DE102011110658A1 - Method for satellite-based determination of position of satellite receiver for mobile device, involves applying distance estimation for position determination on detecting intervisibility between satellite receiver and satellite - Google Patents

Abstract

The method involves detecting whether the satellite receiver has intervisibility with the satellite, and applying distance estimation generated based on a delay locked loop algorithm for position determination when the intervisibility between the satellite receiver and the satellite is detected. The distance estimation generated based on an estimation algorithm, particularly particle receiver algorithm for position determination is applied when the intervisibility between the satellite receiver and the satellite does not exist. The detection of intervisibility with satellites is carried out by measuring the strength of the satellite signal, and the intervisibilty is recorded when the strength of the satellite signal exceeds a predetermined threshold.

Description

The invention relates to a method for satellite-based determination of the position of a satellite receiver.

With satellite receivers, reflections of the signal during transmission on, for example, trees, houses, etc. result in multipath propagation. For their suppression, various methods are used in the prior art. These include estimation algorithms, such as maximum likelihood algorithms or particle receiver algorithms. Furthermore, Delay Locked Loop (DLL) algorithms are known from the state of the art. These can be supplemented to suppress the multipath propagation by various methods, for example by using double delta corrals. The latter DLL algorithms offer the advantage of less complexity.

• 1) Jesus Selva Complexity reduction in the parametric estimation of superimposed signal replicas Signal processing, 2004, Elsevier
• 2) Lentmaier, Michael und Krach, Bernhard Maximum Likelihood Multipath Estimation in Comparison with Conventional Delay Lock Loops The 19th International Technical Meeting of the Institute of Navigation Satellite Division ION GNSS 2006, Proceedings CD-ROM. The Institute of Navigation USA. ION 2006, 2006-09-26–2006-09-29, Fort Worth, TX (USA) .

Maximum likelihood algorithms and DLL algorithms are described in the following publications:

• 1) Jesus Selva Complexity reduction in the parametric estimation of superimposed signal replicas signal processing, 2004, Elsevier
• 2) Lentmaier, Michael and Noise, Bernhard Maximum Likelihood Multipath Estimation in Comparison with Conventional Delay Lock Loops The 19th International Technical Meeting of the Institute of Navigation Satellite Division ION GNSS 2006, Proceedings CD-ROM. The Institute of Navigation USA. ION 2006, 2006-09-26-2006-09-29, Fort Worth, TX (USA) ,

Further algorithms are described in:
Krach, Bernhard and Lentmaier, Michael Efficient Soft-Output GNSS Signal Parameters Estimation using Signal Compression Techniques 3rd ESA Workshop on Satellite Navigation User Equipment Technologies NAVITEC '2006, 2006-12-11-2006-12-13, Noordwijk, Netherlands ,

Particle receiver algorithms are described in the following publication:
Krach, Bernhard and Robertson, Patrick and Weigel, Robert An Efficient Two-Fold Marginalized Bayesian Filter for Multipath Estimation in Satellite Navigation Receivers EURASIP Journal on Advances in Signal Processing, 2010 Hindawi. DOI: 10.1155 / 2010/287215. ISSN 1687-6172 ,

The algorithms mentioned have the disadvantage that they are either very complex (particle receiver algorithms) and thus cause high energy consumption in the terminal or are not particularly powerful (DLL algorithms with double delta correlator), so that depending on the reception situation Multipath suppression is not possible.

The object of the invention is to provide a method for satellite-based determination of the position of a satellite receiver, which allows improved suppression of multipath propagation and / or reduced power consumption.

The object is achieved according to the invention by the features of claim 1.

In the method according to the invention, a reception signal of a satellite navigation receiver is evaluated by the following steps: It is first detected whether the satellite receiver has visual contact with the satellite. Detecting whether there is visual contact with the satellite (that is, whether there is a line of sight connection) can be realized by various methods known in the art. Preferably, for each available satellite, separate detection takes place as to whether there is visual contact with that satellite.

In case the detection shows that there is visual contact between the satellite receiver and the satellite, a distance estimation based on a delay locked loop (DLL) algorithm is used to determine the position.

In the event that the detection shows that there is no visual contact between the satellite receiver and the satellite or that the line-of-sight connection is not longer than for a defined period of time, a distance estimation based on an estimation algorithm is used for the position determination. This estimation algorithm can be, for example, a particle receiver algorithm, a maximum likelihood algorithm or other estimation algorithms known from the prior art. All features illustrated in the present application, which relate, for example, to a particle receiver algorithm, can also be used in other estimation algorithms. The method according to the invention preferably checks on each individual satellite whether there is a line of sight to these satellites. If there is a line-of-sight connection to this satellite, the DLL algorithm can be used for the further processing of the signals received from these satellites, while the more complex estimation algorithms can be used to process the signals of other satellites that are not in line-of-sight ,

The method of the invention offers the advantage that the DLL algorithm can be used with its low complexity as long as there is visual contact between the satellite receiver and the satellite. In this phase, the more complex estimation algorithm, which causes increased power consumption in the terminal in which the satellite receiver is implemented, need not be applied. However, since the DLL algorithm does not provide sufficient suppression of multipath propagation in the event of a non-existent or non-constant line-of-sight connection between the satellite receiver and the satellite, a more complex estimation algorithm will be provided in the event that there is no visual contact or is not long enough used. The increased power consumption at the terminal, which is generated by this more complex estimation algorithm, thus occurs only during the NLOS (non-line of sight) phases, during which there is no visual contact between the satellite receiver and the satellite, or during the phases which this visual contact is not long enough.

According to the invention, an adaptive switching on and off of the complex multipath suppression algorithm thus takes place, depending on the reception situation. The inventive method is particularly suitable for mobile devices, in which thereby the battery life can be increased.

The inventive method also has the advantage that the hardware of the satellite receiver used can be made less complex, since it does not have to be so powerful. In other words, it is possible to design a less complex hardware that provides the same performance for most standard navigation cases and works poorly for only a few, less likely, special cases (eg, very many satellites). For example, a satellite receiver implementing the inventive method may be configured to be capable of processing six visible and six shadowed satellite signals. It could thus have six DLL's and six Particle receivers and still provide a similar performance in most cases due to the inventive adaptive use of the algorithms used, such as a satellite receiver in which twelve Particle Receiver are implemented. Of particular interest is the inventive method when less than four satellites are visible to a satellite receiver.

Furthermore, it is possible to weight the evaluated satellite signals so that more important satellite signals are preferred to less important signals. To assess the importance of a satellite, account is taken of geometric factors relating to the position of the satellite. A satellite with a low G-DOP factor is thus preferred to satellites with a higher G-DOP factor. In other words, the evaluation of unimportant satellites can be omitted unless their evaluation provides a significant improvement in performance. This can save electricity.

The detection of a visual contact between the satellite receiver and the satellite can for example be done by measuring the strength of the satellite signal. Here, a visual contact is assumed if the strength of the satellite signal exceeds a certain threshold.

This threshold can be determined adaptively. For an adaptive determination of the threshold value, for example, the course of the signal strength of the satellite for a certain period of time in the past can be considered. For example, a viewing for the last two hours is possible. Now, a histogram is created showing the frequency at which each signal strength occurred. The threshold value can now be selected such that a certain proportion of the signal strengths from the histogram lies above this threshold value. By way of example, the threshold value can be selected such that 10% of the signal strengths from the time period considered are above the threshold value.

As an alternative method for the adaptive determination of the threshold, this can be taken over by other satellites.

Furthermore, a middle elevation model can be used. This takes into account how strong a signal from a satellite would be at a given elevation. Any signal that is weaker by a certain amount than this expected signal at a given elevation is assumed to be a shadowed signal. For example, a satellite may be considered shadowed if the received signals are 5 dB weaker than the expected satellite signals at the particular elevation.

It is preferred that both the estimation algorithm and the DLL algorithm use correlators that correlate the received signal with a conjugate complex mapping of the transmit sequence. The correlators do not necessarily have to be hardware and can also be software correlators. Such correlators are known from the aforementioned prior art. The output of these correlators is fed to the estimation algorithm and the DLL algorithm (subspace estimation). When switching from the estimation algorithm to the DLL algorithm, a part of the correlators used in the estimation algorithm is turned off, so that only permanent correlators of the DLL algorithm are used. Operation without correlators is also possible.

It is preferred that 60 correlators be used in the estimation algorithm, particularly the particle reveicer algorithm. The particle receiver algorithm preferably uses 5-80, 10-70 or particularly preferably 30-70 correlators. When switching from the estimation algorithm to the DLL algorithm, at least one correlator is switched off. Preferably, a plurality of correlators and in particular the majority of the correlators are switched off, so that only a small number of correlators are used in the DLL algorithm. The number of then still used correlators is preferably less than 10 and more preferably less than 5. For example, double delta correlators can be used. In a particularly preferred embodiment, at least 57 correlators can be switched off when switching from the estimation algorithm to the DLL algorithm, so that in the DLL algorithm only a maximum of three correlators are used.

In a preferred embodiment, the estimation algorithm is used when the line-of-sight connection between the satellite receiver and the satellite has not existed for more than a defined period of time, with the defined period of time being between 0.5-10 seconds, preferably 3-7 seconds to 5 seconds. The switchover from the estimation algorithm to the DLL algorithm is thus delayed. This prevents problems of the DLL algorithm by the so-called "Short Line of Sight Hit" (SLOSH). In this phenomenon, a shadowing phase (NLOS) is followed by a short phase of line of sight (LOS) followed by another shadowing phase. DLL algorithms often react with a lock-out (loss of lock). Since this problem does not occur with estimation algorithms, such as particle receiver algorithms, the o. G. Switching from the estimation algorithm to the DLL algorithm preferably delayed. The resulting unnecessary power consumption is relatively small and can be accepted. Conversely, it is preferred that the change from the DLL algorithm to the estimation algorithm be lost immediately upon loss of line of sight. H. without delay, takes place.

In a further preferred embodiment of the invention, during the NLOS phase, in which there is no line of sight connection between the satellite receiver and the satellite, the storable elements, in particular the integrators of the DLL algorithm, are supplied with the distance estimation of the estimation algorithm. Thus it is achieved that when switching to the LOS phase in which there is a line of sight, the distance estimation of the estimation algorithm is already preloaded in the DLL algorithm and the DLL algorithm compulsorily follows the estimation algorithm. There is thus a support of the DLL algorithm by the estimation algorithm during the NLOS phase. For this purpose, an alternative memory is preferably used, which is set when changing from the NLOS phase to the LOS phase so that the output of the estimation algorithm and the DLL algorithm are identical for this time. Subsequently, the output of the memory is held again.

In the following, preferred embodiments of the invention will be explained with reference to figures.

Show it:

1 Simulation results of the entire reusable channel,

2 Simulation results only for LOS,

3 Block diagram of a satellite receiver in which the inventive method can be realized and

4 Block diagram of a portion of a satellite receiver in a preferred embodiment of the method according to the invention.

1 and 2 show simulation results identified in the ESA project mentioned in the introduction. Various receiver techniques were tested on a multi-path channel model (according to ITO Recommendation 681-7). The evaluation was done in two ways: according to 1 a complete analysis was made. The navigation errors of the receivers were evaluated until the loss of synchronization. A loss of lock occurred in 73.2% of the simulated cases. In the simulation according to 2 Loss of Lock has occurred in 14.2% of simulated cases.

According to 2 a LOS analysis was performed. Thus, of the full analysis data, only those data were used in which the satellite was optically visible, that is, during which there was a line of sight connection between the satellite receiver and the satellite.

Again 2 can be taken, supply the particle receiver algorithm and the Standard DLL algorithm during LOS reception a comparable performance. In the case of full analysis, however, it can be seen that the Particle Receiver algorithm provides much better accuracy than the DLL receiver.

Thus, according to the invention, the more powerful estimation algorithm is only used adaptively.

3 shows an exemplary structure of a satellite receiver, in which the inventive method can be realized. The received signal is first amplified by means of a low noise amplifier and then forwarded to the correlators (KN). These correlators are known in the art, for example Parkinson / Spilker Global Positioning Systems Theory and Applications , The result of the correlators is fed to a DLL algorithm.

A line of sight detector detects whether the receive level is above or below a defined threshold. Thus, the LOS case can be detected.

If the LOS detector detects the LOS case, the output of the DLL algorithm is used as a distance estimate to determine position (position solver not shown). This distance estimation also serves to determine the offset of the correlators and is transmitted to all correlators (see bottom arrow).

During the LOS phase, the correlators 4-60 can be turned off, so the particle receiver algorithm does not have to be executed. It can thus be saved a large amount of energy.

If the LOS detector detects an NLOS phase, the remaining correlators (in this case 4-60) are switched on and the particle receiver algorithm takes over the distance estimation. The power consumption thus increases. The number of in 3 shown correlators is merely exemplary. Importantly, the DLL algorithm uses significantly fewer correlators than the estimation algorithm.

In 4 an advantageous development of the method according to the invention is shown. In this case, the particle receiver algorithm in the NLOS phase supports the DLL algorithm so that when the line of sight returns, the integrators of the DLL are already preloaded. For this purpose, the distance estimation of the Particle receiver instead of the DLL detector output is used to control the DLL. The DLL thus forcibly follows the particle receiver.

The detector shown in the figure is merely exemplary. Alternatively, any other detector (eg Double Delta or similar) may be used. Also, the loop filter is merely exemplary, so any other suitable filter can be used. For example, only one PI element or one PID controller can be used. It is essential that the estimated result of the estimation algorithm, in this case the particle receiver, is fed into the loop filter. Thus, integrators contained herein are fed continuously.

In the embodiment according to 4 an alternative memory is advantageous.

As an alternative LOS detector, the particle receiver can also be used to detect the LOS case. For this purpose, during the NLOS phase, the probability or the variance of the estimate can be taken into account and decided by means of a threshold value. Of course, such detection of the LOS case can only be done during the NLOS phase in which the Particle Receiver is used. In the LOS phase, the detection of the line-of-sight connection is preferably carried out again via the power of the received signal, since in this phase the particle receiver is switched off.

In support of the DLL algorithm according to 4 the probability distribution of the Particle receiver can be taken into account: Combining the probability of estimation with the filing itself (for example, by a multiplication) so the integrators are heavily pre-charged by very probable estimates of filing by the Particle receiver, whereas by improbable estimates weak pre-charging takes place.

As already mentioned, different methods can be used as estimation algorithms. A maximum likelihood method is preferably used as the estimation algorithm for slowly moving receivers.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited non-patent literature

• Jesus Selva Complexity reduction in the parametric estimation of superimposed signal replicas Signal processing, 2004, Elsevier [0003]
• Lentmaier, Michael and Noise, Bernhard Maximum Likelihood Multipath Estimation in Comparison with Conventional Delay Lock Loops The 19th International Technical Meeting of the Institute of Navigation Satellite Division ION GNSS 2006, Proceedings CD-ROM. The Institute of Navigation USA. ION 2006, 2006-09-26-2006-09-29, Fort Worth, TX (USA) [0003]
• Krach, Bernhard and Lentmaier, Michael Efficient Soft-Output GNSS Signal Parameters Estimation using Signal Compression Techniques 3rd ESA Workshop on Satellite Navigation User Equipment Technologies NAVITEC '2006, 2006-12-11-2006-12-13, Noordwijk, Netherlands [0004]
• Krach, Bernhard and Robertson, Patrick and Weigel, Robert An Efficient Two-Fold Marginalized Bayesian Filter for Multipath Estimation in Satellite Navigation Receivers EURASIP Journal on Advances in Signal Processing, 2010 Hindawi. DOI: 10.1155 / 2010/287215. ISSN 1687-6172 [0005]
• Parkinson / Spilker "Global Positioning Systems Theory and Applications" [0034]

Claims (8)

Method for satellite-based determination of the position of a satellite receiver, comprising the steps of: Detecting whether the satellite receiver has visual contact with the satellite, In the case of visual contact between the satellite receiver and the satellite: use of a distance estimation for position determination based on a delay-locked-loop (DLL) algorithm, - in the event that there is no visual contact between the satellite receiver and the satellite or the line of sight is not longer than for a defined period of time: using a based on an estimation algorithm, in particular a Particle Receiver algorithm or maximum likelihood algorithm, created distance estimation for position determination. A method according to claim 1, characterized in that the detection of a visual contact to the satellite is carried out by measuring the strength of the satellite signal, wherein a visual contact is assumed when the strength of the satellite signal exceeds a certain threshold. A method according to claim 2, characterized in that the threshold value is determined adapted. A method according to claim 2 or 3, characterized in that the threshold value is derived from the position, in particular from the elevation, of the satellite. Method according to one of claims 1-4, characterized in that the estimation algorithm and the DLL algorithm correlators are used, which correlate the received signal with a complex conjugate image of the transmission sequence, and the output of the correlators the estimation algorithm and the DLL algorithm is supplied , wherein when switching from the estimation algorithm to the DLL algorithm, a part, in particular at least one of the correlators used in the estimation algorithm is turned off and only permanent correlators of the DLL algorithm are used. A method according to claim 5, characterized in that the estimation algorithm 60 correlators are used, of which at least 57 are turned off when switching from the estimation algorithm to the DLL algorithm, so that the DLL algorithm only more than three correlators are used. A method according to any one of claims 1-6, characterized in that the estimation algorithm is used when the line-of-sight connection between the satellite receiver and the satellite has not existed for more than a defined period of time, the defined time being set at a value between 3 and 7 seconds , so that the switchover from the estimation algorithm to the DLL algorithm is delayed. Method according to one of claims 1-7, characterized in that during the NLOS phase (non-line of sight), in which there is no line of sight connection between the satellite receiver and the satellite, the integrators of the DLL algorithm with the distance estimation of the estimation algorithm are fed such that when switching to the LOS (line of sight) phase in which there is line of sight, the range estimation of the estimation algorithm is already preloaded in the DLL algorithm and the DLL algorithm follows the estimation algorithm.

Images