the following steps：
Step B11, is tracked using the 3rd phaselocked loop to the primary signal, to obtain its phase and filtering signal；
Step B12, the phase to the primary signal carries out fitting of a polynomial, to obtain first polynomial fitting；
Step B13, the first model signals are constructed using first polynomial fitting；
First model signals and the filtering signal of the primary signal are carried out conjugate multiplication, to obtain by step B14
First thermal compensation signal；And
Step B15, carries out down-sampled processing to first thermal compensation signal, obtains the reconstruction signal.
In summary, the present invention treats side primary signal using practical frequency and carries out Doppler effect correction.Signal after compensation
Change is slower, therefore can tentatively be filtered by way of integration, at the same it is down-sampled, greatly reduce data processing amount.Make
When being tracked with PLL, slowly varying signal is beneficial to reduction PLL loop bandwidths；By way of PLL iteration, in compensation portion big absolutely
On the basis of point Doppler, then compensate residual doppler frequencies, make PLL loop bandwidth narrower, filter effect more preferably so that real
The purpose of existing high-acruracy survey.Compared with the method for the prior art, the invention has the advantages that：
1) compared with several weeks counting method, measuring method of the invention does not need the information of any connecting station transmission signal, no
Need initial driving force model, it is adaptable to any observation station, it is adaptable to unidirectional, two-way, three-dimensional Doppler measurement.
2) present invention is carried out on the basis of instantaneous Doppler with PLL, compared to instantaneous Doppler measuring method, sheet
More preferably, measurement accuracy is higher, and can export the frequency of long integration segment for invention filter effect；Measuring method compared to PLL,
Iteration Doppler effect correction and the implementation of filtering, greatly reduce PLL loop bandwidth, improve filter effect.
Brief description of the drawings
Fig. 1 is the flow chart of one embodiment of deep space probe Doppler frequency passive measurement method of the present invention；
Fig. 2 is the flow chart of the instantaneous Doppler measuring process in Fig. 1；
Fig. 3 is the MEX X-band spectrograms of an example of the invention；
Fig. 4 is the preliminary Doppler frequency f between first observation segmental arc of an example of the invention1(t) curve map；
Fig. 5 is the primary signal P (t) and model signals R of an example of the invention1(t) comparison schematic diagram；
Fig. 6 is the Doppler compensated signal C of an example of the invention1(t) spectrogram；
Fig. 7 is the Doppler compensated signal C of an example of the invention1(t) main carrier enlarged drawing；
Fig. 8 is the Doppler compensated signal C of an example of the invention1(t) the segment phase Δ obtained by subsection integral
Φ1(t) spectrogram；
Fig. 9 is the preliminary filtered reconstruction signal L of an example of the invention1(t) spectrogram；
Figure 10 is reconstruction signal L1(t) filtering signal L2(t) spectrogram；
Figure 11 is the model signals R of an example of the invention2(t) spectrogram；
Figure 12 is the filtering signals of the PLL twice L of an example of the invention2And L (t)3(t) spectral contrast figure；
Figure 13 is the flow chart of another embodiment of deep space probe Doppler frequency passive measurement method of the present invention.
Embodiment
Below in conjunction with the accompanying drawings, embodiments of the invention are provided, and are described in detail.
The present invention, i.e. deep space probe Doppler frequency passive measurement method, are received by a survey station receiver first
Primary signal P (t) to be measured, the primary signal P (t) is expressed as：
P (t)=A (t) exp [j Φ (t)]+noise (1),
In formula (1), t represents time series, and A (t) represents the amplitude of primary signal, and Φ (t) represents primary signal
Phase, noise represents the noise of primary signal.
One embodiment of the invention carries out the flow of data processing to P (t) as shown in figure 1, being summarized as follows：First with existing
Some instantaneous Doppler measuring methods obtain preliminary Doppler frequency f1(t), wherein frequency time is at intervals of 1s.Seen according to each
Segmental arc is surveyed, to preliminary Doppler frequency f1(t) fitting of a polynomial is carried out to obtain the first polynomial fitting.Utilize first fitting
The first model signals of fitting of a polynomial coefficients to construct R1(t), wherein, R1(t) Doppler frequency shift of the overwhelming majority is contained, then
Utilize model signals R1(t) Doppler effect correction is carried out to primary signal P (t), you can remove exhausted big in primary signal P (t)
Part Doppler frequency, so as to obtain the first thermal compensation signal C1(t).To C1(t) subsection integral is carried out, to obtain its phase, and profit
A signal, i.e. reconstruction signal L are reconfigured with the phase1(t).Now, because the Doppler frequency changed greatly is removed, L1(t)
Comprising residual doppler frequencies it is slowly varying with the time.Subsection integral is filtering, thus reconstruction signal L1(t) for
It has passed through the signal of down-sampled and preliminary filtering process.Then, by the first phaselocked loop (PLL) to reconstruction signal L1(t) carry out
Tracking filter, to obtain filtered filtering signal L2And phase delta Φ (t)2(t)。ΔΦ2(t) comprising residual doppler frequency
The change of rate, to ΔΦ2(t) it is fitted to obtain the second polynomial fitting, and the second model is constructed using its fitting coefficient
Signal R2(t), then by R2(t) the filtering signal L exported with the first PLL2(t) conjugate multiplication is carried out, to remove L2(t) it is residual in
Remaining Doppler frequency, obtains the second thermal compensation signal C2(t).Then, to the second thermal compensation signal C2(t) the second phaselocked loop tracking is carried out,
Obtain its filtered filtering signal L3(t), phase delta Φ3And frequency Δ f (t)3(t).Finally, with reference to foregoing frequency and phase
Position polynomial fitting, it is final to obtain high-precision doppler frequency f3(t)。
Substep will be carried out to above-mentioned flow below to be discussed in detail：
Step 1, primary signal P (t) is filtered, fitting of a polynomial, Doppler effect correction and down-sampled processing, to obtain
First polynomial fitting and a reconstruction signal, reconstruction signal is in this as preliminary filtering signal.
In the embodiment in figure 1, step 1 is measured by the instantaneous Doppler performed successively, frequency is fitted, tectonic model is believed
Number, five steps of Doppler effect correction and signal reconstruction realize, wherein,
Instantaneous Doppler measuring process is prior art, and its process as described in Figure 2, comprises the following steps (a)-(d)：
(a) Fourier transform is carried out to the primary signal P (t) in a period of time (time span is generally 1s) first, with
Estimate primary signal P (t) original frequency f0, and utilize original frequency f0Construct an initial reference signal R0(t)：
R0(t)=exp (j2 π f0t) (2)。
(b) by initial reference signal R0(t) with primary signal P (t) conjugate multiplications, initial compensation signal C is obtained0(t)：
C0(t)=A (t) exp [j (π of Φ (t) -2 f0t)]+noise (3)。
(c) to the signal C after multiplication0(t) subsection integral, and obtain the segment phase ΔΦ after its subsection integral0(t)。
(d) to segment phase ΔΦ0(t) fitting of a polynomial is carried out, generally 2, is expressed asIts
In, n0Typically take 2, aiRepresent ΔΦ0(t) fitted polynomial coefficients, i=0,1 ... n0If intended using 2 order polynomials
Close, then n0=2；noise1Represent phase delta Φ0(t) noise.
Then the phase parameter Φ (t) in primary signal P (t) is：
(e) derivation is carried out to formula (4), the preliminary Doppler frequency f of any time in this period will be obtained1(t)：
Data to random time section carry out step (a)~(e) processing, so as to obtain first in whole observation period
Walk Doppler frequency f1(t)。
Obtain preliminary Doppler frequency f1(t) after, according to observation segmental arc to preliminary Doppler frequency f1(t) multinomial is carried out
Fitting, obtains the first polynomial fitting (6)：
In formula (6), bkRepresent preliminary Doppler frequency f1(t) coefficient of polynomial fitting, k=0,1 ... n1；Typically
It is fitted using 5 order polynomials, i.e. n1=5, naturally it is also possible to use other number of times.noise2Represent f1(t) it is multinomial relative to being fitted
The noise of formula.
Obtain after the first polynomial fitting, utilize its coefficient of polynomial fitting bkConstruct the first model signals R1(t)：
Then, by the first model signals R1(t) with primary signal P (t) conjugate multiplications, deposited with removing in primary signal P (t)
Most Doppler frequency shifts, so as to obtain the first thermal compensation signal C1(t)：
After Doppler effect correction step, primary signal P (t) most of Doppler frequency shift has been removed, C1(t) with
Time is slowly varying.Thus can be to C1(t) subsection integral is carried out, a series of phase delta Φ are obtained1(t) signal weight, is then carried out
Structure is to set up a reconstruction signal L1(t)：
L1(t)=exp (j ΔΦs1(t)) (9),
When the subsection integral time is 0.25ms, equivalent to down-sampled 2000 times, filtering bandwidth is 4000Hz.
Step 2, using the first phaselocked loop (PLL) to reconstruction signal L1(t) filtering is tracked, to obtain its phase and filter
Ripple signal.In the Doppler effect correction of step 1, primary signal P (t) most of Doppler frequency shift is compensated, L1(t) still wrap
Containing residual doppler frequencies and noise to be canceled, thus the PLL of Selection utilization the first is to its tracking filter.PLL tracking accuracy
Relevant with loop bandwidth (Loop bandwidth, abbreviation BL), BL is smaller, and tracking accuracy is higher, but the time that locking needs gets over
Long, system is more unstable.Relative to primary signal P (t), the L become slowly1(t) it is beneficial to the first PLL of reduction loop bandwidth.Through PLL
Tracking filter, will export L1(t) filtered signal L2(t), phase delta Φ2(t) with frequency Δ f2(t).Compared to ΔΦ1(t),
ΔΦ2(t) random noise is smaller.
According to the first polynomial fitting (6) and Δ f2(t) the Doppler frequency f of the first PLL measurements, is obtained2(t)：
Step 3, to reconstruction signal L1(t) filtered phase delta Φ2(t) fitting of a polynomial is carried out, to obtain the second fitting
Multinomial (11)：
In formula (11), clRepresent ΔΦ2(t) coefficient of polynomial fitting；L=0,1 ... n2；When multinomial using 5 times
When formula is fitted, n2=5, noise3Represent signal delta Φ2(t) noise.
ΔΦ2(t) with Δ f2(t) change of uncompensated residual doppler frequencies is contained.
Step 4, the fitting coefficient c in the second polynomial fitting (11)lConstruct the second model signals R2(t)：
Step 5, by the second model signals R2(t) and through the filtered filtering signal L of the first PLL2(t) conjugate multiplication, to move
Residual doppler frequencies are walked, the second thermal compensation signal C is obtained2(t)：
Step 6, using the second phaselocked loop to the second thermal compensation signal C2(t) it is tracked, obtains its filtered signal L3
(t), phase delta Φ3(t) with frequency Δ f3(t).Wherein, after the residual doppler compensation by step 5, the 2nd PLL loop band
Width can set smaller, and the precision of tracking is higher.ΔΦ3(t) with Δ f3(t) any Doppler will not included, be Gauss white noise
Sound.
Step 7, believed according to first frequency polynomial fitting (6) and second phase polynomial fitting (11) and the second compensation
Number C2(t) filtered frequency Δ f3(t), you can obtain the high-precision Doppler frequency f that the present invention needs to measure3(t)：
In formula (14), first and second is polynomial fitting, and not comprising any random noise, Section 3 is random noise.
Section 3 is integrated, frequency measurement accuracy is also can further improve.
The realization principle of the present invention is such：Measured by instantaneous Doppler, obtain preliminary Doppler frequency f1(t),
Wherein, f1(t) it is Doppler frequency f3(t) a maximum part in；Utilize f1(t) tectonic model signal R1(t), and the mould is made
Type signal R1(t) conjugate multiplication is carried out with measured signal P (t).Because the quick change of measured signal frequency is embodied in noiseless
Model signals in, thus be multiplied after signal C1(t) change slow；Primary signal is directly tracked compared to using phaselocked loop
Method, changes the loop noise bandwidth that slow signal has advantageously reduced phaselocked loop, improves filter effect.Then, utilize
The phase of phaselocked loop output constructs alternate model signal R again2(t), compensated for residual doppler, signal C after compensation2(t)
Frequency it is basic constant with the time, beneficial to the 2nd PLL of reduction loop bandwidth.Finally, accurate Doppler is passed through in the 2nd PLL outputs
The when not varying signal of compensation and filtering, beneficial to further improving signal to noise ratio by way of integration.
Example：
The above method is verified with reference to Mars Express (being abbreviated as MEX) observation example.MEX observation times are
2014.11.28 day, year day of year be 332 days.Observation station is stood for Shanghai 25m She Shan, and data sampling rate is 8Mbps/s, with a width of
4MHz.Table 1 is observation segmental arc description：
The MEX of table 1 observation segmental arc
| Observe segmental arc | Time started~end time |
| First segmental arc (scan 1) | 06：00~06：19 |
| Second segmental arc (scan 2) | 06：20~06：39 |
| 3rd segmental arc (scan 3) | 06：40~06：59 |
| 4th segmental arc (scan 4) | 07：00~07：19 |
| 5th segmental arc (scan 5) | 07：20~07：39 |
| 6th segmental arc (scan 6) | 07：40~07：59 |
Effective observation time of first segmental arc is 06：00：03-06：19：00 (She Shan stations were not observed in first three second), about 19 points
Clock.Illustrated by taking the first segmental arc as an example.
Fig. 3 is 06：00：The spectrogram of 03 second MEX X-band signal, amplitude normalization to 0dB.Main carrier both sides have a lot
Sidetone.When measuring main carrier frequency, to improve measurement accuracy, it is necessary to by sidetone and noise filtering.Explain below in conjunction with the accompanying drawings
Step 1-7 of the present invention concrete operations：
Step 1, first with instantaneous Doppler measuring method, preliminary Doppler frequency is obtained.Fig. 4 is first observation arc
The preliminary Doppler frequency of section.
Secondly, according to the time span of observation segmental arc, 5 order polynomial fittings are carried out to frequency, corresponding frequency is obtained and intends
Close multinomial, fitting coefficient bk(k=0,1,2,3,4,5) is shown in Table 2.
The frequency fitting coefficient of 2 first observation segmental arcs of table
| b0 | 2168665.92036006 |
| b1 | -1.2195252292932570182 |
| b2 | 0.00023239753626135261624 |
| b3 | 2.8452724033926750562e-09 |
| b4 | 1.4013049666167922156e-12 |
| b5 | 2.078971663056222067e-18 |
Said frequencies polynomial fitting is integrated, fit phase is obtained.
Then the phase formation model signals R of fitting is utilized1(t) (such as Fig. 5), measured signal P (t) and model signals R1(t)
Frequency and frequency bandwidth Hz magnitudes overlap.
Again by model signals R1(t) grip and be multiplied altogether with primary signal P (t), remove the Doppler frequency shift of the overwhelming majority, obtain
C1(t), its frequency spectrum is as shown in Figure 6.
In figure 3, main carrier bandwidth is between 2.168610~2.168654MHz, bandwidth about 44Hz.Mended by Doppler
After repaying, its main carrier bandwidth about 2Hz (as shown in Figure 7), it was demonstrated that the validity of Doppler effect correction.
Finally, by the signal C after compensation1(t) integrate, the time of integration is 0.25ms, obtain the phase delta of data after integration
Φ1(t) (such as Fig. 8).
Utilize ΔΦ1(t) signal is reconfigured.Because the phase integral time is 0.25ms, equivalent to down-sampled 2000 times.
After down-sampled, signal sampling rate is 4Kbps/s, bandwidth 2000Hz, as shown in Figure 9.
Step 2, PLL is tracked：Due to L1(t) residual doppler and noise are still included, thus is carried out using the first PLL
Filter again.The PLL loop bandwidths about 4Hz, output signal L2(t) (Figure 10), phase delta Φ2And frequency Δ f (t)2(t).Compare
In L1(t), L2(t) noise reduction about 20dB.
Step 3, phase-fitting：The phase delta Φ of first PLL outputs2(t) residual doppler is still included, 6 are carried out to it
Order polynomial is fitted.Fitting coefficient cl(l=0,1,2,3,4,5,6) is shown in Table 3.
The fitting coefficient c of table 3l
| c0 | -1.70515328150197143 |
| c1 | 0.00596957353780802613 |
| c2 | 1.3703760765913718e-05 |
| c3 | -1.01527152676408948e-07 |
| c4 | 1.95793164865488272e-10 |
| c5 | -1.69255489354493619e-13 |
| c6 | 5.61365859356498717e-17 |
Step 4, construction reference signal R2(t)：Utilize ΔΦ2(t) fitting coefficient clTectonic model signal R2(t) (figure
11)。
Step 5, Doppler effect correction：Utilize R2(t) to L2(t) Doppler effect correction is carried out, C is obtained2(t)。
Step 6, PLL is tracked：Using the 2nd PLL to C2(t) it is tracked, PLL loop bandwidths are 0.04Hz (the first PLL
Loop bandwidth be 4Hz).Figure 12 is the filtered filtering signal L of PLL twice2And L (t)3(t) spectrogram, spectral resolution is equal
For 0.5Hz.Compared to L2(t), L3(t) noise is smaller, and main carrier energy is more concentrated (in 2mHz).
Step 7, the signal exported according to the 2nd PLL obtains last Doppler frequency f3(t)。
The frequency f measured using identical multinomial instantaneous Doppler1(t) (formula 6), the f of the first PLL measurements2(t)
The Doppler frequency f that (formula 10), the 2nd PLL measurement are obtained3(t) (formula 14) progress residual analysis, respectively 11mHz,
7.78mHz、3.73mHz.It can be seen that, after the 2nd PLL filtering, doppler accuracy is significantly improved.
Figure 13 shows another embodiment of step 1 of the present invention, in this embodiment, and step 1 is real as follows
It is existing：
First, side primary signal is directly treated using PLL to be tracked, due to treating that side signal P (t) changes are very fast, the PLL
Loop noise bandwidth it is larger, generally hundreds of Hz.PLL output primary signal P (t) phase Φ '0And filtered filter (t)
Ripple signal L '0(t)。
Secondly, to Φ '0(t) fitting of a polynomial is carried out, the first polynomial fitting (15) is obtained：
In formula (14), b'kFor coefficient of polynomial fitting, noise' is phase-fitting noise.Then, more than first are utilized
Item formula (15) constructs the first model signals R '1(t)：
Recycle R '1(t) to filtered signal L'0(t) Doppler effect correction is carried out, the first thermal compensation signal C ' is obtained1(t)：
Finally, due to which the signal after Doppler effect correction is slowly varying with the time, frequency very little is down-sampled to its, you can
To reconstruction signal L'1(t)。
Reconstruction signal L' is obtained according to above-mentioned steps1(t) after, continue using above-mentioned steps 2-7 to reconstruction signal L'1(t) enter
Row processing.
In previous embodiment, the first polynomial fitting is frequency fitting (formula 6), and in the present embodiment, first intends
Multinomial (formula 15) is closed for phase-fitting, therefore f3(t) calculation formula is different.
More than, only embodiments of the invention are not limited to the scope of the present invention, the above embodiment of the present invention
It can also make a variety of changes.I.e. every claims and description according to the present patent application made it is simple, etc.
Effect change and modification, fall within the claims of patent of the present invention.
Claims (3)
1. a kind of deep space probe Doppler frequency passive measurement method, including：
Step 1, the primary signal that survey station is received is filtered, fitting of a polynomial, Doppler effect correction and down-sampled processing, to obtain
Obtain the first polynomial fitting and a reconstruction signal；Characterized in that, this method is further comprising the steps of：
Step 2, filtering is tracked to the reconstruction signal using the first phaselocked loop, to obtain its filtered phase and filtering
Signal；
Step 3, the filtered phase to the reconstruction signal carries out fitting of a polynomial, to obtain the second polynomial fitting；
Step 4, the second model signals are constructed according to second polynomial fitting；
Step 5, by the filtering signal conjugate multiplication of second model signals and the reconstruction signal, to obtain the second benefit
Repay signal；
Step 6, filtering is tracked to second thermal compensation signal using the second phaselocked loop, to obtain its filtered frequency；
And
Step 7, after according to the filtering of first polynomial fitting, second polynomial fitting and second thermal compensation signal
Frequency acquisition Doppler frequency.
2. deep space probe Doppler frequency passive measurement method according to claim 1, it is characterised in that the step
Rapid 1 comprises the following steps：
Step A11, using instantaneous Doppler measuring method, preliminary Doppler frequency is sought according to the primary signal；
Step A12, carries out fitting of a polynomial, to obtain first polynomial fitting to the preliminary Doppler frequency；
Step A13, one first model signals are constructed according to first polynomial fitting；
First model signals and the primary signal are carried out conjugate multiplication, to obtain the first thermal compensation signal by step A14；
And
Step A15, subsection integral is carried out to first thermal compensation signal, to obtain its segment phase, and is mended according to described first
The segment phase for repaying signal sets up the reconstruction signal.
3. deep space probe Doppler frequency passive measurement method according to claim 1, it is characterised in that the step
Rapid 1 comprises the following steps：
Step B11, is tracked using the 3rd phaselocked loop to the primary signal, to obtain its phase and filtering signal；
Step B12, the phase to the primary signal carries out fitting of a polynomial, to obtain first polynomial fitting；
Step B13, the first model signals are constructed using first polynomial fitting；
First model signals and the filtering signal of the primary signal are carried out conjugate multiplication, to obtain first by step B14
Thermal compensation signal；And
Step B15, carries out down-sampled processing to first thermal compensation signal, obtains the reconstruction signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510471208.6A CN105158817B (en) | 2015-08-04 | 2015-08-04 | Deep space probe Doppler frequency passive measurement method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510471208.6A CN105158817B (en) | 2015-08-04 | 2015-08-04 | Deep space probe Doppler frequency passive measurement method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105158817A CN105158817A (en) | 2015-12-16 |
| CN105158817B true CN105158817B (en) | 2017-08-25 |
Family
ID=54799732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510471208.6A Active CN105158817B (en) | 2015-08-04 | 2015-08-04 | Deep space probe Doppler frequency passive measurement method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105158817B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109682383B (en) * | 2018-11-23 | 2022-11-04 | 中国西安卫星测控中心 | Real-time filtering positioning method for measuring distance and data by using deep space three-way |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5987059A (en) * | 1996-07-12 | 1999-11-16 | General Electric Company | Method for Doppler-replica harmonic avoidance |
| CN101132191A (en) * | 2007-10-15 | 2008-02-27 | 北京航空航天大学 | Baseband signal processing method for GNSS receiver |
| CN102299713A (en) * | 2011-05-17 | 2011-12-28 | 中国电子科技集团公司第十研究所 | Method for tracking signal with extremely low signal to noise ratio and large Doppler frequency change rate |
-
2015
- 2015-08-04 CN CN201510471208.6A patent/CN105158817B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5987059A (en) * | 1996-07-12 | 1999-11-16 | General Electric Company | Method for Doppler-replica harmonic avoidance |
| CN101132191A (en) * | 2007-10-15 | 2008-02-27 | 北京航空航天大学 | Baseband signal processing method for GNSS receiver |
| CN102299713A (en) * | 2011-05-17 | 2011-12-28 | 中国电子科技集团公司第十研究所 | Method for tracking signal with extremely low signal to noise ratio and large Doppler frequency change rate |
Non-Patent Citations (2)
| Title |
|---|
| 深空探测器被动式高精度多普勒测量方法与应用;郑为民 等;《宇航学报》;20131130;第34卷(第11期);第1462-1467页 * |
| 深空探测用数字开环多普勒技术初步研制及其在嫦娥一号探测任务中的应用;简念川 等;《中国科学 G辑：物理学 力学 天文学》;20091031;第39卷(第10期);第1400-1409页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105158817A (en) | 2015-12-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104091064B (en) | PS-DInSAR ground surface deformation measurement parameter estimation method based on optimal solution space search method | |
| CN104898172B (en) | Cross-correlation-based nuclear magnetic resonance full wave signal noise filtering method | |
| CN101556340B (en) | Three-dimensional small surface element magnetotelluric continuous array data acquisition method | |
| CN103869162B (en) | Dynamic signal phasor measurement method based on time domain quasi-synchronization | |
| CN107783200A (en) | Joint EMD and TFPF algorithms a kind of all-wave magnetic resonance signal random noise method for reducing | |
| CN104898176B (en) | A kind of rotating accelerometer gravity gradiometer gravity gradient demodulation method | |
| CN109507704B (en) | Double-satellite positioning frequency difference estimation method based on mutual ambiguity function | |
| CN104199058A (en) | Time scale adjusting algorithm based on Kalman filter real-time prediction value | |
| CN103901442A (en) | Detecting and relieving method for single-frequency multi-path effect | |
| CN104316160B (en) | Underwater signal instantaneous frequency demodulation method based on Wavelet Ridge | |
| CN102269803B (en) | Method for correcting low-frequency components in discrete spectrum based on time delay | |
| CN103018555A (en) | High-precision electric power parameter software synchronous sampling method | |
| CN102928713B (en) | A kind of background noise measuring method of magnetic field antenna | |
| CN105158817B (en) | Deep space probe Doppler frequency passive measurement method | |
| Yuan et al. | An improved noise reduction algorithm based on wavelet transformation for MEMS gyroscope | |
| Yu et al. | Adaptive multiple second-order synchrosqueezing wavelet transform and its application in wind turbine gearbox fault diagnosis | |
| CN103777237B (en) | A kind of earth's surface elevation smoothing method based on space-variant weighting fringing wavenumber domain filtering | |
| Shen et al. | Correlation theory-based signal processing method for CMF signals | |
| CN102565541B (en) | Recursion demodulation method for electrical tomography system | |
| CN102655419B (en) | Calculation method of decision function for captured spread spectrum signals | |
| CN101556173A (en) | Coriolis mass flowmeter digital resolver based on wavelet de-noising and method thereof | |
| Lyubushin et al. | Properties of GPS noise at Japan islands before and after Tohoku mega-earthquake | |
| CN102589551A (en) | Real-time filtering method based on wavelet transformation for ship optical-fiber gyro signals | |
| CN102072987B (en) | Phase estimation method for short-interval sinusoidal signal and experimental device thereof | |
| CN104677486A (en) | Aero-engine vibration signal phase measurement method based on revolving speed pulse reconstruction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |