Navigation system failure detection apparatus and method based on baseband domain assisted raim algorithm

Abstract

The application provides a navigation system fault detection device and method based on a baseband domain auxiliary RAIM algorithm, and relates to the technical field of satellite navigation monitoring.The method comprises the following steps: in the first step, a plurality of correlation values of the nth channel of a receiver baseband domain are acquired, a single-channel multi-path correlation estimation error detection quantity model is established, and then, on the basis of the single-channel multi-path correlation estimation error detection quantity model, an nth-channel multi-path correlation estimation error detection quantity model is established; in the second step, the nth-channel multi-path correlation estimation error detection quantity of the first step is combined with the fusion of the RAIM algorithm, and an extended observation equation of the baseband domain auxiliary RAIM algorithm is established; in the third step, a joint test statistic model of the baseband domain auxiliary RAIM algorithm is established; and in the fourth step, a FPGA and DSP hardware implementation of the baseband domain auxiliary RAIM algorithm is provided.Compared with the traditional algorithm which only exists in theoretical simulation, the application can be implemented by using FPGA and DSP, and has practical application value.

Description

TECHNICAL FIELD

[0001] The present application relates to the satellite navigation monitoring technical field, especially to a navigation system fault detection device and method based on baseband domain assisted RAIM algorithm. BACKGROUND

[0002] Integrity monitoring refers to the ability of the system to give timely alarm information when the satellite positioning error is greater than the set threshold. If the system fails to give such alarm information, it will lead to the reduction of positioning reliability and the occurrence of integrity risk. Receiver autonomous integrity monitoring (RAIM) is a technology for monitoring and identifying satellite faults using redundant observation data.

[0003] With the increasingly wide application of satellite navigation technology, people pay more and more attention to positioning integrity services. At present, RAIM algorithms can be divided into two categories: one is based on filter algorithm, and the other is based on pseudorange residual snapshot algorithm. The basis of existing RAIM research is to base on the measurement and calculation of user receiver measurement domain and navigation calculation domain, and to complete integrity monitoring by consistency judgment. But from the user receiver, integrity monitoring is related to the processing of navigation signal by receiver. In terms of the whole process of receiver processing satellite navigation signal, navigation positioning measurement domain parameters are obtained after navigation baseband signal tracking processing. The existing receiver integrity monitoring algorithm does not consider the influence of receiver intermediate frequency baseband signal processing part. Therefore, the positioning error caused by code phase error of navigation signal in receiver baseband tracking loop is studied, and the baseband tracking processing and integrity monitoring are integrated to improve the integrity monitoring performance of navigation system positioning. The present application assists the fault detection of RAIM algorithm by the monitoring parameters obtained in the receiver intermediate frequency baseband signal processing, establishes a baseband domain assisted RAIM algorithm monitoring model and realizes the hardware, which can improve the sensitivity of fault detection, reliably perform fault alarm and identification, and further improve the integrity performance of navigation positioning. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a navigation system fault detection device and method based on baseband domain assisted RAIM algorithm.

[0005] A navigation system fault detection method based on baseband domain assisted RAIM algorithm, specifically comprising the following steps:

[0006] Step 1: Obtain the multi-path correlation value of the n-th channel in the baseband domain of the satellite navigation receiver, establish an estimation error detection quantity model of single-channel multi-path correlation, and then establish an estimation error detection quantity model of n-channel multi-path correlation based on the estimation error detection quantity model of single-channel multi-path correlation;

[0007] In the satellite navigation receiver, n channels are set, n≥6, and each channel processes the intermediate frequency signal of one satellite. In order to improve the detection accuracy of the code tracking loop of each channel of the receiver, a multi-correlator module is used, which is composed of N correlators, N≥3, and N is an odd number. The detection interval is set as chip, that is, the correlator interval is chip, where The calculated value has only integer, and only the first decimal place is reserved; the multi-path correlation value of the n-th channel obtained by the correlator calculation is According to the slope calculation formula, the correlation value of the navigation baseband signal received by the channel and the multi-path local pseudo code is The slope expression on both sides of the correlation peak is:

[0008]

[0009] Where, k -Δ represents the left slope, k Δ represents the right slope; i=0, 1,..., N-1 represents the index number of the correlator output, and the corresponding multi-path correlation value is That is, when Δ N / 2 =0, it is the instantaneous branch correlation value I p ; when the navigation intermediate frequency signal input by the n-th channel is an ideal signal, the slopes on both sides of the correlation peak are consistent, so the single-channel single-time estimation detection error is established as:

[0010]

[0011] If it is a normal satellite navigation intermediate frequency signal, then If it is an abnormal satellite navigation intermediate frequency signal, then the above formula is used; after removing the coefficients of the estimation detection error and , the single-channel single-time estimation detection error is rewritten as:

[0012]

[0013] The expansion is:

[0014]

[0015] For the estimation detection error of the above formula, the detection relationship between the satellite navigation intermediate frequency signal to be detected and the ideal satellite navigation intermediate frequency signal is established as:

[0016]

[0017] wherein, is the ith detection quantity of the signal to be detected, defined here as the ith detection quantity of the micro-abnormal signal; is the ith detection quantity of the ideal signal, K is the detection quantity of the ideal signal, and P f and the missed detection probability P m is a constant obtained by calculation, σ nor is the standard deviation under the ideal signal; β represents the maximum estimated detection error quantity screened out, and when the detector β is greater than or equal to the set value, it is considered that the signal is abnormal, and the satellite is marked;

[0018] Considering that the baseband domain signal processing time step T1 and the RAIM algorithm monitoring epoch time step T2 are inconsistent, but T2 can be divided by T1, that is, T2 / T1 = λ, and the multi-path correlation estimation error detection quantity β is the monitoring result of a single satellite in the baseband domain, the maximum error quantity of the nth channel multiple estimations, that is, the estimation error detection quantity model α of the single channel multi-path correlation is as follows:

[0019]

[0020] wherein, j = 0, 1,..., λ, representing the number of baseband signal processing under the length of single RAIM detection; β j represents the estimation error detection quantity value of the jth time; α represents the maximum value in the multiple estimations under the T2 detection length;

[0021] Therefore, the estimation error detection matrix of the n channels, that is, the estimation error detection quantity model α of the n-channel multi-path correlation is derived as follows:

[0022]

[0023] wherein, α n represents the maximum estimation error detection quantity of the nth channel;

[0024] Step 2: Combined with the estimation error detection quantity of the n-channel multi-path correlation and the fusion of the RAIM algorithm, the baseband domain assisted RAIM algorithm extended observation equation is established;

[0025] After the receiver completes the carrier and pseudo-code stripping in the baseband domain, the three-dimensional coordinate information (x n , y n , z n), clock error At and pseudo-range error Ap, where n is the satellite number index, because there is only one satellite signal for each channel, so here n represents both the channel number and the number of visible satellites; assuming the number of observable satellites is n, the RAIM observation equation is:

[0026] y = Hx + e

[0027] where y = [Ap 1 Ap 2 ... Ap n ] T , Ap n represents the pseudo-range error of the nth satellite; H n×4 is an n-row 4-column state observation matrix; x = [Ax Ay Az At] T is the receiver horizontal, vertical direction position deviation and clock error; e represents a noise vector subject to Gaussian distribution;

[0028] From the estimated error detection quantity model of step 1 baseband domain multi-correlation, the maximum estimated error of the ranging code phase of each satellite is C dev = [a 1 a 2 ... a n ] T , assuming the pseudo-code rate is c code_rate , then from the definition of pseudo-range, the average pseudo-range deviation of the pseudo-range deviation generated by each satellite within T2 detection time is:

[0029]

[0030] where Ap represents the average pseudo-range deviation caused by the ranging code phase error of n satellites; c represents the speed of light; I 1×n represents a one-dimensional unit row vector of 1 row and n columns, the unit row vector means that the elements are all 1, and here n represents the number of visible satellites; the average pseudo-range deviation vector is constructed as P = [Ap 1 Ap 2 ... Ap n ] T , and the values in the vector P are all Ap, so the reconstructed pseudo-range error quantity after linearization is Z = y-P;

[0031] The extended state observation matrix is reconstructed as:

[0032]

[0033] where w n1 = (x n -x u ) / r n , w n2 = (yn -y u ) / r n , w n3 =(z n -z u ) / r n The three-dimensional position coordinates of the nth satellite are (x n , y n , z n ), the three-dimensional position coordinates of the user receiver are (x u , y u , z u ), and the geometric distance from the nth satellite to the user receiver is The clock difference is Δt = 1; the bias amount is X = [Δx Δy Δz Δt g] T , g = c / c code_rate is an auxiliary observation constant; therefore, the extended observation equation of the baseband domain assisted RAIM algorithm is:

[0034] Z = FX + ε

[0035] wherein ε represents a noise vector subject to a Gaussian distribution.

[0036] Step 3: Calculate the residual amount of the receiver pseudorange observation value by the least square method, and then establish a joint test statistic model of the baseband domain assisted RAIM algorithm;

[0037] The least square solution of the least square method calculation on the extended observation equation of the baseband domain assisted RAIM algorithm in step 2 is:

[0038]

[0039] The pseudorange residual of the user receiver is calculated as:

[0040]

[0041] wherein I is a unit matrix;

[0042] Let U = I - F(F T F) -1 F T , then the pseudorange residual of the user receiver is rewritten as R = Uε; the square sum of the components of the pseudorange residual is calculated as:

[0043] F εε = R T R = ε T Uε

[0044] wherein F εε / σ 2 is subject to a χ 2 distribution with a degree of freedom of (n-4);

[0045] So the joint test statistic model T of baseband domain assisted RAIM algorithm join is:

[0046]

[0047] and the joint test statistic T join The corresponding detection threshold T THR is:

[0048]

[0049] Wherein, σ is the standard deviation of pseudo-range noise; t is the constant calculated by constant false alarm rate P FA ;

[0050] Finally, the joint test statistic T join is compared with the detection threshold T THR If T join <T THR , the satellite navigation system at this moment is fault-free, otherwise, there is a fault, giving an alarm information and fault isolation.

[0051] A navigation system fault detection device based on baseband domain assisted RAIM algorithm, for realizing the navigation system fault detection method based on baseband domain assisted RAIM algorithm, specifically comprising ROM1, ROM2, RAM, FPGA and DSP; the ROM1 and the ROM2 are connected with the FPGA, wherein the ROM1 prewrites the satellite navigation receiver local pseudo code sequence, and the ROM2 prewrites the discrete sequence of the sine function; the RAM is connected with the FPGA and the DSP respectively, and is used for temporarily storing the data amount generated in the FPGA module and the DSP logic calculation process, facilitating subsequent data reading; the FPGA adopts twelve channels, a plurality of correlators in each channel are used for operation, each channel comprises a carrier NCO module, a de-carrier module, a fast capture module, a C / A code generation module and a plurality of correlators, and the FPGA and the DSP are connected through an interactive data bus; the DSP is responsible for the task of logic operation, and comprises text decoding, baseband domain detection quantity calculation, observation data least square calculation, joint detection quantity function calculation, fault detection and fault decision calculation.

[0052] The carrier NCO module calls the sine function information in the ROM2, generates an adjustable single-frequency complex sine signal through the initialization carrier phase provided by the fast capture module, and provides the single-frequency complex sine signal for the de-carrier module;

[0053] The de-carrier module receives the zero intermediate frequency navigation signal and performs complex multiplication operation with the local carrier, moves the intermediate frequency signal to zero frequency, and sends the de-carrier signal into the plurality of correlators.

[0054] The fast capture module captures and discriminates the signal entering the receiver through FFT capture algorithm, and provides the coarse carrier phase and coarse pseudo code phase of the initialization carrier NCO generation module and C / A code generation module;

[0055] The C / A code generation module obtains the pseudo code through the corresponding pseudo code sequence in initialization ROM1, and then generates the local coarse real-time C / A code through the coarse code phase in the fast capture module, and then provides 9 different local time delay C / A codes to the multi-correlator module according to the set correlator interval;

[0056] In the multi-correlator module, the received de-carrier signal is correlated and accumulated with the 9 local time delay C / A codes provided by the C / A code generation module, and 9 correlation values are calculated and stored in RAM1, and the C / A code stripped signal is sent to the DSP for next decoding and decision calculation.

[0057] The beneficial effects produced by the above technical solution are:

[0058] The application provides a navigation system fault detection device and method based on a baseband domain auxiliary RAIM algorithm, which has the following

[0059] Beneficial effects:

[0060] 1. In the integrity monitoring process of the navigation satellite receiver, the influence of the intermediate frequency baseband signal processing part in the receiver is considered. The monitoring parameters in the intermediate frequency baseband signal processing are used to assist the fault detection of the RAIM algorithm, which can improve the sensitivity of fault detection and effectively give alarm information and fault identification.

[0061] 2. The application is no longer limited to theoretical simulation, but can be implemented by FPGA and DSP while providing the theoretical model of the method, and the scheme has practical reference value. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 The detection method of the application is a whole flow chart;

[0063] Figure 2 The monitoring system of the application is a whole block diagram. DETAILED DESCRIPTION

[0064] The specific embodiments of the application will be further described in detail below in combination with the drawings and examples. The following examples are used to illustrate the application, but are not used to limit the scope of the application.

[0065] A navigation system fault detection method based on a baseband domain auxiliary RAIM algorithm, as shown in Figure 1As shown, specifically comprising the following steps:

[0066] Step 1: Obtain the multi-path correlation value of the n-th channel of the satellite navigation receiver baseband domain, establish a single-channel multi-path correlation estimation error detection quantity model, and then establish an n-channel multi-path correlation estimation error detection quantity model based on the single-channel multi-path correlation estimation error detection quantity model;

[0067] In the satellite navigation receiver, n = 1, 2,..., 12 channels are set, and each channel processes the intermediate frequency signal of one satellite. In order to improve the detection accuracy of the code tracking loop of each channel of the receiver, a multi-correlator module is used, which is composed of N (N ≥ 3, and N is an odd number) correlators. The number of correlators depends on the processing capability of FPGA and DSP. The more correlators used, the more conducive to improving the detection accuracy of the code tracking loop. In this embodiment, 9 correlators are taken as an example, wherein the detection interval is Δ i = -0.4:0.1:0.4 chip, that is, the correlator interval is d = 0.1 chip, and the multi-path correlation value of the n-th channel is obtained by correlator calculation According to the slope calculation formula, the correlation value of the navigation baseband signal received by the channel and the multi-path local pseudo code is The slope expression on both sides of the correlation peak is:

[0068]

[0069] Wherein, k -Δ represents the left slope, k Δ represents the right slope; i = 0, 1,..., 8, the index number of the correlator output, and the corresponding multi-path correlation value is That is, when Δ4 = 0, it is the real-time branch correlation value I p ; the correlator interval is d = 0.1 chip; when the navigation intermediate frequency signal input by the n-th channel is an ideal signal, the slopes on both sides of the correlation peak are consistent, so the single-channel single-time estimation detection error is established as:

[0070]

[0071] If it is a normal satellite intermediate frequency signal, then If it is an abnormal satellite intermediate frequency signal, it is the above original formula; after removing the coefficients of the estimation detection error and , the single-channel single-time estimation detection error is rewritten as:

[0072]

[0073] The expansion is:

[0074]

[0075] The estimated detection error of the above formula is established as a detection relationship between the navigation intermediate frequency signal to be detected and the ideal navigation intermediate frequency signal:

[0076]

[0077] wherein, is the ith detection quantity of the signal to be detected, which is defined here as the ith detection quantity of the slight abnormal signal; is the ith detection quantity of the ideal signal, and K is a constant calculated by the false alarm probability P f and the missed detection probability P m , and P f = 1.5 x 10 -7 and P m = 1.0 x 10 -3 , the value of K is 8.35; σ nor is the standard deviation under the ideal signal; and β represents the maximum estimated detection error quantity screened out, and in this embodiment, when the detector β ≥ 1, it is considered that the signal is abnormal, and the satellite is marked;

[0078] The slight abnormal distortion signal: this definition is initially used to describe the abnormal distortion signal generated by the GPS SVN 19 satellite in 1993, and for the convenience of studying the influence of the abnormal distortion of the signal waveform, the International Civil Aviation Organization (ICAO) divides the abnormal distortion signal model into three types: digital distortion (TMA), analog distortion (TMB), and mixed distortion (TMC). Since the signal distortion is less than 1 chip error required by the receiver to pull it into the tracking loop, it is called the satellite slight abnormal distortion signal. The abnormal signal model index is as follows:

[0079] Table 1 ICAO signal abnormal model technical index

[0080] f d (MHz) σ (MNeper / s) Δ (chip) TMA 0 0 -0.12 < Δ < 0.12 TMB 4 < f d ≤ 17 0.8 < σ < 8.8 0 TMC 7.3 < f d ≤ 13 0.8 < σ < 8.8 -0.12 < Δ < 0.12

[0081] Considering that the baseband signal processing time step T1 and the RAIM algorithm monitoring epoch time step T2 are inconsistent, but T2 can be divided by T1, that is, T2 / T1 = λ, and the multiple correlation estimated error detection quantity β is the monitoring result of a single satellite in the baseband domain, the maximum error quantity of the nth channel multiple estimation, that is, the estimated error detection quantity model α of the single channel multiple correlation is as follows:

[0082]

[0083] wherein, j = 0, 1,..., λ, represents the number of baseband signal processing times under the length of a single RAIM detection; β jrepresents the estimated error detection value of the jth; a represents the maximum value of the estimated error in the T2 detection duration;

[0084] Therefore, the estimated error detection matrix of n channels is derived, that is, the estimated error detection value model of n-channel multi-path correlation is:

[0085]

[0086] wherein a n represents the maximum estimated error detection value of the nth channel; n represents the number of receiver channels, n = 1, 2,..., 12.

[0087] Step 2: Combine the estimated error detection value of n-channel multi-path correlation and RAIM algorithm fusion to establish the extended observation equation of baseband domain assisted RAIM algorithm;

[0088] After the receiver completes the carrier and pseudo-code stripping in the baseband domain, the data code is solved to obtain the three-dimensional coordinate information (x n , y n , z n ) of the satellite, clock error Δt and pseudo-range error Δρ, wherein n = 1, 2,..., 12 is the satellite number index number, because each channel has only one satellite signal, so here n represents both the channel number and the number of visible satellites; assuming that the number of observable satellites is n, the RAIM observation equation is:

[0089] y = Hx + ε

[0090] wherein y = [Δρ 1 Δρ 2 ... Δρ n ] T , Δρ n represents the pseudo-range error of the nth satellite; H n×4 is an n-row 4-column state observation matrix; x = [Δx Δy Δz Δt] T is the receiver horizontal and vertical position deviation and clock error; ε represents a noise vector obeying Gaussian distribution;

[0091] From the estimated error detection value model of step 1 baseband domain multi-path correlation, the maximum estimated error of the ranging code phase of each satellite is C dev = [α 1 α 2 ... α n ] T , assuming that the pseudo-code rate is c code_rate , then from the definition of pseudo-range, the average pseudo-range deviation of the pseudo-range deviation generated by each satellite within the T2 detection time is:

[0092]

[0093] wherein, Δp represents the average pseudo-range deviation caused by n satellites ranging code phase error; c represents the speed of light; I 1×n represents a one-dimensional unit row vector of 1 row and n columns, the unit row vector is a vector whose elements are all 1, and n represents the number of visible satellites, n = 1, 2, …, 12; the average pseudo-range deviation vector is constructed as P = [Δp 1 Δp 2 ...Δp n T The values in the vector P are all Δp, so the reconstructed pseudo-range error after linearization is Z = y - P; and the extended state observation matrix is reconstructed as:

[0094]

[0095] wherein, w n1 = (x n -x u ) / r n , W n2 = (y n -y u ) / r n , w n3 = (z n -z u ) / r n , the three-dimensional position coordinates of the n th satellite are (x n , y n , z n ), the three-dimensional position coordinates of the user receiver are (x u , y u , z u ), and the geometric distance from the n th satellite to the user receiver is The clock difference is Δt = 1; the deviation is X = [Δx Δy Δz Δt g] T , g = c / c code_rate is an auxiliary observation constant; therefore, the extended observation equation of the baseband domain assisted RAIM algorithm is:

[0096] Z = FX + ε

[0097] wherein, ε represents a noise vector subject to Gaussian distribution.

[0098] Step 3: Calculate the residual amount of the observation value by the least square method, and then establish the joint test statistic model of the baseband domain assisted RAIM algorithm;

[0099] The least square solution of the least square method for the extended observation equation of the baseband domain assisted RAIM algorithm in step 2 is: ​

[0100]

[0101] The user receiver pseudorange residual is calculated as:

[0102]

[0103] Where I is the unit matrix; U=IF(F T F) -1 F T , then the pseudorange residual of the user receiver is rewritten as: R = Uε; the sum of squares of the components of the pseudorange residual is derived and calculated as:

[0104] F εε =R T R=ε T Uε

[0105] Among them, F εε / σ 2 Obey χ with (n-4) degrees of freedom 2 distributed;

[0106] Therefore, the joint test statistic model T of the baseband-assisted RAIM algorithm is join for:

[0107]

[0108] and the joint test statistic T join The corresponding detection threshold T THR for:

[0109]

[0110] Where σ is the standard deviation of pseudorange noise; t is the constant false alarm rate P FA Calculated constant;

[0111] Finally, the joint test statistic T join and the detection threshold T THR Compare, if T join <T THR , then the satellite navigation system at this moment has no fault, otherwise there is a fault, an alarm message is given and the fault is isolated.

[0112] The application discloses a navigation system fault detection device based on a baseband domain auxiliary RAIM algorithm, which is used for realizing the navigation system fault detection method based on the baseband domain auxiliary RAIM algorithm, and is realized by FPGA and DSP hardware, and specifically comprises ROM1, ROM2, RAM, FPGA and DSP; the ROM1 is connected with the FPGA and the ROM2, wherein the ROM1 is pre-written with a satellite navigation receiver local pseudo code sequence, and the ROM2 is pre-written with a discrete sequence of a sine function; the RAM is connected with the FPGA and the DSP respectively, and is used for temporarily storing data generated in FPGA module and DSP logic calculation processes, so as to facilitate subsequent data reading; the FPGA adopts twelve channels, each channel adopts a 9-way correlator for operation, each channel comprises a carrier NCO module, a de-carrier module, a fast capture module, a C / A code generation module and a multi-way correlator module, and the FPGA and the DSP are connected through an interactive data bus; the DSP is responsible for logic operation tasks, and comprises text decoding, baseband domain detection quantity calculation, observation data least square calculation, joint detection quantity function calculation, fault detection and fault decision.

[0113] The carrier NCO module calls the sine function information in the ROM2, generates an adjustable single-frequency complex sine signal through an initialization carrier phase provided by the fast capture module, and provides the single-frequency complex sine signal for the de-carrier module.

[0114] The de-carrier module receives the zero intermediate frequency navigation signal and performs complex multiplication operation with the local carrier, moves the Doppler frequency signal to zero frequency, and sends the de-carrier signal into the multi-way correlator module.

[0115] The fast capture module captures and discriminates the signal entering the receiver through an FFT capture algorithm, provides a rough carrier phase and a rough pseudo code phase for the initialization carrier NCO generation module and the C / A code generation module.

[0116] The C / A code generation module obtains the pseudo code through the corresponding pseudo code sequence in the initialization ROM1, generates a local rough instant C / A code through the rough code phase in the fast capture module, and provides nine different local time delay C / A codes to the multi-way correlator module according to a set correlator interval.

[0117] In the multi-way correlator module, the received de-carrier signal is correlated and accumulated with the nine local time delay C / A codes provided by the C / A code generation module, the nine correlation values are calculated and stored in the RAM1, and the C / A code stripped signal is sent to the DSP for next decoding and logic calculation.

[0118] The multi-path correlator module in the embodiment utilizes a multi-path correlator detection method: the method utilizes a plurality of different time interval correlators to perform correlation processing on the received data, and judges the state of the receiver data according to the outputs of the correlators, which can not only monitor the case that the output data contains multi-path signals, but also detect the waveform distortion of the satellite micro abnormal signals.

[0119] The algorithm is implemented in FPGA and DSP in this part, and the programming languages used are verilog HDL and C.

[0120] Figure 2 The overall block diagram of each module of the monitoring system of the application is shown in the figure. The design method of twelve channels is adopted in the FPGA part, and each channel adopts a 9-way correlator to perform operation, so as to reduce the hardware resources of the FPGA and transfer most of the logic operation to the DSP processing. In addition, the DSP has good support for some common mathematical formulas, which can make up for the deficiency of the FPGA logic operation. The combination of FPGA and DSP can greatly reduce the difficulty of algorithm transplantation, and also can efficiently run the algorithm. At the same time, the FPGA and DSP have good interactive data bus, which is convenient for data exchange between the two. Figure 2 The RAM is used for temporarily storing the correlation results and intermediate data of the FPGA, including satellite coordinates, pseudorange measurement values, receiver clock differences and multi-path correlation results, etc., which are necessary information required by the DSP algorithm implementation. Figure 2 The ROM is used for storing local data, which is convenient for the FPGA fast capture module, the carrier NCO generation module and the C / A code generation module to quickly access and subsequent operation.

[0121] Because of the modular design of the FPGA, its programs are executed in parallel, such as Figure 2 ;

[0122] The processing flow in the DSP is as follows:

[0123] ① Through the decoding program, the three-dimensional position coordinates of the s-th channel satellite required by the RAIM algorithm, the pseudorange measurement values, the receiver clock differences and other data are decoded from the ephemeris, and the receiver number is s=1, 2, …, 12;

[0124] The correlation values of each channel are read from the RAM, and the correlation calculation in step 1 is completed in the baseband domain detection quantity calculation; the estimated error detection quantity a of the multi-path correlation of each channel and the estimated error detection matrix C of the n-channel multi-path correlation are calculated dev , the average pseudorange deviation Δp of the preprocessed signal and other parameters, wherein the detection error of the ideal signal nor and the K value are directly set in the DSP;

[0125] ② input the decoded observation data and the baseband detection quantity parameter obtained in ① into the baseband domain assisted RAIM algorithm observation equation to complete the data quantity input of the observation equation;

[0126] ③ in the least square calculation of the observation equation, the extended observation equation of the baseband domain assisted RAIM algorithm is calculated to obtain the square sum F of the pseudo-range residual components εε ;

[0127] ④ in the joint detection quantity function calculation, the ratio of F εε and the pseudo-range noise variance σ 2 is used to derive the final joint statistical detection quantity T join , and the corresponding detection threshold T THR is calculated;

[0128] ⑤ in the decision function calculation, the size of the final detection quantity T join and the detection threshold T THR is compared to determine whether there is a fault; if T join <T THR , no fault occurs, and the RAM update flag is directly given to prevent too much data from being temporarily stored and affecting the algorithm process; if T join >T THR , a fault occurs, and the faulty satellite is determined, the satellite number of the faulty satellite is calculated, and then the alarm is output and the faulty satellite is isolated.

[0129] The above description is only the preferred embodiments of the present disclosure and the explanation of the applied technical principles. It should be understood by those skilled in the art that the scope of the application involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or equivalent features without departing from the above inventive concept. For example, the above features are replaced with the technical features disclosed in the embodiments of the present disclosure (but not limited to) having similar functions to form technical solutions.

Claims

1. A method for fault detection in a navigation system based on a baseband domain assisted RAIM algorithm, characterized in that: The method comprises the following steps: Step 1: obtaining the multi-channel correlation value of the n-th channel of the satellite navigation receiver in the baseband domain, establishing an estimated error detection quantity model of single-channel multi-channel correlation, and then establishing an estimated error detection quantity model of n-channel multi-channel correlation on the basis of the estimated error detection quantity model of single-channel multi-channel correlation; Step 2: combining the estimated error detection quantity of n-channel multi-channel correlation and the fusion of the RAIM algorithm to establish an extended observation equation of the baseband-assisted RAIM algorithm; Step 3: calculating the residual quantity of the pseudo-range observation value of the receiver by using the least square method, and then establishing a joint test statistic model of the baseband-assisted RAIM algorithm to realize the fault detection of the satellite navigation system.

2. The navigation system fault detection method based on the baseband domain assisted RAIM algorithm according to claim 1, characterized in that: The step 1 is specifically as follows: n channels are set in the satellite navigation receiver, n≥6, the intermediate frequency signal of each satellite is processed, in order to improve the detection accuracy of the code tracking loop of each channel of the receiver, a multi-channel correlator module is used, the module is composed of N correlators, N≥3, and N is an odd number, the number of correlators depends on the processing capacity of FPGA and DSP, and the more correlators used, the more conducive to improving the detection accuracy of the code tracking loop; Wherein the detection interval is set as Chips, i.e. correlator interval is Chips, wherein The calculated value has integer only integer, no integer only one decimal place after the decimal point; The multipath correlation value of the nth channel is obtained by the correlator calculation According to the slope calculation formula, the correlation value of the navigation baseband signal received by the channel and the multipath local pseudo-code is The slope expression of the correlation peaks on both sides is where k -Δ represents the left slope, k Δ represents the right slope; i = 0, 1, …, N-1 represents the index number of the correlator output, and the corresponding multi-path correlation value is That is, when Δ N / 2 = 0, it is the immediate branch correlation value I p ; when the navigation intermediate frequency signal input by the nth channel is an ideal signal, the slopes on both sides of the correlation peak are consistent, so the single-channel single-time estimation detection error is established as: If it is a normal satellite navigation intermediate frequency signal, then If it is an abnormal satellite navigation intermediate frequency signal, then the above original formula is used; the estimation detection error removal coefficient and After that, the single-channel single-time estimation detection error is rewritten as: The expansion is as follows: The detection relationship between the satellite navigation intermediate frequency signal to be detected and the ideal satellite navigation intermediate frequency signal is established as follows: wherein, is the ith detection amount of the signal to be detected, defined here as the ith detection amount of the micro-abnormal signal; is the ith detection amount of the ideal signal, K is the constant obtained by the false alarm probability P f and the missed detection probability P m is the constant obtained by calculation, σ nor is the standard deviation under the ideal signal; β represents the maximum estimated detection error amount screened out, when the detector β is greater than or equal to the set value, it is considered that the signal is abnormal, and the satellite is marked. Considering that the baseband signal processing time step T1 and the RAIM algorithm monitoring epoch time step T2 are inconsistent, but T2 can be divided by T1, that is, T2 / T1=λ, and the multi-channel correlation estimated error detection quantity β is the monitoring result of the baseband domain of a single satellite, therefore, the maximum error quantity of the n-th channel multiple estimation detection, that is, the estimated error detection quantity model α of single-channel multi-channel correlation is as follows: wherein j = 0, 1, …, λ, represents the number of baseband signal processing times under the length of a single RAIM detection; β j represents the jth estimated error detection value; and α represents the maximum value among the multiple estimated errors under the T2 detection length. Therefore, the estimated error detection matrix of n channels, that is, the estimated error detection quantity model α of n-channel multi-channel correlation is as follows: where α n represents the maximum estimated error detection quantity of the nth channel.

3. The navigation system fault detection method based on the baseband domain assisted RAIM algorithm according to claim 1, characterized in that: The step 2 is specifically as follows: The receiver completes carrier and pseudo code stripping in the baseband domain, and solves data code to obtain three-dimensional coordinate information (x n , y n , z n ) of the navigation satellite, clock error Δt and pseudo range error Δρ, where n is an index number of the satellite quantity, because there is only one signal of one satellite in each channel, so here n represents both the channel number and the number of visible satellites; assuming that the number of observable satellites is n, the RAIM observation equation is: y=Hx+ε where y = [Δρ 1 Δρ 2 …Δρ n ] T , Δρ n represents the pseudo-range error of the nth satellite; H n×4 is an n-row 4-column state observation matrix; x = [Δx Δy Δz Δt] T is the position deviation of the receiver in the horizontal and vertical directions and the clock error; and ε represents a noise vector subject to Gaussian distribution. The estimated error detection quantity model of step 1 baseband domain multipath correlation is obtained, and the maximum estimated error of the ranging code phase of each satellite is C dev = [α 1 α 2 …α n ] T , assuming that the pseudo code rate is c code_rate , then the average pseudo range deviation of the pseudo range deviation generated by each satellite within T2 detection time is obtained from the pseudo range definition: Wherein, Δp represents the average pseudo-range deviation caused by n satellite ranging code phase error; c represents the speed of light; I 1×n represents a one-dimensional unit row vector of 1 row and n columns, and the unit row vector is a vector in which the elements are all 1, and n here represents the number of visible satellites; the average pseudo-range deviation vector is constructed as P = [Δp 1 Δp 2 …Δp n ] T And the values in the vector P are all Δp, so the reconstructed pseudo-range error after linearization is Z = y-P; The extended state observation matrix is reconstructed as follows: wherein w n1 = (x n - x u ) / r n , w n2 = (y n - y u ) / r n , w n3 = (z n - z u ) / r n , the three-dimensional position coordinates of the nth satellite are (x n , y n , z n ), the three-dimensional position coordinates of the user receiver are (x u , y u , z u ), and the geometric distance from the nth satellite to the user receiver is The clock difference is Δt = 1; the bias amount is X = [Δx Δy Δz Δt g] T , g = c / c code_rate is an auxiliary observation constant; therefore, the baseband domain auxiliary RAIM algorithm extended observation equation is: Z=FX+ε Wherein, ε represents a noise vector subject to Gaussian distribution.

4. The navigation system fault detection method based on the baseband domain assisted RAIM algorithm of claim 1, wherein: The step 3 is specifically as follows: The least square solution of the least square method calculation of the baseband-assisted RAIM algorithm extended observation equation in step 2 is as follows: The pseudo-range residual of the user receiver is calculated as follows: Wherein, I is a unit matrix. Let U = I - F(F T F) -1 F T Then the pseudo-range residual of the user receiver is rewritten as: R = Uε; and the square sum of each component of the pseudo-range residual is derived as: F εε = R T R = ε T Uε where F εε / σ 2 obeys a χ 2 distribution with (n-4) degrees of freedom; So the joint test statistic model T of the baseband domain assisted RAIM algorithm is: join T = (T1 + T2) / 2 and the combined test statistic T join The corresponding detection threshold T THR is: where σ is the standard deviation of the pseudorange noise; t is the time of the measurement; and P is the constant false alarm rate FA computed constant; Finally, the joint test statistic T join is compared with the detection threshold T THR If T join < T THR , the satellite navigation system is fault-free at this moment, otherwise, there is a fault, and the alarm information is given and the fault isolation is performed.

5. A navigation system fault detection apparatus based on a baseband domain aided RAIM algorithm, for implementing the navigation system fault detection method based on a baseband domain aided RAIM algorithm according to claim 1, characterized in that: Specifically, it comprises ROM1, ROM2, RAM, FPGA and DSP. The ROM1 and the ROM2 are connected with the FPGA, wherein the ROM1 prewrites the local pseudo code sequence of the satellite navigation receiver, and the ROM2 prewrites the discrete sequence of the sine function; the RAM is connected with the FPGA and the DSP respectively, and is used for temporarily storing the data quantity generated in the FPGA module and the DSP logic calculation process, so as to facilitate subsequent data reading; The FPGA adopts twelve channels, each channel adopts a 9-way correlator for operation, each channel comprises a carrier NCO module, a de-carrier module, a fast capture module, a C / A code generation module and a multi-channel correlator module, and the FPGA and the DSP are connected through an interactive data bus; The ROM1 and the ROM2 are connected with the FPGA, wherein the ROM1 prewrites the local pseudo code sequence of the satellite navigation receiver, and the ROM2 prewrites the discrete sequence of the sine function; the RAM is connected with the FPGA and the DSP respectively, and is used for temporarily storing the data quantity generated in the FPGA module and the DSP logic calculation process, so as to facilitate subsequent data reading; The DSP is responsible for logical operation, including text decoding, baseband domain detection quantity calculation, observation data least square calculation, joint detection quantity function calculation, fault detection and fault decision calculation.

6. The apparatus for detecting faults in a navigation system based on a baseband-aided RAIM algorithm as defined in claim 5, wherein: The carrier NCO module calls the sine function information in the ROM2, and provides the adjustable single frequency complex sine signal through the initialization carrier phase provided by the fast capture module, so as to provide the single frequency complex sine signal for the de-carrier module; The de-carrier module receives the zero intermediate frequency navigation signal and performs complex reset multiplication operation with the local carrier, moves the Doppler frequency signal to zero frequency, and sends the de-carrier signal into the multi-path correlator module; The fast capture module captures and discriminates the signal entering the receiver through the FFT capture algorithm, and provides the coarse carrier phase and coarse code phase of the initialization carrier NCO generation module and the C / A code generation module; The C / A code generation module obtains the code through the corresponding code sequence in the initialization ROM1, generates the local coarse prompt C / A code through the coarse code phase in the fast capture module, and provides 9 different local time delay C / A codes to the multi-path correlator module according to the set correlator interval; In the multi-path correlator module, the received de-carrier signal is correlated and accumulated with the 9 local time delay C / A codes provided by the C / A code generation module, 9 correlation values are calculated and stored in the RAM1, and the C / A code stripped signal is sent to the DSP for next decoding and decision calculation.

Images