CN100386648C - Real-time imaging method and real-time imaging device for synthetic aperture radar - Google Patents
Real-time imaging method and real-time imaging device for synthetic aperture radar Download PDFInfo
- Publication number
- CN100386648C CN100386648C CNB200510011292XA CN200510011292A CN100386648C CN 100386648 C CN100386648 C CN 100386648C CN B200510011292X A CNB200510011292X A CN B200510011292XA CN 200510011292 A CN200510011292 A CN 200510011292A CN 100386648 C CN100386648 C CN 100386648C
- Authority
- CN
- China
- Prior art keywords
- distance
- orientation
- processing node
- node
- data
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The present invention discloses a real-time imaging method and a real-time imaging device for synthetic aperture radars, which are provided with isostructural first and second distance/orientation finishing nodes. The two finishing nodes are provided with a switchable state for carrying out distance finishing with a compressed distance and an orientation finishing state for carrying out orientation focus. The first and/or the second distance/orientation finishing node respectively carries out distance compression for received partial image data in the distance finishing state and respectively copies a distance compression result to the counterpart finishing node. The first and/or the second distance/orientation finishing node are/is switched to the orientation finishing state after the distance compression is completed. The distance compression result of the partial image data of the self part finishing node and the distance compression result of the partial image data of the counterpart finishing node are combined so as to carry out the orientation finishing of one-image data. In addition, an imaging result is output to a main control node for displaying after the orientation finishing is completed. The structure of the imaging device of the present invention is symmetrical, and the imaging device has high upgradeable performance and maintainability.
Description
Technical field
The present invention relates to synthetic aperture radar (SAR), specifically, relate to a kind of real time imagery method and real time imagery device that is used for synthetic-aperture radar.
Background technology
Synthetic aperture radar (SAR) is a kind of imaging radar that high-definition picture can be provided.Be generally and obtain desirable imaging effect and improve picture quality, SAR need handle lot of data and adopt the algorithm of high complexity.Therefore, the real time imagery device of SAR general require that big I/O handling capacity is arranged, wide bus bandwidth and sufficient calculation resource.On the other hand, commercial disposable plates (the Commercial-off-the-shelf of current paralleling multi-processor, COTS) product with good characteristics such as its quick exploitation that has, low cost, upgrading fast, easy exploitings, is adopted by the modern signal processing system gradually.In order to use the signal-processing board of identical configuration, need adopt the system architecture of symmetry and balance usually based on the system of COTS.Otherwise the resource of system may be wasted, and the performance of system also can descend accordingly.
At present, the imaging algorithm of many effective synthetic-aperture radar is as wavenumber domain (W-K) algorithm, the Chirp-Scaling algorithm has been suggested and be applied in the actual system.Yet distance-Doppler (RD) algorithm is with its good imaging performance, moderate computation complexity and be easy to extensively be adopted by the imaging system of many reality institute in conjunction with excellent characteristic such as motion compensation.As well known for one of skill in the art, in the RD algorithm, mainly comprise two parts: distance compression and orientation compression.Suppose to comprise N in the frame imaging sampled data of synthetic-aperture radar
rIndividual range pulse sampled point and N
aIndividual orientation impulse sampling point promptly has N altogether
r* N
aIndividual sampling pulse data; The distance compression of RD algorithm then is to be in the N in same orientation
rIndividual sampling pulse carries out pulse compression, and to obtain range resolution, the orientation compression of RD algorithm then is to be in the N of same distance
aIndividual sampling pulse carries out pulse compression, to obtain azimuthal resolution; After finishing distance compression and orientation compression, from N
r* N
aExtract the pixel that satisfies resolution requirement in the individual pixel and do real-time demonstration.
In the prior art, the real time imagery device of realizing the RD algorithm adopts the isomery flowing structure usually.As shown in Figure 1, this isomery flowing structure is made up of five nodes usually, that is: main controlled node, I/O node, distance to processing node, data transposition (Corner-Turn) node, orientation to processing node.In this structure, main controlled node provides the flow process control of whole system, the I/O node provides the data input, the distance compression of distance in processing node realization RD algorithm, the orientation compression of orientation in processing node realization RD algorithm, the data transposition node then is used for the storage mode of change system data, so that will be with the continuous data conversion of storing of distance for to store continuously with the orientation.The real time imagery device of this isomery flowing structure is when work, and the data that SAR gathers are by I/O node input system.For the pending image of a frame, the data of this two field picture are normally stored in the continuous mode of distance in system, promptly are in same orientation and the data of different distance are stored continuously, and this is to be convenient to very much distance carry out apart from compaction algorithms to processing node.Distance is after processing node is finished the distance compression, its output data is to store continuously in the orientation by data transposition (Corner-Turn) node transformation, be about to be in same distance and the data of different azimuth are stored continuously, facilitate the orientation like this and these data are carried out the orientation compression to processing node.The orientation in the end satisfies imaging resolution requirement data extract and is delivered to main controlled node and finish real-time demonstration after processing node is finished the orientation compression in the imaging results.Obviously, in isomery flowing structure shown in Figure 1, each processing node that is made of some dsp processors and local memory is single on function.
Yet distance is actually diverse to processing and orientation to computational load and the request memory handled system, adopts the isomery mode must rewind the unbalanced of system.Suppose that a frame imaging data comprises N
rIndividual distance sample and N
aIndividual orientation impulse sampling, and after imaging finishes, N is arranged
r* N
aThe individual pixel that satisfies resolution requirement is extracted out does real-time demonstration.Then the difference between handling can be summarized as table 1 to distance to processing and orientation.The floating number operation of mainly having compared different disposal in the table, and 8 point interpolations are used to the range migration correction (RMC) of computer azimuth in handling.Aspect more minimum request memory, for simplicity, only consider the internal memory of intermediate data storage.In addition, the data structure of 2-16bit (Short type) is used to the intermediate storage of a complex data, can guarantee the dynamic range of the 93dB of imaging like this.
Table 1 orientation is to handling and unbalanced (the every frame data) of distance to processing
| Distance is to processing | The orientation is to processing | |
| Floating-point operation | N aN r(10log 2N r+6) | N aN r(5log 2N a+27) |
| Minimum internal memory | 4N r byte | 2N aN r byte |
Obviously, for the isomery flowing structure, computing load and minimum request memory have all shown tangible lack of uniformity.Especially, work as N
aWhen big, the lack of uniformity of request memory aspect can show more outstandingly.Therefore, when the commercial disposable plates of identical configuration be used simultaneously in the orientation to distance to disposable plates, distance will be by serious waste to the resource of disposable plates.And, because two processing nodes adopt different process softwares fully, system development and safeguard the more complicated that also can become.
Summary of the invention
The objective of the invention is to overcome the shortcoming and defect that existing SAR real time imagery device adopts the isomery flowing structure to be brought, thereby a kind of new real time imagery method that is used for synthetic-aperture radar and real time imagery device are provided.
To achieve these goals, the invention provides a kind of real time imagery method that is used for synthetic-aperture radar, can carry out real time imagery to the impulse sampling data of synthetic-aperture radar collection, this method comprises:
The distance/direction that two isomorphisms are provided is to processing node, comprise first distance/direction to processing node and second distance/orientation to processing node, described distance/direction has switchable distance of carrying out the distance compression to treatment state with carry out orientation that the orientation focuses on to treatment state to processing node;
With described impulse sampling data is that unit alternately is sent to first and second distance/direction to processing node with the field;
Described first and/or second distance/orientation respectively its field data of receiving carried out distance in its distance to treatment state to processing node and compress, and will be copied to the other side's processing node respectively apart from compression result;
Described first and/or second distance/orientation after processing node is finished distance compression, switch to the orientation to treatment state, the orientation of carrying out frame data apart from compression result apart from the field data of compression result and the other side's processing node in conjunction with the field data of our processing node is to processing, and finishes the orientation to handling back output imaging results.
In technique scheme, described first and second distance/direction are alternately switched to treatment state to treatment state and orientation to each comfortable distance of processing node.Described first and second distance/direction to processing node replace mutual exclusion ground be in distance to treatment state and orientation to treatment state, when first and second distance/direction in processing node was in distance to treatment state, then another was in the orientation to treatment state.Described first and second distance/direction realize that to the commercial disposable plates of processing node by two identical configurations described each commercial disposable plates comprises a plurality of dsp processors, and each dsp processor all has local memory.When described first and/or second distance/orientation be in distance during to processing node to treatment state, a plurality of processors in the described commercial disposable plates one by one carry out distance with pipeline mode paired pulses sampled data and compress.When described first and/or after the dsp processor of second distance/orientation in processing node finish distance compression, also to will duplicate first and second distance/direction apart from compression result in the local memory of other dsp processor of processing node.When described first and/or second distance/orientation when processing node is in the orientation to treatment state, its data mean allocation that need handle is given a plurality of processors in the described commercial disposable plates, and described a plurality of processor carries out the orientation compression with parallel mode.When described first and/or second distance/orientation when processing node is in the orientation to treatment state, before carrying out the orientation compression, also by described first and/or second distance/orientation treat deal with data to processing node and carry out data transposition so that will by distance continuously the data conversion of storage for to store continuously with the orientation.
The present invention also provides a kind of real time imagery device that is used for synthetic-aperture radar, can carry out real time imagery to the impulse sampling data of synthetic-aperture radar collection, and this real time imagery device comprises:
The I/O node, being used for described impulse sampling data is that unit alternately is sent to first and second distance/direction to processing node with the field;
The distance/direction of two isomorphisms in parallel is to processing node, comprise first distance/direction to processing node and second distance/orientation to processing node, described distance/direction to processing node have switchable distance of carrying out the distance compression to treatment state and the orientation of carrying out the orientation compression to treatment state;
Main controlled node is used to provide the flow process control and the imaging of described each node of real time imagery device to show;
Wherein, described first and/or second distance/orientation respectively its field data of receiving carried out distance in its distance to treatment state to processing node and compress, and will be copied to the other side's processing node respectively apart from compression result; Described first and/or second distance/orientation after processing node is finished distance compression, switch to the orientation to treatment state, the orientation of carrying out frame data apart from compression result apart from the field data of compression result and the other side's processing node in conjunction with the field data of our processing node is to processing, and shows to handling back output imaging results to main controlled node finishing the orientation.
In above-mentioned real time imagery device, described first and second distance/direction are the commercial disposable plates of two identical configurations to processing node.
With the node of available technology adopting special use realize distance to or the orientation different to single function, in the present invention, two identical processing nodes alternately finish distance to handle and the orientation to processing, two processing nodes can be realized with identical software and hardware configuration, realized the symmetry of system architecture, and realized equilibrium aspect utilizing at computational load and internal memory.And, because two nodes can be carried out identical program code, therefore have high upgradability and maintainability.
Description of drawings
Fig. 1 is existing real time imagery device structural representation;
Fig. 2 is the structural representation of real time imagery device of the present invention;
Fig. 3 is the frame pulse sampled data synoptic diagram that synthetic-aperture radar is gathered.
Embodiment
Below in conjunction with the drawings and specific embodiments the present invention is described in further detail.
As shown in Figure 2, real time imagery device of the present invention comprise main controlled node, I/O node, first distance/direction to processing node and second distance/orientation to processing node.In a preferred embodiment of the present invention, but first and second distance/direction realize to each personal commercial disposable plates of processing node, and these two commercial disposable plates preferably have identical hardware and software configuration, and each commercial disposable plates is made of a plurality of dsp processors that have local memory usually.Main controlled node is used to provide the control of imaging flow process and the imaging of real time imagery device to show that the I/O node is used for the data input of real time imagery device.
Fig. 3 shows a frame pulse sampled data synoptic diagram of synthetic aperture radar (SAR), and each stain is represented impulse sampling data that are positioned at a certain distance and an orientation among the figure.As shown in Figure 3, a frame treats that imaging data has N
rIndividual range pulse sampled point and N
aIndividual orientation impulse sampling point, promptly total N
r* N
aIndividual impulse sampling data, hereinafter, also with each N that goes among Fig. 3
rIndividual impulse sampling data are called a pulse, i.e. pulse is by the N of delegation among Fig. 3
rIndividual impulse sampling data are formed.Impulse sampling data as shown in Figure 3 normally enter into the real time imagery device with the continuous mode of distance order by the I/O node, so-called distance is exactly to pass through the I/O node continuously according to from top to bottom order from left to right in Fig. 3 continuously, and, after these data enter into the real time imagery device, also be to store in the continuous mode of distance in each local memory of commercial disposable plates, this is suitable for first distance/direction and carries out apart from squeeze operation to processing node to processing node and second distance/orientation.
In the present invention, the I/O node is being that unit alternately be sent to first and second distance/direction to processing node with the field to first and second distance/direction when processing node transmits the impulse sampling data.As shown in Figure 3, the preceding N in the frame pulse sampled data
a/ 2 row and back N
a/ 2 line data are respectively two field data, and every field comprises N
r* N
a/ 2 impulse sampling data.Table 1 shows in one embodiment the I/O node to the data sending order of first and second distance/direction to processing node.In table 1, the transmitted data sequence of I/O node is shown in first tabulation, secondary series and the 3rd tabulation are shown in sometime, and section sends to the field data of first and second distance/direction to processing node, wherein UHF represents first frame data, LHF represents second frame data, and the subscript of UHF and LHF is represented frame number.
Table 1
| In proper order | First distance/direction is to processing node | Second distance/orientation is to processing node |
| ... | ... | ... |
| I | UHF K | |
| I+1 | LHF K | |
| I+2 | UHF K+1 | |
| I+3 | LHF K+1 | |
| I+4 | UHF K+2 | |
| ... | ... | ... |
The I/O node switches by strict trigger action and step-by-step counting realization state, in case I/O node counts field N
a/ 2 pulses (N in other words
r* N
a/ 2 impulse sampling data), the I/O node is then with new next field N
a/ 2 pulses (N in other words
r* N
a/ 2 impulse sampling data) be delivered to another distance/direction to processing node.
In the present invention, first distance/direction is isomorphism to processing node and second distance/orientation to processing node, and just these two nodes have identical functions and signal processing mode.As shown in Figure 2, first and second distance/direction are provided with in parallel to processing node, and each distance/direction all is connected with the I/O node to processing node, so as according to the order of aforementioned table 1 from I/O node received pulse sampled data.
First and second distance/direction all have two kinds of treatment states to processing node: distance to treatment state and orientation to treatment state, so that distance compression and the orientation finished respectively in the RD algorithm are compressed.And distance/direction can be switched between treatment state to treatment state and orientation in distance to processing node.When a distance/direction is in distance during to treatment state to processing node, this distance/direction is carried out the distance compression to processing node to its field impulse sampling data that receive, and will be copied to another distance/direction to processing node apart from compression result, be copied to the other side's processing node in other words.When a distance/direction then switches to the orientation to treatment state after processing node is finished the distance compression, in conjunction with this processing node after apart from compression the field data and the other side's processing node copy field data after the distance compression of coming finish frame data the orientation to processing, and after handling, export main controlled node to and show finishing the orientation.In the signal processing of real time imagery device, first and second distance/direction each node in processing node all alternately switches between treatment state to treatment state and orientation in distance.And when a distance/direction is in distance during to treatment state to processing node, another distance/direction then is in the orientation to treatment state to processing node; When a distance/direction when processing node is in the orientation to treatment state, another distance/direction to processing node then be in the distance to treatment state.
In order to be illustrated more clearly in the stateful switchover process of first and second distance/direction to processing node, table 2 shows in one embodiment first and second distance/direction to the order of processing node signal Processing.
Table 2
| In proper order | First distance/direction is to processing node | Second distance/orientation is to processing node |
| ... | ... | ... |
| I | R(UHF K) | D[R(UHF K-1)，R(LHF K-1)] |
| I+1 | D[R(LHF K-1)，R(UHF K)] | R(LHF K) |
| I+2 | R(UHF K+1) | D[R(UHF K)，R(LHF K)] |
| I+3 | D[R(LHF K)，R(UHF K+1)] | R(LHF K+1) |
| I+4 | R(UHF K+2) | D[R(UHF K+1)，R(LHF K+1)] |
| ... | ... | ... |
In table 2, the signal processing sequence of first and second distance/direction to processing node shown in first tabulation, first and second distance/direction were to the treatment state of processing node when secondary series and the 3rd tabulation were shown in a certain order, and the UHF of subscripting is identical with table 1 with the implication that LHF represents.Wherein, R (*) expression distance/direction is carried out the distance compression to processing node in its distance under treatment state operation and result, the content representation operand in the bracket; D[*, *] expression distance/direction the orientation squeeze operation from its orientation to processing node and the result that carry out under treatment state in, the content representation operand in the bracket.As can be seen, first and second distance/direction are carried out the distance compression to each comfortable its distance of processing node double frame pulse sampled data under treatment state, and two field impulse sampling data are carried out the orientation compression of frame data under treatment state in its orientation.Particularly, as shown in table 2:
In order I, first distance/direction is to first frame data UHF of processing node to the K frame
KCarry out the distance compression, obtain (UHF apart from compression result R
K), and with R (UHF
K) be copied to second distance/orientation to processing node; And second distance/orientation to processing node to first frame data R (UHF of the K-1 frame of distance compression before
K-1) and second frame data (LHF
K-1) carry out the orientation compression of frame data.
In order I+1, first distance/direction switches to the orientation to treatment state to processing node, the R (UHF that first distance/direction is obtained to processing node
K) and duplicate next R (LHF from second distance/orientation to processing node before
K-1) carry out the orientation compression of frame data; And second distance/orientation switches to distance to treatment state, to second frame data LHF of K frame to processing node
KCarry out the distance compression, obtain (LHF apart from compression result R
K), and with R (LHF
K) be copied to first distance/direction to processing node.
In order I+2, first distance/direction switches to distance to treatment state, to first frame data UHF of K+1 frame to processing node
K+1Carry out the distance compression, obtain (UHF apart from compression result R
K+1), and with R (UHF
K+1) be copied to second distance/orientation to processing node; And second distance/orientation switches to the orientation to treatment state to processing node, the R (LHF that second distance/orientation is obtained to processing node
K) and duplicate next R (UHF from first distance/direction to processing node before
K) carry out the orientation compression of frame data.
In order I+3, first distance/direction switches to the orientation to treatment state to processing node, the R (UHF that first distance/direction is obtained to processing node
K+1) and duplicate next R (LHF from second distance/orientation to processing node before
K) carry out the orientation compression of frame data; And second distance/orientation switches to distance to treatment state, to second frame data LHF of K+1 frame to processing node
K+1Carry out the distance compression, obtain (LHF apart from compression result R
K+1), and with R (LHF
K+1) be copied to first distance/direction to processing node.
In order I+4, first distance/direction switches to distance to treatment state, to first frame data UHF of K+2 frame to processing node
K+2Carry out the distance compression, obtain (UHF apart from compression result R
K+2), and with R (UHF
K+2) be copied to second distance/orientation to processing node; And second distance/orientation switches to the orientation to treatment state to processing node, the R (LHF that second distance/orientation is obtained to processing node
K+1) and duplicate next R (UHF from first distance/direction to processing node before
K+1) carry out the orientation compression of frame data.
And the like, until the impulse sampling data of handling all frames.In said process, directly deliver to main controlled node through the data of orientation compression and show in real time.As known from Table 2, first and second distance/direction are alternately switched to treatment state to treatment state and orientation to each comfortable distance of processing node, and when first and second distance/direction in processing node was in distance to treatment state, then another was in the orientation to treatment state.
In addition, when first and second distance/direction are being carried out the orientation when handling to processing node, be preferably in and carry out treating deal with data before the orientation compression and carry out data transposition, so that will the data conversion of storage is for to store continuously in the orientation continuously by distance, this be the same with processing of the prior art.But, in the present invention, adopt and carry out data transposition, and independent data transposition node is not provided by the program of first and second distance/direction in processing node.
From table 1 and table 2 also as can be known, first and second distance/direction carry out data are carried out having identical workflow when the distance compression is compressed with the orientation in the received pulse sampled data to processing node, therefore, first and second distance/direction can realize that to processing node the commercial disposable plates of for example available two identical configurations realizes with the hardware and software with identical configuration.The real time imagery device structural symmetry of Xing Chenging like this realizes that first and second distance/direction are basic identical to the two commercial disposable plates computational loads and the request memory of processing node.
Be used for realizing that first and second distance/direction generally include a plurality of dsp processors to the commercial disposable plates of processing node, and each dsp processor all there is local memory separately.First or second distance/orientation to processing node its distance when treatment state carries out apart from compaction algorithms, the N if distance samples is counted
rWhen very big, even for state-of-the-art dsp processor, at a pulse recurrence interval T
pIn finish pulse distance compression also be very difficult.Therefore, in the present invention, when distance/direction to processing node in distance during to treatment state, the dsp processor of the N in the commercial disposable plates preferably adopts the pipeline mode paired pulses to carry out distance one by one to compress.Promptly be that (each pulse comprises N in unit with the pulse
rIndividual range pulse sampled data) gives N dsp processor with the pipeline mode order assignment.
Table 3
| ... | ... |
| Pulse I | DSPN makes distance to processing |
| Pulse I+1 | DSP1 makes distance to processing |
| Pulse I+2 | DSP2 makes distance to processing |
| ... | ... |
| Pulse I+N | DSPN makes distance to processing |
| Pulse I+N+1 | DSP1 makes distance to processing |
| ... | ... |
In one embodiment, the corresponding relation of different pulses and dsp processor is as shown in table 3.To be example to I+1 burst process, the DSP1 processor can be finished process of pulse-compression before I+N+1 pulse arrives, i.e. the time of pulse compression can be loosened to N*T
pAfter a pulse was by the distance compression, compressed data not only were stored in the local memory of current processor, also by in the local memory that is delivered to every other processor of part.Suppose that a node has N dsp processor, not only be retained in the local memory of current dsp processor, also be passed in other the local memory of 2*N-1 dsp processor of system by the data of distance compression.After a dsp processor compresses a pulse, obtain N
rIndividual apart from packed data, this N
rIndividually be divided into N part, be stored in respectively in the local memory of N dsp processor of this processing node apart from packed data, and corresponding respectively copying in the local memory of the dsp processor in another processing node, at N
rIndividual as shown in table 4 in the storage of the dsp processor of two processing nodes apart from packed data:
Table 4
| Apart from packed data | |
| 1～Nr/N | Two node DSP1 separately |
| Nr/N+1～2*Nr/N | Two node DSP2 separately |
| ... | ... |
| (N-1)Nr/N+1～Nr | Two node DSPN separately |
Then, as field data, i.e. N
a/ 2 pulses are after the present node compression, in conjunction with another node compression and the next N of transmission
aData just accumulated at current processing node and finish after/2 pulses, a frame pitch tripping were contracted, and this node is directly switch to " orientation is to handling ".Under " orientation is to handling " state, the parallel mode of N DSP employing table 5 is finished the orientation of frame data to processing on this processing node.Wherein each DSP finishes the data processing of Nr/N range unit, and the orientation of promptly finishing 1/N wherein is to Processing tasks.
Table 5
| DSP1 | DSP2 | ... | DSP N |
| Handle the data block of 1～Nr/N range unit correspondence | Handle the data block of Nr/N+1～2*Nr/N range unit correspondence | ... | Handle the data block of (N-1) * Nr/N+1～Nr range unit |
Claims (10)
1. a real time imagery method that is used for synthetic-aperture radar can be carried out real time imagery to the impulse sampling data of synthetic-aperture radar collection, and this method comprises:
The distance/direction that two isomorphisms are provided is to processing node, comprise first distance/direction to processing node and second distance/orientation to processing node, described distance/direction has switchable distance of carrying out the distance compression to treatment state with carry out orientation that the orientation focuses on to treatment state to processing node;
With described impulse sampling data is that unit alternately is sent to first and second distance/direction to processing node with the field;
Described first and/or second distance/orientation respectively its field data of receiving carried out distance in its distance to treatment state to processing node and compress, and will be copied to the other side's processing node respectively apart from compression result;
Described first and/or second distance/orientation after processing node is finished distance compression, switch to the orientation to treatment state, the orientation that the field data of coming apart from compression result and the other side's processing node copy in conjunction with the field data of our processing node are carried out frame data apart from compression result is to processing, and finishes the orientation to handling back output imaging results.
2. the real time imagery method that is used for synthetic-aperture radar according to claim 1 is characterized in that, described first and second distance/direction are alternately switched to treatment state to treatment state and orientation to each comfortable distance of processing node.
3. the real time imagery method that is used for synthetic-aperture radar according to claim 2, it is characterized in that, described first and second distance/direction to processing node replace mutual exclusion ground be in distance to treatment state and orientation to treatment state, when first and second distance/direction in processing node was in distance to treatment state, then another was in the orientation to treatment state.
4. the real time imagery method that is used for synthetic-aperture radar according to claim 1, it is characterized in that, described first and second distance/direction realize to the commercial disposable plates of processing node by two identical configurations, described each commercial disposable plates comprises a plurality of dsp processors, and each dsp processor all has local memory.
5. the real time imagery method that is used for synthetic-aperture radar according to claim 4, it is characterized in that, when described first and/or second distance/orientation be in distance during to processing node to treatment state, a plurality of processors in the described commercial disposable plates carry out distance with pipeline mode paired pulses sampled data and compress.
6. the real time imagery method that is used for synthetic-aperture radar according to claim 5, it is characterized in that, when described first and/or after the dsp processor of second distance/orientation in processing node finish distance compression, also to will copy to first and second distance/direction apart from compression result in the local memory of other dsp processor of processing node.
7. according to claim 4 or the 6 described real time imagery methods that are used for synthetic-aperture radar, it is characterized in that, when described first and/or second distance/orientation when processing node is in the orientation to treatment state, its the data mean allocation that need handle is given a plurality of processors in the described commercial disposable plates, and described a plurality of processor carries out the orientation compression with parallel mode.
8. the real time imagery method that is used for synthetic-aperture radar according to claim 7, it is characterized in that, when described first and/or second distance/orientation when processing node is in the orientation to treatment state, before carrying out the orientation compression, also by described first and/or second distance/orientation treat deal with data to processing node and carry out data transposition so that will by distance continuously the data conversion of storage for to store continuously with the orientation.
9. a real time imagery device that is used for the real time imagery method of synthetic-aperture radar according to claim 1 can carry out real time imagery to the impulse sampling data of synthetic-aperture radar collection, and this real time imagery device comprises:
The I/O node, being used for described impulse sampling data is that unit alternately is sent to first and second distance/direction to processing node with the field;
The distance/direction of two isomorphisms in parallel is to processing node, comprise first distance/direction to processing node and second distance/orientation to processing node, described distance/direction to processing node have switchable distance of carrying out the distance compression to treatment state and the orientation of carrying out the orientation compression to treatment state;
Main controlled node is used to provide the flow process control and the imaging of described each node of real time imagery device to show;
Wherein, described first and/or second distance/orientation respectively its field data of receiving carried out distance in its distance to treatment state to processing node and compress, and will be copied to the other side's processing node respectively apart from compression result; Described first and/or second distance/orientation after processing node is finished distance compression, switch to the orientation to treatment state, the orientation of carrying out frame data apart from compression result apart from the field data of compression result and the other side's processing node in conjunction with the field data of our processing node is to processing, and shows to handling back output imaging results to main controlled node finishing the orientation.
10. real time imagery device according to claim 9 is characterized in that, described first and second distance/direction are the commercial disposable plates of two identical configurations to processing node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200510011292XA CN100386648C (en) | 2005-02-01 | 2005-02-01 | Real-time imaging method and real-time imaging device for synthetic aperture radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200510011292XA CN100386648C (en) | 2005-02-01 | 2005-02-01 | Real-time imaging method and real-time imaging device for synthetic aperture radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1815260A CN1815260A (en) | 2006-08-09 |
| CN100386648C true CN100386648C (en) | 2008-05-07 |
Family
ID=36907538
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200510011292XA Expired - Fee Related CN100386648C (en) | 2005-02-01 | 2005-02-01 | Real-time imaging method and real-time imaging device for synthetic aperture radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100386648C (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101452073B (en) * | 2007-11-30 | 2011-12-28 | 清华大学 | Broadband signal synthesizing method based on multi-sending and multi-receiving frequency division radar |
| CN101216553B (en) * | 2007-12-27 | 2011-03-30 | 南京航空航天大学 | Synthetic aperture radar polar coordinates format image-forming method based on variable metric principle |
| CN101498787B (en) * | 2008-01-29 | 2011-02-09 | 电子科技大学 | Linear array three-dimensional imaging synthetic aperture radar fast imaging method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1148067C (en) * | 2001-06-01 | 2004-04-28 | 中国科学院计算技术研究所 | Data compressing method for complex image of synthetic apertre radar |
| CN1534309A (en) * | 2003-03-31 | 2004-10-06 | 中国科学院电子学研究所 | Synthetic pore diameter radar real time image treatment device distance compression treatment circuit |
-
2005
- 2005-02-01 CN CNB200510011292XA patent/CN100386648C/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1148067C (en) * | 2001-06-01 | 2004-04-28 | 中国科学院计算技术研究所 | Data compressing method for complex image of synthetic apertre radar |
| CN1534309A (en) * | 2003-03-31 | 2004-10-06 | 中国科学院电子学研究所 | Synthetic pore diameter radar real time image treatment device distance compression treatment circuit |
Non-Patent Citations (2)
| Title |
|---|
| 一种合成孔径雷达参数化自聚焦的新方法. 许稼，彭应宁，张砾坪，林彦，夏香根.雷达科学与技术，第1卷第4期. 2003 |
| 一种合成孔径雷达参数化自聚焦的新方法. 许稼，彭应宁，张砾坪，林彦，夏香根.雷达科学与技术，第1卷第4期. 2003 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1815260A (en) | 2006-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mantel | Why stepdown procedures in variable selection | |
| JPS635775B2 (en) | ||
| CN102298139B (en) | Two-dimensional windowing method of synthetic aperture radar (SAR) imaging system based on field programmable gate array (FPGA) | |
| CN101246256B (en) | Scalable high speed wave front slope handler based on FPGA | |
| CN102541858A (en) | Data equality processing method, device and system based on mapping and protocol | |
| CN100386648C (en) | Real-time imaging method and real-time imaging device for synthetic aperture radar | |
| KR19990029407A (en) | CT imaging method and X-ray CT device | |
| Liu et al. | Evsrnet: Efficient video super-resolution with neural architecture search | |
| JPH04167168A (en) | Signal processor | |
| CN110991619A (en) | Neural network processor, chip and electronic equipment | |
| CN110377874B (en) | Convolution operation method and system | |
| CN101256668B (en) | Method for implementing video filtering to working balanced multiple nucleus | |
| CN105044706B (en) | A kind of Adaptive beamformer method | |
| CN104821801A (en) | Moving target detection filtering device and method | |
| CN107678781B (en) | Processor and method for executing instructions on processor | |
| CN106204669A (en) | A kind of parallel image compression sensing method based on GPU platform | |
| CN108960203A (en) | A kind of vehicle checking method based on FPGA Heterogeneous Computing | |
| Srinivasan et al. | High performance scalable FPGA accelerator for deep neural networks | |
| Zhong et al. | Content-adaptive image compressed sensing using deep learning | |
| Kim et al. | A real-time 3D image refinement using two-line buffers | |
| Koch et al. | Adaptive load balancing in a distributed environment | |
| JP2775307B2 (en) | Data collection device for magnetic resonance diagnostic equipment | |
| Adinetz et al. | Computation of mutual information metric for image registration on multiple gpus | |
| CN1672327B (en) | Data processing circuit | |
| RU1810943C (en) | Digital adaptive antenna system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080507 Termination date: 20110201 |