CN103926603A - Extremely-weak signal tracking method of GNSS receiver - Google Patents

Abstract

The invention provides an extremely-weak signal tracking method of a GNSS receiver, and aims at providing a rapid and accurate tracking method of the GNSS receiver under the extremely-weak signal condition. According to the technical scheme, the extremely-weak signal tracking method includes the steps that firstly, frequency mixing multiplying is carried out on a GNSS digital intermediate frequency signal and carrier signals copied by a carrier loop; related operation is carried out on a frequency mixing result of an I sub-loop and ahead C/A codes, immediate C/A codes and lagging C/A codes copied by a code loop, and carried out on a frequency mixing result of a Q sub-loop and the ahead C/A codes, the immediate C/A codes and the lagging C/A codes copied by the code loop respectively, and output six related results are sent into six integration erasers respectively; then six coherent integration values output by the six integration erasers are sent into a capacity Kalman filter to estimate related parameters; finally, the corresponding estimation values are sent into a code loop filter and a carrier loop filter respectively to carry out filtering, and then fed back to a carrier numerically-controlled oscillator and a C/A code numerically-controlled oscillator respectively, then the carrier phase, the carrier frequency, the code phase and the code frequency are adjusted in real time, and finally GNSS signals are rapidly and accurately tracked.

Description

The tracking of GNSS receiver utmost point weak signal

Technical field

The present invention relates to a kind of tracking of GNSS receiver utmost point weak signal, under the weak environment of that the method is mainly used in is indoor, tunnel and geo-stationary orbit and HEO satellite etc. receive navigation signal, GNSS receiver in real time, accurately follow the tracks of fast.

Background technology

GPS (Global Position System) GNSS is widely used in worldwide, but when GNSS operation of receiver is indoor, in the time of in the severe Complex Channel environment of the various signal conditionings such as dense city, GNSS signal power can be subject to serious decay, the impact that signal amplitude and phase place also can be subject to multipath fading changes violent, received signal to noise ratio will severe exacerbation, and common GNSS receiver will be difficult to correct catching and track navigation satellite-signal.Because Direct Sequence Spread Spectrum Signal has the advantages such as highly concealed type and anti-intercepting and capturing, interference performance be strong, carrier-to-noise ratio C/N0 is very common lower than 35dB/Hz feeble signal.The basis that GNSS receiver is realized navigator fix is by the satellite-signal receiving being carried out to a series of signal processing, and then extracts corresponding navigational parameter.Its signal processing generally comprises following several stages, catches, tracking, bit synchronization and frame synchronization.In the signal trace stage, the guestimate value to current satellite signal carrier frequency and code phase that signalling channel obtains from acquisition phase by the progressively meticulous estimation to two signal parameters of track loop, is exported the various GNSS measured values of signal simultaneously.

GNSS track loop comprises two elementary cycles, i.e. code ring and carrier wave ring.Traditional carrier loop adopts standard delay locked loop (DLL) and phaselocked loop (PLL) to realize, although its robustness is better, but there is non-linear factor in Discr. wherein, and the dynamic change of signal phase, make locked loop (being that carrier-to-noise ratio is lower) a little less than pickup electrode in the situation that, the as easy as rolling off a log signal losing lock that occurs receiver.In the situation that carrier-to-noise ratio is lower, easily there is the problem that losing lock even cannot effectively lock in standard DDL and PLL track loop.Some scholars have proposed to adopt EKF to realize the tracking of weak signal for this reason, but the method is deposited deficiency both ways, one, the inearized model of expanded Kalman filtration algorithm is inaccurate, in algorithm model, coefficient calculations is complicated, parameter estimation cumulative errors is large, and needs to calculate comparatively complicated Jacobian matrix; Its two, by linearization of nonlinear system is realized to state estimation, so its estimated accuracy is lower, even there will be and disperses.

Summary of the invention

The object of the invention is the weak point existing for prior art, provide a kind of tracking power strong, tracking accuracy is high, without calculating Jacobian matrix, a little less than pickup electrode in the situation that, can the little quick precision tracking of real-time follow-up utmost point weak signal, to solve tradition (standard DDL and PLL) track loop, in the situation that carrier-to-noise ratio is lower, easily there is the problem that losing lock even cannot effectively lock.

The present invention solves the scheme that prior art problem adopts: a kind of GNSS receiver utmost point weak signal tracking, it is characterized in that comprising the steps: in GNSS receiver tracking circuit, first, the GNSS digital medium-frequency signal sIF (n) of frequency mixer will be passed through, on I branch road, copy carrier multiplication with sine, on Q branch road, copy carrier multiplication with cosine; Then the mixing multiplied result of I branch road and Q branch road is carried out to related operation with leading, the instant and hysteresis C/A code that code ring copies respectively, the multichannel correlated results of related operation output is sent into respectively to multichannel integration-remover and carry out coherent integration; Again the multichannel coherent integration value of multichannel integration-remover output is sent into the estimation that volume Kalman filter is carried out correlation parameter; Finally, corresponding estimated value is sent into respectively to Loop filter and carrier wave ring wave filter carries out filtering, after filtering, feed back to respectively carrier number controlled oscillator and C/A yardage controlled oscillator, and then realize the real-time adjusting of carrier phase and carrier frequency, code phase and code frequency, finally realize the tracking of GNSS signal.

The present invention has following beneficial effect than prior art:

(1) tracking power is strong.Jiang Liu of the present invention road coherent integration value is sent into the estimation that volume Kalman filter is carried out correlation parameter, estimated result carries out filtering through Loop filter and carrier wave ring wave filter respectively, after filtering, as the input of C/A yardage controlled oscillator and carrier number controlled oscillator, realize the quick accurate tracking of carrier frequency and phase place, code frequency and phase place respectively.Than traditional track loop, the method can realize the real-time follow-up that utmost point weak signal is little.

(2) tracking accuracy is high.Code ring of the present invention copies a C/A code sequence consistent with receiving C/A code in signal by its inner code generator, then both carry out related calculation, realize the C/A code of peeling off in GNSS reception signal, also improved the signal to noise ratio (S/N ratio) that is originally submerged in the GNSS signal in noise simultaneously.Carrier wave ring is to be made hand and foot its carrier signal copying and be consistent with the satellite carrier signal receiving by carrier loop, thereby peels off up hill and dale the carrier wave in satellite-signal by mixed batch of mechanism.Between carrier wave ring and code ring, by organically combining, mutually support, jointly complete the tracking of signal and measurement.Than the track loop based on EKF, its tracking accuracy is higher, and without calculating Jacobian matrix.The GNSS weak signal tracking based on volume Kalman filtering proposing, than prior art, volume Kalman filtering of the present invention not only does not need to calculate Jacobian matrix, and its filtering accuracy is higher.

Accompanying drawing explanation

Below in conjunction with drawings and Examples, the present invention is further described.

Fig. 1 is the GNSS receiver tracking theory diagram that the present invention is based on volume Kalman filter.

Fig. 2 is volume Kalman filter principle of work block diagram in Fig. 1.

Embodiment

Consult Fig. 1.As shown in Figure 1, volume Kalman filter wherein realizes by the workflow shown in Fig. 2 track loop of the present invention.First, the carrier signal that the GNSS digital medium-frequency signal by frequency mixer and carrier wave ring are copied is carried out mixing and is multiplied each other, wherein s _iF(n) on I branch road, copy carrier multiplication with sine, on Q branch road, copy carrier multiplication with cosine; Then the mixing multiplied result of I branch road and Q branch road is carried out to related operation with leading, the instant and hysteresis C/A code that code ring copies respectively, the multichannel correlated results of related operation output is sent into respectively to multichannel integration-remover and carry out coherent integration value; Again multichannel coherent integration value is sent into the estimation that volume Kalman filter is carried out correlation parameter; Finally, corresponding estimated value is sent into respectively to Loop filter and carrier wave ring wave filter carries out filtering, after filtering, feed back to respectively carrier number controlled oscillator and C/A yardage controlled oscillator, realize the real-time adjusting of carrier phase and carrier frequency, code phase and code frequency, follow the tracks of GNSS signal.Multichannel described in the present embodiment can Shi Liu road, and six road integration-removers are most preferred embodiments of the present invention.

(1) GNSS digital medium-frequency signal s _iF(n) carry out Frequency mixing processing with the carrier wave circle replication carrier signal of 90 ° of two-way phase differences respectively, wherein GNSS digital medium-frequency signal s _iF(n) shown in being expressed as:

y(t _n)＝A(t _n)C(t _n-τ(t _n))D(t _n-τ(t _n))sin(w _IFt _n-φ(t _n))+v _n

In formula: t _nrepresent that discrete sampling constantly; y(t _n) representative digit intermediate-freuqncy signal s _iF(n); A(t _n) represent the amplitude of carrier wave; C(t _n) represent the C/A code that satellite is broadcast, D (t _n) represent the numeric data code that satellite sends, and both level values may be only ± 1; τ (t _n) represent the propagation delay of signal; w _iFrepresentative digit intermediate frequency angular frequency; φ (t _n) represent the carrier phase of signal; v _nrepresent that average is zero Gaussian sequence.

(2) mixing results of I branch road and Q branch road is carried out related operation with leading, the instant and hysteresis C/A code that code ring copies respectively, that is:

$\left\{\begin{array}{cc}\begin{array}{c}{I}_k\left(\mathrm{\δ}\right)=\underset{n=n_k+1}{\overset{n_k+N_k}{\mathrm{\Σ}}}y\left(t_n\right)\·\left(C_{\mathrm{NCO}}(t_n+\mathrm{\δ}-t_{\mathrm{NCOk}})\sin \right[w_{\mathrm{IF}}{t}_n+{\mathrm{\φ}}_{\mathrm{NCO}}\left(t_n\right)\left]\right)\\ Q_k\left(\mathrm{\δ}\right)=\underset{n=n_k+1}{\overset{n_k+N_k}{\mathrm{\Σ}}}y\left(t_n\right)\·\left(C_{\mathrm{NCO}}(t_n+\mathrm{\δ}-t_{\mathrm{NCOk}})\cos \right[w_{\mathrm{IF}}{t}_n+{\mathrm{\φ}}_{\mathrm{NCO}}\left(t_n\right)\left]\right)\end{array}& \mathrm{\δ}=-\frac{{\mathrm{\Δ}}_{e\_l}}{2}\end{array},,0,\frac{{\mathrm{\Δ}}_{e\_l}}{2}\right.$

In formula: k represents current residing position of the 1ms time interval; N represents the sampling instant in the 1ms time interval; Nk represents the number of samples in every 1ms time interval; C _nCO(t _n) represent the PRN replica code of the tracked satellite-signal produced by receiver; t _nCOkrepresent that code NCO produces the initial time of instantaneous code; φ _nCO(t _n) represent the carrier phase that receiver copies.Δ _{e_l}represent the spacing between leading and lag correlation device; δ represents the phase difference between copied San road C/A code and instantaneous code; I _k(δ) and Q _k(δ) represent respectively the relevant accumulated value of I branch road and Q branch road; Wherein, when time, I _k(δ), Q _k(δ) represent that respectively the mixing results of I branch road and Q branch road and the advanced code that code ring copies carry out related operation; When δ=0, I _k(δ), Q _k(δ) represent that respectively the mixing results of I branch road and Q branch road and the instantaneous code that code ring copies carry out related operation; When time, I _k(δ) represent that the mixing results of I branch road and Q branch road and the hysteresis code that code ring copies carry out related operation.

Correlated results after related operation can be expressed as:

$\left\{\begin{array}{cc}\begin{array}{c}{I}_k\left(\mathrm{\δ}\right)=\frac{N_k{\stackrel{\‾}{A}}_k{D}_m}{2}\cos \left({\mathrm{\Δ\φ}}_k\right)R({\mathrm{\Δt}}_k+\mathrm{\δ})+v_{\mathrm{Ik}}\\ Q_k\left(\mathrm{\δ}\right)=\frac{N_k{\stackrel{\‾}{A}}_k{D}_m}{2}\sin \left({\mathrm{\Δ\φ}}_k\right)R({\mathrm{\Δt}}_k+\mathrm{\δ})+v_{\mathrm{Qk}}\end{array}& \mathrm{\δ}=-\frac{{\mathrm{\Δ}}_{e\_l}}{2}\end{array},0,\frac{{\mathrm{\Δ}}_{e\_l}}{2}\right.$

In formula: the mean value that represents the cumulative time period intercarrier amplitude of whole 1ms, D _mrepresent navigation data bit, Δ φ _krepresent cumulative time period intercarrier phase error phi (the t)-φ of 1ms _nCO(t) mean value, Δ t _k=τ (t _midk)-t _midkthe code phase error that represents 1ms intermediate point in the cumulative time period, and t _midk=(t _nCOk+ t _nCOk+1) 2; v _ikand v _qkrepresent respectively the mutual incoherent Gaussian sequence of zero-mean, its variance is ; In formula, do not have R (Δ t _k+ δ) represent the autocorrelation function of PRN code.

Export Liu road correlated results i _e, i _p, i _l, q _e, q _pand q _lfor:

$\left\{\begin{array}{c}{i}_E=I_k(-\frac{{\mathrm{\Δ}}_{e\_l}}{2})\\ i_P=I_k\left(0\right)\\ i_L=I_k\left(\frac{{\mathrm{\Δ}}_{e\_l}}{2}\right)\\ q_E=Q_k(-\frac{{\mathrm{\Δ}}_{e\_l}}{2})\\ q_P=Q_k\left(0\right)\\ q_L=Q_k\left(\frac{{\mathrm{\Δ}}_{e\_l}}{2}\right)\end{array}\right.$

In formula: i _e, i _pand i _lthe correlation that represents respectively leading on I branch road, instant and hysteresis branch road; q _e, q _pand q _lthe correlation that represents respectively leading on Q branch road, instant and hysteresis branch road.

(3) six road correlated results are exported six tunnel coherent integration values by six road integration-removers.

(4) Jiang Liu road coherent integration value is sent into volume Kalman filter, completes the estimation of correlation parameter, comprising: volume Kalman filter system state equation x _m+1=f _m(x _m, w _m) with the setting up in formula of measurement equation: $x_m={\left[\begin{array}{ccccc}{x}_{{\mathrm{\φ}}_m}& x_{w_m}& x_{{\mathrm{\α}}_m}& A_m& t_{s_m}\end{array}\right]}^T,$ For the state vector of system, m represents integration interval zero hour, and selecting integral time is here 20ms, represent the poor of the m carrier phase that truly carrier phase and receiver carrier wave NCO copy constantly; represent the constantly true carrier doppler drift of m; represent m carrier doppler drift constantly rate of change; A _mrepresent m carrier amplitude constantly; represent m code phase constantly; x _m+1represent m+1 system state vector constantly; f _mrepresent nonlinear state transfer function; w _mrepresent system noise vector.

The expression of state equation is as follows:

${\left[\begin{array}{c}{x}_{\mathrm{\φ}}\\ x_w\\ x_{\mathrm{\α}}\end{array}\right]}_{m+1}=\left[\begin{array}{ccc}1& {\mathrm{\δt}}_m& {\mathrm{\δt}}_m^2/2\\ 0& 1& {\mathrm{\δt}}_m\\ 0& 0& 1\end{array}\right]{\left[\begin{array}{c}{x}_{\mathrm{\φ}}\\ x_w\\ x_{\mathrm{\α}}\end{array}\right]}_m-\left[\begin{array}{c}{\mathrm{\δt}}_m\\ 0\\ 0\end{array}\right]W_{{\mathrm{NCO}}_m}+\left[\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 0\\ 0& 0& 1& 0\end{array}\right]w_{{\mathrm{\φ}}_m}$

$t_{s_{m+1}}=t_{s_m}+\frac{w_{L1}{\mathrm{\δt}}_{\mathrm{nom}}-\left[\begin{array}{cccc}1& 0& 0& 0\end{array}\right]w_{{\mathrm{\φ}}_m}}{w_{L1}+x_{w_m}+0.5{\mathrm{\δt}}_{\mathrm{nom}}{x}_{{\mathrm{\α}}_m}}+w_{{\mathrm{ts}}_m}$

$A_{m+1}=A_m+w_{A_m}$

In formula: represent the cumulative time period; the value that represents the carrier doppler frequency displacement that carrier wave NCO produces within the cumulative time period; WL1 represents L1 carrier angular frequencies; and represent that average is zero, mutual incoherent Gaussian sequence.

The foundation of measurement equation.Here choose the instant coherent integration value of 20ms, 20ms subtracts hysteresis coherent integration value in advance as measurement amount, sets up measurement equation, that is:

$\begin{array}{c}{z}_m=\frac{1}{{\mathrm{\σ}}_n}\sqrt{\frac{2}{N_m}}\left[\begin{array}{c}\underset{k=k_m}{\overset{k_m+19}{\mathrm{\Σ}}}{I}_k\left(0\right)\\ \underset{k=k_m}{\overset{k_m+19}{\mathrm{\Σ}}}{Q}_k\left(0\right)\\ \frac{1}{\sqrt{\mathrm{\η}}}\underset{k=k_m}{\overset{k_m+19}{\mathrm{\Σ}}}[I_k\left(\frac{{\mathrm{\Δ}}_{E\_L}}{2}\right)-I_k(-\frac{{\mathrm{\Δ}}_{E\_L}}{2})]\\ \frac{1}{\sqrt{\mathrm{\η}}}\underset{k=k_m}{\overset{k_m+19}{\mathrm{\Σ}}}[Q_k\left(\frac{{\mathrm{\Δ}}_{E\_L}}{2}\right)-Q_k(-\frac{{\mathrm{\Δ}}_{E\_L}}{2})]\end{array}\right]\\ =D_m{h}_m({\mathrm{\Δ\φ}}_m,{\mathrm{\Δt}}_m,{\stackrel{\‾}{A}}_m)+v_{\mathrm{zm}}\end{array}=\frac{{\stackrel{\‾}{A}}_m{D}_m}{{\mathrm{\σ}}_n}\sqrt{\frac{N_m}{2}}\left[\begin{array}{c}\cos \left({\mathrm{\Δ\φ}}_m\right)R\left({\mathrm{\Δt}}_m\right)\\ \sin \left({\mathrm{\Δ\φ}}_m\right)R\left({\mathrm{\Δt}}_m\right)\\ \frac{1}{\sqrt{\mathrm{\η}}}\cos \left({\mathrm{\Δ\φ}}_m\right)R_{E\_L}\left({\mathrm{\Δt}}_m\right)\\ \frac{1}{\sqrt{\mathrm{\η}}}\sin \left({\mathrm{\Δ\φ}}_m\right)R_{E\_L}\left({\mathrm{\Δt}}_m\right)\end{array}\right]+v_{\mathrm{zm}}$

In formula: every the number of samples of 20ms η=2[1-R (Δ _{e_L})]; the mean value that represents 20ms carrier amplitude; Δ φ _mthe mean value that represents 20ms carrier phase difference; the code phase error that represents intermediate point; $R_{E\_L}\left(\mathrm{\Δt}\right)=R(\mathrm{\Δt}+\frac{{\mathrm{\Δ}}_{E\_L}}{2})-R(\mathrm{\Δt}-\frac{{\mathrm{\Δ}}_{E\_L}}{2})$ Represent to subtract in advance lag correlation function; the related function that represents advanced code; the related function that represents hysteresis code; h _mrepresent to measure function; ν _zmthe Gaussian sequence that represents zero-mean.

Wherein in measurement equation, every expression is as follows:

${\mathrm{\Δ\φ}}_m=\left[\begin{array}{ccc}1& {\mathrm{\δt}}_m/2& {\mathrm{\δt}}_m^2/6\end{array}\right]{\left[\begin{array}{c}{x}_{\mathrm{\φ}}\\ x_w\\ x_{\mathrm{\α}}\end{array}\right]}_m-\frac{{\mathrm{\δt}}_m}{2}+\left[\begin{array}{cccc}0& 0& 0& 1\end{array}\right]w_{\mathrm{\φm}}$

${\mathrm{\Δt}}_m=(t_{s_{m+1}}+t_{s_m})/2-t_{{\mathrm{mid}}_m}=t_{s_m}+\frac{1}{2}\frac{w_{L1}{\mathrm{\δt}}_{\mathrm{nom}}-\left[\begin{array}{cccc}1& 0& 0& 0\end{array}\right]w_{{\mathrm{\φ}}_m}}{w_{L1}+x_{w_m}+0.5{\mathrm{\δt}}_{\mathrm{nom}}{x}_{{\mathrm{\α}}_m}}+\frac{1}{2}{w}_{{\mathrm{ts}}_m}-t_{{\mathrm{mid}}_m}$

${\stackrel{\‾}{A}}_m=(A_{m+1}+A_m)/2=A_m+0.5w_{A_m}$

Design of Mathematical Model volume Kalman filter according to above-mentioned foundation, completes the estimation of correlation parameter, comprising the time, upgrades with measurement and upgrades two primary iteration processes.

Consult Fig. 2, the specific works flow process of volume Kalman filter is as follows:

1. system state and state covariance matrix are carried out to initialization.

2. suppose that k-1 state estimation value and state covariance matrix is constantly respectively with to state covariance matrix P _k-1|k-1carry out factorization, that is:

$P_{k-1|k-1}=S_{k-1|k-1}{S}_{k-1|k-1}^T$

3. computed volume point

$X_{i,k-1|k-1}=S_{k-1|k-1}{\mathrm{\ξ}}_i+{\hat{x}}_{k-1|k-1}$

4. propagate volume point

$X_{i,k-1|k-1}^{*}=f\left(X_{i,k-1|k-1}\right)$

5. computing system state one-step prediction and one-step prediction covariance matrix

${\hat{x}}_{x|k-1}=\frac{1}{2n}\underset{i=1}{\overset{2n}{\mathrm{\Σ}}}{X}_{i,k|k-1}^{*}$

$P_{k|k-1}=\frac{1}{2n}\underset{i=1}{\overset{2n}{\mathrm{\Σ}}}{X}_{i,k|k-1}^{*}{X}_{i,k|k-1}^{*T}-{\hat{x}}_{k|k-1}{\hat{x}}_{k|k-1}^T+Q_{k-1}$

6. to state one-step prediction covariance matrix P _k|k-1carry out factorization, that is:

$P_{k|k-1}=S_{k|k-1}{S}_{k|k-1}^T$

7. computed volume point

$X_{i,k|k-1}=S_{k|k-1}{\mathrm{\ξ}}_i+{\hat{x}}_{k|k-1}$

8. propagate volume point

Z _i,k|k-1＝h(X _i,k|k-1)

9. measuring value prediction

${\hat{z}}_{k|k-1}=\frac{1}{2}\underset{i=1}{\overset{2n}{\mathrm{\Σ}}}{Z}_{i,k|k-1}$

10. newly cease covariance matrix

$P_{k|k-1}=\frac{1}{2n}\underset{i=1}{\overset{2n}{\mathrm{\Σ}}}{Z}_{i,k|k-1}{Z}_{i,k|k-1}^T-{\hat{z}}_{k|k-1}{\hat{z}}_{k|k-1}^T+R_k$

cross-covariance is estimated

$P_{\mathrm{xz},k|k-1}=\frac{1}{2n}\underset{i=1}{\overset{2n}{\mathrm{\Σ}}}{X}_{i,k|k-1}{Z}_{i,k|k-1}^T-{\hat{x}}_{k|k-1}{\hat{z}}_{k|k-1}^T$

filter gain is calculated

$K_k=P_{\mathrm{xz}.k|k-1}{P}_{\mathrm{zz},k|k-1}^{-1}$

state covariance matrix upgrades

$P_{k|k}=P_{k|k-1}-K_k{P}_{\mathrm{zz},k|k-1}{K}_k^T$

calculate k state estimation value constantly

${\hat{x}}_{k|k}={\hat{x}}_{k|k-1}-K_k(z_k-{\hat{z}}_{k|k-1})$

The estimated result of (5) six road correlated results is through carrying out filtering through Loop filter and carrier wave ring wave filter respectively, after filtering, respectively as the input of C/A yardage controlled oscillator and carrier wave number of rings controlled oscillator, and then realize the quick accurate tracking of carrier frequency and phase place, code phase and frequency.

Claims (9)

1. a GNSS receiver utmost point weak signal tracking, is characterized in that comprising the steps: in GNSS receiver tracking circuit, first, will pass through the GNSS digital medium-frequency signal s of frequency mixer _iF(n), on I branch road, copy carrier multiplication with sine, on Q branch road, copy carrier multiplication with cosine; Then the mixing multiplied result of I branch road and Q branch road is carried out to related operation with leading, the instant and hysteresis C/A code that code ring copies respectively, the multichannel correlated results of related operation output is sent into respectively to multichannel integration-remover and carry out coherent integration; Again the multichannel coherent integration value of multichannel integration-remover output is sent into the estimation that volume Kalman filter is carried out correlation parameter; Finally, corresponding estimated value is sent into respectively to Loop filter and carrier wave ring wave filter carries out filtering, after filtering, feed back to respectively carrier number controlled oscillator and C/A yardage controlled oscillator, and then realize the real-time adjusting of carrier phase and carrier frequency, code phase and code frequency, finally realize the tracking of GNSS signal.

2. GNSS receiver utmost point weak signal tracking according to claim 1, is characterized in that: GNSS digital medium-frequency signal s _iF(n) carry out Frequency mixing processing with the carrier wave circle replication carrier signal of 90 ° of two-way phase differences respectively, wherein GNSS digital medium-frequency signal s _iF(n) be expressed as:

y(t _n)＝A(t _n)C(t _n-τ(t _n))D(t _n-τ(t _n))sin(w _IFt _n-φ(t _n))+v _n

In formula: t _nrepresent that discrete sampling constantly; y(t _n) representative digit intermediate-freuqncy signal s _iF(n); A(t _n) represent the amplitude of carrier wave; C(t _n) represent the C/A code that satellite is broadcast, D (t _n) represent the numeric data code that satellite sends, and both level values may be only ± 1; τ (t _n) represent the propagation delay of signal; w _iFrepresentative digit intermediate frequency angular frequency; φ (t _n) represent the carrier phase of signal; v _nrepresent that average is zero Gaussian sequence.

3. GNSS receiver utmost point weak signal tracking according to claim 1, is characterized in that, output multi-channel correlated results, completes in the steps below:

1. the mixing multiplied result of I branch road and Q branch road is carried out to related operation with leading, the instant and hysteresis C/A code that code ring copies respectively:

$\left\{\begin{array}{cc}\begin{array}{c}{I}_k\left(\mathrm{\δ}\right)=\underset{n=n_k+1}{\overset{n_k+N_k}{\mathrm{\Σ}}}y\left(t_n\right)\·\left(C_{\mathrm{NCO}}(t_n+\mathrm{\δ}-t_{\mathrm{NCOk}})\sin \right[w_{\mathrm{IF}}{t}_n+{\mathrm{\φ}}_{\mathrm{NCO}}\left(t_n\right)\left]\right)\\ Q_k\left(\mathrm{\δ}\right)=\underset{n=n_k+1}{\overset{n_k+N_k}{\mathrm{\Σ}}}y\left(t_n\right)\·\left(C_{\mathrm{NCO}}(t_n+\mathrm{\δ}-t_{\mathrm{NCOk}})\cos \right[w_{\mathrm{IF}}{t}_n+{\mathrm{\φ}}_{\mathrm{NCO}}\left(t_n\right)\left]\right)\end{array}& \mathrm{\δ}=-\frac{{\mathrm{\Δ}}_{e\_l}}{2}\end{array},,0,\frac{{\mathrm{\Δ}}_{e\_l}}{2}\right.$

In formula: k represents current residing position of the 1ms time interval; N represents the sampling instant in the 1ms time interval; N _kthe number of samples that represents every 1ms time interval; C _nCO(t _n) represent the PRN replica code of the tracked satellite-signal produced by receiver; t _nCOkrepresent that code NCO produces the initial time of instantaneous code; φ _nCO(t _n) represent the carrier phase that receiver copies.Δ _{e_l}represent the spacing between leading and lag correlation device; δ represents the phase difference between copied San road C/A code and instantaneous code; I _k(δ) and Q _k(δ) represent respectively the relevant accumulated value of I branch road and Q branch road; Wherein, when time, I _k(δ), Q _k(δ) represent that respectively the mixing results of I branch road and Q branch road and the advanced code that code ring copies carry out related operation; When δ=0, I _k(δ), Q _k(δ) represent that respectively the mixing results of I branch road and Q branch road and the instantaneous code that code ring copies carry out related operation; When time, I _k(δ) represent that the mixing results of I branch road and Q branch road and the hysteresis code that code ring copies carry out related operation;

2. the correlated results 1. being carried out after related operation by step is expressed as:

$\left\{\begin{array}{cc}\begin{array}{c}{I}_k\left(\mathrm{\δ}\right)=\frac{N_k{\stackrel{\‾}{A}}_k{D}_m}{2}\cos \left({\mathrm{\Δ\φ}}_k\right)R({\mathrm{\Δt}}_k+\mathrm{\δ})+v_{\mathrm{Ik}}\\ Q_k\left(\mathrm{\δ}\right)=\frac{N_k{\stackrel{\‾}{A}}_k{D}_m}{2}\sin \left({\mathrm{\Δ\φ}}_k\right)R({\mathrm{\Δt}}_k+\mathrm{\δ})+v_{\mathrm{Qk}}\end{array}& \mathrm{\δ}=-\frac{{\mathrm{\Δ}}_{e\_l}}{2}\end{array},0,\frac{{\mathrm{\Δ}}_{e\_l}}{2}\right.$

In formula: the mean value that represents the cumulative time period intercarrier amplitude of whole 1ms, D _mrepresent navigation data bit, Δ φ _krepresent cumulative time period intercarrier phase error phi (the t)-φ of 1ms _nCO(t) mean value, Δ t _k=τ (t _midk)-t _midkthe code phase error that represents 1ms intermediate point in the cumulative time period, and t _midk=(t _nCOk+ t _nCOk+1) 2; v _ikand v _qkrepresent respectively the mutual incoherent Gaussian sequence of zero-mean; R (Δ t _k+ δ) represent the autocorrelation function of PRN code;

3. multichannel correlated results can be expressed as:

$\left\{\begin{array}{c}{i}_E=I_k(-\frac{{\mathrm{\Δ}}_{e\_l}}{2})\\ i_P=I_k\left(0\right)\\ i_L=I_k\left(\frac{{\mathrm{\Δ}}_{e\_l}}{2}\right)\\ q_E=Q_k(-\frac{{\mathrm{\Δ}}_{e\_l}}{2})\\ q_P=Q_k\left(0\right)\\ q_L=Q_k\left(\frac{{\mathrm{\Δ}}_{e\_l}}{2}\right)\end{array}\right.$

In formula: i _e, i _pand i _lthe correlation that represents respectively leading on I branch road, instant and hysteresis branch road; q _e, q _prepresent respectively on Q branch road the correlation of leading, instant and hysteresis branch road with qL.

4. GNSS receiver utmost point weak signal tracking according to claim 3, is characterized in that, multichannel coherent integration value is sent into volume Kalman filter, completes the estimation of correlation parameter, comprising: volume Kalman filter system state equation x _m+1=f _m(x _m, w _m) with the foundation of measurement equation; In formula: $x_m={\left[\begin{array}{ccccc}{x}_{{\mathrm{\φ}}_m}& x_{w_m}& x_{{\mathrm{\α}}_m}& A_m& t_{s_m}\end{array}\right]}^T,$ For the state vector of system, m represents integration interval zero hour, and selecting integral time is here 20ms, represent the poor of the m carrier phase that truly carrier phase and receiver carrier wave NCO copy constantly; represent the constantly true carrier doppler drift of m; represent m carrier doppler drift constantly rate of change; A _mrepresent m carrier amplitude constantly; represent m code phase constantly; x _m+1represent m+1 system state vector constantly; f _mrepresent nonlinear state transfer function; w _mrepresent system noise vector.

5. GNSS receiver utmost point weak signal tracking according to claim 4, is characterized in that, the expression of state equation is:

${\left[\begin{array}{c}{x}_{\mathrm{\φ}}\\ x_w\\ x_{\mathrm{\α}}\end{array}\right]}_{m+1}=\left[\begin{array}{ccc}1& {\mathrm{\δt}}_m& {\mathrm{\δt}}_m^2/2\\ 0& 1& {\mathrm{\δt}}_m\\ 0& 0& 1\end{array}\right]{\left[\begin{array}{c}{x}_{\mathrm{\φ}}\\ x_w\\ x_{\mathrm{\α}}\end{array}\right]}_m-\left[\begin{array}{c}{\mathrm{\δt}}_m\\ 0\\ 0\end{array}\right]W_{{\mathrm{NCO}}_m}+\left[\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 0\\ 0& 0& 1& 0\end{array}\right]w_{{\mathrm{\φ}}_m}$

$t_{s_{m+1}}=t_{s_m}+\frac{w_{L1}{\mathrm{\δt}}_{\mathrm{nom}}-\left[\begin{array}{cccc}1& 0& 0& 0\end{array}\right]w_{{\mathrm{\φ}}_m}}{w_{L1}+x_{w_m}+0.5{\mathrm{\δt}}_{\mathrm{nom}}{x}_{{\mathrm{\α}}_m}}+w_{{\mathrm{ts}}_m}$ f

$A_{m+1}=A_m+w_{A_m}$

In formula: represent the cumulative time period; the value that represents the carrier doppler frequency displacement that carrier wave NCO produces within the cumulative time period; w _l1represent L1 carrier angular frequencies; and represent that average is zero, mutual incoherent Gaussian sequence.

6. GNSS receiver utmost point weak signal tracking according to claim 4, is characterized in that, gets the instant coherent integration value of 20ms, 20ms subtracts hysteresis coherent integration value in advance as measurement amount, sets up measurement equation:

$\begin{array}{c}{z}_m=\frac{1}{{\mathrm{\σ}}_n}\sqrt{\frac{2}{N_m}}\left[\begin{array}{c}\underset{k=k_m}{\overset{k_m+19}{\mathrm{\Σ}}}{I}_k\left(0\right)\\ \underset{k=k_m}{\overset{k_m+19}{\mathrm{\Σ}}}{Q}_k\left(0\right)\\ \frac{1}{\sqrt{\mathrm{\η}}}\underset{k=k_m}{\overset{k_m+19}{\mathrm{\Σ}}}[I_k\left(\frac{{\mathrm{\Δ}}_{E\_L}}{2}\right)-I_k(-\frac{{\mathrm{\Δ}}_{E\_L}}{2})]\\ \frac{1}{\sqrt{\mathrm{\η}}}\underset{k=k_m}{\overset{k_m+19}{\mathrm{\Σ}}}[Q_k\left(\frac{{\mathrm{\Δ}}_{E\_L}}{2}\right)-Q_k(-\frac{{\mathrm{\Δ}}_{E\_L}}{2})]\end{array}\right]\\ =D_m{h}_m({\mathrm{\Δ\φ}}_m,{\mathrm{\Δt}}_m,{\stackrel{\‾}{A}}_m)+v_{\mathrm{zm}}\end{array}=\frac{{\stackrel{\‾}{A}}_m{D}_m}{{\mathrm{\σ}}_n}\sqrt{\frac{N_m}{2}}\left[\begin{array}{c}\cos \left({\mathrm{\Δ\φ}}_m\right)R\left({\mathrm{\Δt}}_m\right)\\ \sin \left({\mathrm{\Δ\φ}}_m\right)R\left({\mathrm{\Δt}}_m\right)\\ \frac{1}{\sqrt{\mathrm{\η}}}\cos \left({\mathrm{\Δ\φ}}_m\right)R_{E\_L}\left({\mathrm{\Δt}}_m\right)\\ \frac{1}{\sqrt{\mathrm{\η}}}\sin \left({\mathrm{\Δ\φ}}_m\right)R_{E\_L}\left({\mathrm{\Δt}}_m\right)\end{array}\right]+v_{\mathrm{zm}}$ In formula: every the number of samples of 20ms $N_m=(N_{k_m}+N_{k_{m+1}}+\·\·\·N_{k_{m+19}});\mathrm{\η}=-2[1-R\left({\mathrm{\Δ}}_{E\_L}\right)];{\stackrel{\‾}{A}}_m$ The mean value that represents 20ms carrier amplitude; Δ φ _mthe mean value that represents 20ms carrier phase difference; the code phase error that represents intermediate point; represent to subtract in advance lag correlation function; the related function that represents advanced code; the related function that represents hysteresis code; h _mrepresent to measure function; ν _zmthe Gaussian sequence that represents zero-mean.

7. GNSS receiver utmost point weak signal tracking according to claim 6, is characterized in that, expression every in measurement equation is:

${\mathrm{\Δ\φ}}_m=\left[\begin{array}{ccc}1& {\mathrm{\δt}}_m/2& {\mathrm{\δt}}_m^2/6\end{array}\right]{\left[\begin{array}{c}{x}_{\mathrm{\φ}}\\ x_w\\ x_{\mathrm{\α}}\end{array}\right]}_m-\frac{{\mathrm{\δt}}_m}{2}+\left[\begin{array}{cccc}0& 0& 0& 1\end{array}\right]w_{\mathrm{\φm}}$

${\mathrm{\Δt}}_m=(t_{s_{m+1}}+t_{s_m})/2-t_{{\mathrm{mid}}_m}=t_{s_m}+\frac{1}{2}\frac{w_{L1}{\mathrm{\δt}}_{\mathrm{nom}}-\left[\begin{array}{cccc}1& 0& 0& 0\end{array}\right]w_{{\mathrm{\φ}}_m}}{w_{L1}+x_{w_m}+0.5{\mathrm{\δt}}_{\mathrm{nom}}{x}_{{\mathrm{\α}}_m}}+\frac{1}{2}{w}_{{\mathrm{ts}}_m}-t_{{\mathrm{mid}}_m}$

${\stackrel{\‾}{A}}_m=(A_{m+1}+A_m)/2=A_m+0.5w_{A_m}.$

8. weak signal GNSS receiver tracking method according to claim 4, it is characterized in that, mathematical model according to said system equation and measurement equation foundation, designs corresponding volume Kalman filter, comprising the time, upgrades with measurement and upgrades two primary iteration processes.

9. weak signal GNSS receiver tracking method according to claim 4, it is characterized in that, the estimated result of multichannel correlated results is through carrying out filtering through Loop filter and carrier wave ring wave filter respectively, after filtering, as the input of C/A yardage controlled oscillator and carrier wave number of rings controlled oscillator, realize the quick accurate tracking of carrier frequency and phase place, code phase and frequency respectively.