MULTIPATH IDENTIFICATION BY COMPARING TWO DIFFERENT GNSS SIGNALS

Abstract

A method for detecting a multipath effect in a GNSS receiver which is designed to receive different signals from a GNSS satellite and includes a parameter which is determined from directly received signals and has a substantially constant target value, including the steps receiving at least two mutually independent signals; determining a current parameter value from at least the first and the second signal; evaluating the parameter value in relation to the target value, and detecting a multipath effect when the parameter value has a deviation (ΔK) which deviates from the already known target value.

Claims

1. A method for detecting a multipath effect in a GNSS receiver which is designed to receive different signals from a GNSS satellite and includes a parameter which is determined from directly received signals and has a substantially constant target value, comprising: receiving at least two mutually independent signals, determining a current parameter value from at least the first and the second signal, evaluating the parameter value in relation to the target value, and detecting a multipath effect when the parameter value has a deviation which deviates from the already known target value.

2. The method according to claim 1, wherein the parameter value includes an atmospherically induced error in the signals.

3. The method according to claim 2, wherein the target value of the parameter is newly determined after expiry of a predefined period, upon the occurrence of a predefined event or upon environmental changes.

4. The method according to claim 1, wherein the deviation includes that the parameter value exceeds a predefined upper and lower threshold value.

5. The method according to claim 1, wherein the deviation includes that the parameter value exceeds a first tolerance range within a predefined time period.

6. The method according to claim 5, wherein the change in the parameter over time has a trend.

7. The method according to claim 5, wherein the first tolerance range is set in relation to a measured value from the first and/or second signal.

8. The method according to claim 5, wherein the deviation includes that the parameter value exceeds a second tolerance range.

9. The method according to claim 9, wherein the second tolerance range is set in relation to the signal strength of the first and/or the second signal.

10. The method according to claim 1, wherein the signals differ in terms of their carrier frequency.

11. The method according to claim 1, wherein a GNSS L1 signal and an L2 or L5 signal or other combinations of the carrier frequencies are compared with one another.

12. The method according to claim 1, wherein the signals differ in terms of their modulation frequency.

13. The method according to claim 1, wherein a C/A signal and P/Y signal are compared with one another.

14. The method according to claim 1, wherein, after detecting a multipath effect, the received GNSS signals are marked as erroneous.

15. A system for a motor vehicle, having a receiver for receiving satellite signals for determining an actual position and for implementing a method according to claim 1, wherein the receiver is designed to receive and to process signals of different frequencies.

16. The method according to claim 6, wherein the first tolerance range is set in relation to a measured value from the first and/or second signal.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The invention is described in more detail below with reference to an exemplary embodiment and drawings, which show:

[0035] FIG. 1 a block diagram of a system according to the invention for implementing the method according to the invention; and

[0036] FIG. 2 an exemplary illustration of a travel-time measurement of a signal.

[0037] FIG. 3a, 3b an exemplary illustration of the parameter determination with reference to the travel-time measurement of two signals.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0038] FIG. 1 shows an exemplary embodiment of a system 1 according to an aspect of the invention, including a receiver or multi-frequency receiver A, which is designed to receive two different mutually independent signals S1, S2 at different frequencies f1, f2. A multi-frequency receiver A of this type is well known from the prior art.

[0039] The choice of signals to be observed can be made depending on the receiver and the availability of the signals. However, it is advantageous to use two signals which differ in terms of their carrier frequency, modulation frequency and/or signal type, for example C/A and P/Y signal. It is also possible to use more than two signals for multipath evaluation.

[0040] FIG. 2 shows the correlation of 3 phase-shifted replicas to a received signal S, namely E (Early), P (Prompt) and L (Late). The time shift is applied to the x-coordinate, wherein the point of origin represents the start of the time mark of the receiver. For brevity of terminology, the common English scientific terms for the replica signals are used below.

[0041] The phase shift from Prompt to the other replicas Early and Late amounts to one nav-chip length in each case, which corresponds to 10 ms for a C/A code and 1 ms for a P/Y code. The illustration in FIG. 2 shows the necessary time shift of the replicas Early and Late so that they correspond with the received signal S. This results in the travel-time measurement with a duration of t2 for the correlation maximum corr_max.

[0042] If, instead of only one signal, two signals S1, S2 are now received and the distance from the satellite or the signal travel-time measurement therefor is determined for both signals in relation to the same instant, for example C/A and P/Y code, the results should be approximately identical since both signals S1, S2 should have the same distance from the satellite at the same instant.

[0043] Reflections have a slightly different effect on the different signals. Added to this are influences resulting from different measuring methods owing to different sampling rates and resolutions.

[0044] A signal which is distorted by multipath would therefore result in the instant t2 at which the replica signal Prompt reaches the correlation maximum corr-max with the received signal being shifted by a certain value to t2′, for example 5% of the nav-chip length. The result of such a shift would be that the multipath influence on a signal would amount to 500 μs=149500 m according to the C/A code and yet only 50 μs=14950 m based on the P/Y code. Since the distance between C/A and P/Y should be approximately identical but, in this exemplary case, there is a mutual deviation of 135 km, the multipath can be detected via the method.

[0045] FIGS. 3a and 3b show the correlation peak and therefore the signal travel time for the signals S1 and S2.

[0046] FIG. 3a shows the offset of both signals under ideal conditions and therefore forms the target value of the parameter. Since, owing to different signal properties (e.g.: different frequencies), the signal travel time of both signals is slightly different—signal 2 is more delayed on the transmission path (e.g. as a result of the ionosphere)—the result is a mutual offset of both signals.

[0047] FIG. 3b shows how the current parameter value changes when multipath is present, which has a different effect on the two signals owing to different signal properties. Signal S1 and S2 are both delayed by the multipath, but signal S2 is more delayed than S1.

[0048] It is therefore possible to detect a multipath effect from the changed offset between the two signals S1 and S2. The target value of the parameter would then be the difference or the offset in the travel-time measurement between the two signals S1, S2, preferably minus the respective distortion of the travel times due to atmospheric factors. The current parameter value would then need to be determined on the basis of the currently received signals, one of which can be falsified by multipath.

[0049] The normal target value of the parameter could have, for example, an offset of 5 m between the signals. If the current parameter value, i.e. offset for a signal pair, jumps to 15 m, it should be assumed that a multipath effect is present. The difference between the two signals can also be determined in other units, for example in nav-chip lengths.

[0050] A further example for a parameter is described below. The influence of atmospheric disturbance effects can be determined from the received signals by means of multi-frequency receivers. This variable can likewise be used as a parameter. It takes into account, amongst other things, the influence of the ionosphere and troposphere on the breaking and absorption of the signals. The parameter depends for the most part on a plurality of factors in the ionosphere and troposphere, such as e.g. time of day, humidity, temperature, cloud cover etc. and can be assumed to be constant over a certain time period. Based on this assumption, the target value of the parameter can also be used as a constant offset. If the target value is subtracted from the current parameter value or the current atmospheric offset, a residual difference remains which is not atmospherically induced and can be attributed to multipath effects. The differential value determined in this way is therefore also suitable for detecting different types of deviation due to multipath.

[0051] A parameter within the context of the invention is therefore preferably an offset between the received signals, which generally has a constant value and can have different deviations from the target value as a result of multipath.

[0052] For the comparison between the target value and the current parameter value, the system 1 has a computing unit C which determines the difference K from the two variables. The system furthermore has a plurality of comparison units 20, 30, 40 which are suitable for detecting unique, creeping multipath effects or an influence on noise behavior of the signals which is changed by multipath based on the difference ΔK.

[0053] The first comparison unit 20 compares the difference ΔK in relation to an upper and lower threshold value s_max, s_min. The threshold values can be specified relatively in relation to the target value of the parameter or in relation to an absolute value. If the difference K exceeds one of these threshold values s_max, s_min, it can be established that one of the signals S1, S2 has not reached the receiver on the direct path and is therefore erroneous and should be discarded. It is thus possible to detect individual runaways in the parameter value owing to multipath effects.

[0054] The second comparison unit 30 checks whether a slow or creeping change in the difference ΔK is present. To this end, a time period t_D is defined, in which the difference ΔK must remain within a tolerance range D. A so-called drift of the signals cased by multipath is thus detectable. The current parameter value, or the offset in the travel-time measurement, could change by several millimeters per second within several seconds. The tolerance range specified for a time period or a time window ensures that such drifts are detected.

[0055] The comparison unit 40 takes into account whether the multipath has a negative effect on the noise behavior. To this end, a further tolerance range D_r is introduced, which is set depending on the signal strength.

[0056] The invention has the advantage that the computing costs for detecting multipath effects are relatively low, since it is only necessary to determine the current parameter value from the two signals and to compare it with a substantially constant target parameter value.