CN107561489B - MLS passive direction finding positioning method based on anomaly detection - Google Patents

MLS passive direction finding positioning method based on anomaly detection Download PDF

Info

Publication number
CN107561489B
CN107561489B CN201710638188.6A CN201710638188A CN107561489B CN 107561489 B CN107561489 B CN 107561489B CN 201710638188 A CN201710638188 A CN 201710638188A CN 107561489 B CN107561489 B CN 107561489B
Authority
CN
China
Prior art keywords
positioning
radiation source
function
result
square
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
Application number
CN201710638188.6A
Other languages
Chinese (zh)
Other versions
CN107561489A (en
Inventor
张怡
梁亚辉
唐成凯
廉保旺
吴建民
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN201710638188.6A priority Critical patent/CN107561489B/en
Publication of CN107561489A publication Critical patent/CN107561489A/en
Application granted granted Critical
Publication of CN107561489B publication Critical patent/CN107561489B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention provides an MLS passive direction finding positioning method based on anomaly detection, which can effectively detect an abnormal radio direction finding result by introducing an anomaly detection function, further provides a basis for correction, and corrects an abnormal point to be within a normal range through a correction function so as to reduce the influence on a positioning result. The invention further improves the convergence rate of the subsequent least square positioning result by processing the abnormal points. The positioning mechanism of the invention depends on the airborne platform, and the positioning point only needs to select two places, thereby improving the positioning speed of the system.

Description

MLS passive direction finding positioning method based on anomaly detection
Technical Field
The invention relates to a passive positioning technology used in the fields of radio monitoring, rescue and the like, in particular to an Anomaly detection mechanism (Anomaly detection mechanism) modified least square MLS (modified least square) passive direction finding positioning algorithm based on nearest neighbor dereference.
Background
The passive positioning technology determines the position of a radiation source by receiving an electromagnetic wave signal radiated by a radiation source through an antenna, is a passive detection technology, is widely applied to the aspects of radio frequency spectrum monitoring, maritime rescue, military detection, navigation and the like, and researchers in various countries in the world carry out a great deal of research on passive positioning technologies of different systems according to actual requirements.
The passive positioning system of the phase difference change rate is researched by Chengyun et al 'single-station passive positioning by utilizing the phase difference change rate', the measurement noise is effectively limited, but the system is sensitive to the base line of an antenna array, and cannot be positioned under the condition of single base line under the influence of phase ambiguity.
Aiming at the passive positioning problem of a single-station external radiation source target, Zhao champion et al provides a Regularization Constraint Total Least Square (RCTLS) positioning algorithm combining arrival angle and time difference information. However, this method needs to solve the problem of clock synchronization of each point in engineering application, which makes the monitoring equipment more complicated.
Among many passive positioning technologies, direction-finding cross positioning is a passive positioning technology developed earlier and researched more mature at present. Dawn et al propose an algorithm that uses sine and cosine values of angles to construct a cross-location equation set that is more accurate than conventional cross-location methods, but is not generic because of the limited distance parameters between the signal source and the observation station.
Shengdan et al have studied the problem of optimal intersection angle of the direction-finding cross-location system under the systematic error, have optimized the sensor configuration, have improved the positioning accuracy, but can't act on the long-distance target, the practical application is limited.
With the rapid development of the unmanned aerial vehicle technology, the passive positioning technology using an airborne platform is a trend in development due to the good electromagnetic environment in the air and the flexible maneuverability of the unmanned aerial vehicle. Some researchers establish an airborne three-dimensional model of a passive positioning system, and verify the effectiveness of a method combining EKF (extended Kalman Filter algorithm) and a control input algorithm on a passive positioning method based on azimuth information in a simulation way, although algorithm fusion improves the positioning precision, the algorithm complexity is increased at the same time, and the airborne rapid positioning is not facilitated.
Disclosure of Invention
Aiming at the problems that the positioning result is abnormal due to the existence of the inherent error of the system, so that the real-time performance and the accuracy of the positioning target are not high, and the positioning is invalid, the method combines the limitation of the existing algorithm, the error interference and the limitation of the technical engineering application, and the applicant uses a multi-rotor unmanned aerial vehicle as a platform and combines the actual requirements, provides a modified least square single-machine passive positioning algorithm based on abnormal detection, obviously improves the positioning precision, shortens the positioning time, and has better stability. The method has a good effect particularly aiming at monitoring of abnormal radio signals between cities, the positioning effect is poor due to the fact that ground monitoring equipment is greatly influenced by multipath, and the position of an abnormal radiation source cannot be found quickly due to the fact that ground monitoring is influenced by terrain. The invention can reduce the influence caused by non-line-of-sight errors to a certain extent through the airborne platform, enables a user to quickly position the position of the radiation source based on the airborne platform, quickly obtains an on-site image through the on-board equipment needing to be reloaded, and provides powerful help for subsequent work decision.
The technical scheme of the invention is as follows:
the MLS passive direction finding positioning method based on anomaly detection is characterized in that: the method comprises the following steps:
step 1: measuring the azimuth angle theta of the carrier relative to the target radiation source at two monitoring points through the carrier sensing equipment at two detection points of the carrier at two different positions in the airspace1,θ2(ii) a And by the formula
Figure BDA0001365324010000021
Performing cross positioning on the target radiation source to obtain a cross positioning result of (x, y), wherein (x)1,y1),(x2,y2) Position coordinates of the two detection points;
step 2: repeating the step 1 for a plurality of times to obtain a plurality of results of cross positioning the target radiation source, and taking the average value of the plurality of results of cross positioning the target radiation source as (X)0,Y0)；
And step 3: establishing an azimuth angle theta of the carrier relative to the target radiation sourceiRelation theta to actual position (X, Y) of target radiation sourcei＝Fi(X,Y)+ωi，
Figure BDA0001365324010000022
ωiIs a direction finding noise error;
and 4, step 4: for non-linear function FiAdvancing linePerforming sexual treatment:
to convert a non-linear function FiIn (X)0,Y0) Is subjected to Taylor series expansion to obtain the relation thetaiLinear function of (c):
Figure BDA0001365324010000031
obtaining an observation equation Z ═ H × W + ω, wherein
Figure BDA0001365324010000032
Figure BDA0001365324010000033
And
Figure BDA0001365324010000034
corresponding to theta obtained when the target radiation source is positioned in the step 2 in a plurality of times in a cross way1And theta2Average value of (d);
and 5: obtaining a least squares estimate of the coordinate error W from the observation equation:
Figure BDA0001365324010000035
then the equation of state transition
Figure BDA0001365324010000036
Obtaining least square estimation of the position of the radiation source at the current k moment by an iterative method
Figure BDA0001365324010000037
Where k is the number of iterations,
Figure BDA0001365324010000038
is the iteration result of the last iteration; obtaining the least square positioning estimation result of the time after the iteration is finished
Figure BDA0001365324010000039
Step 6: repeating the stepsThe steps from 1 to 5 are carried out for N times to obtain the least square positioning estimation result of N times, and the least square estimation result of the nth time is
Figure BDA00013653240100000310
And the n-1 th estimation result is
Figure BDA00013653240100000311
Introducing a canonical detection function H+Sum negative then detect function H-The following were used:
Figure BDA00013653240100000312
Figure BDA00013653240100000313
where std is the detection threshold range std-,std+](ii) a If the detection function is satisfied, the result of the nth least square estimation is
Figure BDA00013653240100000314
Making a correction in which a regular detection function H is satisfied+Then the correction function is:
Figure BDA00013653240100000315
if negative is satisfied, the function H is detected-Then the correction function is:
Figure BDA0001365324010000041
and 7: and averaging the corrected least square positioning estimation results of the N times to obtain a final positioning result (X, Y).
Advantageous effects
According to the method, the abnormal radio direction finding result can be effectively detected by introducing the abnormal detection function based on the theoretical basis of nearest neighbor maximum value removal, so that a basis is provided for correction, and the abnormal point is corrected to be within a normal range through the correction function, so that the influence on the positioning result is reduced. The invention further improves the convergence rate of the subsequent least square positioning result by processing the abnormal points. The positioning mechanism of the invention depends on the airborne platform, and the positioning point only needs to select two places, thereby improving the positioning speed of the system.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic diagram of the single machine direction-finding cross-location of the present invention.
(X, Y) are the actual position coordinates of the radiation source;
(x1,y1) Is the first direction finding point of the unmanned aerial vehicle;
(x2,y2) Is the second direction finding point of the unmanned aerial vehicle;
θ1,θ2is the azimuth angle measured twice;
fig. 2 is a graph of 200 uncorrected positioning errors in an example. Firstly, mean value filtering is carried out, lines with squares are mean value cross positioning errors, and black straight lines are mean values of errors of 200 times; the line with the lower triangle is the uncorrected least squares positioning error, and an outlier appears at point 84.
Fig. 3 is a graph of the positioning error corrected in the example. The same mean value is filtered and input, and the mean value cross difference positioning error is unchanged; as can be seen from the upper right box, the mean value of the corrected least square positioning error is 471m, and the relative positioning error is 2.9%.
Fig. 4 is 400 generic experiments performed by the present invention to verify the validity of the algorithm.
Detailed Description
The following detailed description of embodiments of the invention is intended to be illustrative, and not to be construed as limiting the invention.
The invention aims to provide a modified least square passive direction finding positioning algorithm based on abnormal signal detection, which can identify abnormal data in direction finding data and modify the abnormal data through a modification function, so that a positioning result can be converged more quickly, and a stable positioning result can be obtained. The whole passive direction-finding positioning system consists of an unmanned aerial vehicle unit, a direction-finding antenna unit and a digital signal processing unit. The unmanned aerial vehicle provides the direction finding platform and relies on the airborne sensor to provide local position information for the location, and the radio signal that the radiation source sent is received to the direction finding antenna, sends to digital signal processing unit after through radio frequency processing and carries out the direction finding location.
In this embodiment, based on the MLS passive direction finding positioning algorithm of anomaly detection, a multi-rotor unmanned aerial vehicle is used as a platform, and a detection function model is combined to detect the health of a direction finding angle and correct angle data of a correction function model, so as to obtain a high-precision and stable positioning result. The method specifically comprises the following steps:
step 1: as shown in figure 1, the carrier is at two detection points at different positions in the airspace, and the azimuth angle theta of the carrier relative to a target radiation source at the two monitoring points is measured by the carrier sensing equipment1,θ2(ii) a And by the formula
Figure BDA0001365324010000051
Figure BDA0001365324010000052
Performing cross positioning on the target radiation source to obtain a cross positioning result of (x, y), wherein (x)1,y1),(x2,y2) Position coordinates of two detection points.
Step 2: repeating the step 1 for 100 times to obtain 100 results of cross positioning the target radiation source, and taking the average value of the 100 results of cross positioning the target radiation source as (X)0,Y0)。
And step 3: consider the azimuth angle θiSubject to mean of measurement noise of0, normal distribution with variance δ. Establishing an azimuth angle theta of the carrier relative to the target radiation sourceiRelation theta to actual position (X, Y) of target radiation sourcei＝Fi(X,Y)+ωi，
Figure BDA0001365324010000053
ωiIs the direction finding noise error.
And 4, step 4: due to the existence of measurement errors and noise interference, a plurality of direction lines cannot intersect at one point, and the estimation errors can be reduced by adopting a least square algorithm, so that better position optimal estimation is obtained.
For non-linear function FiCarrying out linearization treatment:
to convert a non-linear function FiIn (X)0,Y0) Performing Taylor series expansion, and reserving the first two terms to obtain the relation thetaiLinear function of (c):
Figure BDA0001365324010000061
obtaining an observation equation Z ═ H × W + ω, wherein
Figure BDA0001365324010000062
Figure BDA0001365324010000063
And
Figure BDA0001365324010000064
corresponding to theta obtained when the target radiation source is positioned in the step 2 in a plurality of times in a cross way1And theta2Average value of (a).
And 5: obtaining a least squares estimate of the coordinate error W from the observation equation:
Figure BDA0001365324010000065
then the equation of state transition
Figure BDA0001365324010000066
Obtaining least square estimation of the position of the radiation source at the current k moment by an iterative method
Figure BDA0001365324010000067
Where k is the number of iterations,
Figure BDA0001365324010000068
is the iteration result of the last iteration; obtaining the least square positioning estimation result of the time after the iteration is finished
Figure BDA0001365324010000069
Step 6: repeating the steps 1 to 5 for 200 times to obtain a least square positioning estimation result of 200 times, wherein the least square estimation result of the nth time is
Figure BDA00013653240100000610
And the n-1 th estimation result is
Figure BDA00013653240100000611
Introducing a canonical detection function H+Sum negative then detect function H-The following were used:
Figure BDA00013653240100000612
Figure BDA00013653240100000613
where std is the detection threshold range std-,std+](ii) a If the detection function is satisfied, the result of the nth least square estimation is
Figure BDA00013653240100000614
Making a correction in which a regular detection function H is satisfied+Then the correction function is:
Figure BDA00013653240100000615
if negative is satisfied, the function H is detected-Then the correction function is:
Figure BDA0001365324010000071
and 7: and averaging the corrected least square positioning estimation results for 200 times to obtain a final positioning result (X, Y).
Performance analysis
Aiming at the problem of the existing least square direction-finding positioning algorithm, the passive direction-finding positioning algorithm based on a single machine system is deeply researched. The theoretical basis of the invention is a passive positioning technology, and the position of a target is obtained by utilizing angle information in a cross positioning mode, as shown in fig. 1. The angle information is obtained by an antenna array by using phase information of radio signals and adopting a direction-finding system of correlation interference. Because the direction-finding error follows normal distribution with the average value of 0, on the basis of obtaining angle measurement data for many times, preliminary data processing is carried out through average value filtering, and the stepping interval is 50. Abnormal values are detected by introducing a nearest neighbor de-maxima detection method, and normal values are used for replacing abnormal points, so that a group of benign positioning results is obtained.
It can be found from the comparison between fig. 2 and fig. 3 that the algorithm provided by the present invention can effectively find and process abnormal data points, thereby greatly improving the positioning accuracy, reducing the original 14.25% relative error to about 2.9%, and simultaneously increasing the stability of the system.
To illustrate the stability of the present invention, a generalised experiment was performed 400 times. The result of fig. 4 shows that the maximum error does not exceed 100m, that is, the stable relative error reaches 0.41%, and it can be seen that the passive positioning algorithm provided by the invention has higher positioning accuracy and better stability, and has better practical application significance.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made in the above embodiments by those of ordinary skill in the art without departing from the principle and spirit of the present invention.

Claims (1)

1. An MLS passive direction finding positioning method based on anomaly detection is characterized in that: the method comprises the following steps:
step 1: measuring the azimuth angle theta of the carrier relative to the target radiation source at two monitoring points through the carrier sensing equipment at two detection points of the carrier at two different positions in the airspace1,θ2(ii) a And by the formula
Figure FDA0001365324000000011
Performing cross positioning on the target radiation source to obtain a cross positioning result of (x, y), wherein (x)1,y1),(x2,y2) Position coordinates of the two detection points;
step 2: repeating the step 1 for a plurality of times to obtain a plurality of results of cross positioning the target radiation source, and taking the average value of the plurality of results of cross positioning the target radiation source as (X)0,Y0)；
And step 3: establishing an azimuth angle theta of the carrier relative to the target radiation sourceiRelation theta to actual position (X, Y) of target radiation sourcei＝Fi(X,Y)+ωi，
Figure FDA0001365324000000012
ωiIs a direction finding noise error;
and 4, step 4: for non-linear function FiCarrying out linearization treatment:
to convert a non-linear function FiIn (X)0,Y0) Is subjected to Taylor series expansion to obtain the relation thetaiLinear function of (c):
Figure FDA0001365324000000013
obtaining an observation equation Z ═ H × W + ω, wherein
Figure FDA0001365324000000014
Figure FDA0001365324000000015
And
Figure FDA0001365324000000016
corresponding to theta obtained when the target radiation source is positioned in the step 2 in a plurality of times in a cross way1And theta2Average value of (d);
and 5: obtaining a least squares estimate of the coordinate error W from the observation equation:
Figure FDA0001365324000000017
then the equation of state transition
Figure FDA0001365324000000018
Obtaining least square estimation of the position of the radiation source at the current k moment by an iterative method
Figure FDA0001365324000000019
Where k is the number of iterations,
Figure FDA00013653240000000110
is the iteration result of the last iteration; obtaining the least square positioning estimation result of the time after the iteration is finished
Figure FDA00013653240000000111
Step 6: repeating the steps 1 to 5 for N times to obtain the least square positioning estimation result of N times, wherein the least square estimation result of the nth time is
Figure FDA0001365324000000021
And the n-1 th estimation result is
Figure FDA0001365324000000022
Introducing a canonical detection function H+Sum negative then detect function H-The following were used:
Figure FDA0001365324000000023
Figure FDA0001365324000000024
where std is the detection threshold range std-,std+](ii) a If the detection function is satisfied, the result of the nth least square estimation is
Figure FDA0001365324000000025
Making a correction in which a regular detection function H is satisfied+Then the correction function is:
Figure FDA0001365324000000026
if negative is satisfied, the function H is detected-Then the correction function is:
Figure FDA0001365324000000027
and 7: and averaging the corrected least square positioning estimation results of the N times to obtain a final positioning result (X, Y).
CN201710638188.6A 2017-07-31 2017-07-31 MLS passive direction finding positioning method based on anomaly detection Expired - Fee Related CN107561489B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710638188.6A CN107561489B (en) 2017-07-31 2017-07-31 MLS passive direction finding positioning method based on anomaly detection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710638188.6A CN107561489B (en) 2017-07-31 2017-07-31 MLS passive direction finding positioning method based on anomaly detection

Publications (2)

Publication Number Publication Date
CN107561489A CN107561489A (en) 2018-01-09
CN107561489B true CN107561489B (en) 2020-05-12

Family

ID=60974761

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710638188.6A Expired - Fee Related CN107561489B (en) 2017-07-31 2017-07-31 MLS passive direction finding positioning method based on anomaly detection

Country Status (1)

Country Link
CN (1) CN107561489B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108535688B (en) * 2018-03-06 2019-12-06 西安大衡天成信息科技有限公司 radiation source positioning method based on radio frequency spectrum monitoring big data processing
CN110929396B (en) * 2019-11-19 2022-06-14 西北工业大学 Electromagnetic situation generation method based on information geometry
CN112954633B (en) * 2021-01-26 2022-01-28 电子科技大学 Parameter constraint-based dual-network architecture indoor positioning method
CN113325406B (en) * 2021-05-24 2023-05-30 哈尔滨工程大学 Regularized constraint weighted least square-based passive positioning method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6407703B1 (en) * 2000-08-07 2002-06-18 Lockheed Martin Corporation Multi-platform geolocation method and system
WO2009142963A3 (en) * 2008-05-23 2010-02-25 Commscope, Inc. Of North Carolina System and method for locating wimax or lte subscriber stations
CN105572635A (en) * 2016-01-25 2016-05-11 西安电子科技大学 Single-station passive quick positioning method based on least square method
CN106597364A (en) * 2016-11-18 2017-04-26 烟台职业学院 Target radiation source initial position estimation method for single-antenna single-station passive positioning

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6407703B1 (en) * 2000-08-07 2002-06-18 Lockheed Martin Corporation Multi-platform geolocation method and system
WO2009142963A3 (en) * 2008-05-23 2010-02-25 Commscope, Inc. Of North Carolina System and method for locating wimax or lte subscriber stations
CN105572635A (en) * 2016-01-25 2016-05-11 西安电子科技大学 Single-station passive quick positioning method based on least square method
CN106597364A (en) * 2016-11-18 2017-04-26 烟台职业学院 Target radiation source initial position estimation method for single-antenna single-station passive positioning

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
《基于正则化约束总体最小二乘的单站DOA-TDOA 无源定位算法》;赵拥军;《电子与信息学报》;20160930;第38卷(第9期);2336-2342 *
《基于递推总体最小二乘的机载单站无源定位算法》;吴昊;《空军工程大学学报（自然科学版）》;20130228;第14卷(第1期);62-65 *

Also Published As

Publication number Publication date
CN107561489A (en) 2018-01-09

Similar Documents

Publication Publication Date Title
CN107561489B (en) MLS passive direction finding positioning method based on anomaly detection
Zhou et al. An exact maximum likelihood registration algorithm for data fusion
CN110058205B (en) Warning radar system error correction method based on iterative closest point algorithm
CN109917333B (en) Passive positioning method integrating AOA observed quantity and TDOA observed quantity
CN103047982B (en) Adaptive target tracking method based on angle information
CN104869541A (en) Indoor positioning tracking method
CN108761387B (en) Double-station time difference and frequency difference combined positioning method for fixed radiation source
CN111536967A (en) EKF-based multi-sensor fusion greenhouse inspection robot tracking method
CN113342059B (en) Multi-unmanned aerial vehicle tracking mobile radiation source method based on position and speed errors
CN112346104A (en) Unmanned aerial vehicle information fusion positioning method
CN104363649A (en) UKF (unscented Kalman filter)-based WSN (wireless sensor network) node location method with constraint conditions
Fang et al. Robust node position estimation algorithms for wireless sensor networks based on improved adaptive Kalman filters
CN114137562B (en) Multi-target tracking method based on improved global nearest neighbor
CN114509069B (en) Indoor navigation positioning system based on Bluetooth AOA and IMU fusion
CN110187337B (en) LS and NEU-ECEF space-time registration-based high maneuvering target tracking method and system
CN103499809B (en) A kind of Pure orientation double computer cooperation target following location path planing method
CN112240957B (en) Method for correcting amplitude-phase characteristics of antenna in satellite navigation interference direction finding
CN110779544B (en) Double-task deep matching method for self-positioning and target positioning of multiple unmanned aerial vehicles
CN109375159B (en) Pure orientation weighting constraint total least square positioning method
CN112333634A (en) Hybrid node positioning method based on UAV
CN115792800A (en) Grid search-based double-station three-dimensional cross positioning method
CN115633306A (en) Positioning correction method and device for multi-region UWB (ultra Wide band) signals
CN115307644A (en) Three-dimensional positioning model based on UWB
CN113933798A (en) Global sensor system error partition registration algorithm based on similarity principle
CN109884582B (en) Method for rapidly determining three-dimensional coordinates of target by utilizing one-dimensional direction finding

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20200512

Termination date: 20210731