CN105259538B - One kind is based on the convergent signal quality evaluating method of signal characteristic and device - Google Patents
One kind is based on the convergent signal quality evaluating method of signal characteristic and device Download PDFInfo
- Publication number
- CN105259538B CN105259538B CN201510712858.5A CN201510712858A CN105259538B CN 105259538 B CN105259538 B CN 105259538B CN 201510712858 A CN201510712858 A CN 201510712858A CN 105259538 B CN105259538 B CN 105259538B
- Authority
- CN
- China
- Prior art keywords
- pulse signal
- pulse
- signal
- value
- feature
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/36—Means for anti-jamming, e.g. ECCM, i.e. electronic counter-counter measures
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses one kind to be based on the convergent signal quality evaluating method of signal characteristic and device, is related to the field of target recognition of electronic reconnaissance specialty.Technical key point：Aerial radar pulse signal is received, the detection parameters of each radar pulse signal are measured, obtains the detection parameters of each pulse signal；The pulse signal received is clustered according to the detection parameters of pulse signal, the pulse signal that same radar signal source is sent under same mode of operation is classified as one group；Characteristics extraction is carried out to each pulse signal in each pulse signal group, obtains N number of feature value vector Vi；Calculate the Euclidean distance value d of the feature value vector of any two pulse signal in each pulse groupi；Calculate the value C for reacting each distance value average sizeD；If described value CDReact pulse signal characteristic convergence in the pulse group.
Description
Technical field
The present invention relates to the field of target recognition of electronic reconnaissance specialty.
Background technology
Usually need to scout the radar pulse signal in objective emission to environment in target identification technology field, carry out
Various interference signals in the actual environment of radar signal reconnaissance be present so that it is sufficiently complex so that signal transactings at different levels to scout environment
Condition often do not reach perfect condition.Such as scout environment in exist reflection, multipath, low signal-to-noise ratio, receiver burr signal,
Situations such as spurious signal, these can all have detrimental effect to signal characteristic parameter extraction, and system generation may finally be caused wrong
Information by mistake.
Existing application system is simply simply judged signal amplitude, not to signal due to receiving environment and biography
The factors such as defeated approach bring quality to influence effectively to be assessed.
The content of the invention
The technical problems to be solved by the invention are：For above-mentioned problem, there is provided one kind is received based on signal characteristic
The signal quality evaluating method and device held back.
It is provided by the invention to be based on the convergent signal quality evaluating method of signal characteristic, including：
Step 1：Aerial radar pulse signal is received, the detection parameters of each radar pulse signal are measured, obtained
To the detection parameters of each pulse signal；
Step 2：The pulse signal received is clustered according to the detection parameters of pulse signal, by same radar signal
The pulse signal that emission source is sent under same mode of operation is classified as one group；
Step 3：To in each pulse signal group each pulse signal carry out characteristics extraction, obtain N number of characteristic value to
Measure Vi, ViRepresent the feature value vector of i-th of pulse；Wherein, i take 1,2,3 ..., N；N is the pulse signal in pulse signal group
Sum；
Step 4：Calculate the Euclidean distance value d of the feature value vector of any two pulse signal in each pulse groupi；Wherein
I takes 1,2,3 ...,
Calculate reactionThe value C of individual distance value average sizeD；If described value CDReact pulse signal characteristic in the pulse group
Convergence.
Step 4 further comprises：
Step 41：To ViBeing normalized makes each of which element for the numerical value between 0~1；Wherein, i take 1,
2、3、…、N；
Step 42：Calculate the feature value vector of any two pulse signal in each pulse group Euclidean distance be worth to away from
From matrix D,Wherein,dijFor ith feature value to
Measure the Euclidean distance value to j-th of feature value vector, vikFor k-th of element of ith feature value vector, vjkFor j-th of feature
It is worth k-th of element of vector；M is characterized the sum of element in value vector；
Step 43：Calculation formulaWherein i takes 1,2,3 ..., N, j take 1,2,3 ..., N, and i ≠
j。
Further, the detection parameters include at least one of following parameter：Pulse signal arrival time, pulse signal
Frequency, pulse signal pulsewidth, pulse amplitude, pulse signal orientation.
Further, the characteristic value includes at least one of following characteristics value：Characteristic parameters of spectra, the waveform of pulse signal
Characteristic value, intrapulse modulation characteristic value.
Present invention also offers one kind to be based on the convergent Signal quality assessment device of signal characteristic, including：
Detection parameters measuring unit, for receiving aerial radar pulse signal, the detection to each radar pulse signal
Parameter measures, and obtains the detection parameters of each pulse signal；
Cluster cell, the pulse signal received is clustered for the detection parameters according to pulse signal, will be same
The pulse information that radar signal source is sent under same mode of operation is classified as one group；
Characteristics extraction unit, for carrying out characteristics extraction to each pulse signal in each pulse signal group, obtain
To N number of feature value vector Vi, ViRepresent the feature value vector of i-th of pulse；Wherein, i take 1,2,3 ..., N；N is pulse signal group
In pulse signal sum；
Pulse signal feature convergence acquiring unit, for calculating the feature of any two pulse signal in each pulse group
It is worth the Euclidean distance value d of vectori；Wherein i takes 1,2,3 ...,
And reacted for calculatingThe value C of individual distance value average sizeD；If described value CDReact pulse in the pulse group
Signal characteristic convergence.
Pulse signal feature convergence acquiring unit further comprises：
Normalization unit, for ViBeing normalized makes each of which element for the numerical value between 0~1；Its
In, i takes 1,2,3 ..., N；
Distance matrix acquiring unit, for calculating the Europe of the feature value vector of any two pulse signal in each pulse group
Formula distance is worth to Distance matrix D,Wherein,dijFor
Ith feature value vector is to the Euclidean distance value of j-th of feature value vector, vikFor ith feature value vector k-th of element,
vjkFor k-th of element of j-th of feature value vector；M is characterized the sum of element in value vector；
Pulse signal feature convergency value acquiring unit, for calculation formulaObtain pulse signal
Feature convergency value CD, wherein i takes 1,2,3 ..., N, j take 1,2,3 ..., N, and i ≠ j.
In summary, by adopting the above-described technical solution, the beneficial effects of the invention are as follows：
In Practical Project it was found that, signal characteristic abstraction error mainly comes from three aspect reasons：First, pulse is adopted
Concentrate because the factors such as multipath, burr cause feature to be distorted；Second, when target pulse signal to noise ratio is weaker, feature is not enough received
Hold back；Third, when signal sorting has leakage batch, other signals of incorporation interfere to echo signal feature extraction.These three situations
It can cause the convergence of eigenmatrix relatively low, then influence follow-up identifying processing.
, can be from each category feature of extraction the invention provides a kind of convergent signal quality evaluating method of feature based
In, select a part of feature of better astringency to reduce above-mentioned three kinds of non-idealities as far as possible as final feature extraction result
Influence to signal quality, improve the accuracy of follow-up signal identification.
Brief description of the drawings
Examples of the present invention will be described by way of reference to the accompanying drawings, wherein：
Fig. 1 is the inventive method flow chart.
Embodiment
All features disclosed in this specification, or disclosed all methods or during the step of, except mutually exclusive
Feature and/or step beyond, can combine in any way.
Any feature disclosed in this specification, unless specifically stated otherwise, can be equivalent by other or with similar purpose
Alternative features are replaced.I.e., unless specifically stated otherwise, each feature is an example in a series of equivalent or similar characteristics
.
During electronic reconnaissance, signal quality quality has tremendous influence to feature extraction and identification.The present invention carries
Go out a kind of convergent signal estimation method of feature based, judge signal quality using the signal characteristic degree of convergence.Signal characteristic is received
Degree of holding back more high RST quality is better, and the signal characteristic degree of convergence more low signal quality is poorer.Specific steps such as Fig. 1, including：
Step 1：Aerial radar pulse signal is received, the detection parameters of each radar pulse signal are measured, obtained
To the detection parameters of each pulse signal.Detection method commonly used in the art has tim e- domain detection and frequency domain detection, detection parameters bag
Include but be not limited to arrival time of pulse signal, frequency, pulsewidth, amplitude, orientation etc..
Step 2：The pulse signal received is clustered according to the detection parameters of pulse signal, the purpose of cluster is
The pulse signal that same radar signal source is sent under same mode of operation is classified as one group.Cluster side commonly used in the art
Method has：CDIF and SDIF clustering methods based on PRI or based on parameter Euclidean distance clustering method etc..
Step 3：To in each pulse signal group each pulse signal carry out characteristics extraction, obtain N number of characteristic value to
Measure Vi, ViRepresent the feature value vector of i-th of pulse；Wherein, i take 1,2,3 ..., N；N is the pulse signal in pulse signal group
Sum.Characteristic value includes but is not limited to spectrum signature, wave character and intrapulse modulation characteristic of pulse signal etc..
Step 4：Feature convergence judges, by calculating signal characteristic convergence, judges whether signal characteristic is stablized.
Feature convergence computational methods include：
Calculate the Euclidean distance value d of the feature value vector of any two pulse signal in each pulse groupi；Wherein i takes 1,
2、3、…、Euclidean distance can be with measuring vector ViWith vectorial VjBetween uniformity.
Calculate reactionThe value C of individual distance value average sizeD；Described value CDPulse signal characteristic in the pulse group is reacted to receive
Holding back property, if value CDIt is smaller, then it is assumed that the better astringency of this feature value, it is more accurate to react pulse signal with this feature value.
Value C in one embodimentDForThe arithmetic mean of instantaneous value of individual distance value.In a preferred embodiment, value CDIt is logical
Cross following steps acquisition：
Step 41：To ViBeing normalized makes each of which element for the numerical value between 0~1；Wherein, i take 1,
2、3、…、N；The mode of normalized has a variety of, and one of which is to calculate vectorial ViIn each element summation, then use summation
Remove vectorial ViIn each element obtain the value after the normalization of the element.
Step 42：Calculate the feature value vector of any two pulse signal in each pulse group Euclidean distance be worth to away from
From matrix D,Wherein,dijFor ith feature value vector
To the Euclidean distance value of j-th of feature value vector, vikFor k-th of element of ith feature value vector, vjkFor j-th of characteristic value
K-th of element of vector；M is characterized the sum of element in value vector.
Step 43：Calculation formulaWherein i takes 1,2,3 ..., N, j take 1,2,3 ..., N, and i ≠
j。
Due to vectorial ViIt has passed through normalized, therefore dijIt is the numerical value between 0~1, dijIt is smaller, represent from
I pulse with from j-th of DISCHARGE PULSES EXTRACTION to feature it is more consistent, otherwise more dissipate.By dijAvailable N × N Distance matrix Ds.
It is not difficult to learn, the element on diagonal in Distance matrix D is 0, and element dijEqual to element dji, i ≠ j.
Generally, the accurate mark of feature extraction is, to the feature phase of each DISCHARGE PULSES EXTRACTION of same target
Mutually convergence, i.e. dijIt is as small as possible.Theoretically, the spy of signal of the same radar signal source radiation under same mode of operation
Sign is constant, and it is because environment, radiation, the influence of receiving channel various factors why to change, it is considered that these influences
It is typically random.It is demonstrated experimentally that the factor of various influences is smaller, feature more restrains.Therefore, feature is received defined in the present embodiment
Holding back property is：
Wherein i takes 1,2,3 ..., N, j take 1,2,3 ..., N, and i ≠ j.Characteristic vector between pulse more restrains, CDValue
Closer 1, otherwise closer 0, therefore CDCan be as the evaluation index of feature extraction quality.Technical staff can be special according to signal
Levy convergency value CDOptimal feature is chosen to characterize signal, improves the degree of accuracy of follow-up signal identification.
The invention is not limited in foregoing embodiment.The present invention, which expands to, any in this manual to be disclosed
New feature or any new combination, and disclose any new method or process the step of or any new combination.
Claims (6)
1. one kind is based on the convergent signal quality evaluating method of signal characteristic, it is characterised in that including：
Step 1：Aerial radar pulse signal is received, the detection parameters of each radar pulse signal are measured, is obtained each
The detection parameters of individual pulse signal；
Step 2：The pulse signal received is clustered according to the detection parameters of pulse signal, by same radar signal
The pulse information that source is sent under same mode of operation is classified as one group；
Step 3：Characteristics extraction is carried out to each pulse signal in each pulse signal group, obtains N number of feature value vector Vi,
ViRepresent the feature value vector of i-th of pulse；Wherein, i take 1,2,3 ..., N；N is the total of the pulse signal in pulse signal group
Number；
Step 4：Calculate the Euclidean distance value d of the feature value vector of any two pulse signal in each pulse groupi；Wherein i takes
Calculate what reactions steps 4 obtainedThe value C of individual distance value average sizeD；If described value CDPulse in the pulse group is reacted to believe
Number feature convergence；
Step 4 further comprises：
Step 41：To ViBeing normalized makes each of which element for the numerical value between 0~1；Wherein, i take 1,2,
3、…、N；
Step 42：The Euclidean distance for calculating the feature value vector of any two pulse signal in each pulse group is worth to apart from square
Battle array D,Wherein,dijFor ith feature value vector to
The Euclidean distance value of j feature value vector, vikFor k-th of element of ith feature value vector, vjkFor j-th of feature value vector
K-th of element；M is characterized the sum of element in value vector；
Step 43：Calculation formulaWherein i takes 1,2,3 ..., N, j take 1,2,3 ..., N, and i ≠ j.
2. one kind according to claim 1 is based on the convergent signal quality evaluating method of signal characteristic, it is characterised in that institute
Stating detection parameters includes at least one of following parameter：Pulse signal arrival time, pulse signal frequency, pulse signal arteries and veins
Width, pulse amplitude, pulse signal orientation.
3. one kind according to claim 1 is based on the convergent signal quality evaluating method of signal characteristic, it is characterised in that institute
Stating characteristic value includes at least one of following characteristics value：Characteristic parameters of spectra, wave character value, the intra-pulse modulation of pulse signal are special
Value indicative.
4. one kind is based on the convergent Signal quality assessment device of signal characteristic, it is characterised in that including：
Detection parameters measuring unit, for receiving aerial radar pulse signal, to the detection parameters of each radar pulse signal
Measure, obtain the detection parameters of each pulse signal；
Cluster cell, the pulse signal received is clustered for the detection parameters according to pulse signal, by same radar
The pulse information that signal emitting-source is sent under same mode of operation is classified as one group；
Characteristics extraction unit, for carrying out characteristics extraction to each pulse signal in each pulse signal group, obtain N number of
Feature value vector Vi, ViRepresent the feature value vector of i-th of pulse；Wherein, i take 1,2,3 ..., N；N is in pulse signal group
The sum of pulse signal；
Pulse signal feature convergence acquiring unit, for calculate the characteristic value of any two pulse signal in each pulse group to
The Euclidean distance value d of amounti；Wherein i takes
And reacted for calculatingThe value C of individual distance value average sizeD；If described value CDReact pulse signal in the pulse group
Feature convergence；
Pulse signal feature convergence acquiring unit further comprises：
Normalization unit, for ViBeing normalized makes each of which element for the numerical value between 0~1；Wherein, i takes
1、2、3、…、N；
Distance matrix acquiring unit, for calculate the feature value vector of any two pulse signal in each pulse group it is European away from
From being worth to Distance matrix D,Wherein,dijFor i-th
Feature value vector is to the Euclidean distance value of j-th of feature value vector, vikFor k-th of element of ith feature value vector, vjkFor
K-th of element of j-th of feature value vector；M is characterized the sum of element in value vector；
Pulse signal feature convergency value acquiring unit, for calculation formulaObtain pulse signal feature
Convergency value CD, wherein i takes 1,2,3 ..., N, j take 1,2,3 ..., N, and i ≠ j.
5. one kind according to claim 4 is based on the convergent Signal quality assessment device of signal characteristic, it is characterised in that institute
Stating detection parameters includes at least one of following parameter：Pulse signal arrival time, pulse signal frequency, pulse signal arteries and veins
Width, pulse amplitude, pulse signal orientation.
6. one kind according to claim 4 is based on the convergent Signal quality assessment device of signal characteristic, it is characterised in that institute
Stating characteristic value includes at least one of following characteristics value：Characteristic parameters of spectra, wave character value, the intra-pulse modulation of pulse signal are special
Value indicative.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510712858.5A CN105259538B (en) | 2015-10-28 | 2015-10-28 | One kind is based on the convergent signal quality evaluating method of signal characteristic and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510712858.5A CN105259538B (en) | 2015-10-28 | 2015-10-28 | One kind is based on the convergent signal quality evaluating method of signal characteristic and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105259538A CN105259538A (en) | 2016-01-20 |
| CN105259538B true CN105259538B (en) | 2018-02-02 |
Family
ID=55099299
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510712858.5A Active CN105259538B (en) | 2015-10-28 | 2015-10-28 | One kind is based on the convergent signal quality evaluating method of signal characteristic and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105259538B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106056098B (en) * | 2016-06-23 | 2019-07-02 | 哈尔滨工业大学 | A kind of pulse signal cluster method for separating based on categories combination |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5999373A (en) * | 1982-11-30 | 1984-06-08 | Mitsubishi Electric Corp | Radar equipment |
| CN101762808A (en) * | 2010-01-15 | 2010-06-30 | 山东大学 | Method for extracting radar pulse based on self-adaption threshold value |
| CN103839073A (en) * | 2014-02-18 | 2014-06-04 | 西安电子科技大学 | Polarization SAR image classification method based on polarization features and affinity propagation clustering |
-
2015
- 2015-10-28 CN CN201510712858.5A patent/CN105259538B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5999373A (en) * | 1982-11-30 | 1984-06-08 | Mitsubishi Electric Corp | Radar equipment |
| CN101762808A (en) * | 2010-01-15 | 2010-06-30 | 山东大学 | Method for extracting radar pulse based on self-adaption threshold value |
| CN103839073A (en) * | 2014-02-18 | 2014-06-04 | 西安电子科技大学 | Polarization SAR image classification method based on polarization features and affinity propagation clustering |
Non-Patent Citations (2)
| Title |
|---|
| 《复杂电磁环境下雷达信号分选技术》;易冰歆 等;《电子信息对抗技术》;20141115;第57-59页 * |
| 《通信对抗中的现代信号处理技术应用研究》;姜园;《中国优秀博硕士学位论文全文数据库 信息科技辑》;20040915;第91-95页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105259538A (en) | 2016-01-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10429488B1 (en) | System and method for geo-locating and detecting source of electromagnetic emissions | |
| Gross et al. | Maximum-likelihood power-distortion monitoring for GNSS-signal authentication | |
| US8537050B2 (en) | Identification and analysis of source emissions through harmonic phase comparison | |
| Kawalec et al. | Specific emitter identification using intrapulse data | |
| CN108988987B (en) | Method for verifying credibility of complex modulation signal processing result based on extreme value likelihood ratio | |
| Shangguan et al. | OTrack: Towards order tracking for tags in mobile RFID systems | |
| CN111965632A (en) | Radar target detection method based on Riemann manifold dimensionality reduction | |
| US11470451B2 (en) | System and method for automated data collection and anchor location evaluation | |
| CN106486769A (en) | Spatial interpolation methods for linear phased array antenna and equipment | |
| CN106443660B (en) | A kind of Modulation Continuous Wave Radar multiple mobile object matching process | |
| CN106168662A (en) | The error registration method of passive sensor based on Maximum-likelihood estimation and device | |
| RU2380719C2 (en) | Method for location finding of radiation sources at one frequency | |
| Hindson et al. | A large-scale, low-frequency murchison widefield array survey of galactic H II regions between 260< l< 340 | |
| CN106802406A (en) | A kind of radiation source correlating method for passive radar | |
| Riseley et al. | Deep observations of the Super-CLASS supercluster at 325 MHz with the GMRT: the low-frequency source catalogue | |
| CN110837079B (en) | Target detection method and device based on radar | |
| Rojas et al. | Transceiver inphase/quadrature imbalance, ellipse fitting, and the universal software radio peripheral | |
| CN105259538B (en) | One kind is based on the convergent signal quality evaluating method of signal characteristic and device | |
| Marchili et al. | A seasonal cycle and an abrupt change in the variability characteristics of the intraday variable source S4 0954+ 65 | |
| Li et al. | Range-angle dependent detection for FDA-MIMO radar | |
| CN108226851A (en) | For the method and goniometer of direction finding | |
| CN103076595B (en) | Abnormal type identifying method for multivariate discrete radar emitter signal | |
| CN105068053A (en) | Method for extracting background signal from radar cross section (RCS) measurement data | |
| CN104502889B (en) | Positioning credibility computational methods based on reference point ultimate range in fingerprint location | |
| CN105008950A (en) | Measuring device and measuring method for direction finding and direction uncertainty determination |
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 |