KR100438396B1 - Realtime multipath detection method in Global navigation satellite system and positioning apparatus utilizing the same - Google Patents

Abstract

The present invention relates to a positioning device using satellites, and in particular, to such a positioning device, a method for avoiding multipath errors.

A positioning device according to the present invention comprises: a plurality of satellite receivers; A calculation matrix storage unit for storing a calculation matrix for calculating a degree of change in the permutation of the difference between the pseudo-distance time increment and the cumulative phase time increment for a plurality of measurements received from each of the channels of the plurality of receivers; An observer calculator for calculating observer values by calculating in real time a plurality of measurements received from the matrix stored in the storage unit and the channel of the receiver unit; And a plurality of selectors for restricting whether to select a corresponding channel according to a detection result of the observer calculator.

Description

Real-time multipath detection method of positioning receiver using satellite and positioning device using same method {real-time multipath detection method in Global navigation satellite system and positioning apparatus utilizing the same}

The present invention relates to a positioning device using satellites, and in particular, to such a positioning device, a method for avoiding multipath errors.

GPS signals, which are propagated using radio waves, contain various error components due to ionospheric delay, tropospheric delay, receiver clock bias, thermal noise, and multipath. do. Double ionospheric delay, tropospheric delay, clock bias, and thermal noise error components are always included even in normal GPS operation, and much research has been conducted to mitigate their effects. Thus, ionospheric delay and tropospheric delay can be reduced by the approximation by modeling. In the case of receiver clock bias, if more than four satellites are observed, the effects of the receiver as well as the location of the receiver as well as the unknown variable and thermal noise can be greatly reduced by various low-pass filters. Using DGPS with two receivers also reduces the components of error caused by ionospheric delay, tropospheric delay, and clock bias.

Multipath errors have different characteristics from those listed above. The error factors listed above are always present even when the GPS is operating normally. Therefore, the accuracy of positioning due to their influence is statistically announced as the performance of the GPS. Multipath error is an abnormal error factor that occurs when the signal reaching the receiver and the signal arriving directly from the satellite are combined with the signal tracking unit inside the receiver. Therefore, separation and removal are not possible using conventional methods using DGPS and low pass filtering. In order to eliminate the multipath error with this characteristic, the researches to date have been mainly performed at the signal tracer level inside the receiver. Representative studies include A. J. van Dierendonck, P. Fenton, and T. Ford, Theory and Performance of Narrow Correlator Spacing in a GPS Receiver, Navigation: Journal of the Institute of Navigation, Vol. 39, pp. Narrow Correlator ™ method described in 265-283, 1992, and B. Townsend and P. Fenton, A Practical Approach to the Reduction of Pseudorange Multipath Errors in a L1 GPS Receiver, Proceedings of ION GPS 94, Salt Lake City, pp. METTM method described in 143 138, 1994, and in R. D. J. van Nee, The Multipath Estimating Delay Lock Loop, Proceedings of Int. Symp. On Spread Spectrum Techniques and Applications, pp. 39-42, 1992, and R.D.J. van Nee, J. Siereveld, P. C. Fenton, and B. R. Townsend, The Multipath Estimating Delay Lock Loop: Approaching Theoretical Accuracy Limits, Proceedings of IEEE Position Location and Navigation Symposium, pp. MEDLL ™ described in 246 251, 1994, etc. are mentioned. Using this method can greatly reduce the error due to multipath, but it requires a lot of cost because it requires a change in the receiver hardware, and there is a disadvantage that the application is fundamentally impossible after the hardware of the receiver is determined.

The point of reference for the results of the signal tracer level is that the tendency of the received measurements due to the multipath error is irregular but periodic and continuously changes within an abnormally large maximum value. Therefore, it is difficult to identify the signs of multipath by instantaneous measurement and can be identified only by observing the trend of measurement over a certain period. As an observer outside the receiver that can indicate the existence of multipath, a method using a signal to noise ratio signal and a signal having a code and phase difference is known. Until now, a method of identifying a multipath using such an observer has been known, such as a method using an instantaneous measurement value, a method of installing a plurality of antennas, using a constraint condition, and using a long-term observer's tendency. The method of using the instantaneous measurement with one antenna has a high possibility of failing to identify the multipath error due to the lack of information, and the method of installing a plurality of antennas and using the constraint condition is somewhat expensive. In addition, the method of using the observer's tendency over a long time has a disadvantage in that it is difficult to apply to real-time navigation.

Another prior art is described in US Pat. No. 5,927,445, issued June 29, 1999 to Schipper et al. This technique compares the magnitudes of pseudoranges and cumulative phase differences with reference values to determine the presence of multipath errors. This method is based on multiple antennas and receivers, and has the advantage of detecting multipath errors in real time, but has a disadvantage of low detection reliability because only a limited number of samples are used. In addition, since the ionospheric delay is represented by the opposite sign in the pseudorange and the cumulative phase, there is a problem that this method alone cannot be free from the influence of the ionospheric delay. In addition, there is a problem in that an unspecified integer for converting the cumulative phase into a distance concept is assumed as a predetermined value.

Another prior art is described in US Pat. No. 6,172,638 to Jan. 9, 2001, Arrethens et al. This method takes advantage of the rate of change in pseudorange time increments and the difference in relative speed between the satellite and the receiver, calculated from Doppler measurements. This method judges the presence of multipath error by comparing the permutation of the moving average value during the long sample value of this difference or the fast Fourier transform of these moving average values with a reference value. This method has a problem that it is difficult to apply when the receiver is moving because it uses the average relative speed between the satellite and the receiver, and also difficult to implement in real time because it performs a complex operation on a large number of samples. In addition, these prior arts do not all present any definite criteria for the setting of the reference value which is the standard of judgment.

An object of the present invention is to provide a multi-path error detection method that can be applied to a moving receiver.

Furthermore, an object of the present invention is to provide a multipath error detection method applicable to a single receiver.

Furthermore, an object of the present invention is to provide a method capable of detecting a multipath error in real time.

In addition, an object of the present invention is to provide a multi-path error detection method that can minimize the effect of the ionospheric delay by itself.

Furthermore, an object of the present invention is to provide a multi-path error detection method that can be applied without changing the hardware structure of a conventional GPS receiver.

Furthermore, an object of the present invention is to provide a multipath error detection method capable of providing a basis for determining a reference value for determining whether a multipath error exists.

1 is a block diagram showing the configuration of main parts of a positioning device according to an embodiment of the present invention.

2 is a block diagram showing the configuration of main parts of a positioning apparatus according to another embodiment of the present invention.

3 to 5 show the test values extracted using the patterning technique and the method according to the invention, respectively.

FIG. 6 illustrates a position distribution between 0 seconds and 150 seconds when satellites 20 and 22 are ignored in FIG. 5.

Multipath error detection method according to an aspect of the present invention for achieving the above object comprises: providing a plurality of consecutive pseudorange and cumulative phase measurements, pseudo distance time increment and cumulative phase time for the plurality of measurements Calculating a permutation of the incremental difference, and determining whether multipath noise exists from the degree of change of the permutation.

By detecting the degree of permutation of the difference between the pseudorange time increment and the cumulative phase time increment, the multipath noise detection method according to the present invention can be applied to a moving receiver and can be applied to a single receiver.

Furthermore, the multipath error detection method according to the present invention converts the permutations into orthogonal permutations, calculates a sum of squares of values obtained by dividing each of the orthogonal permutations by an error covariance, and the calculated observer When the value is larger than the reference value, it may be determined that the channel has a multipath error.

The observed observer values are normalized independent random variables and have a chi square distribution. Accordingly, the multi-path error detection method according to the present invention can accurately determine the reference value of the error detection according to this statistical feature.

That is, in the chi square distribution of the observer values, a reference value of error detection may be determined according to a given false alarm probability and an undetected probability.

According to another aspect of the present invention, the present invention provides a positioning device using a satellite to which the multi-path error detection method is applied.

A positioning device according to the first aspect of the present invention comprises: a plurality of satellite receivers; A plurality of filters for filtering to increase the accuracy of a signal received from each of the reception channels of the plurality of receivers; Calculating a permutation of the difference between pseudorange time increments and cumulative phase time increments for a plurality of measurements received from each of the plurality of channels, and determining the presence or absence of multipath noise in the corresponding channel from the degree of change of the permutations. A multipath detector of; A plurality of selectors for limiting selection of a corresponding channel according to a detection result of the weighted path detector; Characterized in that it comprises a.

By effectively excluding a channel determined to have a multipath error by the multipath detector, only a channel having a high degree of signal is selected, thereby enabling high-precision position tracking.

According to another aspect of the present invention, a positioning device includes: a plurality of satellite receivers; A calculation matrix storage unit for storing a calculation matrix for calculating a degree of change in the permutation of the difference between the pseudo-distance time increment and the cumulative phase time increment for a plurality of measurements received from each of the channels of the plurality of receivers; An observer calculator for calculating observer values by calculating in real time a plurality of measurements received from the matrix stored in the storage unit and the channel of the receiver unit; A plurality of selectors for restricting whether to select a corresponding channel according to a detection result of the observer calculator; Characterized in that it comprises a.

According to this aspect of the invention, it is possible to avoid repeating complex operations every time by using a pre-stored determinant, thus making the real-time implementation easier.

According to a further aspect of the present invention, the calculation matrix storage unit is characterized in that for storing a matrix for orthogonalizing and normalizing the permutation of the difference between the pseudo-distance time increment and the cumulative phase time increment received from each channel.

These and further aspects of the invention will be described in more detail below through the preferred embodiments described with reference to the accompanying drawings.

Typical measurements provided by GPS receivers are scalar quantities provided by satellite code and phase tracking circuits, which include pseudo range and accumulated carrier phase. There is speed solution. Among them, the scalar pseudorange and cumulative phase are expressed as follows.

Doctor's Distance:

Cumulative Phase:

here

Is the pseudorange from the j-th satellite, measured by the receiver at the k-th time.

is the actual distance between the receiver of the k-th time and the j-th satellite

: Receiver clock bias of k-th time

is the orbital error between the receiver at k-th time and the j-th satellite.

is the ionospheric delay between the receiver at k-th time and the j-th satellite.

is the convective delay between the receiver at k-th time and the j-th satellite.

pseudorange multipath error between the k-th receiver and the j-th satellite

: cumulative phase multipath error between the k-th receiver and the j-th satellite

Pseudo-distance thermal noise between the k-th receiver and the j-th satellite

: cumulative phase thermal noise between the k-th receiver and the j-th satellite

In the above equation, it is assumed that the thermal noises appearing in the pseudorange and cumulative phase measurements have the following Gaussian distribution and are not correlated.

The receiver clock bias, satellite orbit error, ionospheric delay, and convective delay that GPS measurements include are typically less instantaneous than multipath error and thermal noise. Using this, the following equations can be obtained.

, i = k-b + 1, k-b + 2, Λ, k

here,

B: number of measurements used for multipath error detection

Multipath error does not exist There is a correlation in permutations. For efficient fault detection statistics Orthogonalized as Convert to

Permutations with Stochastic Orthogonalization Has a Gaussian distribution:

here,

Therefore, multipath test value using the most recent (B + 1) pseudoranges and cumulative phase If you define

The condition of multipath detection using is based on the following hypotheses.

That is, this permutation follows the distribution of chi squares. That is, according to an aspect of the present invention, the present invention reconstructs such a permutation to have a chi-square distribution by normalizing a permutation of the difference between pseudorange time increment and cumulative phase time increment, and normalizing it by the value divided by the error covariance. do. Accordingly, by using these statistical distribution characteristics, it is possible to reasonably determine a reference value, which is a judgment of multipath error.

: Χ-square distribution with asymmetric variable λ and n degrees of freedom

The test values described above are suitable for strongly coupled GPS / SDINS configurations because they can detect multipath errors included in pseudoranges and cumulative phases for each satellite.

Hereinafter, another embodiment of the present invention suitable for detecting a multipath error present in a position and velocity solution of a vector form provided by a GPS receiver will be described. This method can also be detected in a similar way to the previous case. For this purpose, the position and velocity solutions are expressed as

(k-1), Velocity error due to receiver clock bias change rate of k-th time

: (k-1), Velocity error due to change rate of satellite orbit error in k-th time

: (k-1), Velocity error due to rate of ionospheric delay change at k-th time

: (k-1), Velocity error due to change rate of convective delay of k-th time

: Velocity error due to multipath included in cumulative phase of k-th time

: Velocity error due to thermal noise included in the cumulative phase of the k-th time

It is assumed that the thermal noise components located in the equation above and included in the velocity solution have Gaussian distributions and are not correlated with each other.

As with the error characteristics of each channel, the error components generated due to the receiver clock bias, satellite orbit error, ionospheric delay, and convective delay and velocity solution are momentary compared to the error components due to multipath error and thermal noise. Using this, we can obtain a fault detection equation using two consecutive positional and velocity solutions as follows:

For efficient fault detection statistics Orthogonalized as Convert to

here,

Thereby permutation Has the following statistical characteristics:

Therefore, multipath test using the most recent (B + 1) position and velocity solution If you define

This calculation can facilitate real-time implementation by precomputing, storing, and executing the middle matrix term.

The condition of multipath detection using is based on the following hypotheses.

here,

According to this embodiment It is suitable for weakly coupled GPS / SDINS configurations because it can detect multipath errors using the time series of position and velocity solutions provided by conventional GPS receivers. So far, we have proposed a new multipath detection technique based on the principle of transfer measurement convergence. Since the proposed technique uses the basic GPS receiver's measurements, it can be applied to many fields at low cost and because it does not use the information by the auxiliary filter, it is independent of the multipath error included in each channel or located and velocity solution in the GPS receiver. It can be detected as an advantage. In general, the reliability of a test with a chi-square distribution is proportional to the degrees of freedom. The proposed technique can improve the reliability because the degree of freedom can be controlled by the size B of the measurement accumulation interval.

The multipath detection method described above uses successive measurements provided for each channel for reliable fault diagnosis. Therefore, a complex delay phenomenon of a plurality of consecutive measurements is inevitably generated for measurement verification and complete acquisition of measurement information. In order to carry out the proposed fault diagnosis technique under the dynamic movement of the receiver and to perform real-time navigation, an efficient convergence technique is required. 1 shows an embodiment of a positioning apparatus using a satellite according to an aspect of the present invention to which this method is applied.

Each channel filter shown in FIG. 1 includes one compressed pseudorange whose accuracy is improved by recursive calculation using continuously provided pseudoranges and cumulative phases while troubleshooting using the measured time series is performed. ) For each channel. This is expressed as follows.

reset :

Projection:

Compression:

For the fault diagnosis shown in FIG. 1, the multipath observer described above is accumulated by recursive calculation and then compared with a predetermined threshold. At this time, the threshold of the multipath observer is calculated by the statistical characteristics of the error factors included in each pseudorange and cumulative phase. If the value of the multipath observer is smaller than the threshold, the channel is determined to be unaffected by the multipath. When the channel of the multipath observer is smaller than the threshold value is determined, the main filter uses the cumulative phase difference of the channels determined to be safe for the time transmission between the last measurement update and the present time as follows.

First, calculate the following for each channel:

And store the following values:

Transfer of position and clock bias:

: selection matrix for screening only satellite measurements not affected by failure in the global satellite measurement vector from (kB) -th to k-th time

here,

= 0: first

, etc

The main filter, which performs time propagation using the cumulative phase of the multipath-independent channels, updates the measurement using the compressed pseudorange. At this time, the compressed pseudorange includes continuous pseudorange and cumulative phase information for a predetermined period before the present time. In addition, since the main filter has already used the cumulative phase between the last measurement update time and the present time, there is a correlation between the main filter and the compressed pseudo distance. In order to consider this correlation, the main filter, unlike the existing Kalman filter, first forms an indirect measurement as follows and then performs a compression filter algorithm.

First, calculate the following for each channel:

After that, the following values are calculated and stored.

Then update the measurements as follows:

First, calculate the gain as shown below.

Next, update the position and clock bias.

The receiver filter structure described so far uses a plurality of channel-specific scalar filters to fuse pseudorange and cumulative phase information for a relatively short period, and uses one main filter to present the current after the start of the system. Only the information of all the channels whose safety has been verified up to the point of time is fused to generate consecutive positions.

Hereinafter, a multipath error detection method and a positioning device according to the method according to another aspect of the present invention will be described. The compression filter described above enables a multipath observer to be efficiently calculated using recursive computational structures over time. On the other hand, the compression filter described below has a form different from the conventional filter equation that uses the measurement at every moment in order to take into account the influence of the measurement of the delayed number of measurements. In the situation where the existing filter structure cannot be changed, permutation of the past double difference by the degree of freedom B of the multipath observer at the present time (j = k) Should be stored. Stored bidifferential permutation Permutation of Orthogonalization Orthogonal to The following relationship is established between them.

And Dual difference permutation Permutation of Orthogonalization Orthogonal to As a cumulative vector of, each defined as

and The following relationship is established between them.

Also, and The following relationship holds between

and The following relationship is established between them.

Matrix from the above expression Wow Since is determined by the statistical characteristics of pseudorange and cumulative phase measurements, it is fixed unless the statistical characteristics of the measurements change. Therefore, precomputation and storage at the beginning of the navigation system need not be recalculated unless the statistical properties of subsequent measurements change. Therefore, according to the advantageous aspect of the present invention, by using a pre-calculated matrix value, the amount of calculation is reduced, thereby facilitating real-time implementation and simplifying the system configuration. Dual difference permutations occurring at this point in the system and past points up to (B-1) -Phase Prepare a memory to store the data, and double-permutation permutation stored in memory the moment a new (k + 1) -th measurement occurs in And remove If you insert, the matrix Is fixed, Can be computed every minute with a simple matrix operation. Figure 2 is a block diagram showing the main part of the positioning device according to another aspect of the present invention, showing the configuration of the system when using the proposed method for the filter of the existing concept. Each observer calculation unit includes a calculation matrix storage unit in which the matrix values calculated in advance are stored.

A standalone experiment was performed to analyze the performance and characteristics of the proposed multipath detection technique in real time navigation. In the experiment, the reflector was approached and fixed after about 100 seconds by the receiver installed in the fixed position. The receiver is in a fixed position, but the navigation algorithm uses a dynamic algorithm that can be used while the receiver is in motion. The dynamic situation of the receiver makes it easier to screen multipath errors because there are more changes in the path between the receiver and each satellite than in the static situation. In addition, when the DGPS is performed, the error except for the multipath error and the thermal noise is commonly removed, so it is relatively easy to identify the multipath error. Therefore, the environment used in the experiment can be regarded as a situation where it is relatively difficult to identify multipaths.

In order to compare the performance of the proposed scheme, a parity technique and a general GPS Kalman filter, which are relatively easy to apply in the standalone GPS situation where inertial navigation system or DR cannot be used, are used. The Kalman filter associated with the parity technique and the main filter of the proposed technique both use position error only as the state variable.The distribution of pseudo-range measurement error is 7.5 m, and the difference of differential phase measurement error is 0.71 m /. sec was used. The parity technique assumes that the variance of pseudorange measurement errors at each moment is 7.5m under normal conditions, and the proposed technique assumes that the variances of thermal noise included in the pseudorange and cumulative phase are 1.5 m and 0.025 m, respectively. The reason that the parity technique and the proposed fault detection technique use different measurement error variances is that the measurement surplus used in the parity technique includes all of the measurement errors including satellite orbit errors, tropospheric delay, ionospheric delay, and thermal noise. The technique eliminates most of the other sources of error except for thermal noise in the process of double difference.

The experiment was carried out for 10 minutes and the receiver secured 7 satellites in all sections. When using the parity technique, the degree of freedom of the associated c2-distribution is determined by the number of visible satellites as (visibility number) -4. Therefore, the parity technique has a disadvantage that it is not fundamentally applicable when the number of visible satellites is 4 or less. In this experiment, we have 7 satellites, so the degree of freedom of c2-distribution associated with the parity technique is fixed to 7-4 = 3. The proposed channel-specific fault detector can be applied to any number of visible satellites, and the associated c2-distribution can be adjusted according to the number of consecutive measurements used for fault detection. In the experiment, five consecutive measurements are used every 5 seconds, so the degrees of freedom of the associated c2-distribution are five.

3 and 4 show test values extracted by using a patter technique and a proposed technique, respectively. Comparing FIG. 3 with FIG. 4, it can be seen that the proposed technique is easier to distinguish between normal and abnormal situations because the sensitivity of the (maximum test value) / (minimum test value) is greater than that of the parity technique. Therefore, it can be seen that the proposed scheme is not significantly affected by the determination of the channel where the failure occurs even if the threshold value changes significantly compared to the parity scheme.

5 shows a failure detection result determined by a parity technique using a threshold of 0.58 (90% probability of false warning) and a method according to the present invention applying a threshold of 11.10 (5% probability of false warning). The reason why the false alarm rate is set high in the parity technique is to equalize the distribution of fault satellites detected by the two methods with different sensitivity to faults. This is because when used, it is difficult to detect errors caused by multipath. For reference, if the probability of false warning is set to 10% in the parity technique, the threshold value becomes 6.25 in 3 degrees of freedom, and thus no failure is detected as shown in FIG. 3. As shown in FIG. 5, the two detection methods were differently determined to be satellites 20 and 22, respectively, at the beginning of the experiment at 114 seconds and 144 seconds.

Satellites 20 and 22 were excluded to identify failures caused by real multipaths and channels, and they were placed for 150 seconds without applying specific failure detection techniques. 6 shows its result. As shown in FIG. 6, when the location is generated except for satellite 20, the location is densely distributed in a certain area, whereas when the location is created except for satellite 22, the location is maintained in a specific direction and is 1 to 2 m each. You can find phenomena peculiar to multipaths that gradually move. Therefore, it can be concluded that the pseudoranges of satellite 20 detected by the proposed technique include multipath errors.

7 is located in the range of 0 seconds to 600 seconds, 1) least squares using instantaneous measurements without separate fault detection technique, 2) located by Kalman filter associated with parity technique, and 3) located by estimated technique. Each distribution is shown. When the position is calculated in the drawing without applying the failure detection technique, it is possible to find a phenomenon unique to the multipath in which the position is moved in the ECEF y-axis direction. In addition, it can be seen that the Kalman filter associated with the parity technique and the proposed technique generate trajectories of almost the same location in the range of 0 seconds to 110 seconds where the failure is not detected by both failure detection techniques. On the other hand, comparing the estimated trajectories of the located positions after the 110-second interval where the first multipath error occurs, it can be seen that the proposed method does not change much in the direction of the multipath generation compared to the Kalman filter associated with the parity method. Therefore, it can be concluded that under the multipath environment, the proposed scheme is more efficient in detecting / separating and positioning the faulty satellite than the Kalman filter associated with the parity technique.

The present invention proposes an efficient technique that can simultaneously perform real-time navigation and code multipath error detection regardless of the static or dynamic environment, regardless of the type of receiver, in conjunction with the use of a compression filter. It was. The proposed code multipath error detection technique uses the difference between the current and previous code transfer values as an observer and uses the observer's time series instead of the instantaneous observer to improve the reliability of fault detection. The proposed technique uses only the basic measurements provided by the GPS receiver without using additional information by a separate auxiliary sensor, and detects the abnormality of GPS measurements by multipath independently for each channel while the receiver is moving. As a result, it is expected to be applicable to many fields.

The present invention proposes a multipath error detection technique using pseudo-ranges and cumulative phase time series for each channel and a compression filter structure for real-time navigation. In the standalone GPS experiment, the proposed multipath error detection method can improve the sensitivity of fault detection by separating the errors caused by satellite orbit error, tropospheric delay, and ionospheric delay from multipath error and thermal noise. In addition, the proposed filter structure is estimated to be similar to Kalman filter under normal conditions without multipath error, and to be less affected by multipath than Kalman filter associated with parity technique under multipath error. I could check the zoom.

The proposed technique enables the detection of various signs of failures regardless of the dynamic / static conditions of the receiver by using only the pseudorange and cumulative phase measurements that can be provided by a general GPS receiver. It is expected. In addition, it is expected that it will be easily applied to various applications such as DGPS and GPS / INS combined navigation through some changes.

In addition, the present invention enables adjustment between detection reliability and calculation amount by determining the number of samples under consideration.

Although the invention has been described with reference to the examples, many various modifications are possible to those skilled in the art without departing from the scope of the invention. Such obvious modifications of the invention are intended to be covered by the appended claims.

Claims (13)

a) providing a plurality of consecutive pseudoranges and cumulative phase measurements provided by a GPS receiver; b) calculating a permutation of the difference between pseudorange time increments and cumulative phase time increments for the plurality of measurements; c) determining the presence of multipath noise from the degree of change in the permutation; Multipath error detection method of a positioning receiver using a satellite comprising a. The process of claim 1, wherein step c) is: c1) converting a permutation orthogonal to the permutation; c2) calculating a square sum of each of the orthogonal permutations divided by the error covariance; c3) determining a channel having a multipath error according to a result of comparing the calculated observer value with a reference value; Multipath error detection method of a positioning receiver using a satellite comprising a. The method of claim 2, wherein the reference value of step c3) is: The method of detecting a multipath error of a positioning receiver using a satellite, characterized in that calculated using the statistical distribution characteristics of the observer. The method of claim 3, wherein the reference value is: A method for detecting a multipath error of a positioning receiver using a satellite, characterized in that the observer values are calculated using a characteristic having a chi square distribution. The method of claim 4, wherein the reference value is: A method for detecting a multipath error of a positioning receiver using a satellite, characterized in that it is determined according to a given false alarm probability and an undetected probability in a chi squared distribution of observer values. A plurality of satellite receivers; A plurality of filters for filtering to increase the accuracy of a signal received from each of the reception channels of the plurality of receivers; Compute a permutation of the difference between the pseudorange time increment and the cumulative phase time increment for a plurality of measurements corresponding to the pseudorange and cumulative phase measurements received from each of the plurality of channels, and from the degree of change of the permutation, A plurality of multipath detectors for determining the presence or absence of path noise; A plurality of selectors for limiting selection of a corresponding channel according to a detection result of the weighted path detector; Positioning device using a satellite, characterized in that it comprises a. 7. The apparatus of claim 6, wherein the multipath detector is: And determining a channel having a multipath error according to a result of comparing a square sum of values obtained by dividing each orthogonal permutation by an error covariance with a reference value. The method of claim 7, wherein the reference value is: Positioning device using a satellite, characterized in that the observer values are calculated using a characteristic having a chi (χ) square distribution. The method of claim 8, wherein the reference value is: A device for positioning satellites, characterized in that it is determined according to a given false alarm probability and an undetected probability in a chi squared distribution of observer values. A plurality of satellite receivers; A calculation matrix for storing a calculation matrix for calculating a degree of permutation of the difference between the pseudo distance time increment and the cumulative phase time increment for a plurality of measurements corresponding to the pseudo distance and cumulative phase measurements received from each of the channels of the plurality of receivers; A storage unit; An observer calculator for calculating observer values by calculating in real time a plurality of measurements received from the matrix stored in the storage unit and the channel of the receiver unit; A plurality of selectors for restricting whether to select a corresponding channel according to a detection result of the observer calculator; Positioning device using a satellite, characterized in that it comprises a. The method of claim 10, wherein the calculation matrix storage unit: And a matrix for orthogonalizing and normalizing a permutation of a difference between pseudorange time increments and cumulative phase time increments received from each channel. The method of claim 10, wherein the reference value is: Positioning device using a satellite, characterized in that the observer values are calculated using a characteristic having a chi (χ) square distribution. The method of claim 12, wherein the reference value is: A device for positioning satellites, characterized in that it is determined according to a given false alarm probability and an undetected probability in a chi squared distribution of observer values.