CN106873003A - A kind of dual mode receiver RAIM methods based on the accumulation of many epoch parity vectors - Google Patents

Abstract

The invention discloses a kind of dual mode receiver RAIM methods based on the accumulation of many epoch parity vectors, the method constructs the observational equation of the Big Dipper/GPS dual-mode receiver first, and solve parity vector using typical odd_even adjudgement rule, then by the quadratic sum of many moment parity vectors in the sliding window that adds up, new statistic mixed-state amount is constructed with this, finally set fault threshold to judge new statistic mixed-state amount, be out of order so as to detect, and then recognize out of order satellite pseudorange.Dual mode receiver RAIM method of the present invention based on the accumulation of many epoch parity vectors, can effectively reject the Big Dipper/small soft fault of GPS dual-mode receiver, be that follow-up high accuracy positioning provides safeguard.

Description

A kind of dual mode receiver RAIM methods based on the accumulation of many epoch parity vectors

Technical field

The present invention relates to a kind of dual mode receiver RAIM methods based on the accumulation of many epoch parity vectors, belong to dual mode reception Machine fault detection technique field.

Background technology

The conventional single star fault detection algorithm based on dual mode receiver has at present：Odd_even adjudgement rule, least square method etc. " snapshot " formula detection method, such method has good Detection results for larger pseudo range observed quantity deviation, but for small Soft fault Detection results are not obvious.And there is little deviation in single satellite pseudorange, though will not as satellite catastrophe failure when occur Very big pseudorange biases cause user's positioning result completely unavailable, but can also make position error precise decreasing, influence positioning precision. Therefore in Satellite Navigation Technique, dual mode receiver is very important with rejecting to small pseudorange Fault Identification.

The content of the invention

The technical problems to be solved by the invention are：A kind of dual mode receiver based on the accumulation of many epoch parity vectors is provided RAIM methods, using typical odd_even adjudgement rule produce parity vector, add up sliding window in many moment parity vectors square With, new statistic mixed-state amount is constructed with this, can effectively reject the Big Dipper/small soft fault of GPS dual-mode receiver.

The present invention uses following technical scheme to solve above-mentioned technical problem：

A kind of dual mode receiver RAIM methods based on the accumulation of many epoch parity vectors, comprise the following steps：

Step 1, according to the characteristic of the Big Dipper/GPS dual-mode receiver, sets up the observation at the Big Dipper/GPS dual-mode receiver each moment Equation；

Step 2, by the observational equation at step 1 each moment, calculates the odd even arrow at the Big Dipper/GPS dual-mode receiver each moment Amount；

Step 3, the time span for setting sliding window is t, the satellite that the kth moment Big Dipper/GPS dual-mode receiver is received Number is n, if kth -1, k-2 ..., k-t reception to number of satellite be n, then by kth -1, k-2 ..., the k-t moment The cumulative summation of parity vector, obtain the detection statistic at current time；If kth -1, k-2 ..., k-i, 1≤i≤t connect The number of satellite for receiving is n, and the kth-i-1 moment is not n, then by kth -1, k-2 ..., k-i, the odd even of 1≤i≤t The cumulative summation of vector, obtains the detection statistic at current time；The detection statistic at current time is carried out into measuring noise square difference After normalization, breakdown judge detection limit is obtained；

Step 4, sets fault threshold, and breakdown judge detection limit is judged according to fault threshold, when breakdown judge inspection When measurement is more than or equal to fault threshold, represent that detection is out of order, while being identified to fault satellites；When breakdown judge detection limit During less than fault threshold, expression is not detected is out of order.

As a preferred embodiment of the present invention, the observational equation at the Big Dipper described in step 1/GPS dual-mode receiver each moment Formula is：

y_k=H_kx_k+ε

Wherein, y_kFor the kth moment observes pseudorange and calculates the difference of pseudorange, n n dimensional vector ns；H_kIt is kth moment observed differential square Battle array, n*5 dimensions；x_kIt is the kth moment Big Dipper/GPS dual-mode receiver state amount；ε is n dimension observation pseudorange noise vectors；N be the Big Dipper/ The number of satellite that GPS dual-mode receiver is received.

Used as a preferred embodiment of the present invention, the detailed process of the step 2 is：

To kth moment observed differential matrix H_kDo QR decomposition：

H_k=Q_kR_k

Wherein, Q_kIt is kth moment orthogonal matrix, n*n dimensions；R_kIt is kth moment upper triangular matrix, n*5 dimensions；

Above formula is substituted into observational equation and premultiplication Q_k ^T, obtain Q_k ^Ty_k=R_kx_k+ ε, T represent transposition；

To Q_k ^TPiecemeal is carried out by row,Wherein,It is Q_k ^TPreceding 5 row,It is Q_k ^TN-5 rows,For strange Even transformation matrix；

Pass throughBy y_kIt is converted into odd even parity vector p spatially_k：

Wherein,It is matrixThe s column vectors of pseudorange biases, 1≤s≤n, B_sIt is failure constant, H₀、H₁Difference table Show fault-free, faulty.

As a preferred embodiment of the present invention, the detection statistic at current time is carried out into measurement noise described in step 3 After normalized square mean, the formula for obtaining breakdown judge detection limit is：

Wherein, fd represents breakdown judge detection limit, sum_pIt is the detection statistic at current time, t is the sliding window of setting The time span of mouth, σ²It is measuring noise square difference.

Used as a preferred embodiment of the present invention, the formula of fault threshold is described in step 4：

Wherein, fd represents breakdown judge detection limit, p_rIt is fault threshold, a is setting probability, and x connects for the Big Dipper/GPS dual-mode Receipts machine quantity of state, n is the number of satellite that the Big Dipper/GPS dual-mode receiver is received, P_FAIt is false-alarm probability.

The present invention uses above technical scheme compared with prior art, with following technique effect：

1st, the present invention improves sensitivity of the dual mode receiver for the slowly varying detection of small pseudorange, strange using many epoch The cumulative stability that can improve system of even vector, is that follow-up high accuracy positioning calculating is laid a good foundation.

2nd, the Big Dipper/GPS dual-mode receiver that the present invention is used, its certain moment usable satellite quantity, positioning precision are than list One receiver is outstanding, and odd_even adjudgement rule availability is higher.

3rd, applicability of the present invention extensively, is not limited by carrier.

Brief description of the drawings

Fig. 1 is the principle flow chart of dual mode receiver RAIM method of the present invention based on the accumulation of many epoch parity vectors.

Specific embodiment

Embodiments of the present invention are described below in detail, the example of the implementation method is shown in the drawings.Below by The implementation method being described with reference to the drawings is exemplary, is only used for explaining the present invention, and is not construed as limiting the claims.

The present invention on the basis of original dual mode receiver typical case's RAIM algorithms, for the spy of the Big Dipper/GPS dual-mode receiver Property, it is proposed that the method that many epoch parity vectors of the one kind based on the Big Dipper/GPS dual-mode receiver are accumulated, its object is to improve The Big Dipper/GPS dual-mode receiver is for the sensitivity of gradual pseudorange fault detect, and specific steps are as shown in Figure 1.

Based on the dual mode receiver RAIM methods of many epoch parity vectors accumulation, main devices are that the Big Dipper/GPS dual-mode is received Machine, specifically comprises the following steps：

Step 1, constructs the observational equation of the Big Dipper/GPS dual-mode receiver.

Because the Big Dipper that dual mode receiver is received is different from GPS satellite navigation system clock correction, therefore can be by the two clock correction Equivalent respective distance is set to quantity of state x=[Δ x, Δ y, Δ z, δ_b,δ_g], Δ x, Δ y, Δ z represents that the Big Dipper/GPS dual-mode connects The differential of receipts machine coordinate, δ_b、δ_gRespectively represent the Big Dipper, GPS kind it is poor.Assuming that the satellite number that certain moment receives is n, The observational equation of the Big Dipper/GPS dual-mode receiver is as follows：

y_k=H_kx_k+ε (1)

Wherein, y_kFor the kth moment observes pseudorange and calculates the difference of pseudorange, n n dimensional vector ns, x_kIt is the kth moment Big Dipper/GPS dual-mode Receiver state amount, ε is n dimension observation pseudorange noise vectors, H_kFor n*5 dimension observed differential matrix expressions are：

Wherein, it is direction cosines value between user and satellite that preceding 3 row are corresponding, the 4th with the 5th arrange respectively the corresponding Big Dipper, The receiver clock-offsets coefficient correlation of GPS navigation system.Assuming that observing j Beidou navigation satellite, n-j gps satellite.In formula Respectively satellite j measurement vector length at receiver The value of [x, y, z] place local derviation, andIt is user's estimated location to the several of satellite j What distance.

The observational equation that can be set up according to step 1, solves the parity vector at per moment in subsequent step 2.

Step 2, solves the Big Dipper/GPS dual-mode receiver per moment parity vector.

According to the receiver observational equation that step 1 is set, the Vector Message in odd even spatially is built according to below equation, To front construction observed differential matrix H_kQR decomposition is carried out, formula is as follows：

H_k=Q_kR_k (3)

Wherein, Q_kIt is kth moment orthogonal matrix, n*n dimensions；R_kIt is kth moment upper triangular matrix, n*5 dimensions；

According to Q_kThe orthogonality of battle array, formula (1) and premultiplication Q are substituted into by formula (3)_k ^T, formula (1) can be changed into formula (4)：

Q_k ^Ty_k=R_kx_k+ε (4)

To transposed matrix Q_k ^TPiecemeal is carried out, is tieed up by Factorization algorithm n and observation space is divided into 5 dimension state spaces and n-5 Dimension odd even space：

Wherein,It is Q_k ^TPreceding 5 row,It is Q_k ^TN-5 rows,It is parity transformation matrix；

Pass throughN dimension measurement vectors are converted into odd even parity vector pk spatially：

In formula,It is matrixThe s column vectors of pseudorange biases, 1≤s≤n, B_sIt is failure constant, H₀、H₁Difference table Show fault-free, faulty, it can thus be concluded that the parity vector p for going out_k, it is observation error ε projections in odd even spatially, can be anti- The information such as failure deviation are reflected, general tradition odd_even adjudgement rule is by building per moment parity vector | p_k|²Construction detection statistic 's.

Step 3, the parity vector accumulated in the corresponding time builds detection statistic.

The present invention constructs the sliding window of certain hour length, and added up corresponding parity vector with this, according to window The statistical property of interior parity vector sum come determine window end measure fault message.Here length of window sets according to the actual requirements Put.

In actual conditions, many times the cumulative time do not arrive the length of window for setting, or during cumulative, no Different number of satellite are observed in the same time, such situation, that is, the corresponding time odd even arrow before the number of satellite that adds up change occur Amount, subsequent time carries out the fault detect of downstream, and specific formula is as follows：

After new parity vector is normalized according to measuring noise square difference, fault statistics function is built：

Wherein, fd represents breakdown judge detection limit, sum_pIt is the detection statistic at current time, t is the sliding window of setting The time span of mouth, σ²It is measuring noise square difference.

Step 4, sets fault threshold, judges whether failure, and detection is identified to fault satellites.

According to the fault detect statistic tried to achieve in step 3, fault detect thresholding is built.Broken down after normalization The Big Dipper/GPS dual-mode receiver detection probability is distributed as the non-central χ of free degree n-5²Distribution, due to no matter window size how much, Detection statistic obeys χ²Distribution, so the detection threshold under its permanent false alarm rate is also consistent, i.e. false-alarm detection threshold is：

Wherein, fd represents breakdown judge detection limit, p_rIt is fault threshold, a is setting probability, and x connects for the Big Dipper/GPS dual-mode Receipts machine quantity of state, n is the number of satellite that the Big Dipper/GPS dual-mode receiver is received, P_FAIt is false-alarm probability.

If fd >=p_r, represent that detection is out of order, it is necessary to further recognize out of order satellite pseudorange；If fd<p_r, represent not Detection is out of order, into follow-up dual mode receiver positioning flow.

Parity transformation matrix Q is tried to achieve in step 2^pIn, each column vector one satellite of correspondence, the referred to as spy of the satellite Vector is levied, the parity vector sum that step 3 statistics is obtained_pProject on every characteristic vector of satellite, deviate farthest being Fault satellites：

Wherein, D is observation vector, its pseudorange biases that can be used to count each satellite as detection limit, in observation vector The maximum element of numerical value corresponds to the satellite number for breaking down, Q^pIt is parity transformation matrix, σ is poor to measure noise criteria.

Above example is only explanation technological thought of the invention, it is impossible to limit protection scope of the present invention with this, every According to technological thought proposed by the present invention, any change done on the basis of technical scheme each falls within the scope of the present invention Within.

Claims (5)

1. it is a kind of based on many epoch parity vectors accumulation dual mode receiver RAIM methods, it is characterised in that comprise the following steps：

Step 1, according to the characteristic of the Big Dipper/GPS dual-mode receiver, sets up the observation side at the Big Dipper/GPS dual-mode receiver each moment Journey；

Step 2, by the observational equation at step 1 each moment, calculates the parity vector at the Big Dipper/GPS dual-mode receiver each moment；

Step 3, the time span for setting sliding window is t, the number of satellite that the kth moment Big Dipper/GPS dual-mode receiver is received Be n, if kth -1, k-2 ..., k-t reception to number of satellite be n, then by kth -1, k-2 ..., the k-t moment it is strange The cumulative summation of even vector, obtains the detection statistic at current time；If kth -1, k-2 ..., k-i, 1≤i≤t are received Number of satellite be n, and the kth-i-1 moment is not n, then by kth -1, k-2 ..., k-i, the parity vector of 1≤i≤t Cumulative summation, obtains the detection statistic at current time；The detection statistic at current time is carried out into measuring noise square difference normalizing After change, breakdown judge detection limit is obtained；

Step 4, sets fault threshold, breakdown judge detection limit is judged according to fault threshold, when breakdown judge detection limit During more than or equal to fault threshold, represent that detection is out of order, while being identified to fault satellites；When breakdown judge detection limit is less than During fault threshold, expression is not detected is out of order.

2. the dual mode receiver RAIM methods for being accumulated based on many epoch parity vectors according to claim 1, it is characterised in that The observational equation formula at the Big Dipper described in step 1/GPS dual-mode receiver each moment is：

y_k=H_kx_k+ε

Wherein, y_kFor the kth moment observes pseudorange and calculates the difference of pseudorange, n n dimensional vector ns；H_kIt is kth moment observed differential matrix, n*5 Dimension；x_kIt is the kth moment Big Dipper/GPS dual-mode receiver state amount；ε is n dimension observation pseudorange noise vectors；N is the Big Dipper/GPS dual-mode The number of satellite that receiver is received.

3. the dual mode receiver RAIM methods for being accumulated based on many epoch parity vectors according to claim 2, it is characterised in that The detailed process of the step 2 is：

To kth moment observed differential matrix H_kDo QR decomposition：

H_k=Q_kR_k

Wherein, Q_kIt is kth moment orthogonal matrix, n*n dimensions；R_kIt is kth moment upper triangular matrix, n*5 dimensions；

Above formula is substituted into observational equation and premultiplication Q_k ^T, obtain Q_k ^Ty_k=R_kx_k+ ε, T represent transposition；

To Q_k ^TPiecemeal is carried out by row,Wherein,It is Q_k ^TPreceding 5 row,It is Q_k ^TN-5 rows,For odd even becomes Change matrix；

Pass throughBy y_kIt is converted into odd even parity vector p spatially_k：

$p_k=Q_k^p{y}_k=\left\{\begin{array}{cc}{Q}_k^p\mathrm{\&epsiv;}& H_0\\ Q_{k:,s}^p{B}_s+Q_k^p\mathrm{\&epsiv;}& H_1\end{array}\right.$

Wherein,It is matrixThe s column vectors of pseudorange biases, 1≤s≤n, B_sIt is failure constant, H₀、H₁Indicate respectively without It is failure, faulty.

4. the dual mode receiver RAIM methods for being accumulated based on many epoch parity vectors according to claim 1, it is characterised in that After the detection statistic at current time is carried out into measuring noise square difference normalization described in step 3, breakdown judge detection limit is obtained Formula is：

$fd=\frac{{\left|{\mathrm{sum}}_p\right|}^2}{{\mathrm{t\&sigma;}}^2}$

Wherein, fd represents breakdown judge detection limit, sum_pBe the detection statistic at current time, t be the sliding window of setting when Between length, σ²It is measuring noise square difference.

5. the dual mode receiver RAIM methods for being accumulated based on many epoch parity vectors according to claim 1, it is characterised in that The formula of fault threshold described in step 4 is：

$p_r(fd<a)=\underset{0}{\overset{a}{\&Integral;}}{f}_{x^2(n-5)}\left(x\right)dx=1-P_{FA}$

Wherein, fd represents breakdown judge detection limit, p_rIt is fault threshold, a is setting probability, and x is the Big Dipper/GPS dual-mode receiver shape State amount, n is the number of satellite that the Big Dipper/GPS dual-mode receiver is received, P_FAIt is false-alarm probability.