CN103412285B - The preprocess method of lunar surface rover moon sight radar data - Google Patents

Abstract

The invention discloses a kind of preprocess method of lunar surface rover moon sight radar data, the method comprising the steps of: step 1, the raw data that real-time reception moon sight radar passes down; Step 2, performs frame synchronization to the raw data that passes down, goes to disturb and RS decoding process obtains Framed Data; Step 3, carries out shunt solution frame to described Framed Data and subpackage process is wrapped to obtain moon sight radar data source; Step 4, unpacks process to generate moon sight radar data block to described moon sight radar data source bag execution; Step 5, converts the moon sight radar data with actual physics implication to the moon sight radar data block of described generation; Step 6, eliminates the DC component in every track data; Step 7, carries out bandpass filtering to every track data, removes high and low frequency composition.Utilize method of the present invention can process exactly the data of lunar surface rover moon sight radar, generate the moon remotely-sensed data with actual physics meaning that can issue.

Description

The preprocess method of lunar surface rover moon sight radar data

Technical field

The present invention relates to data processing field, being specifically related to a kind of preprocess method of data of lunar surface rover moon sight radar, for realizing the rough handling to lunar surface rover moon sight radar data.

Background technology

Moon exploration program is the systematization engineering of environment for detecting the moon, resource.Moon exploration program system mainly comprises the parts such as satellite, rocket, launching site, observing and controlling, ground application system, and wherein ground application system is responsible for reception, the Storage and Processing of detector data.

Usual lunar orbiter comprises lander and lunar surface rover, is equipped with multiple science load, for detecting various environment, the resource data of moonscape.As one of Main Load, wherein moon sight radar is a kind of time domain detection radar worked in without carrier frequency millimicrosecond pulse state, adopts bistatic antenna.Its principle of work is: moon sight radar transmitter produce ultra broadband without carrier frequency millimicrosecond pulse, through emitting antenna radiation under lunar surface/coupling super-broadband electromagnetic impulse signal, signal is in the communication process of lunar soil and lunar crust rock medium, if run into the targets such as uneven layer, different medium interface, lava tube, erratic boulder, by the reflection of the signal that generates electromagnetic waves and scattering.After moon sight radar receiving antenna receives this reflection and scattered signal, corresponding detection data is obtained after receiver amplification, sampling, by analyzing detection data, processing and imaging, obtain the information such as lunar soil thickness in rover travel region and distribution thereof, the distribution of erratic boulder and lava tube and lunar crust time top layer rock geologic structure.

The science load that lunar surface rover is installed is the instrument with scientific exploration ability.The pre-service this instrument being detected to the science data obtained is one of task of ground application system, is also that data publication is to the precondition of numerous scientific research personnel in order to research.The pre-service of radar data mainly processes the former bag of scientific exploration instrument downlink data, resolve science data and project data, correction data, eliminate the various deviations existed in detection data and project data, finally become and can directly use, the science data product of reflection lunar surface remote sensing characteristic.

After ground application system receives rover data, in order to reduce the science data of science load (as moon sight radar) truly, remove instrument self interference, reflection lunar surface truth, will carry out 0 grade, 1 grade, 2 DBMS process to data usually.Wherein 0 grade of disposal route is source bag process, is extracted by the detection data of each load, forms file separately, and each load source bag data spliced according to the sampling period.1 DBMS is physical quantity conversion, and each load true form data are converted to the numerical value with actual physics meaning.2 DBMS process then carry out calibration process, geometry location process to data, generate the moon remotely-sensed data with locus had nothing to do with load.

But the disposal route of existing ultra wide band pulsed radar, mainly for the treatment of the data of ground ground penetrating radar, cannot meet the data handling requirements of moon sight radar.

Summary of the invention

The object of this invention is to provide a kind of preprocess method of lunar surface rover moon sight radar data, described method can process exactly to the data of lunar surface rover moon sight radar, generates the moon remotely-sensed data with actual physics meaning that can issue.

According to an aspect of the present invention, provide a kind of preprocess method of lunar surface rover moon sight radar data, described method comprises: receive raw data, the raw data that real-time reception moon sight radar passes down; Framed Data process, performs frame synchronization to the raw data that passes down, goes to disturb and RS decoding process obtains Framed Data; Channel Processing, carries out shunt solution frame to described Framed Data and subpackage process is wrapped to obtain moon sight radar data source; Data block is extracted, and unpacks process to generate moon sight radar data block to described moon sight radar data source bag execution; Physical quantity is changed, and the moon sight radar data block of described generation is converted to the moon sight radar data with actual physics implication; DC processing, eliminates the DC component in every track data; Bandpass filtering treatment, carries out bandpass filtering to every track data, removes high and low frequency composition.

By the above-mentioned data processing method that the present invention proposes, conventional processing can be carried out to the data that lunar surface rover moon sight radar obtains, remove correlation noise, use for numerous scientific research personnel so that issue final 2B DBMS in time.

Accompanying drawing explanation

Fig. 1 shows the process flow diagram processed moon sight radar data.

Embodiment

For making the object, technical solutions and advantages of the present invention clearly understand, below in conjunction with embodiment also with reference to accompanying drawing, the present invention is described in more detail.

First the composition and working principle of lunar surface rover moon sight radar is introduced.

Moon sight radar is a kind of time domain detection radar worked in without carrier frequency millimicrosecond pulse state, adopts bistatic antenna.Its principle of work is: moon sight radar transmitter produce ultra broadband without carrier frequency millimicrosecond pulse, through emitting antenna radiation under lunar surface/coupling super-broadband electromagnetic impulse signal, signal is in the communication process of lunar soil and lunar crust rock medium, if run into the targets such as uneven layer, different medium interface, lava tube, erratic boulder, by the reflection of the signal that generates electromagnetic waves and scattering.After moon sight radar receiving antenna receives this reflection and scattered signal, corresponding detection data is obtained after receiver amplification, sampling, by analyzing detection data, processing and imaging, obtain the information such as lunar soil thickness in rover travel region and distribution thereof, the distribution of erratic boulder and lava tube and lunar crust time top layer rock geologic structure.

First passage pulse transmitter produces satisfactory pulse signal under the transmit clock 1 that radar master controller exports triggers, and then Differential Output radiate to first passage emitting antenna.After first passage receiving antenna receives buried target echoed signal, give first passage receiver, first passage receiver amplifies echoed signal according to certain rule under radar main controller controls, and the first passage then giving radar master controller carries out digitizing.

Second channel pulse transmitter produces satisfactory pulse signal under the transmit clock 2 that radar master controller exports triggers, and then Differential Output radiate to second channel emitting antenna.After second channel receiving antenna receives buried target echoed signal, give second channel pulse receiver, second channel receiver amplifies echoed signal according to certain rule under radar main controller controls, and the second channel then giving radar master controller carries out digitizing.

Usually, the real time data content that moon sight radar obtains comprises science data and project data, and the science data of moon sight radar are the science amounts with obvious physical significance, are collected by moon sight radar.In the present invention, the science data that moon sight radar collects are the reflex response of moon top layer to radar pulse signal.

Fig. 1 is the process flow diagram of lunar surface rover moon sight radar data preprocess method of the present invention.As shown in Figure 1, the method comprises the steps:

Step 1, the original bit stream data that land station's real-time reception lunar surface rover passes down.

Original bit stream data of the present invention comprise science data, science data be receive after moon sight radar transmitted pulse electromagnetic wave from the ELECTROMAGNETIC REFLECTION signal below lunar surface and top layer.Described track data refers to that the science data that radar obtains within a sampling period, multiple tracks data form a complete radar data file.

Step 2, carries out Framed Data process to the original bit stream data received, obtains multiple Framed Data, Framed Data process comprises frame synchronization, goes to disturb, RS decoding and data framing.

Wherein step 2 is further comprising the steps: step 2.1, frame synchronization: search the frame synchronization head " 1ACFFC1D " in original bit stream data by bit (bit), intercepts 1024 bytes and form multiple initial data frame from frame synchronization head; Step 2.2, goes to disturb: carry out exclusive-OR operation one by one, to go to disturb to initial data frame by frame synchronization 992 later bytes in the scrambler of 992 bytes and initial data frame; Step 2.3, RS decoding: the RS code spending last 128 bytes in the initial data frame after disturbing (1024 bytes) carries out RS decoding to initial data frame, and records the data after RS decoding; Step 2.4, data framing: the data quality information of the Frame after RS decoding and described two bytes is formed the Framed Data that size is 1024 bytes.

Step 3, Channel Processing, according to the virtual channel identification of carrying in described Framed Data, is extracted into the data of specific virtual channel in frame data, and the valid data byte of taking out wherein forms virtual channel data.

In this step, DPS system is carried out shunt according to the identifier of load to the Framed Data after process and is separated frame and subpackage, sort out the data belonging to target load, to adopt the method adapted with this load to carry out data processing, obtain the data source bag of science load thus, in the present invention, be LPR data source bag.

Wherein, separate frame along separate routes and refer to according to the pseudo channel information in data frame structure, determine the pseudo channel of Frame, the byte valid data in Frame are taken out, form virtual channel data, and form the statistical information of Frame.Subpackage and separate loading source bag, refer to input channel configuration information, decompose according to useful load application process identifier, by the detector data of channel composition binary data source APMB package, and the data quality information to each source bag interpolation one byte, form source bag data.

Step 4, separate loading source is wrapped: accord with according to the useful load application identities of arranging in advance the data source APMB package data of moon sight radar being extracted from virtual channel data composition binary format specific as follows:

Virtual channel data after separating frame according to useful load application identities symbol to shunt decomposes and extracts, and described useful load application identities symbol, for indicating the load source of data, namely indicates that these data come from moon sight radar, or other load.The moon sight radar data extracted from pseudo channel is formed the data source APMB package of binary format, form source bag data.The source bag data of moon sight radar are divided into section's cosmogony bag data and engineering source bag data according to data identifier.The data now generated are 0A DBMS.

Wherein, the form of moon sight radar science source bag data (0A DBMS) is:

The form of moon sight radar engineering source bag data is:

In step 4, the process that unpacks of section's cosmogony bag data refers to and splices section's cosmogony bag data (or optimize after source bag data), form the data division of moon sight radar (LPR) science data block: be stitched together, the data in X packet sequence control code continuous print section cosmogony bag data as the data division of a LPR science data block.Each source bag data are identified by the grouping mark " * * " in packet sequence control code: in packet sequence control code continuous print situation, grouping mark " * * " is first source bag that " 01 " is expressed as in LPR science data block in X conjoint data source bag, " 00 " is expressed as removing first in X conjoint data source bag and wraps with the source that is connected of the centre of last, and " 10 " are expressed as last source bag in X conjoint data source bag.Multiple connected science data sources bag is spliced into a complete science data block (also referred to as a track data).The data now generated are moon sight radar 0B DBMS, and table 1 shows the data layout explanation of the complete science data block (track data) of of moon sight radar.

Table 1 moon sight radar science data layout

Like this, source bag data, after unpacking process, form corresponding science data block.

Step 5, carries out physical quantity conversion to the data block that step 4 obtains.

Because the data stored in science data block are through the binary data after A/D conversion, therefore, need to carry out physical quantity conversion to the road science data in science data block.

During physical quantity conversion, first binary number is converted to metric magnitude of voltage, then according to following formula, the road science data of signless 16bit is converted into signed number according to accumulative frequency:

X _(m)=S _(m)-128×(A+1)（1）

Wherein, m be in the ordinal number of sampled point, X is the signed integer of 16bit, and S is 16bit signless integer, and A is the accumulative frequency of first passage in parameter.The data now generated are moon sight radar 1 DBMS.

Step 6, (1 DBMS refers to the raw voltage data that instrument exports to carry out the process of 2A DBMS to the data after described physical quantity conversion; The process of 2A level refers to does calibration process to instrument, eliminates instrument itself to the impact of data, makes data representation have the observed quantity of actual physics meaning; The process of 2B level refers to does geometry location process to data, adds spatial information to data), direct current, bandpass filtering treatment are gone to radar data.

The process of 2A DBMS still with " road " for unit carries out.In 2A DBMS processing procedure, data validity judgement, DC processing, bandpass filtering step are carried out to 1 DBMS product, generate 2A DBMS, and generate corresponding report.

The process of first passage 2A DBMS and second channel A road and B road 2A DBMS treatment step are that the same (A road and B road refer to two receiving antennas of second channel, one two mode detection received taked by second channel), just configuration parameter is different, and therefore they can represent by same flow process.

Process idiographic flow:

(1) read data files:

Read moon sight radar first passage 1 DBMS or second channel A road 1 DBMS or second channel B road 1 DBMS.

(2) according to the respective channel in step (1), a track data is read according to table 5-8, table 5-9 or table 5-10 data layout;

(3) data validity judgement is carried out;

Read 1 byte quality state of 1 grade of this track data, to bit0 ~ bit5 position according to inclusive-OR operation process, obtain result of calculation (0: represent normal, 1: represent abnormal), result of calculation is write the bit6 position of 2A DBMS quality state, bit7 inherits position 1 grade of value, bit0 ~ bit5 zero setting.

(4) DC processing is carried out

First arithmetic mean is got to per pass clinical practice data, then the data in this road is deducted mean value:

$Y_{\left(n\right)}=X_{\left(n\right)}-\frac{1}{N}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{X}_{\left(n\right)}$

Wherein Y (n) is for removing the data value after direct current, and X (n) is the data value before removing direct current, and n represents the ordinal number of sampled point in this road, and N represents Dao Nei effective sampling points sum (reading in road header), and padding data does not process.

(5) bandpass filtering treatment is carried out

A. calculated rate interval, formula is:

$\mathrm{\Δf}=\frac{1}{T_{\mathrm{sample}}\×N}$

Wherein T _samplerepresent sampling interval, (first passage is 2.5ns, and second channel is 0.3125ns), N represents that efficiently sampling is counted.

B. calculate the position ordinal number of key frequency point, formula is:

$N_{\mathrm{lowcut}}=\frac{f_{\mathrm{lowcut}}}{\mathrm{\Δf}}$

$N_{\mathrm{low}}=\frac{f_{\mathrm{low}}}{\mathrm{\Δf}}$

$N_{\mathrm{high}}=\frac{f_{\mathrm{high}}}{\mathrm{\Δf}}$

$N_{\mathrm{highcut}}=\frac{f_{\mathrm{highcut}}}{\mathrm{\Δf}}$

Wherein f _lowcutrepresent low-end cut-off frequency, f _highcutrepresent high end cut-off frequencies, f _lowrepresent low-pass cut-off frequencies, f _highrepresent high pass cut off frequency.Region outside cutoff frequency is called stopband; Region between flow to fhigh is called passband.(f _lowcut, f _highcut, f _low, f _highread in configuration file)

C. be Fast Fourier Transform (FFT) FFT to this track data, obtain intermediate data T, formula is:

$T_{\left(k\right)}=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{e}^{-i\frac{2\mathrm{\π}}{N}(n-1)(k-1)}{Y}_{\left(n\right)},k=\mathrm{1,2},......,N$

Wherein T _(k)represent the data value after Fourier transform, Y _(n)for removing the data value after direct current, k is the ordinal number of the data after conversion.

D. use ordinal number Nlowcut, Nlow, Nhigh, Nhighcut of obtaining in step B, by the data value zero setting in the stopband range of T, formula is:

$\left\{\begin{array}{cc}{T}_{\left(k1\right)}=0& k1=\mathrm{1,2},......,N_{\mathrm{lowcut}}-1\\ T_{\left(k2\right)}=0& k2=N_{\mathrm{highcut}}+1,......,N\end{array}\right.$

Wherein k1, k2 are data ordinal.

And transitional zone adopts Hanning window level and smooth:

$\left\{\begin{array}{cc}{T}_{\left(k3\right)}=0.5[1-\cos \left(\frac{2\mathrm{\π}{m}_1}{M_1+1}\right)]T_{\left(k3\right)}& m1=\mathrm{1,2},......,\frac{M_1}{2}\\ T_{\left(k4\right)}=0.5[1-\cos \left(\frac{2\mathrm{\π}(m_2+\frac{M_2}{2})}{M_2+1}\right)]T_{\left(k4\right)}& m2=\mathrm{1,2},......,\frac{M_2}{2}\end{array}\right.$

Wherein ordinal number k3=N _lowcut-1+m ₁, M ₁=2 × (N _low-N _lowcut)

And ordinal number k4=N _high+ m ₂, M ₂=2 × (N _highcut-N _high).

E. the T after D step process is carried out inversefouriertransform IFFT and obtain Z, complete the bandpass filtering of this track data, formula is:

$Z_{\left(n\right)}=\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{e}^{i\frac{2\mathrm{\π}}{N}(n-1)(k-1)}{T}_{\left(k\right)},n=\mathrm{1,2},......,N$

Wherein Z _(n)represent the data value after inversefouriertransform, T _(k)data value after bandpass filtering, n is the ordinal number of the data after inverse transformation.

(6) process that next track data is carried out in step (2) ~ (5) is repeated, until process into all track datas in file.

Step 7, goes the data after direct current, bandpass filtering treatment to carry out the process of 2B DBMS to described, carries out geometry location to 2A DBMS, generates 2B DBMS.Namely latitude and longitude coordinates information is added to moon sight radar data.

When moon sight radar raw data is after above-mentioned a series of process, namely eliminate the data error that moon sight radar itself brings, reflect real lunar surface situation, 2B DBMS can be used for issuing, for data user in order to scientific research.

Institute's above-mentioned embodiment of the present invention that it should be understood that only for exemplary illustration or explain principle of the present invention, and is not construed as limiting the invention.Therefore, any amendment made when without departing from the spirit and scope of the present invention, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.In addition, claims of the present invention be intended to contain fall into claims scope and border or this scope and border equivalents in whole change and modification.

Claims (6)

1. a preprocess method for lunar surface rover moon sight radar data, the method comprising the steps of:

Step 1, the raw data that real-time reception moon sight radar passes down;

Step 2, performs frame synchronization to the raw data that passes down, goes to disturb and RS decoding process obtains Framed Data;

Step 3, carries out shunt solution frame to described Framed Data and subpackage process is wrapped to obtain moon sight radar data source;

Step 4, unpacks process to generate moon sight radar data block to described moon sight radar data source bag execution;

Step 5, converts moon sight radar raw voltage data to the moon sight radar data block of described generation;

Step 6, eliminates the DC component in every track data;

Step 7, carries out bandpass filtering to every track data, removes high and low frequency composition;

Step 5 comprises further: carry out physical quantity conversion to the road science data in science data block, first binary number is converted to metric magnitude of voltage, then according to following formula, the road science data of signless 16bit is converted into signed number according to accumulative frequency:

X _(m)＝S _(m)-128×(A+1)(1)

Wherein, m be in the ordinal number of sampled point, X is the signed integer of 16bit, and S is 16bit signless integer, and A is the accumulative frequency of first passage in parameter;

Step 6 comprises further: carry out the process of 2A DBMS to the data after described physical quantity conversion, in 2A DBMS processing procedure, data validity judgement, DC processing, bandpass filtering step are carried out to the data after physical quantity conversion, generate 2A DBMS, and generate corresponding report, wherein, the process of 2A DBMS refers to does calibration process to instrument, eliminate instrument itself to the impact of data, make data representation have the observed quantity of actual physics meaning;

Described bandpass filtering comprises further: calculated rate interval, calculate the position ordinal number of key frequency point, Fast Fourier Transform (FFT) FFT is done to this track data, obtain intermediate data T, use the ordinal number obtained by the data value zero setting in the stopband range of T, T is after treatment carried out inversefouriertransform IFFT and obtains Z, complete the bandpass filtering of this track data.

2. method according to claim 1, is characterized in that, step 2 comprises further:

Step 2.1, searches the frame synchronization head in original bit stream data by bit, intercept 1024 bytes and form multiple initial data frame from frame synchronization head;

Step 2.2, carries out exclusive-OR operation, to go to disturb to initial data frame one by one by frame synchronization 992 later bytes in the scrambler of 992 bytes and initial data frame;

Step 2.3, the RS code spending last 128 bytes in the initial data frame after disturbing carries out RS decoding to initial data frame, and records the data after RS decoding;

Step 2.4, forms the data quality information of the Frame after RS decoding and two bytes the Framed Data that size is 1024 bytes.

3. method according to claim 2, it is characterized in that, shunt solution frame in step 3 refers to according to the pseudo channel information in data frame structure, determine the pseudo channel of Frame, byte valid data in Frame are taken out, form virtual channel data, and form the statistical information of Frame, subpackage and separate loading source bag, refer to input channel configuration information, decompose according to useful load application process identifier, by the detector data of channel composition binary data source APMB package, and the data quality information to each source bag interpolation one byte, formation source bag data.

4. method according to claim 3, it is characterized in that, step 4 comprises further: the virtual channel data after separating frame according to useful load application identities symbol to shunt decomposes and extracts, described useful load application identities symbol is for indicating the load source of data, the moon sight radar data extracted from pseudo channel is formed the data source APMB package of binary format, formation source bag data, the source bag data of moon sight radar are divided into section's cosmogony bag data and engineering source bag data according to data identifier.

5. method according to claim 4, is characterized in that, in step 4, the process that unpacks of section's cosmogony bag data refers to and splices section's cosmogony bag data, forms the data division of LPR science data block.

6. method according to claim 1, is characterized in that, described DC processing comprises further: first get arithmetic mean to per pass clinical practice data, then the data in this road are deducted mean value:

$Y_{\left(n\right)}=X_{\left(n\right)}-\frac{1}{N}\underset{n=1}{\overset{N}{\Σ}}{X}_{\left(n\right)}$

Wherein Y (n) is for removing the data value after direct current, and X (n) is the data value before removing direct current, and n represents the ordinal number of sampled point in this road, and N represents Dao Nei effective sampling points sum, and padding data does not process.