CN102636795A

CN102636795A - Multi-receiver networked wireless positioning method

Info

Publication number: CN102636795A
Application number: CN2012101283371A
Authority: CN
Inventors: 陆建华; 陈曦; 张中华; 高文云
Original assignee: Tsinghua University
Current assignee: Tsinghua University
Priority date: 2012-04-27
Filing date: 2012-04-27
Publication date: 2012-08-15
Anticipated expiration: 2032-04-27
Also published as: CN102636795B

Abstract

The invention discloses a multi-receiver networked wireless positioning method which is characterized in that multiple receivers form a wireless network for intercommunication, the receivers receive information of a global navigation satellite system to obtain a first measured value, and simultaneously, neighboring receivers in the wireless network measure the distance between the receivers to obtain a second measured value. Multiple receivers realize positioning resolving in a distributed manner by virtue of the wireless network among the receivers; each receiver firstly obtains a first positioning result, and broadcasts the first positioning result; each receiver calculates a own Kth positioning result after receiving the (K-1)th positioning result of the neighboring receiver; after convergence of the positioning result, the receiver performs independent integrity detection and measured value modification, and after modification, calculates again by using the modified measured value and the method, until the positioning result converges. The method has the following advantages: a new positioning freedom of distance measurement between the neighboring receivers is introduced, a distributed positioning algorithm which can be used for saving wireless network bandwidth resources is used, and the usability and precision of positioning are improved.

Description

Multi-receiver networked wireless positioning method

Technical Field

The invention relates to a global navigation positioning method, in particular to a positioning method for mutual assistance positioning between receivers through a wireless network, belonging to the technical field of computer information.

Background

The Global Navigation Satellite System (GNSS) is a generic name for all Navigation Satellite systems, and currently mainly includes the Global Positioning System (GPS) in the united states, the Global Navigation Satellite System (GLONASS) in russia, the Galileo System in europe, and the Compass in china (Compass). The basic principle of operation of a global navigation satellite system receiver is: and receiving a radio signal sent by a navigation satellite, extracting pseudo ranges, and calculating the position of the navigation satellite in a geographic coordinate system according to more than 4 pseudo ranges, wherein common calculation algorithms comprise a least square method and a Kalman filtering method. Since the navigation satellite continuously works in the space environment with space irradiation, thermal vacuum and thermal cycle, short-term or long-term faults are inevitable. Although each constellation of navigation satellites is provided with a dedicated ground monitoring system, it usually takes tens of minutes or even hours for the ground system to find a satellite signal error. Although navigation systems of different safety and accuracy levels have different requirements for integrity, it is a general requirement that the process of detecting a fault to alerting the user is completed within 10 seconds or even 3 seconds. Therefore, the advanced global navigation satellite system has an autonomous Integrity Monitoring (RAIM) function while performing position solution, and common methods thereof include a pseudo-range comparison method, a least square residual method, a parity vector method, and the like. Although the position can be resolved by 4 visible satellites, at least 5 visible satellites are required for autonomous integrity detection, and more than 6 satellites are required for determining which visible satellite has a problem in signal. In receivers common to modern global navigation satellite systems, solution and autonomous integrity checking are typically integral.

Although the space-based navigation system has the advantages of all-weather global coverage and the like, the space-based navigation system is difficult to realize indoor coverage due to the inherent characteristics of the space-based signals, and the space-based navigation system cannot be positioned due to the small number of visible stars in scenes such as urban canyons. In order to overcome the above problems, a hybrid positioning method combining a ground network and a space-based network has been invented, as disclosed in patent application nos. 00818807.6 and 200410042932.9, which is a new positioning method for providing auxiliary positioning information and positioning signals to a ground mobile communication network to implement combined use of a space-based navigation system and a ground mobile network. The positioning technology is characterized by relying on a central ground infrastructure and being suitable for mobile equipment with ground mobile communication functions, such as mobile phones.

In recent years, with the development of wireless communication technology, direct wireless ranging technology between receivers has also been commercialized, such as using signal strength indication received by a network node to obtain the distance between the node and a wireless access point in a wireless local area network; also as the nanoLOC technique using Time of flight (Time of flight) positioning in a sensor network. The problem with this type of technique is that locating an anchor point that depends on a known location, such as in a wireless sensor network where it is generally necessary to know the location of at least three reference sensors, assuming that the geographical location of the access point is known.

The invention discloses a multi-receiver networked wireless positioning method and a receiver autonomous integrity detection method applied to a wireless self-organizing network. From an application perspective, the advantages are that no anchor points are needed and no ground infrastructure wireless support is needed. The disclosed multi-receiver networked wireless positioning method and the receiver autonomous integrity detection method realize the utilization of the ranging information and the space-based navigation system information among the receivers in the wireless self-organizing network, and improve the positioning availability and the positioning precision compared with the existing system.

Disclosure of Invention

In the networked positioning, a plurality of receivers form a wireless network to communicate with each other, the receivers receive global navigation satellite system information to obtain a first measured value (pseudo range or phase), and meanwhile adjacent receivers in the wireless network measure the distance between the adjacent receivers to obtain a second measured value; the distributed positioning is realized by utilizing a wireless communication network between receivers, and particularly, the positioning method comprises the following steps:

suppose satellite S_iAnd a receiver R_iA pseudorange measurement ofReceiver R_iAnd R_jThe distance between is measured as

And

the observation equation of (a) is:

d_{R_{i} R_{j}} = \sqrt{{(x_{R_{i}} - x_{Rj})}^{2} + {(y_{R_{i}} + y_{R_{j}})}^{2} + {(z_{R_{i}} - z_{R_{j}})}^{2}}

wherein (x)_iy_iz_i) Representing a satellite S_iIn the position of (a) in the first,indicating a receiver R_iPosition estimate, t_RRepresenting the advance value of the receiver relative to the satellite system. The differential is taken on both sides of the square path,

from the above differential equation, the following equation can be given:

z＝Hx

its weighted least squares solution is: x ═ H^TWH)^-1H^TWz and W are weight matrices, which can be selected according to design requirements, or unit matrices.

Wherein,

the initial value of the current position may be any value, wherein

To

All pseudorange differences for all receivers are included,to

The ranging differences between all receivers and adjacent receivers are included;

x is the correction amount, i.e.

H is an observation matrix, which is defined as,

the invention provides a distributed algorithm for realizing the resolving process, which comprises the following steps:

step 1, at each epoch, each receiver at least collects pseudo range, visible satellite position and ranging information of adjacent receivers;

step 2, a receiver with 4 or more than 4 pseudo ranges performs position resolving by using the pseudo ranges, and broadcasts the first positioning result;

step 3, the receivers with less than 4 pseudo ranges receive the first positioning results from the adjacent receivers, once the sum (namely X + Y) of the number X of the collected pseudo ranges and the number Y of the positioning results of the adjacent receivers is more than or equal to 4, the first positioning results of the receivers are determined at least according to the pseudo ranges, the positions of the visible satellites, the first positioning results of the adjacent receivers and the ranging results of the adjacent receivers, and the first positioning results of the receivers are broadcasted to the adjacent receivers;

step 4, repeating the step 3 until no receiver broadcasts the first positioning result;

step 5, the receiver which does not determine the first positioning result randomly initializes the position of the receiver to obtain the first positioning result;

step 6, resolving again by all receivers according to at least all received pseudo ranges, visible satellite positions, first positioning results of adjacent receivers and ranging results of adjacent receivers to obtain self second positioning results, and broadcasting the self second positioning results to all receivers;

step 7, all receivers with own Kth 1 positioning results perform resolving again according to at least all received pseudo ranges, visible satellite positions, adjacent receiver Kth 1 positioning results and ranging results of adjacent receivers to obtain own Kth positioning results, and broadcast own Kth positioning results to all receivers;

step 8, all receivers repeat step 7 until the error between the Kth positioning result and the Kth-1 positioning result is less than a given threshold;

step 9, all receivers carry out autonomous integrity detection operation of the receivers, and problems possibly existing in pseudo-range and ranging information are checked;

step 10, if the autonomous integrity detection operation check passes, resolving is completed; otherwise, the pseudo range and the ranging information are corrected according to the autonomous integrity detection operation check result, and the step 6 is returned.

The measured value correction method is that the measured value of the ith row is selected as a suspected object, and [ I-HAW ] z is calculated to be f, so that S is the sum of [ I-HAW ] ith row elements, and f (I) is the ith element of f, and the corrected value of the corresponding measured value is e (I) ═ f (I)/S.

One way to select the measurement value in row i as the suspect is to consider the measurement value in row i as problematic if the mean of the e (i) values of the previous epochs is around 0, and consider the measurement value in row i as problematic if there is a mean that is apparently not 0; and in the other method, the measured value in the ith row is selected as a suspected object optionally, the measured value is corrected and relocated, if the relocated result has no problem in the autonomous integrity detection of a plurality of continuous epochs, the measured value in the ith row has a problem, otherwise, a new row is selected as a new measured value suspected object until the problem measured value is found or the autonomous integrity detection does not give an alarm any more.

According to the present invention, the weight W is calculated by,

W = C_{i}^{- 1},

the more the noise term therein coincides with reality, the more efficient the algorithm. For each satellite, C_iThe method for determining each diagonal variance term comprises the following steps:

(1) for satellite pseudoranges, the variance is the sum of the satellite clock and ephemeris variance, the ionosphere delay variance, the troposphere delay variance, the receiver multipath error variance, and the receiver noise variance;

(2) the inter-receiver ranging variance is set to a mapping inversely proportional to the signal-to-noise ratio of the received signal, i.e.

The SNR may be obtained from a communication receiver receiving a signal strength indication.

Compared with the existing algorithm, the networked positioning calculation algorithm has the following characteristics:

(1) by adopting a distributed algorithm, the network service flow required by network resolving is reduced to the maximum extent by initializing the first position information of the receiver in a grading way, so that the influence of network packet loss, transmission delay and the like on a positioning result is reduced, and the power consumption is reduced;

(2) in the RAIM algorithm of the receiver, a horizontal positioning error corresponding to ranging information in a network is defined, and autonomous integrity detection of the receiver can be more effectively carried out;

(3) in the calculation of the weight, the mapping relation that the square of the error of the measurement distance related item between the receivers is in inverse proportion to the signal-to-noise ratio of the received signal is utilized, and the SNR is obtained by utilizing the strength indication of the received signal of the receiver, so that more accurate weight can be obtained.

The algorithm of the patent is more suitable for realizing networking due to the characteristics, and the accuracy can be further improved compared with the existing algorithm in a networking environment.

Drawings

Fig. 1 is a typical scenario of networked positioning according to the present invention.

Fig. 2 is a flow chart of a networked distributed positioning algorithm according to the present invention.

Fig. 3 is an example of a measurement scenario for a networked distributed positioning system according to the present invention.

Fig. 4 is a flow chart of a receiver autonomous integrity detection operation according to the present invention.

Detailed Description

The invention discloses a networked hybrid positioning calculation method applied to a wireless self-organizing network and a receiver autonomous integrity detection method thereof.

Fig. 1 is a typical scenario of networked positioning according to the present invention. As shown in fig. 1, a plurality of user receivers, i.e., navigation receivers a21-a25, form a wireless ad hoc network, and these receivers can perform not only wireless communication with each other but also ranging of neighboring receivers. Meanwhile, the receivers can receive positioning signals of global navigation satellite systems (such as GPS and Beidou). According to the invention, these radio receivers are positioned according to the method of the invention by jointly using global navigation satellite system signals and ranging of neighbouring receivers. The multiple receivers form a wireless network to communicate with each other, the receivers receive global navigation satellite system information to obtain a first measurement value, and meanwhile, adjacent receivers in the wireless network measure the distance between the adjacent receivers to obtain a second measurement value.

In a conventional navigation system, assume that the known satellite position is P_sThe receiver position is P_RWhen the receiver time is t and the satellite signal observation amount rho, the basic positioning equation is rho + epsilon | | | P_s-P_RAnd | wherein ε is the observation error correction. In a conventional GPS receiver, for three-dimensional positioning, more than 4 satellite views are required due to the presence of four unknowns of three-axis position and timeThe measurement can be used for positioning the receiver, and a least square algorithm and a Kalman filtering algorithm are commonly used for a positioning algorithm.

Unlike conventional navigation systems, in the networked positioning of the present invention, there are two types of positioning observations for each receiver, which are GPS observations (pseudoranges or phases) and inter-receiver ranging information. Assume a pseudorange measurement between the satellite Si and the receiver Ri of

Receiver R_iAnd Rj is measured as a distance

And

the observation equation of (a) is:

d_{R_{i} R_{j}} = \sqrt{{(x_{R_{i}} - x_{Rj})}^{2} + {(y_{R_{i}} + y_{R_{j}})}^{2} + {(z_{R_{i}} - z_{R_{j}})}^{2}}

wherein (x)_iy_iz_i) Representing a satellite S_iIn the position of (a) in the first,

representing an estimate of the receiver Ri position, t_RRepresenting the advance value of the receiver relative to the satellite system. The differential is taken on both sides of the square path,

from the above differential equation, the following equation can be given:

z＝Hx

Wherein,

the initial value of the current position may be any value, wherein

To

All pseudorange differences for all receivers are included,

to

x is the correction amount, i.e.

H is an observation matrix, defined as:

in a centric calculation, at each epoch, all receivers send their measurements to the central receiver over the wireless network between them, with x ═ H being performed by the central receiver^TWH)^-1H^TAnd Wz operation, and the central receiver sends the operation result back to each receiver through a wireless network between the receivers. The advantage of the central calculation is that the global optimum positioning can be realized, the disadvantage is that a central receiver with stronger calculation capability must be arranged, and the other disadvantage is that the wireless network communication cost is higher and grows linearly with the network scale. When the network is very large, neither the central receiver nor the wireless network can meet the requirement of networked positioning.

Fig. 2 is a flow chart of a networked distributed positioning algorithm according to the present invention. In view of the disadvantages of the above-mentioned central algorithm, we propose a distributed algorithm that implements the above-mentioned solution process. As shown in fig. 2, the algorithm includes the following steps:

As shown in FIG. 3, in the network, there is a total of R₁-R₆Six user receivers, a total of 5 navigation satellites S₁-S₅. The process according to the invention:

step 1, at each epoch, each receiver collects at least pseudoranges, visible satellite positions and ranging information with neighboring receivers. At the time shown in fig. 3, the observed quantity of each receiver is:

receiver R₁：

Receiver R₂：

Receiver R₃：

Receiver R₄：

Receiver R₅：

Receiver R₆：

(d_{R_{6} R_{1}}, d_{R_{6} R_{2}}, d_{R_{6} R_{3}}, d_{R_{6} R_{4}}, d_{R_{6} R_{5}});

And 2, utilizing the pseudo ranges to carry out position calculation by receivers with 4 or more than 4 pseudo ranges, and broadcasting a first positioning result. Due to the receiver R₁And a receiver R₂Having 4 pseudoranges, measured as values, according to a conventional solving algorithm, such as least squares, R₁And R₂Respectively solve their first positioning results

And

R₁will be provided with

Broadcast to adjacent receivers R₂、R₃、R₆；R₂Will be provided with

Broadcast to adjacent receivers R₁、R₅And R₆；

And 3, receiving the first positioning results from the adjacent receivers by receivers with less than 4 pseudo ranges, determining the first positioning results according to at least the pseudo ranges, the positions of the visible satellites, the first positioning results of the adjacent receivers and the ranging results of the adjacent receivers once the sum (namely X + Y) of the number X of the collected pseudo ranges and the number Y of the positioning results of the adjacent receivers is more than or equal to 4 or Y is more than or equal to 3, and broadcasting the first positioning results to the adjacent receivers. In this step, the receiver R₅Due to the reception of R₂First positioning result of

Thus, X is 3, Y is 1, X + Y is 4, it usesSolve its own first positioning result

Receiver 5 broadcastTo R₂、R₄And R₆；

And 4, repeating the step 3 until no receiver broadcasts the first positioning result. In this step, R₆First because R is received₁、R₂、R₅The first positioning result of (a) realizes that Y is 3, which utilizes

Determine its first positioning result

Then the result is compared

Broadcast to adjacent receivers R₁-R₅；R₃Receive from

Then X is 2, Y is 2, and X + Y is 4, thus usingObtaining a first positioning result through resolving

R₃Broadcasting

To adjacent receivers R₁、R₄、R₆And then R is₄By using X-1, Y-3, X + Y-4Obtain the first positioning result

And 5, the receiver which does not determine the first positioning result randomly initializes the position of the receiver to obtain the first positioning result. In this example, no remaining receivers are able to determine their own first positioning result. Suppose R in FIG. 3₅And R₆Is a neighboring receiver, then R₆、R₃And R₄The first positioning result of the receiver cannot be determined, and then RX and R₃And R₄The position of the user can be randomly initialized to obtain a first positioning result, and the better method is that R is₆、R₃And R₄Predicting the position of the current epoch according to the position of the previous epoch to be used as an initial position, or selecting the average value of the positions of adjacent receivers with known positions of the current epoch as a first positioning result of the current epoch;

step 6, all receivers with own Kth 1 positioning results perform resolving again according to at least all received corrected pseudo moments, adjacent receiver Kth 1 positioning results, visible satellite positions and distance measurement results of the adjacent receivers corrected by the correction, so that own Kth positioning results are obtained, and own Kth positioning results are broadcasted to all the receivers;

step 7, all receivers repeat step 6 until the error between the Kth positioning result and the Kth-1 positioning result of all receivers is less than a given threshold;

step 8, all receivers carry out autonomous integrity detection operation of the receivers, and problems possibly existing in pseudo-range and ranging information are checked;

step 9, if the autonomous integrity detection operation check passes, resolving is completed; otherwise, the pseudo range and the ranging value are respectively corrected according to the correction quantity obtained by the autonomous integrity detection operation check, and the step 6 is returned.

Fig. 4 is a flow chart of a receiver autonomous integrity detection method according to the present invention. Each receiver with X + Y > 4 according to the invention performs a receiver autonomous integrity check, comprising the steps of:

step 8.1, setting lambda min, initializing detection threshold

N (n is X + Y) is the total number of the pseudo range and the ranging result which are available for the receiver; specific values for the given level alarm thresholds HAL,. lambda.min and HAL may be referred to in the relevant GPS literature;

step 8.2, the adjacent receivers exchange the positioning result of the Kth time, determine an own observation matrix H according to the own positioning result of the Kth time and the positioning result of the adjacent receivers, and calculate the final residual error epsilon ═ z and A ═ H^TWH)^-1 H^T；

Step 8.3, determining detection statistics:

step 8.4, calculating slope of ith pseudo range

K_{i} = \sqrt{(A_{w, 1 i}^{2} + A_{w, 2 i}^{2}) (n - 4) / S_{w, ii} W_{ii}},

For the ith ranging value, the ranging value,

K_{i} = \sqrt{(A_{w, 1 i}^{2} + A_{w, 2 i}^{2} + A_{w, 1 j}^{2} + A_{w, 2 j}^{2}) (n - 4) / S_{w, ii} W_{ii}};

calculating the level protection threshold HPL ═ max (K)_i)T_w，min；

Step 8.5, if the HPL is larger than the HAL, the autonomous integrity detection is finished, and the detection result is a problem-free measurement value; otherwise, r is compared_wAnd T_w，minIf r is_w＞T_w，minThere is a problem measurement, otherwise there is no problem measurement.

The measured value correction method comprises the following steps: selecting the measured value of the ith row as a suspected object, calculating [ I-HAW ] z as f, and making s be the sum of the elements of the ith row of [ I-HAW ], and f (I) be the ith element of f, so that the corrected value of the corresponding measured value is e (I) ═ f (I)/s.

One method for selecting the measurement value of the ith row as the suspected object is that if the e (i) value of the previous multi-epoch is near 0, the measurement value of the ith row has no problem, and if there is a mean value obviously not being 0, the measurement value of the ith row is considered to have a problem; and in the other method, the measured value in the ith row is selected as a suspected object optionally, the measured value is corrected and relocated, if the relocated result has no problem in the autonomous integrity detection of a plurality of continuous epochs, the measured value in the ith row has a problem, otherwise, a new row is selected as a new measured value suspected object until the problem measured value is found or the autonomous integrity detection does not give an alarm any more.

According to the present invention, the weight W is calculated by,

W = C_{i}^{- 1},

(1) for satellite pseudoranges, the variance is the sum of the satellite clock and ephemeris variance, the ionosphere delay variance, the troposphere delay variance, the receiver multipath error variance, and the receiver noise variance. These variance terms are well described in the GPS-related literature. Ratio ofSuch as satellite clock and ephemeris error RMS 6; multipath error (0.13+0.53 e)^-Ei/10)²(ii) a Tropospheric error (0.12m (Ei))²，

Ei is the satellite elevation angle in degrees. The ionospheric delay error is somewhat complex and can be referred to by the reference RTCA DO-229C]U.S.A: RTCA, inc, 2001, which is not given here. The receiver noise variance may be 0.01;

The SNR is typically obtained by the communication receiver receiving a signal strength indication.

Claims

1. A multi-receiver networked wireless positioning method is characterized in that: the method comprises the following steps that multiple receivers form a wireless network to communicate with each other, the receivers receive global navigation satellite system information to obtain a first measurement value, and meanwhile adjacent receivers in the wireless network measure the distance between the adjacent receivers to obtain a second measurement value; the method for realizing the positioning of the receiver in a distributed way by utilizing the wireless communication network between the receivers comprises the following steps:

step 1, at each epoch, each receiver at least collects pseudo range, visible satellite position and ranging information of a neighbor receiver;

step 2, a receiver with 4 or more than 4 pseudo ranges performs position resolving by using the pseudo ranges, and broadcasts a first positioning result;

step 3, receivers with less than 4 pseudo ranges receive first positioning results from neighbor receivers, once the sum (namely X + Y) of the number X of the collected pseudo ranges and the number Y of the positioning results of the neighbor receivers is more than or equal to 4, the first positioning results of the receivers are determined at least according to the pseudo ranges, the first positioning results of the neighbor receivers, the positions of the visible satellites and the ranging results of the neighbor receivers, and the first positioning results of the receivers are broadcasted to the neighbor receivers;

step 6, all receivers with own Kth 1(K is more than 1) positioning results perform resolving again according to at least all received pseudo ranges, neighbor receiver Kth 1 positioning results, visible satellite positions and ranging results of adjacent receivers to obtain own Kth positioning results, and broadcast own Kth positioning results to all receivers;

step 7, all receivers repeat step 6 until the difference between the Kth positioning result and the Kth-1 positioning result is less than a given threshold;

and 8, outputting the Kth positioning result as a final positioning result.

2. The positioning method according to claim 1, wherein the following expansion steps are present after step 7 of claim 1 and before step 8 is started:

step A, all receivers carry out autonomous integrity detection operation of the receivers, and problems possibly existing in pseudo-range and ranging information are checked;

step B, if the autonomous integrity detection operation check passes, resolving is completed; otherwise, the measured value is corrected according to the autonomous integrity detection result, and the step 6 of the right 1 is returned.

3. The method of claim 2, wherein the autonomous integrity checking method comprises the steps of: step A1, setting lambda min, initializing detection threshold

step A2, exchanging the K-th positioning result by the adjacent receiver, determining the self observation matrix H according to the K-th positioning result and the K-th positioning result of the adjacent receiver, and calculating the final residual error epsilon-z and A-H^TWH)^-1H^T；

Step A3, determining detection statistics:

step A4, calculating slope for ith pseudorange

For the ith ranging value, the ranging value,

K_{i} = \sqrt{(A_{w, 1 i}^{2} + A_{w, 2 i}^{2} + A_{w, 1 j}^{2} + A_{w, 2 j}^{2}) (n - 4) / S_{w, ii} W_{ii}};

calculating the level protection threshold HPL ═ max (K)_i)T_w，min；

Step A5, if the HPL is larger than the HAL, the autonomous integrity detection is completed, and the detection result is a problem-free measurement value; otherwise, r is compared_wAnd T_w，minIf r is_w＞T_w，minThere is a problem measurement, otherwise there is no problem measurement.

4. The method of claim 2, wherein the method of correcting the measurement comprises the steps of: selecting the measured value of the ith row as a suspected object, calculating [ I-HAW ] z as f, and making s be the sum of the elements of the ith row of [ I-HAW ], and f (I) be the ith element of f, so that the corrected value of the corresponding measured value is e (I) ═ f (I)/s.

5. The method of claim 4, wherein the method of selecting the measurement value of the ith row as the suspect object comprises: if the average of the e (i) values of the previous epoch is around 0, the ith row measurement value is not problematic, and if there is an average value that is significantly different from 0, the ith row measurement value is considered to be problematic.

6. The method according to claim 4, wherein the method for selecting the measured value in the ith row as the suspect object selects the measured value in the ith row as the suspect object, and performs the measured value correction and the relocation, if the relocation result shows that the autonomous integrity detection of a plurality of epochs is not problematic continuously, the measured value in the ith row has a problem, otherwise selects a new row as a new measured value suspect object, until a problem measured value is found or the autonomous integrity detection does not alarm any more.

7. The positioning method according to claim 1, wherein the weight W is calculated by:

W = C_{i}^{- 1},

for each satellite, C_iThe method for determining each diagonal variance term comprises the following steps:

(1) for the weight of the satellite pseudo range, the corresponding diagonal variance is the sum of the satellite clock and ephemeris variance, the ionosphere delay variance, the troposphere delay variance, the receiver multipath error variance and the receiver noise variance;

(2) corresponding to the weight of the distance measurement correlation between the receivers, the corresponding diagonal variance term is set to be in inverse proportion to the mapping relation of the signal-to-noise ratio of the received signal, i.e. the mapping relation