CN103116169B - Anti-inference method based on vector tracking loop - Google Patents

Abstract

The invention relates to the field of satellite navigation, in particular to a global position system (GPS) receiver anti-interference method based on a vector tracking loop. The anti-inference method based on the vector tracking loop includes inputting signals into a C/A code phrase discriminator and a carrier wave frequency discriminator at a k moment, obtaining a pseudo-range error which is tested by a receiver and a pseudo-range rate error which is tested by the receiver, estimating a real pseudo-range error and a real pseudo-range rate error between a carrier and a satellite at a k+1 moment; calculating a position error and a speed error of the receiver at the k+1 moment, obtaining position and speed of the carrier at the k+1 moment and updating a unit observation vector. The anti-inference method based on the vector tracking loop utilizes a method that various track channels distribute various weight coefficients, enables channel information with a high signal to noise radio to assist channel information with a low signal to noise radio, improves the whole tracking ability of the tracking loop and inhibits influence of interference signals on the tracking loop.

Description

A kind of anti-interference method based on vector tracking loop

Technical field

The present invention relates to satellite navigation field, relate in particular to a kind of based on vector tracking loop GPS(GPS) receiver anti-interference method.

Background technology

In tradition GPS receiver tracking loop circuit, satellite of each channels track, is separate work between each tracking channel, and the information between passage can not exchange mutually, and such track loop is commonly called scalar track loop.GPS receiver scalar track loop hardware configuration is easily realized, and algorithm is simple, therefore widespread use in practice.

Xie Gang " GPS principle and Receiver Design " (Electronic Industry Press. Beijing .2009) describe traditional GPS receiver tracking loop circuit in detail in a book, its principle of work for: intermediate-freuqncy signal that receiver receives first multiplies each other with the carrier frequency mixing that carrier wave ring copies, wherein on I (in the same way) branch road, copy carrier multiplication with sine, on Q (orthogonal) branch road, copy carrier multiplication with cosine, then, the mixing results signal on I branch road and Q branch road carries out related calculation with leading, the instant and three parts of C/A codes that lag behind that code ring copies respectively; Then, correlated results is exported respectively coherent integration value after integration is cumulative; Again, the coherent integration value on instant branch road is taken as the input of carrier wave ring Discr., and the coherent integration value in other two associated branch is as a yard input for ring Discr.; Finally, carrier wave ring and code ring carry out filtering to their Discr. output valve respectively, and filtering result is used for regulating carrier number controlled oscillator separately and the state such as output phase and frequency of C/A yardage controlled oscillator, the carrier wave that carrier wave ring is copied is consistent with reception carrier, while makes again the C/A code that code ring copies be consistent with receiving C/A code, to ensure that carrier wave and C/A code in next moment reception signal are still thoroughly peeled off in track loop.In service in track loop, carrier signal State-output Doppler shift, integrated Doppler and carrier phase measurement value that carrier wave ring copies according to it, the C/A coded signal State-output code phase and the pseudo range measurement value that copy according to it with time-code ring, and carrier wave ring Discr. can also demodulate the navigation message data bit on satellite-signal extraly.

But in this track loop, gain and the bandwidth of loop filter are fixed, to all phase errors all according to identical weight processing, thereby variation that cannot adaptation signal carrier-to-noise ratio; On the other hand, because of separate between different tracking channels, existing navigation information cannot be fully used.Therefore, some scholars have proposed vector tracking loop pattern, by some technology by the information fusion of each tracking channel together.

The Wang Xinlong of BJ University of Aeronautics & Astronautics, in clean in Chinese inertial technology journal (the 17th the 6th phase of volume, in Dec, 2009, p710-717.) delivered one section " the SINS/GPS deep integrated navigation method based on vector tracking " above, article designs vector tracking loop under the dark group frame of GPS/SINS (Global Navigation System/strapdown inertial navitation system (SINS)), the carrier positions that inertial navigation system is calculated, speed etc. are converted into pseudorange, pseudorange rates measurement information, the information that these measurement informations and GPS receiver are measured combines, and the catching and track loop of feedback compensation receiver, improve precision and the antijamming capability of receiver.Article is pointed out, this SINS/GPS deep integrated navigation method based on vector tracking is mainly used on beyond visual range air to air missile, various tactical missile and the contour maneuvering-vehicle of fighter plane, the tracking performance of this vector tracking loop depends on the precision of inertia device to a great extent, often needs to reach Tactics-level precision and just can be applied on guided missile, the contour dynamic carrier of fighter plane.In addition, because receiver has been introduced inertial navigation measurement information, need to consider the stationary problem of two systems, just can make combined system work reliably.Therefore, the hardware configuration of this receiver and algorithm all can more complicated, and cost is also higher, are not suitable as the product for civilian use and use.

The non-high dynamic carrier such as object boats and ships, vehicle that the present invention considers, adopt independent GPS receiver, do not introduce external information, directly change its track loop structure, by extended filtering device by the information fusion of each tracking channel together, filter parameter is set, make the stronger signal of power by the weak signal of unit measurement vector auxiliary power, increase compared with the signal interference ratio of weak signal (ratio of desired signal power and interfering signal power), to improve the overall antijamming capability of track loop.Meanwhile, the too low signal of power also will likely affect other useful signals, need to wave filter wherein setting signal power threshold filter out the too low signal of power.The present invention has designed a kind of vector tracking loop structure on the scalar track loop basis of traditional GPS receiver, and comparative analysis the performance of vector tracking loop and scalar track loop.

Summary of the invention

The object of the invention is the antijamming capability in order to propose a kind of GPS of raising receiver, based on the anti-interference method of vector tracking loop.

The object of the present invention is achieved like this:

The present invention includes following steps:

(1) in the k moment, after the two-way orthogonal signal mixing that receiver mixer produces the intermediate-freuqncy signal receiving and local carrier generator, leading C/A code, the instant C/A code copying on I branch road and orthogonal Q branch road in the same way with C/A code generator and hysteresis C/A code does correlation computations and integration is cumulative, by the leading in the same way tributary signal I obtaining _e, the tributary signal I that lags behind in the same way _l, orthogonal leading tributary signal Q _ewith orthogonal hysteresis tributary signal Q _linput C/A code phase Discr., instant in the same way tributary signal I _pwith orthogonal instant tributary signal Q _pincoming carrier frequency discriminator;

(2) the C/A code phase error C/A code phase Discr. of vector tracking loop passage being obtained, the carrier frequency error that carrier frequency Discr. obtains are converted into pseudorange error and the pseudorange rates error that receiver measures;

(3), using pseudorange error and pseudorange rates error as measurement information, actual pseudorange error, pseudorange rates error is that quantity of state is input in expansion Kalman wave filter, estimates real pseudorange error and pseudorange rates error between k+1 moment carrier and satellite;

(4) utilize between k+1 moment carrier and satellite the unit measurement vector between real pseudorange error and pseudorange rates error and receiver and satellite wherein with for the component of the unit's of being respectively measurement vector X-axis, Y-axis and Z-direction in the body-fixed coordinate system of the earth's core, calculate receiver at the site error δ in k+1 moment p, velocity error δ v;

(5) utilize unit measurement vector between k moment carrier and satellite to calculate site error and the velocity error of k+1 moment carrier, obtain position and the speed of k+1 moment carrier;

(6) upgrade unit measurement vector, estimate pseudorange and the pseudorange rates of k+1 moment carrier, re-execute step (1).

C/A code phase error described carrier frequency error δ f=[I _p(k-1) Q _p(k)-I _p(k) Q _p(k-1)]/T, wherein T is post detection integration, gets T=20ms here.

The pseudorange error that receiver measures the pseudorange rates error that described receiver measures wherein f _c/a=1.023MHz is C/A bit rate, and c is the light velocity, f _l1=1575.42MHz is L1 carrier frequency.

The pseudorange error that k moment receiver is measured and pseudorange rates error be input in described kalman wave filter measurement equation, estimate pseudorange error and pseudorange rates error between carrier and the satellite in k+1 moment its median filter measurement equation is:

$\tilde{Z}=I_{2n\×2n}X+V$

I _{2n × 2n}be 2n × 2n unit matrix, V is that average is 0, and variance is white Gaussian noise, be 2n × 1 matrix; Wherein filter status equation is:

M=[1T; 01], T is post detection integration T=20ms, η _bfor clock jitter, η _dfor clock correction drift, η _x, η _y, η _zrespectively that pseudorange rates is at unit measurement vector x, y, z direction on dynamic process noise.

Utilize between k+1 moment carrier and satellite the unit measurement vector between real pseudorange error and pseudorange rates error and receiver and satellite calculate described receiver at the site error δ in k+1 moment p, velocity error δ v:

$\left[\begin{array}{c}{\mathrm{\δ\ρ}}^1\\ \·\\ \·\\ \·\\ {\mathrm{\δ\ρ}}^n\\ \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}^1\\ \·\\ \·\\ \·\\ \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}^n\end{array}\right]=\left[\begin{array}{cccccccc}{e}_x^1& e_y^1& e_z^1& 0& 0& 0& -1& 0\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ e_x^n& e_y^n& e_z^n& 0& 0& 0& -1& 0\\ 0& 0& 0& e_x^1& e_y^1& e_z^1& 0& -1\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ 0& 0& 0& e_x^n& e_y^n& e_z^n& 0& -1\end{array}\right]\left[\begin{array}{c}\mathrm{\δx}\\ \mathrm{\δy}\\ \mathrm{\δz}\\ \mathrm{\δ}\stackrel{\·}{x}\\ \mathrm{\δ}\stackrel{\·}{y}\\ \mathrm{\δ}\stackrel{\·}{z}\\ \mathrm{\δt}\\ \mathrm{\δ}\stackrel{\·}{t}\end{array}\right]+\mathrm{\ζ},$

Wherein (δ is z) component of δ p in x, y, z direction for δ x, δ y, for the component of δ v in x, y, z direction, ζ is pseudorange noise and the pseudorange rates noise that measures passage, and ζ is 2n × 1 matrix.

The position of k+1 moment carrier the speed of described k+1 moment carrier

Beneficial effect of the present invention is: the present invention adopts different tracking channels to distribute different way of weight coefficient, make the auxiliary lower passage of signal to noise ratio (S/N ratio) of channel information that signal to noise ratio (S/N ratio) is higher, improve track loop entirety tracking power, thereby suppress the impact of undesired signal on track loop.In addition, adopt vector tracking method, the dynamic property of carrier is to connect by the dynamic of signal in measurement vector and each tracking channel, the renewal that is receiver location and speed will have influence on the signal trace of each tracking channel by measurement vector, this track loop model more approaches the real duty of receiver.

Brief description of the drawings

Fig. 1 is a kind of enforcement block diagram of GPS antijam receiver vector tracking loop design process;

Fig. 2 is the scalar track loop structural drawing of GPS receiver;

Fig. 3 is the reduced graph of vector tracking structure loop;

Fig. 4 is vector tracking loop structure figure of the present invention;

Fig. 5 (a) is satellites in view number and PDOP (geometry of position dilution of precision) value analogous diagram;

Fig. 5 (b) is under different carrier-to-noise ratios, the pseudorange root-mean-square error analogous diagram of scalar method and vector method;

Fig. 6 (a) does not add while interference, the root-mean-square error analogous diagram of scalar method and pseudorange that vector method is predicted;

Fig. 6 (b) is to be under 20dB at jamming-to-signal ratio, the root-mean-square error analogous diagram of scalar method and pseudorange that vector method is predicted.

Embodiment

Below in conjunction with accompanying drawing, the present invention will be further described.

Traditional GPS receiver adopts scalar track loop pattern, and satellite of each channels track, does not exist information interchange between each passage.Discr. in each passage calculates after C/A code phase error and carrier frequency error, these errors directly feed back to code/carrier number controlled oscillator (NCO) of this passage after after filtering, NCO regulates local C/A code and carrier wave to produce frequency, make code phase, the carrier frequency of local replica consistent with code phase, carrier frequency in gps signal, thereby peel off the C/A code and the carrier wave that receive in signal completely, solution expands navigation data signal.Due to the impact of external disturbance and receiver internal thermal noise, in the gps signal of code phase, carrier frequency and the reception of local replica there is error in code phase, carrier frequency all the time, track loop is the process in dynamic adjustment all the time, and code phase error and carrier phase error are kept within limits.Receiver can obtain by code phase and carrier phase with respect to pseudorange and the pseudorange rates of satellite, thereby calculates the Position And Velocity of receiver.

Vector tracking loop in the present invention is converted into the code phase error τ of each channel code phase-shift discriminator output the pseudorange error of measurement the carrier frequency error δ f of carrier frequency Discr. output is converted into the pseudorange rates error of measurement the pseudorange error measuring and pseudorange rates error estimate actual pseudorange error δ ρ and pseudorange rates error as the measurement information of expansion kalman wave filter actual pseudorange error and pseudorange rates error are calculated the navigation information such as position, speed of receiver as intermediate variable, predict next moment this locality pseudorange with the navigation information after upgrading and pseudorange rates again with the pseudorange receiving and the pseudorange rates information closed track loop that carried out related calculation at lower a moment.Adopt actual satellite ephemeris to test, tracking power and the interference free performance of simulation results show vector tracking loop are better than scalar track loop.

The present invention proposes a kind of anti-interference method based on vector tracking loop, be specifically related to integration remover, C/A code phase Discr., carrier frequency Discr., expansion kalman wave filter, navigation calculation module, measurement vector update module, code/carrier frequency fallout predictor, code/carrier number controlled oscillator, code/carrier wave correlator.Key step is as follows:

1) each tracking channel subtracts rear power-type code phase Discr. before adopting calculate code phase error τ, each tracking channel adopts cross product method carrier frequency Discr. (δ f=[I _p(k-1) Q _p(k)-I _p(k) Q _p(k-1)]/T, k represents k sampling instant, T represents that sampling interval obtains carrier frequency error δ f;

2) pseudorange error that each tracking channel measures with pseudorange rates error calculated by τ and δ f

$\mathrm{\δ}\widetilde{\mathrm{\ρ}}=\mathrm{\τ}\·c/f_{c/a}---\left(1\right)$

$\mathrm{\δ}\widetilde{\stackrel{\·}{\mathrm{\ρ}}}=-\mathrm{\δf}\·c/f_{L1}---\left(2\right)$

In formula (1), f _c/a=1.023MHz is C/A bit rate, and c is the light velocity, in formula (2), and f _l1=1575.42MHz is L1 carrier frequency.

3) with with for measurement amount, with actual pseudorange error δ ρ, pseudorange rates error set up expansion kalman wave filter for quantity of state, can estimate the quantity of state of k+1 moment wave filter in the k moment, in the time that the k+1 moment, measurement information entered wave filter, this quantity of state is upgraded, and provides filter equation below:

If there be n satellites in view, pseudorange error that each passage measures, pseudorange rates error are as measurement information actual pseudorange error, pseudorange rates error are as status information error in measurement can be understood as pseudorange, the pseudorange rates of the local prediction generating of receiver with actual pseudorange, pseudorange rates between difference, error in measurement equals true error and adds measurement noise,

Wave filter measurement equation is

$\tilde{Z}=I_{2n\×2n}X+V---\left(3\right)$

In formula (3), I _{2n × 2n}be 2n × 2n unit matrix, V is that average is 0, and variance is white Gaussian noise, be 2n × 1 matrix.

Filter status equation is

In formula (4), m=[1T; 01], T is post detection integration T=20ms, η _band η _drespectively clock jitter and clock correction drift, η _x, η _y, η _zrespectively that pseudorange rates is at measurement vector dynamic process noise (in the body-fixed coordinate system of the earth's core) in direction.

4) if there be n satellites in view, utilize δ ρ, and unit measurement vector between receiver and satellite calculate receiver at the site error δ in k+1 moment p, velocity error δ v, clock jitter δ t

$\left[\begin{array}{c}{\mathrm{\δ\ρ}}^1\\ \·\\ \·\\ \·\\ {\mathrm{\δ\ρ}}^n\\ \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}^1\\ \·\\ \·\\ \·\\ \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}^n\end{array}\right]=\left[\begin{array}{cccccccc}{e}_x^1& e_y^1& e_z^1& 0& 0& 0& -1& 0\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ e_x^n& e_y^n& e_z^n& 0& 0& 0& -1& 0\\ 0& 0& 0& e_x^1& e_y^1& e_z^1& 0& -1\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ 0& 0& 0& e_x^n& e_y^n& e_z^n& 0& -1\end{array}\right]\left[\begin{array}{c}\mathrm{\δx}\\ \mathrm{\δy}\\ \mathrm{\δz}\\ \mathrm{\δ}\stackrel{\·}{x}\\ \mathrm{\δ}\stackrel{\·}{y}\\ \mathrm{\δ}\stackrel{\·}{z}\\ \mathrm{\δt}\\ \mathrm{\δ}\stackrel{\·}{t}\end{array}\right]+\mathrm{\ζ}---\left(5\right)$

In formula (5), (δ is z) component of δ p in x, y, z direction for δ x, δ y, for the component of δ v in x, y, z direction, ζ is pseudorange noise and the pseudorange rates noise that measures passage, and ζ is 2n × 1 matrix.

5) utilize the position p of receiver in the site error δ in k+1 moment p, velocity error δ v and k moment _k, speed v _k, estimation receiver is in the position in k+1 moment with speed

${\hat{p}}_{k+1}={\mathrm{\δp}}_{k+1}+p_k---\left(6\right)$

${\hat{v}}_{k+1}=\mathrm{\δ}{v}_{k+1}+v_k---\left(7\right)$

6) utilize the position of the calculation of parameter visible satellite such as true anomaly of ascending node of orbit right ascension, orbit inclination, the argument of perigee, major radius, excentricity and satellite in satellite ephemeris, then calculate the measurement vector in k+1 moment in conjunction with the receiver location after upgrading

7) again utilize formula (5), dope each passage pseudorange of k+1 moment, pseudorange rates;

8), by formula (1) and (2), dope code phase and the carrier frequency of k+1 each passage of moment;

9) local digital controlled oscillator regulates the frequency reproduction of C/A code and carrier wave, carries out related calculation with the signal receiving in the k+1 moment, completes whole vector tracking loop.

10) tracking performance of analysis scalar track loop and vector tracking loop

If there be n visible satellite, in scalar track loop, measurement equation is

$\mathrm{\δ}\widetilde{\mathrm{\ρ}}=I_{n\×n}\mathrm{\δ\ρ}+V_{\mathrm{\ρ}}---\left(8\right)$

In formula, for the pseudorange error of each tracking channel measurement the vector forming, the vector that the pseudorange error δ ρ that δ ρ is each tracking channel reality forms, I _{n × n}for n × n unit matrix, V _ρfor the vector that each passage pseudorange noise forms, it is n × 1 matrix.

Utilize least square method, can obtain

$\mathrm{\δ\ρ}={(I_{n\×n}^T\·R_{\mathrm{\ρ}}^{-1}\·I_{n\×n})}^{-1}\·I_{n\×n}^T\·R_{\mathrm{\ρ}}^{-1}\·\mathrm{\δ}\widetilde{\mathrm{\ρ}}---\left(9\right)$

Suppose that each passage measures pseudorange noise variance equate, in formula, for R _ρinverse matrix.

The object of track loop makes the pseudorange error of prediction equal actual pseudorange error δ ρ, covariance matrix for

$Q_{\mathrm{\δ}\widehat{\mathrm{\ρ}}}={(I_{n\×n}^T\·R_{\mathrm{\ρ}}^{-1}\·I_{n\×n})}^{-1}=I_{n\×n}\·{\mathrm{\δ}}_v^2---\left(10\right)$

Can find out from formula (10), the prediction pseudorange variance of each passage equates.

In vector tracking loop, only consider the pseudorange information in formula (5),

$\left[\begin{array}{c}{\mathrm{\δ\ρ}}^1\\ \·\\ \·\\ \·\\ {\mathrm{\δ\ρ}}^n\end{array}\right]=\left[\begin{array}{cccc}{e}_x^1& e_y^1& e_z^1& -1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ e_x^n& e_y^n& e_z^n& -1\end{array}\right]\left[\begin{array}{c}\mathrm{\δx}\\ \mathrm{\δy}\\ \mathrm{\δz}\\ \mathrm{\δt}\end{array}\right]+V_{\mathrm{\ρ}}---\left(11\right)$

Order $H_{n\×n}=\left[\begin{array}{cccc}{e}_x^1& e_y^1& e_z^1& -1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ e_x^n& e_y^n& e_z^n& -1\end{array}\right],$ $\mathrm{\δX}=\left[\begin{array}{c}\mathrm{\δx}\\ \mathrm{\δy}\\ \mathrm{\δz}\\ \mathrm{\δt}\end{array}\right],$

Formula (11) can be written as

δρ=H _n×nδX+V _ρ (12)

Utilize least square method, can obtain

$\mathrm{\δX}={(H_{n\×4}^T\·R_{\mathrm{\ρ}}^{-1}\·H_{n\×4})}^{-1}\·I_{n\×4}^T\·R_{\mathrm{\ρ}}^{-1}\·\mathrm{\δ}\widetilde{\mathrm{\ρ}}---\left(13\right)$

The covariance matrix Q of δ X _{δ X}for

$Q_{\mathrm{\δX}}={(H_{n\×4}^T\·R_{\mathrm{\ρ}}^{-1}\·H_{n\×4})}^{-1}={\mathrm{\δ}}_v^2{(H_{n\×4}^T\·H_{n\×4})}^{-1}---\left(14\right)$

δ X predicts the pseudorange error in next moment as intermediate variable

$\mathrm{\δ}\widehat{\mathrm{\ρ}}=H_{n\×n}\mathrm{\δX}---\left(15\right)$

Therefore, covariance matrix be

$Q_{\mathrm{\δ}\widehat{\mathrm{\ρ}}}={\mathrm{\δ}}_v^2\·M---\left(16\right)$

In formula (16), can find out, the prediction pseudorange variance of each passage equals be multiplied by respectively the diagonal entry of Metzler matrix, the element in Metzler matrix is to be made up of measurement vector in H matrix.

11) interference free performance of analysis scalar track loop and vector tracking loop

Measurement pseudorange noise variance in formula (10) and (16) relevant with the carrier-to-noise ratio of signal

${\mathrm{\δ}}_v^2={\left(\frac{c}{R_c}\right)}^2\frac{{4\mathrm{Fd}}^{2}B}{{10}^{{(C/N_0)}_{\mathrm{eq}}/10}}[2(1-d)+\frac{4\mathrm{Fd}}{{10}^{{(C/N_0)}_{\mathrm{eq}}/10}T}]---\left(24\right)$

In formula, R _cfor C/A code spreading rate, F is the code phase Discr. type factor, and d is correlator interval, and T is the coherent integration time, and B is that code endless belt is wide, (C/N ₀) _eqthe carrier-to-noise ratio of equivalence while interference for existing, (C/N ₀) _eqwith in signal, do not exist disturb time carrier-to-noise ratio C/N ₀pass is

${(C/N_0)}_{\mathrm{eq}}=-10\log ({10}^{-(C/N_0)/10}+{10}^{(J/S)/10}/Q{R}_c)---\left(25\right)$

In formula, J/S is undesired signal and the power ratio that receives signal, and Q is for adjusting coefficient (it is 1 that arrowband disturbs, and broadband interference is 2).

The enforcement block diagram of a kind of GPS antijam receiver vector tracking loop design process that Fig. 1 designs for the present invention, has considered tracking link and the navigation calculation link of signal, and two parts have been included in vector tracking loop.Tracking channel Discr. obtains, after code phase error and carrier frequency error, no longer the margin of error directly being fed back to respective channel NCO as control information, but calculates pseudorange, the pseudorange rates of receiver with respect to satellite.Then estimate position and the speed in next moment of receiver by wave filter, utilize the Position And Velocity after upgrading to calculate the unit measurement vector in next moment, thereby dope code phase error and the carrier frequency error in next moment.In figure, provide the key step that builds vector tracking loop, finally needed the vector tracking loop in the present invention to carry out tracking accuracy and anti-jamming Performance Analysis.

Fig. 2 is GPS scalar track loop structural drawing (list of references: Yu Hailiang, based on the research (D) of the auxiliary GPS receiver of INS and tracking technique, master thesis, Changsha: the National University of Defense technology, 2007,31-34), this track loop mainly comprises that the parts such as local carrier generator, frequency mixer, local C/A code generator, correlator, integration totalizer, code ring/carrier wave ring Discr., code ring/carrier wave ring wave filter form.Local carrier generator is at the upper carrier wave reproducing signals that produces 90 ° of phase phasic differences of branch road (i branch road) and quadrature branch (q branch road) in the same way, local code C/A code generator is branch road and quadrature branch produce respectively in advance in the same way, instant and hysteresis replica code, intermediate-freuqncy signal respectively simultaneously by a frequency mixer and local carrier generator in the same way with the carrier wave reproducing signals multiplicative mixing of orthogonal two branch roads generations, the mixing results obtaining is leading what branch road and quadrature branch produced respectively in the same way with local code C/A code generator again, immediately, hysteresis replica code carries out the related operation of time, and the signal obtaining is added up by an integration totalizer respectively, obtain the leading of in-phase branch, the relevant output I of instant and hysteresis branch road _e, I _pand I _l, and the relevant output Q of leading, the instant and hysteresis branch road of quadrature branch _e, Q _pand Q _l, I _pand Q _pincoming carrier Discr. obtains carrier phase error, in the local carrier generator entering through carrier wave ring wave filter, and the frequency that regulates local replica carrier wave to produce.I _e, I _l, Q _eand Q _larticle four, branch road enters a yard ring Discr. and obtains code phase error, in the local code generator entering through Loop filter, and the frequency that regulates local replica C/A code to produce.

Fig. 3 is the reduced graph of vector tracking structure loop.Gps signal becomes intermediate-freuqncy signal after radio-frequency front-end carries out down-converted, if receiver has N tracking channel, intermediate-freuqncy signal in each passage obtains pseudorange error after signal is processed, pseudorange rates error, these margins of error are not re-used as control information and directly feed back to respective channel NCO (digital controlled oscillator), but be input to and in wave filter, estimate the pseudorange between next moment carrier and satellite as measurement information, pseudorange rates, in next moment, carry out related operation with the signal receiving, obtain the pseudorange error in next moment, pseudorange rates error, form closed track loop.

Fig. 4 is the specific works process flow diagram flow chart of vector tracking loop, has related to integration remover, C/A code phase Discr., carrier frequency Discr., expansion kalman wave filter, navigation calculation module, measurement vector update module, code/carrier frequency fallout predictor, code/carrier number controlled oscillator, code/carrier wave correlator.Specifically comprise the steps:

12) each tracking channel subtracts rear power-type code phase Discr. before adopting calculate code phase error τ, each tracking channel adopts cross product method carrier frequency Discr. (δ f=[I _p(k-1) Q _p(k)-I _p(k) Q _p(k-1)]/T, k represents k sampling instant, T represents sampling interval) obtain carrier frequency error δ f;

13) pseudorange error that each tracking channel measures with pseudorange rates error calculated by τ and δ f

$\mathrm{\δ}\widetilde{\mathrm{\ρ}}=\mathrm{\τ}\·c/f_{c/a}---\left(1\right)$

$\mathrm{\δ}\widetilde{\stackrel{\·}{\mathrm{\ρ}}}=-\mathrm{\δf}\·c/f_{L1}---\left(2\right)$

In formula (1), f _c/a=1.023MHz is C/A bit rate, and c is the light velocity, in formula (2), and f _l1=1575.42MHz is L1 carrier frequency.

14) with with for measurement amount, with actual pseudorange error δ ρ, pseudorange rates error set up expansion kalman wave filter for quantity of state, can estimate the quantity of state of k+1 moment wave filter in the k moment, in the time that the k+1 moment, measurement information entered wave filter, this quantity of state is upgraded, and provides filter equation below:

If there be n satellites in view, pseudorange error that each passage measures, pseudorange rates error are as measurement information actual pseudorange error, pseudorange rates error are as status information error in measurement can be understood as pseudorange, the pseudorange rates of the local prediction generating of receiver with actual pseudorange, pseudorange rates between difference, error in measurement equals true error and adds measurement noise,

Wave filter measurement equation is

$\tilde{Z}=I_{2n\×2n}X+V---\left(3\right)$

In formula (3), I _{2n × 2n}be 2n × 2n unit matrix, V is that average is 0, and variance is white Gaussian noise, be 2n × 1 matrix.

Filter status equation is

In formula (4), m=[1T; 01], T is post detection integration T=20ms, η _band η _drespectively clock jitter and clock correction drift, η _x, η _y, η _zrespectively that pseudorange rates is at measurement vector dynamic process noise (in the body-fixed coordinate system of the earth's core) in direction.

15) if there be n satellites in view, utilize δ ρ, and unit measurement vector between receiver and satellite calculate receiver at the site error δ in k+1 moment p, velocity error δ v, clock jitter δ t

$\left[\begin{array}{c}{\mathrm{\δ\ρ}}^1\\ \·\\ \·\\ \·\\ {\mathrm{\δ\ρ}}^n\\ \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}^1\\ \·\\ \·\\ \·\\ \mathrm{\δ}{\stackrel{\·}{\mathrm{\ρ}}}^n\end{array}\right]=\left[\begin{array}{cccccccc}{e}_x^1& e_y^1& e_z^1& 0& 0& 0& -1& 0\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ e_x^n& e_y^n& e_z^n& 0& 0& 0& -1& 0\\ 0& 0& 0& e_x^1& e_y^1& e_z^1& 0& -1\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ \·& \·& \·& \·& \·& \·& \·& \·\\ 0& 0& 0& e_x^n& e_y^n& e_z^n& 0& -1\end{array}\right]\left[\begin{array}{c}\mathrm{\δx}\\ \mathrm{\δy}\\ \mathrm{\δz}\\ \mathrm{\δ}\stackrel{\·}{x}\\ \mathrm{\δ}\stackrel{\·}{y}\\ \mathrm{\δ}\stackrel{\·}{z}\\ \mathrm{\δt}\\ \mathrm{\δ}\stackrel{\·}{t}\end{array}\right]+\mathrm{\ζ}---\left(5\right)$

In formula (5), (δ is z) component of δ p in x, y, z direction for δ x, δ y, for the component of δ v in x, y, z direction, ζ is pseudorange noise and the pseudorange rates noise that measures passage, and ζ is 2n × 1 matrix.

16) utilize the position p of receiver in the site error δ in k+1 moment p, velocity error δ v and k moment _k, speed v _k, estimation receiver is in the position in k+1 moment with speed

${\hat{p}}_{k+1}={\mathrm{\δp}}_{k+1}+p_k---\left(6\right)$

${\hat{v}}_{k+1}=\mathrm{\δ}{v}_{k+1}+v_k---\left(7\right)$

17) utilize the position of the calculation of parameter visible satellite such as true anomaly of ascending node of orbit right ascension, orbit inclination, the argument of perigee, major radius, excentricity and satellite in satellite ephemeris, then calculate the measurement vector in k+1 moment in conjunction with the receiver location after upgrading

18) again utilize formula (5), dope each passage pseudorange of k+1 moment, pseudorange rates;

19), by formula (1) and (2), dope code phase and the carrier frequency of k+1 each passage of moment;

20) local digital controlled oscillator regulates the frequency reproduction of C/A code and carrier wave, carries out related calculation with the signal receiving in the k+1 moment, completes whole vector tracking loop.

21) tracking performance of analysis scalar track loop and vector tracking loop

If there be n visible satellite, in scalar track loop, measurement equation is

$\mathrm{\δ}\widetilde{\mathrm{\ρ}}=I_{n\×n}\mathrm{\δ\ρ}+V_{\mathrm{\ρ}}---\left(8\right)$

In formula, for the pseudorange error of each tracking channel measurement the vector forming, the vector that the pseudorange error δ ρ that δ ρ is each tracking channel reality forms, I _{n × n}for n × n unit matrix, V _ρfor the vector that each passage pseudorange noise forms, it is n × 1 matrix.

Utilize least square method, can obtain

$\mathrm{\δ\ρ}={(I_{n\×n}^T\·R_{\mathrm{\ρ}}^{-1}\·I_{n\×n})}^{-1}\·I_{n\×n}^T\·R_{\mathrm{\ρ}}^{-1}\·\mathrm{\δ}\widetilde{\mathrm{\ρ}}---\left(9\right)$

Suppose that each passage measures pseudorange noise variance equate, in formula, for R _ρinverse matrix.

The object of track loop makes the pseudorange error of prediction equal actual pseudorange error δ ρ, covariance matrix for

$Q_{\mathrm{\δ}\widehat{\mathrm{\ρ}}}={(I_{n\×n}^T\·R_{\mathrm{\ρ}}^{-1}\·I_{n\×n})}^{-1}=I_{n\×n}\·{\mathrm{\δ}}_v^2---\left(10\right)$

Can find out from formula (10), the prediction pseudorange variance of each passage equates.

In vector tracking loop, only consider the pseudorange information in formula (5),

$\left[\begin{array}{c}{\mathrm{\δ\ρ}}^1\\ \·\\ \·\\ \·\\ {\mathrm{\δ\ρ}}^n\end{array}\right]=\left[\begin{array}{cccc}{e}_x^1& e_y^1& e_z^1& -1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ e_x^n& e_y^n& e_z^n& -1\end{array}\right]\left[\begin{array}{c}\mathrm{\δx}\\ \mathrm{\δy}\\ \mathrm{\δz}\\ \mathrm{\δt}\end{array}\right]+V_{\mathrm{\ρ}}---\left(11\right)$

Order $H_{n\×n}=\left[\begin{array}{cccc}{e}_x^1& e_y^1& e_z^1& -1\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ \·& \·& \·& \·\\ e_x^n& e_y^n& e_z^n& -1\end{array}\right],$ $\mathrm{\δX}=\left[\begin{array}{c}\mathrm{\δx}\\ \mathrm{\δy}\\ \mathrm{\δz}\\ \mathrm{\δt}\end{array}\right],$

Formula (11) can be written as

δρ=H _n×nδX+V _ρ (12)

Utilize least square method, can obtain

$\mathrm{\δX}={(H_{n\×4}^T\·R_{\mathrm{\ρ}}^{-1}\·H_{n\×4})}^{-1}\·I_{n\×4}^T\·R_{\mathrm{\ρ}}^{-1}\·\mathrm{\δ}\widetilde{\mathrm{\ρ}}---\left(13\right)$

The covariance matrix Q of δ X _{δ X}for

$Q_{\mathrm{\δX}}={(H_{n\×4}^T\·R_{\mathrm{\ρ}}^{-1}\·H_{n\×4})}^{-1}={\mathrm{\δ}}_v^2{(H_{n\×4}^T\·H_{n\×4})}^{-1}---\left(14\right)$

δ X predicts the pseudorange error in next moment as intermediate variable

$\mathrm{\δ}\widehat{\mathrm{\ρ}}=H_{n\×n}\mathrm{\δX}---\left(15\right)$

Therefore, covariance matrix be

$Q_{\mathrm{\δ}\widehat{\mathrm{\ρ}}}={\mathrm{\δ}}_v^2\·M---\left(16\right)$

In formula (16), can find out, the prediction pseudorange variance of each passage equals be multiplied by respectively the diagonal entry of Metzler matrix, the element in Metzler matrix is to be made up of measurement vector in H matrix.

22) interference free performance of analysis scalar track loop and vector tracking loop

Measurement pseudorange noise variance in formula (10) and (16) relevant with the carrier-to-noise ratio of signal

${\mathrm{\δ}}_v^2={\left(\frac{c}{R_c}\right)}^2\frac{{4\mathrm{Fd}}^{2}B}{{10}^{{(C/N_0)}_{\mathrm{eq}}/10}}[2(1-d)+\frac{4\mathrm{Fd}}{{10}^{{(C/N_0)}_{\mathrm{eq}}/10}T}]---\left(24\right)$

In formula, R _cfor C/A code spreading rate, F is the code phase Discr. type factor, and d is correlator interval, and T is the coherent integration time, and B is that code endless belt is wide, (C/N ₀) _eqthe carrier-to-noise ratio of equivalence while interference for existing, (C/N ₀) _eqwith in signal, do not exist disturb time carrier-to-noise ratio C/N ₀pass is

${(C/N_0)}_{\mathrm{eq}}=-10\log ({10}^{-(C/N_0)/10}+{10}^{(J/S)/10}/{\mathrm{QR}}_c)---\left(25\right)$

In formula, J/S is undesired signal and the power ratio that receives signal, and Q is for adjusting coefficient (it is 1 that arrowband disturbs, and broadband interference is 2).

Fig. 5 (a) is the analogous diagram of satellites in view number and PDOP (geometry of position dilution of precision) value, in the 08:32:35 moment on June 4th, 2012, download the satellite almanac of Yuma form, the position of a satellites in view of every 5 minutes records, continues 12 hours altogether.Receiver location is arranged on Harbin Engineering University, and longitude is 126.6 ° of E, and latitude is 45.8 ° of N, and it is satellites in view that the elevation angle is greater than the satellite of 15 °, and Fig. 5 (a) is satellites in view number and the corresponding PDOP value of observing in 12 hours.In the time of 9:30, have 10 visual

Satellite, Fig. 5 (b) has provided now scalar track loop and the pseudorange root-mean-square error (RMS) of vector tracking loop under different carrier-to-noise ratios.Analyze as formula (10) and (16), the pseudorange RMS of the each passage of scalar loop equates, as shown in solid line in figure.The pseudorange RMS of each passage of vector loop is different, but all little than the pseudorange RMS of scalar loop, as shown in phantom in FIG..

Fig. 6 is for adding as disturbed front and back, the root-mean-square error contrast simulation figure of scalar method and pseudorange that vector method is predicted.

In scalar track loop, receiver carrier-to-noise ratio thresholding is set as 28dB/Hz conventionally, as can be seen from the figure works as C/N ₀when=28dB/Hz, pseudorange RMS=8.58m.For vector tracking loop, as pseudorange RMS=8.58m, the upper and lower bound of the signal carrier-to-noise ratio minimum value that each passage can trace into is respectively 25.8dB/Hz and 22.8dB/Hz.In the time of J/S=20dB, in Fig. 6 (b), provide RMS maximal value, the minimum value of RMS and the vector tracking loop prediction pseudorange of scalar track loop prediction pseudorange.For scalar track loop, as pseudorange RMS=8.58m, the minimum 29.7dB/Hz of being of signal carrier-to-noise ratio that receiver can be followed the tracks of, and the upper and lower bound of the signal carrier-to-noise ratio minimum value that vector tracking loop can be followed the tracks of is respectively 26.6dB/Hz and 23dB/Hz.

While not adding undesired signal, the maximal value of vector tracking loop carrier-to-noise ratio thresholding and minimum value are respectively than the low 2.2dB/Hz of the threshold value of scalar tracking loop and 5.2dB/Hz, add after J/S=20dB, the maximal value of vector tracking loop carrier-to-noise ratio thresholding and minimum value are respectively than the low 3.1dB/Hz of the threshold value of scalar tracking loop and 6.7dB/Hz, illustrate and added after undesired signal, vector tracking loop is lower with respect to the carrier-to-noise ratio of scalar track loop, in the environment of undesired signal, vector tracking loop can better embody with respect to the tracking performance of scalar track loop.

Claims (6)

1. the anti-interference method based on vector tracking loop, is characterized in that, comprises the steps:

(1) in the k moment, after the two-way orthogonal signal mixing that receiver mixer produces the intermediate-freuqncy signal receiving and local carrier generator, leading C/A code, the instant C/A code copying on I branch road and orthogonal Q branch road in the same way with C/A code generator and hysteresis C/A code does correlation computations and integration is cumulative, by the leading in the same way tributary signal I obtaining _e, the tributary signal I that lags behind in the same way _l, orthogonal leading tributary signal Q _ewith orthogonal hysteresis tributary signal Q _linput C/A code phase Discr., instant in the same way tributary signal I _pwith orthogonal instant tributary signal Q _pincoming carrier frequency discriminator;

(2) the C/A code phase error C/A code phase Discr. of vector tracking loop passage being obtained, the carrier frequency error that carrier frequency Discr. obtains are converted into pseudorange error and the pseudorange rates error that receiver measures;

(3), using pseudorange error and pseudorange rates error as measurement information, actual pseudorange error, pseudorange rates error is that quantity of state is input in expansion Kalman wave filter, estimates real pseudorange error and pseudorange rates error between k+1 moment carrier and satellite;

(4) utilize between k+1 moment carrier and satellite the unit measurement vector between real pseudorange error and pseudorange rates error and receiver and satellite , wherein with for the component of the unit's of being respectively measurement vector X-axis, Y-axis and Z-direction in the body-fixed coordinate system of the earth's core, calculate receiver at the site error δ in k+1 moment p, velocity error δ v;

(5) utilize unit measurement vector between k moment carrier and satellite to calculate site error and the velocity error of k+1 moment carrier, obtain position and the speed of k+1 moment carrier;

(6) upgrade unit measurement vector, estimate pseudorange and the pseudorange rates of k+1 moment carrier, re-execute step (1).

2. a kind of anti-interference method based on vector tracking loop according to claim 1, is characterized in that: described C/A code phase error described carrier frequency error δ f=[I _p(k-1) Q _p(k)-I _p(k) Q _p(k-1)]/T, wherein T is post detection integration, gets T=20ms here.

3. a kind of anti-interference method based on vector tracking loop according to claim 2, is characterized in that: the pseudorange error that described receiver measures the pseudorange rates error that described receiver measures wherein f _c/a=1.023MHz is C/A bit rate, and c is the light velocity, f _l1=1575.42MHz is L1 carrier frequency.

4. a kind of anti-interference method based on vector tracking loop according to claim 3, is characterized in that: the pseudorange error that k moment receiver is measured and pseudorange rates error be input in described kalman wave filter measurement equation, estimate pseudorange error and pseudorange rates error between carrier and the satellite in k+1 moment its median filter measurement equation is:

$\tilde{Z}=I_{2n\×2n}X+V$

I _{2n × 2n}be 2n × 2n unit matrix, V is that average is 0, and variance is white Gaussian noise, be 2n × 1 matrix; Wherein filter status equation is:

M=[1T; 01], T is post detection integration T=20ms, η _bfor clock jitter, η _dfor clock correction drift, η _x, η _y, η _zrespectively that pseudorange rates is at unit measurement vector x, y, z direction on dynamic process noise.

5. a kind of anti-interference method based on vector tracking loop according to claim 4, is characterized in that: utilize between k+1 moment carrier and satellite the unit measurement vector between real pseudorange error and pseudorange rates error and receiver and satellite calculate described receiver at the site error δ in k+1 moment p, velocity error δ v:

$\left[\begin{array}{c}{\mathrm{\δ\ρ}}^1\\ .\\ .\\ .\\ {\mathrm{\δ\ρ}}^n\\ {\mathrm{\δ}\stackrel{\·}{\mathrm{\ρ}}}^1\\ .\\ .\\ .\\ {\mathrm{\δ}\stackrel{\·}{\mathrm{\ρ}}}^n\end{array}\right]=\left[\begin{array}{cccccccc}{e}_x^1& e_y^1& e_z^1& 0& 0& 0& -1& 0\\ .& .& .& .& .& .& .& .\\ .& .& .& .& .& .& .& .\\ .& .& .& .& .& .& .& .\\ e_x^n& e_y^n& e_z^n& 0& 0& 0& -1& 0\\ 0& 0& 0& e_x^1& e_y^1& e_z^1& 0& -1\\ .& .& .& .& .& .& .& .\\ .& .& .& .& .& .& .& .\\ .& .& .& .& .& .& .& .\\ 0& 0& 0& e_x^n& e_y^n& e_z^n& 0& -1\end{array}\right]\left[\begin{array}{c}\mathrm{\δx}\\ \mathrm{\δy}\\ \mathrm{\δz}\\ \mathrm{\δ}\stackrel{\·}{x}\\ \mathrm{\δ}\stackrel{\·}{y}\\ \mathrm{\δ}\stackrel{\·}{z}\\ \mathrm{\δt}\\ \mathrm{\δ}\stackrel{\·}{t}\end{array}\right]+\mathrm{\ζ},$

Wherein (δ is z) component of δ p in x, y, z direction for δ x, δ y, for the component of δ v in x, y, z direction, ζ is pseudorange noise and the pseudorange rates noise that measures passage, and ζ is 2n × 1 matrix.

6. a kind of anti-interference method based on vector tracking loop according to claim 5, is characterized in that: the position of described k+1 moment carrier the speed of described k+1 moment carrier