CN102176010B - Multiple-output-single-input-based passive radar positioning tracking system and positioning tracking method - Google Patents

Abstract

The invention discloses a multiple-output-single-input-based passive radar positioning tracking system and a multiple-output-single-input-based passive radar positioning tracking method, which mainly aims to solve the problem of low target positioning tracking accuracy of a passive radar under the conditions of external radiation source distribution non-uniformity and relatively greater distance measurement errors. The system consists of a reference antenna, a signal antenna, a receiver, a signal processor, a constant false alarm detection and flight path processing module and a positioning tracking module, wherein the signal antenna adopts a horizontally polarized antenna, and receives echoes and a multipath signal from a target; the received signal is filtered and amplified by the receiver; AD conversion, digital down-conversion, clutter cancellation and a range-doppler two-dimensional correlation operation are performed on an analogue signal by the signal processor, and the target on an original flight path or a new flight path is detected from an operation result; and finally multi-station information pairing, target positioning tracking and display are performed. By the system and the method, the target can be accurately positioned and tracked under the conditions of external radiation source distribution non-uniformity and relatively greater distance measurement errors.

Description

Passive radar locating and tracking system and positioning and tracing method based on multiple illuminators and single receiver

Technical field

The invention belongs to the Radar Technology field, relate to radar fix, a kind of external radiation source radar fix system and method based on " multiple illuminators and single receiver " system specifically, can be under the environment that a plurality of irradiation source stochastic distribution and strong direct wave and multipath disturb, accurately being positioned of realize target followed the tracks of.

Technical background

Along with the development of modern war technology, the threat of the aspect such as radar is faced with that target is stealthy, antiradiation missile, low-level penetration and electronic interferences.How making radar system finish detection to unfriendly target, location and follow the tracks of under strongly disturbing background, and avoid simultaneously killing and wounding of enemy's anti-radiation weapon, is effectively protection oneself, strike enemy's key.Passive Radar System is because it has not detectable property, the characteristics such as simple in structure and cheap are subject to extensive concern.Passive Radar System itself does not need to transmit, but the electromagnetic signal of the non-cooperating type radiation source of receiving target reflection passively positions and follows the tracks of target.

According to the difference of detected object, passive radar is divided into and utilizes self radiation of detected target to survey and follow the tracks of, and utilizes the electromagnetic wave of external radiation source emission to survey and follow the tracks of two large classes.Utilize the direct electromagnetic radiation energy of target of the passive radar detection of external radiation source emitting electromagnetic wave itself, but at passive radar in when work, by the non-collaborative radiation source of antenna reception from the outside, it is third-party direct wave, and the reflection wave or the scattering wave that form after this foreign radiation sources irradiation target, utilize its Doppler shift that carries, multistation to receive the information such as mistiming of signal and the angle of arrival, extract target information and eliminate garbage and interference, thereby finish detection, location and tracking to target.Yet because the characteristic of uncontrollable, the unpredictable and non-cooperation in external radiation source, the detection performance of external radiation source radar is often poor than the normal radar system, therefore needs to adopt more complicated signal processing means, improves whole system and detects and tracking performance.

At present both at home and abroad the radar system based on the external radiation source of research mainly contains following two kinds: a kind ofly be based on single irradiation source and position tracking, it is to utilize two antennas to receive respectively the direct wave of irradiation source and the echoed signal of target, then disappear mutually, mate and detect, the method not only investigative range is less, and bearing accuracy is lower; Another kind is the target measurement information that merges a plurality of cell sites, it is the multiple illuminators and single receiver pattern, can utilize at receiving end the echo information establishment target RCS flicker of different cell sites, thereby the more effective target detection of finishing, improve bearing accuracy, yet distribute irregular and range measurement accuracy when low at irradiation source, this method will cause bearing accuracy to worsen rapidly so that be difficult to satisfy practical application.

Summary of the invention

The object of the invention is to overcome the deficiency of above-mentioned prior art, a kind of passive radar locating and tracking system and positioning and tracing method based on multiple illuminators and single receiver proposed, with under abominable cloth station condition and low range measurement accuracy, range information measurement fusion by repeatedly carries out filter tracking, improve target location accuracy, satisfy the requirement of practical application.

For achieving the above object, the passive radar locating and tracking system based on multiple illuminators and single receiver provided by the invention comprises:

Reference arm and signal branch, reference arm comprise reference antenna and receiver A, are used for receiving respectively the direct wave D from a plurality of illuminators of opportunities _iSignal branch comprises receiver B, is used for the echoed signal of receiving target; It is characterized in that signal branch also comprises:

Signal antenna receives the FM stations signal R by the moving target reflection _iAnd multipath signal Z _i

Signal processor is used for the analog if signal of receiver A and receiver B output is transformed to baseband digital domain, obtains respectively digital baseband direct-path signal ref _iWith digital baseband echoed signal sig _i+ Z _i, and utilize this direct-path signal ref _iWith echoed signal sig _i+ Z _iCarry out successively adaptive clutter cancellation, matched filtering and the computing of distance-Doppler two-dimensional correlation, the operation result y of this two-dimensional correlation computing _i(f _d, t) output to CFAR detection and flight path processing module;

CFAR detection and flight path processing module are used for the distance-Doppler two-dimensional correlation operation result y to signal processor output _i(f _d, t) carry out CFAR detection and flight path and process, obtain target range Doppler flight path information; And flight path information carried out correction-compensation, and the result after the compensation is exported to the location tracking module, described flight path is processed, and comprises track initiation, and the some mark is related with flight path, the flight path extinction;

The location tracking module, the targetpath information of utilizing CFAR detection and flight path processing module to obtain, finish successively multistation information pairing, target position information resolves and tracking initiation, and carries out track and localization by the EKF method, obtains the accurate positional information of target

Described passive radar locating and tracking system, wherein signal processor comprises:

The analog to digital conversion submodule is used for the analog if signal of receiver output is carried out digitizing, is transferred to Digital Down Convert and processes submodule;

Submodule is processed in Digital Down Convert, is used for the intermediate-freuqncy signal that digitizing is later and is transformed into base band, and direct-path signal and echoed signal are transferred to mutually slake distance-Doppler two-dimensional correlation submodule of clutter;

Clutter phase slake distance-Doppler two-dimensional correlation submodule is used for that the data that signal antenna receives are carried out clutter and disappears mutually, to eliminate direct wave and multipath signal Z wherein _i, and utilize the as a result sig that disappears mutually _iWith direct wave ref _iCarry out the computing of distance-Doppler two-dimensional correlation, the operation result y of this two-dimensional correlation computing _i(f _d, t) send CFAR detection and flight path processing module to.

Described CFAR detection and flight path processing module comprise:

Point mark and track association submodule are with the as a result y that exports after the distance-Doppler two-dimensional correlation _i(f _d, it is related that the point that surpasses given detection threshold in t) and original targetpath carry out, and is in the correlation threshold of setting if change the time, then be judged as the target that detects on original flight path, and the result is transported in Doppler's correction-compensation module, otherwise, impact point is delivered to the track initiation submodule;

Track initiation submodule, the impact point that adopts sliding window method that a mark and track association submodule are sent here carry out track initiation to be judged, if judgement have new Track forming then the export target flight path to Doppler's correction-compensation submodule, otherwise, impact point done gives up processing;

Flight path extinction submodule, which is to need to continue effective flight path of following the tracks of be used for to judge the flight path set up, which is the flight path that finishes of lose objects and needing; When certain bar flight path lose continuously count above the user set cancel thresholding the time, think that target exceeds investigative range, flight path is cancelled; Otherwise flight path is kept;

Doppler's correction-compensation submodule, utilize continuity and the approximate constant characteristic of short time internal object speed of range information, utilize the range Doppler information of previous moment to calculate disappearance distance value constantly, to remedy the of short duration interruption of target range Doppler flight path, the flight path of having set up is carried out correction-compensation.

Described location tracking module comprises:

Multistation metrical information pairing submodule utilizes the proximity of doppler information that revised distance-Doppler flight path information is matched, and finishes resolving of target location, the result is flowed to the location follow the tracks of submodule;

Submodule is followed the tracks of in the location, is used for finishing the location tracking of target, is about to each measurement data that obtains and inputs to filtering algorithm, estimates the positional information of target , and the form with a mark shows in tracking results figure.

For achieving the above object, the passive radar positioning and tracing method based on multiple illuminators and single receiver provided by the invention may further comprise the steps:

(1) points to respectively each frequency modulation broadcasting base station with reference to antenna, be used for receiving direct-path signal D _iSignal antenna is pointed to the target pattern direction, be used for receiving the FM stations signal R of moving target reflection _i, FM stations direct wave and multipath signal Z _i

(2) signal that receives with reference to antenna and signal antenna is respectively through obtaining analog if signal behind receiver A and receiver B mixing and the filter and amplification, and gives signal processor with this analog if signal;

(3) in signal processor, the simulating signal of receiver A and receiver B transmission is carried out analog to digital conversion and Digital Down Convert processing, obtain digital baseband direct-path signal ref _iWith echoed signal sig _i+ Z _i, and with digital baseband direct-path signal ref _iWith echoed signal sig _i+ Z _iCarry out adaptive clutter cancellation, the direct wave of filtering FM stations and multipath signal Z _i, the echoed signal sig after obtaining to purify _iThen use digital baseband direct-path signal ref _iWith the echoed signal sig after the purification _iDo passive relevant processing according to following formula:

$y_i(f_d,n)={{\mathrm{ref}}_i}^{*}\left(n\right)\&CenterDot;{\mathrm{sig}}_i(n-\mathrm{\&tau;})e^{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000044815140000051.png,HDA0000044815140000052.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{f}_d(n-\mathrm{\&tau;})}$

Wherein, y _i(f _d, the range Doppler information that n) obtains for passive relevant processing, f _dThe expression Doppler shift, n represents time series, and τ represents time delay, and * represents conjugate transpose, sig _iThe complex envelope of (n-τ) expression echoed signal;

(4) the doppler information y that adjusts the distance _i(f _d, n) carry out the target double threshold and detect track association, and according to following formula flight path information is done correction-compensation:

ρ(k+1)＝ρ(k)+f _d(k)λT

Wherein ρ (k+1) for replenish the k+1 obtain constantly distance and, ρ (k) be known k distance constantly and, f _d(k) be k doppler values constantly, λ is the emission wavelength in radio station, and T is the some mark time interval;

(5) the multistation target information is matched, judges whether to detect target:

(5a) a plurality of platforms of simulation calculation are observed same target, obtain Doppler and match threshold delta f _dWith distance pairing threshold delta ρ;

The doppler values f that (5b) i cell site is recorded _DiThe doppler values f that records with j cell site _DjSubtract each other, if Doppler difference is matched threshold delta f less than Doppler _d, i.e. f _Di-f _Dj≤ Δ f _d, and doppler values f _Di, f _DjNon-vanishing, then carry out (5c);

The distance value ρ that (5c) i cell site is recorded _iThe distance value ρ that records with j cell site _jSubtract each other, if apart from difference less than distance pairing threshold delta ρ, i.e. ρ _i-ρ _j≤ Δ ρ, and described distance value ρ _iAnd ρ _jNon-vanishing, then judge to detect target, and execution in step (5d);

(5d) utilize the range information of target, resolve and obtain target position information (x, y, z), and make x _k=[x, y, z, 0,0,0] ^T, k=1,2, LM ₀, M ₀For the target that detects is counted;

(6) with front 9 the result of calculation x that calculate in the step (5) ₁, x ₂..., x ₉Average, obtain Initial state estimation

(7) with Initial state estimation

As the filtering initial value, carry out EKF, obtain target position information , and in three-dimensional coordinate figure, show this target location, obtain targetpath.

The present invention has the following advantages:

(1) the present invention can obtain purer direct wave owing to adopting horizontally-polarized antenna to point to respectively each cell site and target direction, reduces simultaneously the multipath clutter in the echoed signal; Because the present invention does not adopt antenna array, system's operand is little in addition.

(2) bearing accuracy is high.The present invention is owing to adopting the adaptive clutter cancellation algorithm, can the establishment multipath clutter on the impact of measurement result; Because the present invention fully utilizes a plurality of cell sites co-located, accurately the information of measurement target has obviously improved the target locating precision simultaneously.

(3) cell site's distribution is not required.Because the present invention introduced expanded Kalman filtration algorithm, compare the method for directly resolving, the bearing accuracy when having improved cell site's non-uniform Distribution, and under the prerequisite that guarantees high position precision, reduced requirement to the cloth station, so feasibility is higher.

(4) the present invention can effectively suppress target false-alarm point owing to adopting the target double threshold to detect Data Association, and the target when reducing track association is failed to report, and accurately detects the range Doppler flight path of target.

Description of drawings

Fig. 1 is system chart of the present invention;

Fig. 2 is positioning and tracing method general flow chart of the present invention;

Fig. 3 is that the target double threshold detects track association processing sub-process figure among the present invention;

Fig. 4 is that target double threshold of the present invention detects track association result figure;

Fig. 5 is target range Doppler's track plot that the present invention obtains with the pairing of three stations;

Fig. 6 is target range Doppler's track plot that the present invention obtains with two optimum station pairings;

Fig. 7 is the surface level two-dimensional coordinate figure that the present invention adopts " three one receipts " mode tracking target to obtain;

Fig. 8 is the surface level two-dimensional coordinate figure that the present invention adopts " two one receipts " mode tracking target to obtain;

Fig. 9 is the Target Location Error distribution plan that the present invention traces into.

Embodiment

Describe content of the present invention and effect in detail below in conjunction with accompanying drawing.

With reference to Fig. 1, radar system of the present invention mainly is comprised of reference antenna, signal antenna, receiver A, receiver B, signal processor, CFAR detection and flight path processing module and location tracking module.Wherein, reference antenna connects receiver A, and signal antenna connects receiver B, and the signal processor input end connects receiver A and receiver B, output terminal connects CFAR detection and track association module, and CFAR detection is connected the location tracking module with track association module output terminal.

Described reference antenna and signal antenna all adopt the horizontal polarization oriented antenna, and the beam angle of individual antenna is 22.5 °, and the reception frequency range is 87-110MHz; The Main Function of reference antenna is to utilize main lobe to point to the f-m broadcast station cell site in system work process, obtains the direct-path signal of f-m broadcast station; The signal antenna main lobe points to the target pattern direction, and receiving target echoed signal and multipath disturb.

Described receiver A and receiver B, the echoed signal that the direct-path signal that respectively reference antenna is received and signal antenna receive carries out mixing and filter and amplification obtains analog if signal, and the signal processor that makes the rear end can be operated in the best range of linearity, and the result gives signal processor.

Described signal processor carries out the A/D conversion with the analog if signal of receiver A and receiver B output and Digital Down Convert is processed, and obtains respectively digital baseband direct-path signal ref _iWith digital baseband echoed signal sig _i+ Z _i, and utilize this direct-path signal and echoed signal to carry out successively adaptive clutter cancellation and the computing of distance-Doppler two-dimensional correlation, it mainly is comprised of following submodule:

The analog to digital conversion submodule, the analog if signal that receiver is exported carries out digitizing, and system adopts the ICS554 data collecting card, but is not limited to this capture card, and sampling rate is 100MHz, and sends the data that sampling obtains to Digital Down Convert processing submodule;

Submodule is processed in Digital Down Convert, adopt the Digital Down Convert function of ICS554 capture card, but be not limited to this product, the intermediate-freuqncy signal that digitizing is later is down-converted to 250Khz, and the direct-path signal after the down coversion and echoed signal are transferred to mutually slake distance-Doppler two-dimensional correlation submodule of clutter;

Clutter phase slake distance-Doppler two-dimensional correlation submodule, the data that signal antenna is received are carried out clutter and are disappeared mutually, to eliminate direct wave and multipath signal Z wherein _i, and utilize the as a result sig that disappears mutually _iWith direct wave ref _iCarry out the computing of distance-Doppler two-dimensional correlation, the operation result y of this two-dimensional correlation computing _i(f _d, t) send CFAR detection and flight path processing module to.

Described location tracking module is used for the target of Doppler's correction-compensation is positioned tracking; At first carry out the pairing of multistation information, and the positional information of solving target, obtain the EKF initial value, then carry out EKF, obtain the positional information of target

, realize that accurately follow the tracks of the location, and the form with a mark shows in tracking results figure.

Described CFAR detection and flight path processing module are used for the two-dimensional correlation signal of signal processor is carried out CFAR detection and flight path processing, and to obtain targetpath information, it mainly is comprised of following submodule:

Point mark and track association submodule, the point that the result who is used for exporting after the distance-Doppler two-dimensional correlation surpasses given detection threshold carries out related with original targetpath, if this point is in the correlation threshold of setting, then be judged as the target that detects on original flight path, and the result is transported in Doppler's correction-compensation module, otherwise, impact point is delivered to the track initiation submodule;

The track initiation submodule, the impact point that adopts sliding window method that a mark and track association submodule are transported carries out the track initiation judgement, if judge have new Track forming then the export target flight path to Doppler's correction-compensation submodule, otherwise, impact point done gives up processing;

Flight path extinction submodule, which is to need to continue effective flight path of following the tracks of be used for to judge the flight path set up, which is the flight path that finishes of lose objects and needing; When certain bar flight path lose continuously count above the user set cancel thresholding the time, think that target exceeds investigative range, flight path is cancelled; Otherwise flight path is kept;

Doppler's correction-compensation submodule, continuity and the approximate constant characteristic of short time internal object speed according to range information, extrapolation disappearance distance value constantly to remedy the of short duration interruption of target range Doppler flight path, carries out correction-compensation to the flight path of having set up.

In the system work process, reference antenna receives direct-path signal, and sends direct-path signal to receiver A, signal antenna receiving target echoed signal, and send target echo signal to receiver B; Receiver A and receiver B carry out filter amplifying processing to direct-path signal and echoed signal respectively, and send the filter amplifying processing result to signal processor; The simulating signal that signal processor sends receiver is finished the A/D conversion, Digital Down Convert is processed, adaptive clutter cancellation and the computing of range Doppler two-dimensional correlation, and send range Doppler two-dimensional correlation operation result to CFAR detection and flight path processing module, finish the detection of target and flight path and process, and send the target that detects to the location tracking module, the line trace filtering of going forward side by side of solving target position, obtain target position information, and in surface level two-dimensional coordinate figure, show.

With reference to Fig. 2, passive radar positioning and tracing method of the present invention, implementation step is as follows:

Step 1 is adjusted each reference antenna direction, makes main lobe point to respectively each cell site, broadcasting station, to obtain pure as far as possible direct-path signal D _i, the signal antenna course-and-bearing that aims at the mark, the multipath signal Z that receives to reduce signal antenna _i, the synchronous signal antenna reception is by the moving target reflection FM stations R that transmits _i

Step 2, the signal that reference antenna and signal antenna receive obtain analog if signal after passing through receiver A and receiver B mixing and filter and amplification respectively, and give signal processor with this analog if signal.

Step 3, the signal that successively two receivers is sent in signal processor carry out analog to digital conversion and Digital Down Convert DDC processes, and obtains digital baseband direct-path signal ref _iWith echoed signal sig _i+ Z _i

Step 4, adopting exponent number is 100 the variable step LMS algorithm that disappears mutually, to the direct-path signal ref that obtains _iWith echoed signal sig _i+ Z _iCarry out the clutter processing that disappears mutually, the direct wave of filtering FM stations and multipath signal Z _i, the echoed signal sig after obtaining to purify _i

This step is implemented as follows:

(4a) setting disappear the mutually dimension of device of variable step LMS is 100, the weight vector initial value

Counter n=1, adaptive clutter cancellation device weight vector initial value w (0) dimension is 100 * 1;

The adaptive clutter cancellation device that (4b) calculates the n step is exported outp (n): outp (n)=w (n-1) ^TR _i(n), w (n-1) wherein ^TBe the transposition that n-1 goes on foot the sef-adapting filter weight vector that obtains, R _i(n)=[ref _i(n), ref _i(n+1), L, ref _i(n+K)] ^T, ref _i(n) expression direct-path signal ref _iIn n data;

(4c) calculate the n adaptive clutter cancellation device output error e (n) in step: e (n)=[sig _i(n)+Z _i(n)]-outp (n), wherein sig _i(n)+Z _i(n) expression comprises the target echo signal sig of multipath clutter _i+ Z _iIn n data;

(4d) upgrade step factor μ (n):

μ wherein ₀Be the step factor initial value of setting, ε ₀Be the step factor adjustment factor, * represents to get conjugation;

(4e) upgrade adaptive clutter cancellation device weight vector w (n): w (n)=w (n-1)+μ (n) e (n) R _i(n);

(4f) whether judge n greater than L-100, if n greater than L-100 filtering finish, otherwise counter n adds 1, and forwards step (4b) to and continue operation, the direct-path signal ref of L for receiving _iWith the target echo signal sig that comprises multipath clutter _i+ Z _iData length.

Step 5 is used direct-path signal ref _iAnd the target echo signal sig after purifying _iDo passive relevant processing according to following formula:

$y_i(f_d,n)={{\mathrm{ref}}_i}^{*}\left(n\right)\&CenterDot;{\mathrm{sig}}_i(n-\mathrm{\&tau;}){\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="HDA0000044815140000051.png,HDA0000044815140000052.png"\; state="\{\{state\}\}">e</figure-callout>}}^j$

Wherein, y _i(f _d, the range Doppler information that n) obtains for passive relevant processing, f _dThe expression Doppler shift, n represents time series, and τ represents time delay, and * represents conjugate transpose, sig _iThe complex envelope of (n-τ) expression echoed signal.

Step 6 is to the range Doppler information y that obtains in the step 5 _i(f _d, n) carry out the target double threshold and detect track association, obtain the less range Doppler flight path information of false-alarm point.

With reference to shown in Figure 3, this step is implemented as follows:

(6a) establishing i is cell site's numbering, and initial value is 1;

(6b) choose as a result y after the distance-Doppler two-dimensional correlation of i cell site _i(f _d, t), and make its maximal value be: m=max (y _i(f _d, t)), if m is greater than given detection threshold d ₁, then change step (6c) over to, otherwise change step (6f) over to;

(6c) m and former targetpath are carried out related, if the difference of the predicted value of m and former flight path is less than correlation threshold d ₂, judgement detects target on the former flight path, the number S zero setting of point that simultaneously will this former track loss, and forward step (6f) to; Otherwise, S is added 1, and changes step (6d) over to;

(6d) adopt sliding window method to detect new flight path, namely use 4/6 logic initial, any four frames are in correlation threshold d in six continuous frame data ₂When interior, then thinking has new track initiation, and preserves the m value; Otherwise change step (6f) over to;

(6e) flight path is withered away and is judged: set flight path extinction thresholding and be T _cIf, the number S of the point of former track loss 〉=T _cThe time, think that target exceeds investigative range, flight path is cancelled; Otherwise flight path is kept; Change step (6f) over to;

(6f) cell site is numbered i and add 1, establishing N is total number of cell site, if i≤N then change step (6b) over to finishes otherwise detect.

Step 7, the targetpath that step 6 detects is done the range Doppler correction-compensation by following formula:

ρ(k+1)＝ρ(k)+Vρ＝ρ(k)+f _d(k)λT

Wherein ρ (k+1) for replenish the k+1 obtain constantly distance and, ρ (k) be known k distance constantly and, f _d(k) be k doppler values constantly, λ is the emission wavelength in radio station, and T is the some mark time interval.

Step 8 is matched to the multistation target information, judges whether to detect target.

This step is implemented as follows:

(8a) a plurality of platforms of simulation calculation are observed same target, obtain Doppler and match threshold delta f _dWith distance pairing threshold delta ρ;

The doppler values f that (8b) i cell site is recorded _DiThe doppler values f that records with j cell site _DjSubtract each other, if Doppler difference is matched threshold delta f less than Doppler _d, i.e. f _Di-f _Dj≤ Δ f _d, and doppler values f _Di, f _DjNon-vanishing, execution in step (8c) then;

The distance value ρ that (8c) i cell site is recorded _iThe distance value ρ that records with j cell site _jSubtract each other, if apart from difference less than distance pairing threshold delta ρ, i.e. ρ _i-ρ _j≤ Δ ρ, and described distance value ρ _iAnd ρ _jNon-vanishing, then judge to detect target, and execution in step (8d);

(8d) utilize the range information of target, solving equations:

$\left\{\begin{array}{c}{r}_{\mathrm{Ti}}=\sqrt{{(x-x_{\mathrm{Ti}})}^2+{(y-y_{\mathrm{Ti}})}^2}\\ r_R=\sqrt{{(x-x_R)}^2+{(y-y_R)}^2}\\ {\mathrm{\&rho;}}_i=r_{\mathrm{Ti}}+r_R-L_i\end{array}\right.$

(x wherein _R, y _R, z _R) expression receiving station coordinate, (x _Ti, y _Ti, z _Ti) expression i cell site coordinate, r _RThe distance of expression target and receiving station, r _TiThe expression target is the distance of cell site therewith, L _iExpression receiving station is the parallax range between the cell site therewith; Resolve and obtain target location coordinate (x, y, z), and make x _k=[x, y, z, 0,0,0] ^T, k=1,2, L M ₀, M ₀For the target that detects is counted.

Step 9 is with front 9 the result of calculation x in the step 8 ₁, x ₂..., x ₉Average, obtain Initial state estimation

Step 10 is with Initial state estimation

As the filtering initial value, carry out EKF, obtain target position information , and in three-dimensional coordinate figure, show this target location, obtain targetpath.

This step is implemented as follows:

(10a) make counter k=1;

(10b) calculate the k observing matrix in step:

$H_k=\left[\begin{array}{cccccc}\frac{\hat{x}-x_R}{r_R}+\frac{\hat{x}-x_1}{r_1}& \frac{\hat{y}-y_R}{r_R}+\frac{\hat{y}-y_1}{r_1}& \frac{\hat{z}-z_R}{r_R}+\frac{\hat{z}-z_1}{{\mathrm{<figure-callout\; id="1"\; label="r"\; filenames="HDA0000044815140000031.png,HDA0000044815140000052.png"\; state="\{\{state\}\}">r</figure-callout>}}_1}& 0& 0& 0\\ \frac{\hat{x}-x_R}{r_R}+\frac{\hat{x}-x_2}{r_2}& \frac{\hat{y}-y_R}{r_R}+\frac{\hat{y}-y_2}{r_2}& \frac{\hat{z}-z_R}{r_R}+\frac{\hat{z}-z_3}{{\mathrm{<figure-callout\; id="3"\; label="r"\; filenames="HDA0000044815140000061.png"\; state="\{\{state\}\}">r</figure-callout>}}_3}& 0& 0& 0\\ M& M& M& M& M& M\\ \frac{\hat{x}-x_R}{r_R}+\frac{\hat{x}-x_N}{r_N}& \frac{\hat{y}-y_R}{r_R}+\frac{\hat{y}-y_N}{r_N}& \frac{\hat{z}-z_R}{r_R}+\frac{\hat{z}-z_{\mathrm{<figure-callout\; id="0"\; label="N\; r\; N"\; filenames="HDA0000044815140000031.png,HDA0000044815140000051.png"\; state="\{\{state\}\}">N</figure-callout>}}}{r_N}& 0& 0& 0\end{array}\right]$

Wherein

Be k-1 step filtering output

In coordinates of targets, N is total number of cell site, (x ₁, y ₁, z ₁) (x ₂, y ₂, z ₂) L (x _N, y _N, z _N) represent respectively the coordinate of each cell site, r ₁, r ₂, L, r _NThe distance that represents respectively target and each cell site;

(10c) calculate the predicted value that k step filtering is exported: Wherein

I is 3 * 3 for unit matrix, and T is the parameter acquiring time interval,

Be the filtering output in k-1 step;

(10d) calculate the predicted value that k goes on foot the error covariance matrix:

P wherein _K-1Be the filtering covariance matrix that obtains in the k-1 step, Q _kBe the process noise covariance matrix;

(10e) calculate the residual error covariance matrix:

R wherein _kKnown observation noise covariance matrix;

(10f) calculate the output of k step wave filter:

Z wherein _k=[ρ ₁(k), ρ ₂(k), L, ρ _N(k)] ^T, N is total number of cell site;

(10g) upgrade the filtering covariance matrix:

(10h) mobility value of calculating target:

The mobility value J of the target that (10i) obtains according to step (10h) and the mobility value J of setting ₀Magnitude relationship modify steps (10d) in Q _kIf, the mobility value J＜J of target ₀, then adjudicate target do not have motor-driven, observation noise Q _k=Q ₀If the mobility value J 〉=J of target ₀, then target have motor-driven, Q _k=α * Q ₀, wherein α is greater than 10 ⁴, Q ₀Be known process noise covariance matrix, the mobility value J that wherein sets ₀Be fixing empirical value 500;

(10j) wave filter is exported

In three-dimensional coordinate figure, show, obtain targetpath; And judge that counter k is whether greater than the target that the detects M that counts ₀If, k＜M ₀, then k is added 1, and goes to step, 10b); Otherwise filtering finishes.

Effect of the present invention can further specify by following experimental result:

1) experiment condition:

Set up coordinate system according to receiving station's position relationship, establish receiver and be positioned at true origin, the surface level coordinate of each cell site is respectively T1 (32.5-33.2) km, T2 (0 ,-45.2) km, T3 (0 ,-50.5) km.In this field trial, the main lobe of receiving antenna points to 45 degree by east, is mainly boarded a plane in this direction course line and carries out detecting and tracking; The tracking that positions of " two one receipts " pattern and " three one receipts " pattern is adopted in experiment.

2) experiment effect:

This experiment obtains the target double threshold and detects the track association result as shown in Figure 4, wherein horizontal ordinate represents Doppler, and unit is Hz, and Doppler is for just representing that target is for flying nearly radar receiving station, Doppler then is expressed as the receiving station of leaving away for negative, and ordinate represents range unit; As can be seen from Figure 4 this experiment detects two targets, and target one is away from receiving station, is intercepted and captured by the A1 point, tracks the B1 position always; Target two is near receiving station, is intercepted and captured by the A2 point, tracks the B2 position always.

Target range Doppler's track plot that this experiment pairing obtains as shown in Figure 5, as shown in Figure 5, the maximum target of " three one receipts " passive location distance depends on that three stop spacings are from the minimum value of measurement.But owing to wherein two stand in the far field and can detect preferably target, so we can use in the far field localization method of " two one receipts ".

Fig. 6 is that the pairing of range Doppler information is carried out at two best stations of detection performance among use Fig. 5, and as can be seen from Figure 6, the pairing of two stations can utilize metrical information more fully, but increases the orientation distance of target.

The surface level two-dimensional coordinate figure that this experiment utilizes " three one receipts " mode tracking gained target as shown in Figure 7, wherein horizontal ordinate represents the x axle, ordinate represents the y axle, unit is km.T1, T2, T3 represent the position of cell site, and R represents the position of receiving station; Solid line represents the flight path that Passive Tracking obtains target among Fig. 7, and the target position information of the some mark that disperses for directly resolving out.Occur two targets among Fig. 7, target one traces into the E1 point near receiving station from S1 point position; Target two traces into the E2 point away from receiving station from the S2 point; " three the one receipts " pattern of utilization can trace into the target apart from receiving station 100km place.

The surface level two-dimensional coordinate figure of this experiment utilization " two one receipts " mode tracking gained target than Fig. 7, utilizes " two one receipts " method tracking range farther as shown in Figure 8, can trace into the target apart from receiving station 200km place.

Fig. 9 is the Target Location Error distribution plan that the present invention traces into, and horizontal ordinate is the distance of target and receiver, and unit is km, and ordinate is error, and unit is km; Can find out from Fig. 9 aircraft pursuit course, the error that the method for directly resolving positions is several kilometers magnitude, and at a distance even reach nearly 5 kilometers, Passive Tracking filtering has then improved precision greatly, error can be controlled in the 0.5km, on average about 300m; The standard deviation of the error of Passive Tracking filtering is 0.3714km, and the error to standard deviation that directly resolves is 0.9411km.Therefore, in cell site's non-uniform Distribution situation, with respect to the method for directly resolving, Passive Tracking has improved the precision of location significantly, and the target location accuracy in far field improves particularly remarkable.

Claims (8)

1. the passive radar locating and tracking system based on multiple illuminators and single receiver comprises reference arm and signal branch, and reference arm comprises reference antenna and receiver A, is used for receiving respectively the direct wave D from a plurality of illuminators of opportunities _iSignal branch comprises receiver B, is used for the echoed signal of receiving target; It is characterized in that signal branch also comprises:

Signal antenna receives the FM stations signal R by the target reflection _iAnd multipath signal Z _i

Signal processor is used for the analog if signal of receiver A and receiver B output is transformed to baseband digital domain, obtains respectively digital baseband direct-path signal ref _iWith digital baseband echoed signal sig _i+ Z _i, and utilize this direct-path signal ref _iWith echoed signal sig _i+ Z _iCarry out successively adaptive clutter cancellation, matched filtering and the computing of distance-Doppler two-dimensional correlation, the operation result y of this two-dimensional correlation computing _i(f _d, t) output to CFAR detection and flight path processing module, f _dThe expression doppler values;

CFAR detection and flight path processing module are used for the distance-Doppler two-dimensional correlation operation result y to signal processor output _i(f _d, t) carry out CFAR detection and flight path and process, obtain target range Doppler flight path information, f _dThe expression doppler values; And flight path information carried out correction-compensation, the result after the compensation is exported to the location tracking module;

Described CFAR detection and flight path processing module comprise:

Point mark and track association submodule are with the as a result y that exports after the distance-Doppler two-dimensional correlation _i(f _dT) point that surpasses given detection threshold in carries out related with original targetpath, if this point is in the correlation threshold of setting, then be judged as the target that detects on original flight path, and the result is transported in Doppler's correction-compensation submodule, otherwise, impact point is delivered to the track initiation submodule;

Track initiation submodule, the impact point that adopts sliding window method that a mark and track association submodule are sent here carry out track initiation to be judged, if judgement have new Track forming then the export target flight path to Doppler's correction-compensation submodule, otherwise, impact point done gives up processing;

Flight path extinction submodule, which is to need to continue effective flight path of following the tracks of be used for to judge the flight path set up, which is the flight path that finishes of lose objects and needing; When certain bar flight path lose continuously count above the user set cancel thresholding the time, think that target exceeds investigative range, flight path is cancelled; Otherwise flight path is kept;

Doppler's correction-compensation submodule, utilize continuity and the approximate constant characteristic of short time internal object speed of range information, utilize the range Doppler information of previous moment to calculate disappearance distance value constantly, to remedy the of short duration interruption of target range Doppler flight path, the flight path of having set up is carried out correction-compensation;

The location tracking module, the targetpath information of utilizing CFAR detection and flight path processing module to obtain, finish successively multistation information pairing, target position information resolves and tracking initiation, and carries out track and localization by the EKF method, obtains the accurate positional information of target

2. passive radar locating and tracking system according to claim 1 is characterized in that the equal usage level polarization of reference antenna and signal antenna oriented antenna, and the beam angle of individual antenna is 22.5 °, and the reception frequency range is 87-110MHz; The reference antenna main lobe points to each broadcasting platform, and the signal antenna main lobe points to the target pattern direction.

3. passive radar locating and tracking system according to claim 1 is characterized in that signal processor comprises:

The analog to digital conversion submodule is used for the analog if signal of receiver output is carried out digitizing, is transferred to Digital Down Convert and processes submodule;

Submodule is processed in Digital Down Convert, and be used for the intermediate-freuqncy signal that digitizing is later and be transformed into base band, and with direct-path signal ref _iWith digital baseband echoed signal sig _i+ Z _iBe transferred to clutter phase slake distance-Doppler two-dimensional correlation submodule;

Clutter phase slake distance-Doppler two-dimensional correlation submodule is used for the data that signal antenna receives are carried out adaptive clutter cancellation, to eliminate direct wave D wherein _iWith multipath signal Z _i, and utilize the as a result sig that disappears mutually _iWith direct wave ref _iCarry out the computing of distance-Doppler two-dimensional correlation, the operation result y of this two-dimensional correlation computing _i(f _d, t) send CFAR detection and flight path processing module, f to _dThe expression doppler values.

4. passive radar locating and tracking system according to claim 1 is characterized in that locating tracking module and comprises:

Multistation metrical information pairing submodule utilizes the proximity of doppler information that revised range Doppler flight path information is matched, and finishes resolving of target location, the result is flowed to the location follow the tracks of submodule;

Submodule is followed the tracks of in the location, is used for finishing the location tracking of target, is about to each measurement data that obtains and inputs to filtering algorithm, estimates the positional information of target And the form with a mark shows in tracking results figure.

5. passive radar positioning and tracing method based on multiple illuminators and single receiver may further comprise the steps:

(1) points to respectively each frequency modulation broadcasting base station with reference to antenna, be used for receiving direct-path signal D _iSignal antenna is pointed to the target pattern direction, be used for receiving the FM stations signal R of moving target reflection _i, FM stations direct wave and multipath signal Z _i

(2) signal that receives with reference to antenna and signal antenna is respectively through obtaining analog if signal behind receiver A and receiver B mixing and the filter and amplification, and gives signal processor with this analog if signal;

(3) in signal processor, the analog if signal that receiver A and receiver B are transmitted carries out analog to digital conversion and Digital Down Convert is processed, and obtains digital baseband direct-path signal ref _iWith echoed signal sig _i+ Z _i, and with digital baseband direct-path signal ref _iWith echoed signal sig _i+ Z _iCarry out adaptive clutter cancellation, the direct wave of filtering FM stations and multipath signal Z _i, the echoed signal sig after obtaining to purify _iThen use digital baseband direct-path signal ref _iWith the echoed signal sig after the purification _iDo passive relevant processing according to following formula:

$y_i(f_d,n)={\mathrm{ref}}_i^{*}\left(n\right)\&CenterDot;{\mathrm{sig}}_i(n-\mathrm{\&tau;})e^{j2\mathrm{\&pi;}{f}_d(n-\mathrm{\&tau;})}$

Wherein, y _i(f _d, the range Doppler information that n) obtains for passive relevant processing, f _dThe expression doppler values, n represents time series, τ represents time delay, ^*The expression conjugate transpose, sig _iThe complex envelope of (n-τ) expression echoed signal, ref _i(n) representative digit base band direct-path signal ref _iIn n data;

(4) the doppler information y that adjusts the distance _i(f _d, n) carry out the target double threshold and detect track association, and according to following formula flight path information is done correction-compensation:

ρ(k+1)＝ρ(k)+f _d(k)λT

Wherein ρ (k+1) for replenish the k+1 obtain constantly distance and, ρ (k) be known k distance constantly and, f _d(k) be k doppler values constantly, λ is the emission wavelength in radio station, and T is the some mark time interval;

(5) the multistation target information is matched, judges whether to detect target:

(5a) a plurality of platforms of simulation calculation are observed same target, obtain Doppler and match threshold delta f _dWith distance pairing threshold delta ρ;

The doppler values f that (5b) i cell site is recorded _DiThe doppler values f that records with j cell site _DjSubtract each other, if Doppler difference is matched threshold delta f less than or equal to Doppler _d, i.e. f _Di-f _Dj≤ Δ f _d, and doppler values f _Di, f _DjNon-vanishing, then carry out (5c);

The distance value ρ that (5c) i cell site is recorded _iThe distance value ρ that records with j cell site _jSubtract each other, if apart from difference less than or equal to distance pairing threshold delta ρ, i.e. ρ _i-ρ _j≤ Δ ρ, and described distance value ρ _iAnd ρ _jNon-vanishing, then judge to detect target, and execution in step (5d);

(5d) utilize the range information of target, resolve and obtain target location coordinate (x, y, z), and make x _k=[x, y, z, 0,0,0] ^T, k=1,2 ... M ₀, M ₀For the target that detects is counted;

(6) with front 9 the result of calculation x that calculate in the step (5) ₁, x ₂..., x ₉Average, obtain Initial state estimation

(7) with Initial state estimation

As the filtering initial value, carry out EKF, obtain target position information

And in three-dimensional coordinate figure, show this target location, obtain targetpath.

6. passive radar positioning and tracing method according to claim 5, wherein step (3) is described with digital baseband direct-path signal ref _iWith echoed signal sig _i+ Z _iCarry out adaptive clutter cancellation, carry out as follows:

(3a) dimension of setting adaptive clutter cancellation device is V, the weight vector initial value

Counter n=1, adaptive clutter cancellation device weight vector initial value w (0) dimension is V * 1;

The adaptive clutter cancellation device that (3b) calculates the n step is exported outp (n): outp (n)=w (n-1) ^TR _i(n), w (n-1) wherein ^TBe the transposition that n-1 goes on foot the sef-adapting filter weight vector that obtains, R _i(n)=[ref _i(n), ref _i(n+1) ..., ref _i(n+K)] ^T, ref _i(n) expression direct-path signal ref _iIn n data;

(3c) calculate the n adaptive clutter cancellation device output error e (n) in step: e (n)=[sig _i(n)+Z _i(n)]-outp (n), wherein sig _i(n)+Z _i(n) expression echoed signal sig _i+ Z _iIn n data;

(3d) upgrade step factor μ (n):

μ wherein ₀Be the step factor initial value of setting, ε ₀Be the step factor adjustment factor, ^*Conjugation is got in expression;

(3e) upgrade adaptive clutter cancellation device weight vector w (n): w (n)=w (n-1)+μ (n) e (n) R _i(n);

(3f) whether judge n greater than L-V, if n greater than L-V filtering finish, otherwise counter n adds 1, and forwards step (3b) to and continue operation, the direct-path signal ref of L for receiving _iWith echoed signal sig _i+ Z _iData length.

7. passive radar positioning and tracing method according to claim 5, the wherein described doppler information y that adjusts the distance of step (4) _i(f _d, n) carry out the target double threshold and detect track association, carry out as follows:

(4a) establishing i is cell site's numbering, and initial value is 1;

(4b) choose as a result y after the distance-Doppler two-dimensional correlation of i cell site _i(f _d, t), and make its maximal value be: m=max (y _i(f _d, t)), if m is greater than given detection threshold d ₁, then change step (4c) over to, otherwise change step (4f) over to;

(4c) m and former targetpath are carried out related, if the difference of the predicted value of m and former flight path is less than correlation threshold d ₂, judgement detects target on the former flight path, the number S zero setting of point that simultaneously will this former track loss, and forward step (4f) to; Otherwise, S is added 1, and changes step (4d) over to;

(4d) adopt sliding window method to detect new flight path, namely use 4/6 logic initial, any four frames are in correlation threshold d in six continuous frame data ₂When interior, then thinking has new track initiation, and preserves the m value; Otherwise change step (4f) over to;

(4e) flight path is withered away and is judged: set flight path extinction thresholding and be T _cIf, the number S of the point of former track loss 〉=T _cThe time, think that target exceeds investigative range, flight path is cancelled; Otherwise flight path is kept; Change step (4f) over to;

(4f) cell site is numbered i and add 1, establishing N is total number of cell site, if i≤N then change step (4b) over to finishes otherwise detect.

8. a kind of passive radar positioning and tracing method according to claim 5, the described EKF of step (7) wherein, carry out as follows:

(7a) make counter k=1;

(7b) calculate the k observing matrix in step:

$H_k=\left[\begin{array}{cccccc}\frac{\hat{x}-x_R}{r_R}+\frac{\hat{x}-x_1}{r_1}& \frac{\hat{y}-y_R}{r_R}+\frac{\hat{y}-y_1}{r_1}& \frac{\hat{z}-z_R}{r_R}+\frac{\hat{z}-z_1}{r_1}& 0& 0& 0\\ \frac{\hat{x}-x_R}{r_R}+\frac{\hat{x}-x_2}{r_2}& \frac{\hat{y}-y_R}{r_R}+\frac{\hat{y}-y_2}{r_2}& \frac{\hat{z}-z_R}{r_R}+\frac{\hat{z}-z_3}{r_3}& 0& 0& 0\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;& \&CenterDot;\\ \frac{\hat{x}-x_R}{r_R}+\frac{\hat{x}-x_N}{r_N}& \frac{\hat{y}-y_R}{r_R}+\frac{\hat{y}-y_N}{r_N}& \frac{\hat{z}-z_R}{r_R}+\frac{\hat{z}-z_N}{r_N}& 0& 0& 0\end{array}\right]$

Wherein

Be k-1 step filtering output

In coordinates of targets, (x _R, y _R, z _R) expression receiving station coordinate, N is total number of cell site, (x ₁, y ₁, z ₁) (x ₂, y ₂, z ₂) ... (x _N, y _N, z _N) represent respectively the coordinate of each cell site, r _RThe distance of expression target and receiving station, r ₁, r ₂..., r _NThe distance that represents respectively target and each cell site;

(7c) calculate the predicted value that k step filtering is exported:

Wherein

I is 3 * 3 for unit matrix, and T is the parameter acquiring time interval, Be the filtering output in k-1 step;

(7d) calculate the predicted value that k goes on foot the error covariance matrix:

P wherein _K-1Be the filtering covariance matrix that obtains in the k-1 step, Q _kBe the process noise covariance matrix;

(7e) calculate the residual error covariance matrix: R wherein _kBe known observation noise covariance matrix;

(7f) calculate the output of k step wave filter: ${\hat{x}}_k={\hat{x}}_{k/k-1}+P_{k/k-1}{H}_k^T{S}_k^{-1}(z_k-H_k\&CenterDot;z_{k-1}),$ Z wherein _k=[ρ ₁(k), ρ ₂(k) ..., ρ _N(k)] ^T, N is total number of cell site;

(7g) upgrade the filtering covariance matrix: $P_k=P_{k/k-1}-P_{k/k-1}{H}_k^T{S}_k{\left(P_{k/k-1}{H}_k^T\right)}^T;$

(7h) mobility value of calculating target: $J={[z_k-H_k\&CenterDot;z_{k-1}]}^T{S}_k^{-1}[z_k-H_k\&CenterDot;z_{k-1}];$

The mobility value J of the target that (7i) obtains according to step (7h) and the target maneuver value J of setting ₀Magnitude relationship modify steps (7d) in Q _kIf, the mobility value J＜J of target ₀, then adjudicate target do not have motor-driven, observation noise Q _k=Q ₀If the mobility value J 〉=J of target ₀, then target have motor-driven, Q _k=α * Q ₀, wherein α is greater than 10 ⁴, Q ₀Be known process noise covariance matrix;

(7j) judge that counter k is whether greater than the target that the detects M that counts ₀If, k＜M ₀, then k is added 1, and go to step (7b), otherwise filtering finishes.

Images