Methods, systems, and computer program products for performing correlation in a positioning system

Abstract

A method performed in a positioning system is provided. The method comprises: (a) receiving, at a receiver, signal data from one or more remote sources; (b) measuring or assuming movement of the receiver; (c) determining a first direction in which signals received at the receiver are desired to be enhanced; (d) determining a second direction in which signals received at the receiver are desired to be suppressed; (e) obtaining a first and second sequence of phasors respectively indicative of the measured or assumed movement of the receiver in the first and second directions, the first and second sequence of phasors each comprising one or more phasors, the phasors comprising an amplitude and / or an angle; (f) generating a third sequence of phasors based on a weighted combination of the first and second sequence of phasors in accordance with the determined first and second directions; (g) providing a local signal; and (h) providing a correlation signal using the third sequence of phasors, wherein providing the correlation signal comprises correlating the local signal with the received signal data and combining at least one of the local signal, the received signal data and a result of the correlation with the third sequence of phasors such that signals received in the second direction are suppressed relative to signals received in the first direction. A corresponding positioning system is also disclosed.

Description

TECHNICAL FIELD

[0001] The present invention relates to a method, system and computer program product usable for performing correlation in a positioning system. More specifically, the correlation performed using the present invention enables a significant improvement in the accuracy of determining a position based on received positioning signals and measurements or assumptions of movement of the receiver. BACKGROUND

[0002] In Global Navigation Satellite System (GNSS) based positioning systems, such as the Global Positioning System (GPS), GLONASS (an acronym for Global Navigation Satellite System in Russian), Galileo, a receiver can determine its global position after receiving and analyzing signals received from satellites of a relevant constellation.

[0003] The signals transmitted by the satellites contain a pseudo random noise (PRN) code that allows the identification of the satellite. The receiver generates a local copy of the transmitted signal comprising the PRN code of the satellite and generates a correlation signal by correlating the local copy signal with the received signal. Multiple correlation measurements can be performed and the highest correlation peak can be used to determine the phase of the received code compared to the phase of the local copy code and thus the delay of the received signal compared to the local clock on the receiver. The time delay can be used to calculate the distance ("pseudo-range") from the receiver to the satellite. By performing this procedure for four satellites, the global position of the receiver can be determined.

[0004] The correlation procedure is performed when the receiver "locks" to a satellite (acquisition phase) and the subsequent tracking phase, wherein, as long as the acquired satellite is in use, the signal from the acquired satellite is tracked. During the tracking phase, a delay lock loop is typically used to adjust the offset of the local copy code, while a frequency lock loop and a phase lock loop can be used to adjust the frequency and phase of the local carrier signal in order to closely match the received signal over time, thus providing an accurate positioning solution.

[0005] However, this "lock" can not be present at the acquisition phase or be lost, for example, due to temporary signal or low signal to noise ratio (SNR). This results in a decrease in the positioning accuracy.

[0006] Further problems arise when the receiver is in an environment where multipath effects are prevalent. Multipath effects are where the signal from a satellite takes an indirect non-straight-line (NSL) trajectory from the satellite to the receiver, for example due to reflections off tall buildings in so-called urban canyons. Multipath effects cause two main problems. Firstly, the reflected signal from the satellite can have a higher absolute power than the expected straight-line (SL) signal, in which case the receiver can lock to the NSL signal. Due to the extra path length of the NSL signal compared to the expected SL trajectory, the pseudorange calculation is wrong. Secondly, the NSL signal can interfere (e.g. constructively or destructively) with the SL signal at the receiver, adversely affecting the receiver's ability to accurately process the received signal.

[0007] WO2017 / 163042 describes a method for providing so-called "motion compensation" to the correlation process in order to closely match the expected signal in the SL direction. The motion compensation comprises applying a phasor sequence indicative of the motion of the receiver to a local replica signal (carrier and PRN code) in order to provide a motion compensated local signal that can be correlated with the received signal. In particular, if the motion compensation is performed in the direction of the SL signal, then the highest correlation for the SL signal can be achieved (i.e. the receiver can "lock" to the expected SL signal) even if the absolute power of the SL is significantly less than the absolute power of the NSL signal to which a conventional receiver would lock.

[0008] However, even with the technique described in WO'042, if the expected SL signal has a very low absolute power, then the increased gain of the expected SL signal can not be sufficient to increase its signal-to-noise ratio above the NSL signal, in which case the receiver can still lock to the NSL signal. As a further problem, the spoofing signal can be significantly stronger than the SL signal, and the motion compensation can not be sufficient to prevent unintentional locking to the stronger spoofing signal.

[0009] Controlled Reception Pattern Antennas (CRPAs) exploit the fact that by implementing multiple antennas, spatial discrimination of the received signal is provided, typically receiving the undesired signal from different directions from the desired (e.g. SL) signal. However, such CRPA systems are bulky and expensive, especially for low-cost positioning devices such as smartphones and other hand-held devices.

[0010] Accordingly, there is a need for continued improvements in positioning systems. SUMMARY

[0011] According to a first aspect of the application, there is provided a method performed in a positioning system, comprising: (a) receiving, at a receiver, signal data from one or more remote sources; (b) measuring or assuming movement of the receiver; (c) determining a first direction in which it is desired to enhance signals received at the receiver along the first direction; (d) determining a second direction in which it is desired to suppress signals received at the receiver along the second direction; (e) obtaining a first sequence of phasors and a second sequence of phasors indicative of the measured or assumed movement of the receiver in the first and second directions respectively, the first and second sequences of phasors each comprising one or more phasors, the phasors comprising an amplitude and / or an angle; (f) generating a third sequence of phasors based on a weighted combination of the first and second sequences of phasors in accordance with the determined first and second directions; (g) providing a local signal; and (h) providing a correlation signal using the third sequence of phasors, wherein providing the correlation signal comprises correlating the local signal with the received signal data and combining at least one of the local signal, the received signal data and the result of the correlation with the third sequence of phasors such that signals received along the second direction are suppressed relative to signals received along the first direction.

[0012] The application comprises obtaining a first sequence of phasors and a second sequence of phasors indicative of the measured or assumed movement of the receiver in a first (i.e. desired) direction and a second (i.e. undesired) direction and generating a third sequence of phasors based on a weighted combination of the first and second sequences of phasors. The correlation is then performed using the third sequence of phasors. The inventors have realised that by performing the correlation using a weighted combination of sequences of phasors indicative of the movement of the receiver in the desired direction and the undesired direction, not only can a preferential gain be obtained along the desired first direction, but energy received at the receiver from the undesired direction can be actively attenuated. Thus, by combining at least one of the local signal, the signal data and the result of the correlation with the third sequence of phasors to provide the correlation signal, signals received along the second direction can be described as being actively suppressed relative to signals received along the first direction. This provides a significant advantage over prior art methods in which the correlation is performed using a sequence of phasors for the desired direction only, in which case signals from the undesired direction are not actively attenuated and in some cases can still (undesirably) result in the highest correlation peak.

[0013] The first direction can typically be any direction in which a signal is desired to be enhanced, by which we mean a preferential gain or signal-to-noise ratio compared to signals received from other directions, such that the highest correlation peak in the relevant signal is for a signal received along the first direction. The first direction is typically a straight-line (SL) direction from the remote source to the receiver. The straight-line direction can be defined as the direction along the shortest (straight-line) path between the receiver and the remote source. In some cases in practice, the straight-line path can pass through a building or other object. In indoor or urban canyon environments, the straight-line path can pass through several objects between the remote source and the receiver. The straight-line direction can typically be referred to as the "line of sight" (LOS) direction.

[0014] The remote source from which the signal is received along the first direction is typically a trusted remote source (e.g. a GNSS positioning satellite) from which the data received can be trusted, i.e. considered to be correct. However, in some embodiments, the remote source from which the signal is received along the first direction can be an untrusted source, for example in the case that a so-called "spoofing" device is to be detected.

[0015] The second direction is typically any direction in which a signal received therealong is desired to be actively suppressed below the level of the signal received along the first direction. Typically, the second direction is a non-straight-line (NSL) direction from a remote source to the receiver, or a straight-line direction from an untrusted remote source (e.g. a spoofing device) to the receiver. The non-straight-line direction is any path taken by a signal, for example due to reflection by a building or other object, which is not a straight-line path between the remote source and the receiver. (It will be appreciated that a signal received along a non-straight-line direction can have straight-line trajectories between reflections.) Thus, the first and second directions can refer to signals received from the same remote source, the first direction referring to the straight-line direction and the second direction referring to a reflected non-straight-line direction. The non-straight-line direction can typically be referred to as a "non-line-of-sight" (NLOS) direction.

[0016] It will be appreciated that the method can comprise obtaining a first sequence of phasors for one or more first directions, and obtaining a second sequence of phasors for one or more second directions, performing the weighting of the sequences of phasors to generate a third sequence of phasors. In typical cases, there is a single first direction (e.g. a SL direction to a positioning satellite) and multiple second directions (e.g. NSL directions to the positioning satellite and other remote sources). In this way, the present application is particularly advantageous in mitigating multipath effects to provide a higher-accuracy positioning solution.

[0017] The present invention can advantageously increase the coherent integration period of received signal data, thereby enhancing the ability to detect very weak signals (e.g. GNSS signals received indoors). For detecting weak signals, integration periods of about 1 second or more can be required. Here, the term "signal data" is used to denote the (typically multiple) received signals received at the receiver from one or more remote sources. The signal data comprises a first signal received along a first direction and a second signal received along a second direction.

[0018] The first and second sequences of phasors each comprise one or more phasors indicative of amplitude and / or phase variations introduced into the received signal data due to measured or assumed motion of the receiver. Each phasor comprises at least one of an amplitude and an angle describing a measured or assumed movement of the receiver in the respective direction.

[0019] Typically, the first and second sequences of phasors are derived from measured or assumed movements of the receiver over time. For example, each phasor in the sequence can be indicative of a measured or assumed movement during a particular time interval. The resulting sequence of phasors is thus indicative of (e.g. corresponds to) measured or assumed movements of the receiver during a time period consisting of the individual time intervals.

[0020] The sequences of phasors can reflect detailed movements of the receiver over time. For example, multiple phasors in the sequence of phasors can reflect movements of the receiver while being placed in a user's pocket during a jog, walk, run or some other repetitive motion of the user. In this example, the receiver can perform a cyclic motion with acceleration peaks corresponding to each foot ground contact.

[0021] The movements of the receiver can be measured or assumed. The movements of the receiver can be measured using data from one or more sensors configured to obtain data from which the movements can be measured. The one or more sensors can comprise inertial sensors, such as accelerometers and gyroscopes. Alternatively or additionally, the one or more sensors can comprise a barometric pressure sensor for indicating an altitude of the receiver, a geomagnetic sensor for indicating an orientation of the receiver, a visual odometry system and other sensors as will be appreciated by the skilled person.

[0022] In cases where it is not possible to measure the movements of the receiver (e.g. no motion sensor output from a motion sensor is available), the movements of the receiver can be assumed based on a pattern of receiver movements in a previous epoch.

[0023] In step (f), a third sequence of phasors is generated based on a weighted combination of the first and second sequences of phasors in dependence on the determined first and second directions. In other words, the weighting is such that when a correlation process is performed using the third sequence of phasors, a preferential gain is provided in the desired first direction, and active suppression is provided for signals received from the undesired second direction. Conversely, a correlation process performed using the first or second sequence of phasors will only provide gain in the respective direction, without providing active suppression in other directions in which signals are received at the receiver.

[0024] The third sequence of phasors is typically generated using an estimation process. The inventors have appreciated that an estimation or "fitting" technique performed based on the first and second sequences of phasors and a desired sensitivity profile in respect of the first and second directions provides a greatly improved preferential gain in the first direction and active attenuation of energy received from the second direction compared to a linear "addition" or "subtraction" of the first and second sequences of phasors. Thus, the estimation process is preferably based on (e.g. preferably "fits to") a desired enhancement of signals received in the first direction and a desired suppression of signals received in the second direction. The desired sensitivity profile can be represented in the form of weights indicative of the desired enhancement of signals received in the first direction and the desired suppression of signals received in the second direction, e.g. in the form of a vector comprising a series of "1"s and "0"s corresponding to the first and second directions respectively. Thus, the estimation process is preferably based on weights indicative of the desired enhancement of signals received in the first direction and the desired suppression of signals received in the second direction.

[0025] Preferably, the estimation process used to generate the third sequence of phasors is a least squares fitting process. Thus, in preferred embodiments, the third sequence of phasors is generated using a least squares fitting process (e.g. using a pseudo-inverse technique such as the Moore-Penrose pseudo-inverse) based on the desired enhancement of signals received in the first direction and the desired suppression of signals received in the second direction. However, other estimation or "fitting" techniques can be used to generate the third sequence of phasors, such as Bayesian inference or a least absolute deviation method.

[0026] Preferably, the third sequence of phasors W is generated by:

[0027] W = S + Z

[0028] where S is a (typically two-dimensional) matrix representing the first and second sequences of phasors, and Z is a vector representing the weights of the first and second directions. S + is a pseudo-inverse of the matrix S, which is typically non-square. Preferably, S +is the Moore-Penrose pseudo-inverse of matrix S. Vector Z typically comprises a series of "1"s and "0"s corresponding to the first and second directions respectively.

[0029] However, other ways of providing the third phasor sequence based on a weighted combination of the first and second phasor sequences are envisaged, such as genetic algorithms, look-up tables, simulations, brute force trials, Monte Carlo simulations, deep learning or cost functions.

[0030] In step (h) of the method, the third phasor sequence is used to provide a correlation signal, wherein providing the correlation signal comprises correlating the local signal with the received signal data and combining at least one of the local signal, the received signal data and the correlation result with the third phasor sequence. As described above, using the third phasor sequence in the correlation process advantageously actively suppresses one of the signals received along the second direction relative to the signal received along the first direction. Using the third phasor sequence in the correlation process is often referred to as "motion compensation" and the resulting correlation signal can be referred to as a "motion compensated" correlation signal.

[0031] It will be appreciated that the received signal data is typically made up of a plurality of signals received from a plurality of different directions. The received signals can comprise any known or unknown pattern of transmitted information, either digital or analogue, which can be found in the broadcast signal data by a cross-correlation process using a local copy of the same pattern. The received signals can be encoded with a chipping code which can be used for ranging. An example of such received signals comprises GPS signals which comprise a Gold Code encoded in the radio transmission. Another example is the spreading training sequence used in Global System for Mobile Communications (GSM) cellular transmissions.

[0032] Typically, phase variations in the received signal caused by variations in the straight-line path between the receiver and the remote source are considered to be a nuisance which reduces the accuracy of the positioning. The counter-intuitive approach of the present invention can actually exploit these phase variations to improve the recognition of the straight-line signal from the remote source, thereby improving the accuracy of the positioning.

[0033] The third phasor sequence can be combined with the local signal prior to the correlation so that it more closely matches the received signal. This is referred to as providing motion compensation to the local signal. In another arrangement, inverse motion compensation can be applied to the received signal data prior to the correlation to reduce the effect of the receiver motion on the received signal data. Similar results can be achieved by providing partial motion compensation to both the local signal and the received signal. These techniques allow relative motion compensation to be applied between the local signal and the received signal. In some embodiments, the motion compensation can be performed in parallel with the correlation. Motion compensation can also be applied directly to the result of the correlation.

[0034] In practice, the received signal can be processed as a complex signal comprising an in-phase component and a quadrature component. Similarly, the local signal can be complex. The correlation can provide a correlation signal, which can also be complex, and can be used as a measure of the correlation between these complex signals.

[0035] High positioning accuracy can be achieved by using the third sequence of phasors to provide motion compensation of at least one of the local signal and the received signal data. In practice, when applied to GNSS signals, the local signal and the received signal can be encoded with a periodically repeating code. For a GPS L1 C / A code, for example, the local signal and the received signal can comprise 1023 pseudo-random number chips. The local signal and the received signal can be analog waveforms, which can be digitised to provide values at the radio sampling rate, meaning there can be millions of values in a 1 ms time period. The correlation between the local signal digital values and the received signal digital values can be calculated, first using motion compensation phasors (of the third sequence of phasors) to correct either set of values for the correlation time period. These data points can then be summed over the time period. In practice, although it can require a large amount of computation, this can produce accurate results because it works at the radio sampling frequency.

[0036] Lower positioning accuracy can be achieved by providing motion compensation of the results of the correlation (i.e. combining the third sequence of phasors with the “initial” correlation signal produced by correlating the local signal with the received signal data to obtain a motion compensated correlation signal). In the above example, when applied to a GPS L1 C / A code, the correlation can be performed independently for each of the approximately 1000 pseudo-random number chips to produce approximately 1000 complex correlator signal outputs. Motion compensation phasors (of the third sequence of phasors) can then be applied to these approximately 1000 correlation signal components. Finally, the motion compensated correlation signals can be summed to produce a measure of the correlation. Thus, motion compensation of the results of the correlation of the local signal and the received signal data can produce an approximation of the results that would be achieved by motion compensating the local signal and the received signal prior to the correlation. However, for some applications, the loss of accuracy can be negligible and can be acceptable because it is able to reduce the computational load. Motion compensation of the correlation results can be successfully applied at a much lower rate than the chip rate of GPS (approximately 1 MHz). For example, motion compensation can be successfully applied over the length of one or more code words (<1 kHz) and can be applied to correlation outputs of length 5 ms (at a rate of 200 Hz).

[0037] The determination of the first direction(s) and the determination of the second direction(s) can be made in a number of different ways. The first direction can be determined based on a known or estimated position of the remote source. In a preferred embodiment where the first direction is a straight line direction of a GNSS positioning satellite, the position of the relevant positioning satellite is typically known, for example from a broadcast ephemeris. An initial position of the receiver can be measured or assumed, which in some cases can be quite coarse - for example, a city or region can be known based on terrestrial radio signals or a last known position.

[0038] The determination of the second direction can be based on an analysis of the received signal data. In general, the received signal data can be analysed to determine one or more second directions in which reception of signals is expected to be suppressed (e.g. reflected NSL directions). In particular, the received signal data can be analysed to determine the presence of reflected signals, and then active suppression of the reflected signals is expected.

[0039] In general, such an analysis can comprise generating a respective sequence of phasors for the one or more directions based on a measured or assumed movement of the receiver in the respective direction(s); for each of the directions, using the respective sequence of phasors to provide a direction-dependent signal, wherein providing the direction-dependent signal comprises correlating the local signal with the received signal data and correlating at least one of the local signal, the received signal data and a result of the correlation with the respective sequence of phasors, and determining the second direction based on an analysis of the respective one or more direction-dependent signals. In other words, the result of the motion compensation performed in the one or more directions can be indicative of the presence of signals received along the second direction, which are expected to be suppressed.

[0040] In one embodiment, the sequence of phasors can be generated for a straight line direction between the receiver and the remote source, and wherein the analysis further comprises determining whether the received signal comprises a component received in a direction different from the straight line direction, wherein the determination is based on a signal strength of the received signal data from the remote source and a signal-to-noise ratio of the direction-dependent signal. For example, if the signal-to-noise ratio of the direction-dependent signal (here using motion compensation in the SL direction) is relatively low, but the signal strength of the signal received from the respective satellite is high, this indicates the presence of a reflected signal, and the reflected signal dominates the signal received from that satellite.

[0041] In another embodiment, a "brute force" search of the sky across all elevation and azimuth pairs can be performed in order to determine the second direction. In such an embodiment, a respective sequence of phasors can be generated for a plurality of directions distributed over substantially all possible directions from which a signal can be received at the receiver, and wherein the analysis further comprises determining whether a signal is received along each of the plurality of directions based at least on a signal-to-noise ratio of the respective correlation signal. For example, if the signal-to-noise ratio of the correlation signal is about 1 or higher, this indicates that a signal is received along the respective phasor direction. Since the first direction is typically known, it is generally desirable to suppress such a recognition signal in the sky search.

[0042] The determination of the second direction can be based on a terrain model of the environment in which the receiver is located, wherein the model is used to predict the presence of a reflected signal received at the receiver. For example, such a terrain model can be a three-dimensional model of the city or region in which the receiver is located. The terrain model can be used prior to the received signal data in order to predict the directions in which a reflected signal can be received. In other words, the terrain model can be used to determine the second direction without the need to analyse the received data.

[0043] Alternatively, the terrain model can be used during the analysis of the received data, for example to estimate the direction of a reflected signal (e.g. from a building) after inferring that a reflected signal has been received at the receiver.

[0044] The determination of the first direction and the determination of the second direction can be based on prior knowledge of the environment in which the receiver is located. For example, it is most likely that a spoofing device is observed at low elevation (near the horizon).

[0045] By generating the sequence of phasors after determining the first direction and the second direction, at least one sequence of phasors in step (e) can be obtained.

[0046] In general, all of the following steps can be performed using the same signal data received for the respective correlation time period: determining the first direction and the second direction, obtaining the first sequence of phasors and the second sequence of phasors indicative of a measured or assumed movement of the receiver along the respective first direction and second direction, generating the third sequence of phasors and providing the correlation signal. However, in certain embodiments, knowledge from a previous time period can be reused in order to reduce power consumption and improve battery performance.

[0047] For example, when the analysis of the received signal data in order to determine the second direction comprises generating one or more respective sequences of phasors, the method can further comprise storing a sequence of phasors in addressable memory, the sequence of phasors corresponding to a measured or assumed movement of the receiver in the determined second direction. Thus, in embodiments, the at least one sequence of phasors obtained in step (e) can be obtained from the addressable memory. This advantageously reduces processing power and battery consumption when performing the steps of the method.

[0048] In embodiments, at least one of the determined first direction and the determined second direction and / or the at least one sequence of phasors obtained in step (e) can be stored in an addressable memory. Thus, if it can be assumed that the first direction and the second direction do not change from one time period to the next (e.g. the receiver is located on a slowly walking pedestrian), then any of these parameters can be obtained from the addressable memory.

[0049] The method can further comprise storing the generated third sequence of phasors in an addressable memory. Thus, the stored third sequence of phasors can be used to provide the correlation signal during a subsequent time period, rather than generating a new third sequence of phasors, thereby advantageously reducing the required processing power and battery consumption. Such reuse of the third sequence of phasors is beneficial when the measured or assumed movement of the receiver during the subsequent time period is substantially the same as the measured or assumed movement of the receiver during the time period in which the third sequence of phasors was constructed, and the first direction and the second direction in the subsequent time period are substantially the same as the first direction and the second direction during the time period in which the third sequence of phasors was constructed. In other words, the third sequence of phasors can advantageously be reused from one (e.g. correlation) time period to another when the "pattern" of the received signals in the two time periods is substantially the same. For example, such a scenario can be assumed for a slowly walking pedestrian, in which the first direction and the second direction of the received signals are unlikely to vary significantly between correlations.

[0050] Preferably, the method further comprises determining a position of the receiver based on the correlation signal generated in step (h). This can be achieved in a known manner by establishing distances to at least three positioning sources and determining the position using a mathematical filter. The calculated position can be used for various applications known in the art.

[0051] In preferred embodiments, the method can further comprise providing a local frequency or phase reference using a local oscillator; determining an offset between the local frequency or phase reference and a received frequency or received phase of a first reference signal received from at least one remote source, the first reference signal having a known or predictable frequency or phase; and wherein at least one of the first sequence of phasors and the second sequence of phasors obtained in step (e) is indicative (e.g. varies over time) of the determined offset. The amplitudes and / or angles of the phasors in the first sequence of phasors and the second sequence of phasors can be adjusted based on the determined offset (or a time series of offsets). Such embodiments advantageously help to remove errors introduced into the correlation signal by instabilities in the local oscillator. This is particularly advantageous in implementations of the local oscillator that are simple and low cost, such as a quartz crystal (e.g. a smartphone).

[0052] The error in the local oscillator can be isolated by cancelling the effects introduced to the received phase or received frequency based on the relative movement of the receiver to a remote source of the received reference signal ("reference source") along a vector between the two. Thus, the method can comprise determining a component of the motion of the receiver and the reference source along a straight line direction between the two ("line of sight"). By compensating for the offset between the local frequency or phase provided by the local oscillator and the reference signal, the moving receiver can provide a longer coherent integration than would otherwise be possible. The received signal can be coherently integrated over a period of 1 second or more. This means that the sensitivity of the receiver is improved, such that in combination with the use of the weighted phasor sequence discussed above, weaker positioning signals can be detected and used in the positioning calculations.

[0053] The reference source can be a terrestrial transmitter. For example, the reference source can be a cellular transmitter or a Digital Audio Broadcasting (DAB), DAB-Terrestrial (DVB-T) or analogue broadcast. The reference source can be a satellite, for example a GNSS satellite with a highly stable atomic local oscillator. The reference source can be a remote source in one of the first direction or the second direction. Importantly, the local oscillator in the remote source should be at least as stable as the local oscillator.

[0054] As discussed, the determined offset can be represented in the first phasor sequence and the second phasor sequence used to generate the third weighted phasor sequence. Alternatively, the method can further comprise providing a local frequency or phase reference using the local oscillator; determining an offset between the local frequency or phase reference and a received frequency or received phase of a first reference signal received from at least one remote source, the first reference signal having a known or predictable frequency or phase; and using the offset to provide a local signal. In such embodiments, the local signal is generated for correlation using the local frequency or phase reference from the local oscillator and the determined offset. In theory, the correction can be applied to at least one of the local signal, the received signal data and the correlation signal.

[0055] The method can comprise determining a sequence of offsets between the local frequency or phase reference and the received frequency or received phase of the first reference signal over time, and using the sequence of offsets to provide the local signal. Thus, the sequence of offsets can be represented as a sequence of phasors having an amplitude and / or angle representative of the determined offset over time.

[0056] According to a second aspect of the application, there is provided a computer program product comprising executable instructions which, when executed by a processor in a positioning system, cause the processor to perform the steps described above in relation to the first aspect of the application.

[0057] According to a third aspect of the application, there is provided a positioning system comprising: a receiver, a motion module, a direction determination unit, a local signal generator, a phasor generation unit and a correlation unit. The receiver is configured to receive signal data from one or more remote sources; the motion module is configured to provide a measured or assumed movement of the receiver; the direction determination unit is configured to determine a first direction and a second direction, wherein an enhancement of signals received at the receiver along the first direction is desired and an inhibition of signals received at the receiver along the second direction is desired; the local signal generator is configured to provide a local signal; the phasor generation unit is configured to: (i) obtain a first phasor sequence and a second phasor sequence indicative of the measured or assumed movement of the receiver in the first direction and second direction respectively, the first phasor sequence and second phasor sequence each comprising one or more phasors, the phasors comprising an amplitude and / or an angle, and (ii) generate a third phasor sequence based on a weighted combination of the first phasor sequence and second phasor sequence in accordance with the determined first direction and second direction; and the correlation unit is configured to provide a correlation signal using the third phasor sequence, wherein providing the correlation signal comprises correlating the local signal with the received signal data and combining at least one of the local signal, the received signal data and a result of the correlation with the third phasor sequence such that signals received along the second direction are inhibited relative to signals received along the first direction.

[0058] Thus, the positioning system of the third aspect of the application provides all the advantages described above in relation to the first aspect. Generally, the positioning system of the third aspect of the application can be configured to perform any of the steps of the first aspect of the application, for example by using a suitably programmed processor.

[0059] As with the first aspect of the application, the first direction is generally a straight line direction from the remote source to the receiver. The second direction is generally a non-straight line direction from the remote source to the receiver, or a straight line direction from an untrusted remote source to the receiver.

[0060] Generally, the third phasor sequence is generated using an estimation process, which is preferably a least squares fitting process. Generally, the estimation process is based on the desired enhancement of signals received along the first direction and the desired inhibition of signals received along the second direction.

[0061] In a preferred embodiment, the third phasor sequence W is generated by:

[0062] W = S + Z

[0063] where S is a matrix representing the first phasor sequence and the second phasor sequence, S +is the pseudo-inverse of S, and Z is a matrix representing the weighting of the first and second directions.

[0064] The receiver can comprise an antenna and electronics for processing received signal data. Preferably, the motion module is configured to provide a measured or assumed movement of the antenna. The present invention finds particular advantage in embodiments in which the receiver comprises exactly one (i.e. a single) antenna for receiving signal data. Thus, use of the present invention in devices having only one antenna (e.g. a smartphone) can result in an improved positioning solution. This contrasts with conventional solutions in which multiple antennas are used to determine different directions of received signals, e.g. using a Controlled Reception Pattern Antenna (CRPA). However, it is envisaged that the present invention can also be used in positioning systems having two or more antennas.

[0065] The positioning system preferably further comprises a positioning unit configured to determine a position of the receiver based on the correlation signals.

[0066] The motion module typically comprises at least one inertial sensor, e.g. an accelerometer or a gyroscope. However, it is envisaged that other sensors can be used to measure movement of the receiver, e.g. a barometric pressure sensor, a geomagnetic sensor or a visual odometry unit. The motion module can typically be or comprise an inertial measurement unit (IMU). Alternatively, the motion module can assume movement of the receiver based on a pattern of movement in a previous time period.

[0067] In preferred embodiments, the positioning system can further comprise an addressable memory configured to store at least one of the phasor sequence generated by the phasor generation unit and / or the first and second directions determined by the direction determination unit. This advantageously means that, if appropriate (e.g. if the measured or assumed movement of the receiver and the first and second directions are considered to be substantially the same in two separate time periods), the stored phasor sequence can be reused during a subsequent time period. This reuse of the phasor sequence advantageously reduces the required computational power and battery resources.

[0068] In a preferred embodiment, the positioning system can further comprise a local oscillator configured to provide a local frequency or phase reference; and a local oscillator offset determination unit configured to determine an offset between the local frequency or phase reference and a reception frequency or reception phase of a first reference signal received from at least one remote source, the first reference signal having a known or predictable frequency or phase, and wherein at least one of the first and second sequence of phasors obtained by the phasor generation unit is indicative of the determined offset. Such an embodiment advantageously allows for longer coherent integration times of 1 second or more, as explained above with respect to the first aspect of the application.

[0069] In an embodiment, the positioning system can further comprise a local oscillator configured to provide a local frequency or phase reference; and a local oscillator offset determination unit configured to determine an offset between the local frequency or phase reference and a reception frequency or reception phase of a first reference signal received from at least one remote source, the first reference signal having a known or predictable frequency or phase, and wherein the local signal generator is configured to provide the local signal using the local frequency or phase reference from the local oscillator and the determined offset. Preferably, the local oscillator offset determination unit is configured to compute a sequence of offsets between the local frequency or phase reference and the reception frequency or reception phase of the first reference signal as a function of time.

[0070] The positioning system is typically provided on a single (e.g. GNSS) positioning device. Such a single positioning device can be provided in an electronic user device such as a smartphone. Alternatively, various modules of the positioning system (e.g. the correlation unit, the direction determination unit and the phasor generation unit) can be provided separately, so that the positioning system is distributed (i.e. configured as a distributed system). For example, certain computations, such as those performed by the phasor generation unit and / or the correlation unit, can be performed by a processor in the network. Thus, for efficiency reasons, the electronic user device can offload computations to other processors in the network where appropriate.

[0071] In any aspect of the application, the receiver is typically a GNSS receiver. The at least one remote source typically comprises at least one GNSS satellite.

[0072] The methods and positioning systems of the application are typically useful for determining a position. However, the application can be used for determining other metrics, such as time and frequency. The metrics determined by the application can be used for navigation or tracking applications. BRIEF DESCRIPTION OF DRAWINGS

[0073] Preferred embodiments of the application will now be described with reference to the accompanying drawings, in which:

[0074] Figure 1 is a schematic diagram illustrating an exemplary environment in which the method and positioning system of the present application can be used;

[0075] Figure 2 is a schematic diagram illustrating a relevant portion of a positioning system according to an embodiment of the present application;

[0076] Figure 3 is a flowchart outlining the main steps of a preferred embodiment of the present application;

[0077] Figure 4 schematically illustrates how motion compensation is applied during the relevant process;

[0078] Figure 5 schematically outlines the effect of the present application on the sensitivity of signal reception from different directions;

[0079] Figure 6 is a flowchart outlining the steps of an exemplary embodiment of the present application; and

[0080] Figure 7 schematically illustrates a second direction of a candidate. DETAILED DESCRIPTION

[0081] Figure 1 is a schematic diagram illustrating an exemplary environment in which the method and positioning system of the present application can be used. Here, a positioning device 1000 is located within a "urban canyon" environment between high-rise buildings that can reflect radio signals, including GNSS signals from remote sources. Such "urban canyon" environments are typically challenging environments for a positioning system to accurately determine a position solution.

[0082] In the example of Figure 1 , the positioning device 1000 receives a weak straight line (SL) signal X1 from a first positioning satellite 1000a, a relatively strong reflected non-straight line (NSL) signal X2 from the same satellite 1000a, and a strong non-straight line signal X3 from a second remote source 1000b, which in this example is a spoofer. The relatively weaker SL signal X1 compared to signals X2 and X3 is schematically illustrated by the dashed line of signal X1 due to the SL signal X1 passing through a high-rise on its path from the satellite to the receiver.

[0083] Signal X1 is received at the receiver along direction D1 ; signal X2 is received at the receiver along direction D2; and signal X3 is received at the receiver along direction D3.

[0084] Figure 2is a schematic diagram showing relevant parts of a positioning system according to the application. The receiver 100 comprises an antenna 2 for receiving radio signals such as GNSS signals. In this example, the receiver 100 comprises a single antenna and can be part of a handheld electronic device such as a smartphone. Typically, the broadcast signals received at the antenna are analog signals and the broadcast signals are amplified, down-converted to baseband or a lower frequency and converted to digital form by an analog-to-digital converter; these processes take place in a receiver front-end block 3. The digitized signals are then processed (correlated) as will be discussed below.

[0085] The received signals are correlated in a correlation unit 1212 with a local copy of the signal generated by a local signal generator 8. The correlation unit comprises a correlator. The local signal generator 8 is configured to generate a local copy of a known correlation sequence (such as a pseudo-random number (PRN) code of a GNSS satellite) using a frequency or phase reference of a local oscillator 10.

[0086] The motion module 4 comprises sensors capable of measuring motion of the receiver 100, in particular of the antenna 2. The motion module 4 can comprise inertial sensors such as accelerometer and gyroscope sensors, the data of which can be used to infer motion of the receiver. The motion module 4 typically comprises an inertial measurement unit (IMU) using inertial sensors, although other (non-GNSS) means of determining receiver motion can be used instead or in addition, such as a barometer, a magnetometer and a visual odometry system such as Google System. In embodiments, the motion module can be in a closed loop system with the positioning unit.

[0087] The phasor generation unit 20 derives motion-compensated phasors indicative of the motion of the receiver measured by the motion module 4. The motion-compensated phasors can be applied to at least one of the local signals from the local signal generator 8, the received signal data and the results of the correlation (e.g. the initial output from the correlator within the correlation unit 12). The phasors generated by the phasor generation unit 20 can be stored in the local memory 16.

[0088] Each phasor comprises at least one of an amplitude and a (phase) angle.

[0089] The direction determination unit (DDU) 6 is operable to determine in which direction or directions the received signal should be enhanced and in which direction or directions the received signal should be actively suppressed. The determined directions can be stored in the local memory 16.

[0090] The local oscillator offset determination unit (LO unit) 5 is operable to determine an offset ("error") between the reference frequency or phase provided by the local oscillator 10 and the frequency or phase of a reference signal received from a remote reference source having a known or predictable frequency or phase. In this way, the accuracy of the local oscillator 10 can be matched to that of the reference source. These techniques will be described in more detail below.

[0091] After correlation by the correlation unit 12, the positioning unit 14 calculates the position of the receiver based on the results of the correlation and in combination with the phasors generated by the phasor generation unit. In other embodiments, different physical metrics can be determined, for example time or frequency.

[0092] Each of the above-described units of the positioning system are in logical communication with the processor 1, which is operable to control the operation of the individual units in accordance with the software or firmware executed. In the currently illustrated embodiment, the modules are provided within a single positioning device, although in alternative embodiments they can be provided in a distributed manner across a network.

[0093] Figure 3 is a flowchart illustrating the main steps of a preferred embodiment of the present application, and will be described with reference to the receiver 100 and environment shown in Figure 1 and Figure 2 At step S100, the receiver 100 receives signal data from remote sources 1000a and 1000b. It will be appreciated that the signal data comprises signals X1, X2 and X3 (and other signals not shown in the present example, for example NSL signals from further reflections of the satellite 1000a or signals from other remote sources).

[0094] The positioning satellite 1000a is a trusted source, for example a GNSS positioning satellite. Figure 1 One form of noise that can arise in the positioning system is illustrated, known as multipath effects. Here, the SL signal X1 and NSL "multipath" signal X2 come from the same remote source 1000a, but arrive at the receiver 100 via different paths. These signals are therefore received at different times, and can have different attenuation and phase characteristics. As a result, the signals X1 and X2 can act as noise to each other (for example, through destructive and / or constructive interference), leading to significant problems with positioning accuracy. Furthermore, as will be described, the SL signal X1 is attenuated on its path between the remote source 1000a and the receiver 100, resulting in a lower absolute power than the NSL signal X2. Conventionally, therefore, the receiver can "lock" to the reflected signal X2 having the higher absolute power, leading to incorrect pseudorange calculations due to the additional path length. As will be described, the present application helps to mitigate these problems. Figure 1 As shown schematically, the SL signal X1 has a lower absolute power than the NSL signal X2 due to attenuation of the signal X1 on its path between the remote source 1000a and the receiver 100. Conventionally, therefore, the receiver can "lock" to the reflected signal X2 having the higher absolute power, leading to incorrect pseudorange calculations due to the additional path length. As will be described, the present application helps to mitigate these problems.

[0095] In this example, the second remote source 1000b is an untrusted remote source, for example a spoofing device. The signal X3 is received in a straight line direction D3 from the spoofing device, although the signal X3 contains incorrect positioning data. Thus, processing of the signal X3 by the positioning system will result in an undesired incorrect positioning solution.

[0096] At step S102, the motion module 4 determines the motion of the receiver, for example using data obtained from the IMU. Alternatively, at step S102, the receiver can assume movement of the antenna based on previously detected movement patterns. For example, if previous measurements by the motion module 4 indicate that the receiver is moving in a constant direction and at a constant speed, it can be assumed that the current movement is the same as the movement in the previous period. Steps S100 and S102 are typically performed in a continuous manner while the position of the receiver is being calculated.

[0097] At step S104, the DDU 6 determines one or more first directions in which the signal is desired to be enhanced and one or more second directions in which the signal is desired to be actively suppressed. Typically, the direction determination module 6 determines a single first direction in which the signal is desired to be enhanced and multiple directions in which the received signal is desired to be suppressed. In Figure 1 In the example shown, the first direction is the SL direction D1 between the positioning satellite 1000a and the receiver 100, i.e. the first direction is where the signal X1 is desired to be enhanced. The second directions in which the received signal is desired to be suppressed are the NSL direction D2 between the positioning satellite 1000a and the receiver 100 and the SL direction D3 between the spoofing device 1000b and the receiver 100. In other words, the second directions are where the signals X2 and X3 are desired to be actively suppressed below the power of the weak SL signal X1.

[0098] As in the present example, the first direction is typically a SL direction pointing towards a trusted source such as a positioning satellite and can be known or estimated from broadcast ephemeris of a satellite constellation. A number of different techniques can be used to determine the second directions. These techniques can include:

[0099] • Scanning the sky in almost all elevations and azimuths to measure the direction of the received signal and using a predefined criterion to determine whether the received signal is a reflected (i.e. undesired NSL) signal.

[0100] • Using a three-dimensional map to assist in estimating the direction of the reflected (i.e. undesired) signal.

[0101] • Selecting specific directions based on prior knowledge, for example, spoofers are most likely to be observed near the horizon (with a small elevation).

[0102] These techniques will be further described herein in relation to Figure 6 Further details regarding these techniques can be found in the published document WO2019 / 058119.

[0103] The determined first and second directions can be stored in the local memory 16 (step S104’) and reused in a suitable subsequent time period, wherein the motion of the receiver and the first and second directions are assumed to remain substantially constant throughout the time period.

[0104] The method can optionally comprise a step S105 of determining an offset of the frequency or phase reference of the local oscillator 10 with respect to a reference frequency or phase of a reference signal received from a reference source. This is performed by the local oscillator offset determination unit 5. First, a reference source is selected based on the received signal data. The reference source should have a highly stable local oscillator that is at least more stable than the local oscillator 10 in the receiver. The reference source can be a satellite or a terrestrial transmitter, the reference signal preferably being received by the receiver along a straight line path in a direct line of sight. For the purpose of this example, we assume that the reference source is a satellite in the constellation Galileo® (not shown) that has an unobstructed SL with the receiver. Figure 1

[0105] Next, the local oscillator offset determination unit 5 determines the component of the measured movement of the receiver (obtained in step S102) in the direction of the selected reference source. The local oscillator offset determination unit further determines the motion of the reference source. In particular, the local oscillator offset determination unit 5 determines the component of the motion of the selected reference source along the straight line between the receiver and the reference source. Thus, the local oscillator offset determination unit can determine the relative movement of the receiver and the selected reference source along the vector connecting them.

[0106] Thus, the local oscillator offset determination unit 5 can calculate the frequency or phase error introduced to the received reference signal due to the relative movement of the reference source and the receiver 100. The received reference signal is provided by the reference source at a known and stable frequency or phase. Thus, once the Doppler error is removed, any remaining difference between the known frequency or phase of the reference source and the actually received frequency or phase can be attributed to an error in the local frequency or phase reference provided by the local oscillator 10. On this basis, the local oscillator offset determination unit 5 is configured to calculate the amount of offset of the frequency or phase reference provided by the local oscillator 10. More details on determining the error in the frequency or phase provided by the local oscillator 10 can be found in the published documents WO2019 / 008327 and WO2019 / 063983.

[0107] At step S106, the phasor generation unit 20 generates a sequence of phasors from the first and second directions determined by the direction determination unit 6. More specifically, the phasor generation unit generates a respective sequence of phasors indicative of the motion of the receiver along each of the signal directions Di, D2, and D3.

[0108] ​Each phasor sequence φ comprises a plurality of phasors, each phasor typically having the same time duration as a sample of the received signal. During the time period in which the received signal data and the movement of the measurement receiver are measured, there are typically the same number of N phasors φ in the generated phasor sequence φ because there are samples of the received signal and samples of the local signal i (i = 1..N). Each phasor φ i represents a phase and / or amplitude compensation based on the motion of the receiver at time t, such that the phasor sequence consisting of a plurality of phasors indicates the motion of the receiver along a particular direction over time. Thus, the phasor sequence can be referred to as a "motion compensated" phasor sequence.

[0109] In embodiments in which the local oscillator offset is determined in step S105, each phasor sequence φ generated at step S106 can further represent a phase and / or amplitude compensation based on the offset determined by the local oscillator offset determination unit 5. This is represented by step S105a in Figure 3

[0110] The phasor φ i is a transformation in the phase space and is complex-valued, via its real value a motion compensated phasor sequence's in-phase component is generated and via its imaginary value a motion compensated phasor sequence's quadrature phase component is generated. The phasor φ i is typically a cyclic phasor and can be represented in a number of different ways, for example as a clockwise rotation from the real axis or as an anticlockwise rotation from the imaginary axis. As described above, each directional phasor sequence indicates the measured (or assumed) movement of the receiver along that direction.

[0111] The determined phasor sequences for the respective first and second directions can be stored in local memory 16 (step S106') and reused in appropriate subsequent time periods, in which the motion of the receiver and the first and second directions are assumed to remain substantially constant throughout the time period.

[0112] Once the phasor sequences for each direction (both desired SL directions and undesired NSL / untrusted SL directions) have been obtained, a weighted phasor sequence is generated based on a weighted combination of the individual directional phasor sequences (step S107). The weighted phasor sequence vector W is given by the following equation:

[0113] W = S + Z,

[0114] where S is a matrix representing the individual phasor sequences corresponding to the signal directions D1, D2 and D3 and Z is a weighting vector.

[0115] S + denotes the pseudo-inverse of the matrix S and can be obtained using the Moore-Penrose pseudo-inverse or similar. + ​To solve the inverse of a non-square matrix.

[0116] We now describe an exemplary process for generating the weighted phasor sequence vector W using the following pseudocode.

[0117] a = {number of elements in the weighted phasor sequence vector W};

[0118] dx = {change in receiver position along the x-axis};

[0119] dy = {change in receiver position along the y-axis};

[0120] dz = {change in receiver position along the z-axis};

[0121] f = {radio frequency of the received signal};

[0122] b = {angle of incidence of the received signal};

[0123] A = {vector of angles of incidence to be suppressed};

[0124] c = {speed of the signal carrier in the medium (e.g., speed of light)};

[0125] Z = {desired gain vector};

[0126] i = imaginary operator;

[0127] We first construct the sequence of receiver positions over time during the creation of the vector W:

[0128] for J = 1 : a

[0129] d(J) = sqrt(dx(J)^2 + dy(J)^2 + dz(J)^2);

[0130] end

[0131] Next, we generate the phasor sequence φ for the direction of arrival of the signal of interest (e.g., SL signal), noting that without loss of generality, we can select to construct the weighted phasor sequence vector W from K first phasor sequences φ, as long as K < a, i.e., there are K signal directions of interest, and the receiver moves relative to the signal of interest and / or signal reflections. Typically, K = 1 (single line-of-sight path for a trusted satellite).

[0132]

[0133] Note that for the case where the receiver moves along a straight line and the total path length (|d(end) - d(1)|) is much smaller than the distance to any reflector or signal source, then b(I,J) = b(I) = a constant in the given calculation of the weighted phasor sequence vector W.

[0134] Next, the phase sequence φ at which the signal of interest is expected to arrive is generated:

[0135]

[0136]

[0137] Finally, the weighted phasor sequence vector W is solved by performing the following matrix operation:

[0138] W = (pinv(S) * Z)

[0139] where pinv refers to the Moore-Penrose pseudo-inverse or similar method for solving the inverse of a non-square matrix, and Z is a vector of desired weightings for the different received signal directions (1 indicates a desired signal direction to be enhanced, and 0 indicates a signal direction to be suppressed).

[0140] The resulting weighted phasor sequence vector W can then be applied to coherently integrate the received signal data.

[0141] Continuing with the example environment described above with respect to Figure 1 For simplicity, assume that there are four elements in the weighted phasor sequence vector W, i.e., a = 4, the phasor sequence along direction D1 can be represented as:

[0142] φ1 = (φ(1,1) φ(1,2) φ(1,3) φ(1,4));

[0143] the phasor sequence along direction D2 can be represented as:

[0144] φ2 = (φ(2,1) φ(2,2) φ(2,3) φ(2,4));

[0145] and the phasor sequence along direction D3 can be represented as:

[0146] φ3 = (φ(3,1) φ(3,2) φ(3,3) φ(3,4)).

[0147] Thus, S can be written as:

[0148]

[0149] Since the signal received along direction D1 is desired to be enhanced and the signals received along directions D2 and D3 are actively suppressed, the vector Z takes the form:

[0150] Z = (1, 0, 0).

[0151] Thus,

[0152]

[0153] The generated weighted phasor sequence vector W can be stored in local memory 16 (step S107') and reused in a suitable subsequent time period, with the assumption that the motion of the receiver and the first and second directions remain substantially constant throughout the time period.

[0154] Returning to Figure 3 In step S108, the correlation unit 12 correlates the local signal from the local signal generator 8 with the received signal data, with motion compensation applied by combining the weighted phasor sequence vector W with at least one of the local signal and the received signal data prior to correlation. Alternatively or additionally, the weighted phasor sequence vector W can be combined with the result of the correlation. The result of this correlation operation, combined with the weighted phasor sequence W, advantageously provides enhanced gain in the SL direction D1 and maximum suppression of unwanted energy from directions D2 and D3, since energy from these unwanted directions is actively attenuated by the weighting of the individual phasor sequences as described above. This technique significantly improves the ability to detect and use straight path positioning signals in challenging environments such as urban canyons and even indoor environments as shown. Figure 1 In the example shown, taking into account the active attenuation of energy received along directions D2 and D3, this can mean that a SL signal X1 which is significantly attenuated by a building can be used for positioning calculations. Figure 1 In the example shown, taking into account the active attenuation of energy received along directions D2 and D3, this can mean that a SL signal X1 which is significantly attenuated by a building can be used for positioning calculations.

[0155] As mentioned above, in embodiments where the local oscillator offset is determined in step S105, the phasor sequence generated in step S106 can represent phase compensation based on the frequency or phase offset determined by the local oscillator offset determination unit 5 (step S105a). Alternatively, in step S108, the local signal generator 8 can generate the local signal using the frequency or phase reference provided by the local oscillator 10 and the offset determined by the local oscillator offset determination unit 5. This is represented by step S105b in Figure 3 The accuracy of the local oscillator can therefore be matched to the accuracy of the local oscillator, and the local signal can therefore be provided with greater stability.

[0156] The use of the phase or frequency offset determined by the local oscillator offset determination unit 5 (introduced via S105a or S105b) in the correlation process can advantageously allow coherent integration of received positioning signals over a period of 1 second or more, without introducing errors due to any inherent instability in the local oscillator 10.

[0157] The weighted phasor sequence vector W can be applied to at least one of the local signal, the received signal data and the initial output from the correlation unit 12. Figure 4is a schematic diagram showing how the weighted phasor sequence vector W is applied to the local signal in order to provide motion compensation, the equivalent process being performed when the weighted phasor sequence vector W is applied to the received signal data or the result of the correlation. As explained above, the phasor generation unit 20 generates a weighted motion compensated phasor sequence vector W comprising an in-phase component I and a quadrature phase component Q. Both the in-phase component I and the quadrature phase component Q are mixed (22) with the same correlation code generated by the local signal generator 8 to generate a motion compensated correlation code as the in-phase component I and the quadrature phase component Q. The correlation unit 12 mixes (24) the in-phase component of the motion compensated correlation code with the received signal X (in digital form) and performs integration and dumping 13 on the result to produce the in-phase correlation result 30. The correlation unit 12 mixes (24) the quadrature phase motion compensated correlation code with the same received signal X and performs integration and dumping 13 on the result to produce the quadrature phase correlation result 32.

[0158] As already explained herein, in general, the weighted phasor sequence vector W can be applied to at least one of the local signal, the received signal data and the result of the correlation in order to perform motion compensation. More details on the generation of the motion compensated phasor sequence and its use in correlation can be found in WO2017 / 163042.

[0159] At step S110, the position of the receiver is calculated based on the result of the correlation process, as known in the art.

[0160] Reference will now be made to Figure 5 to describe the effect of correlating using the weighted phasor sequence vector W. Figure 5 is a plot of the sensitivity (vertical axis) of the incident signal with respect to the angle of arrival (horizontal axis) when the correlation is performed using only the phasor sequence generated for the motion of the receiver along the desired SL signal direction D1 (dashed line in Figure 5 and when the correlation is performed using the weighted phasor sequence vector W according to the present application (solid line in Figure 5 For the purpose of this example, we consider a simple linear receiver path, the SL signal direction D1 having an angle of arrival Θ at the receiver. The plot shows that the sensitivity of the correlation result is higher for the weighted phasor sequence vector W than for the phasor sequence generated for the motion of the receiver along the desired SL signal direction D1.

[0161] Referring first to the dashed line (motion compensation only along D1), we can see a peak at angle θ (shown at 50°) corresponding to the desired signal direction D1 and multiple secondary peaks corresponding to the undesired signal angle. In some cases, the combination of the sensitivity of the secondary peaks and the signal strength received along the corresponding angle can result in a higher received signal strength for the undesired signal compared to the desired signal, and the receiver may unintentionally lock onto the undesired signal (e.g., reflected signal X2 or spoofing signal X3). As mentioned above, this is a particular problem if the desired SL signal is particularly weak and / or the undesired signal has a high signal strength.

[0162] The solid lines schematically illustrate the sensitivity when the weighted phasor sequence vector W is used to correlate with the same signal data received at the antenna. The sensitivity pattern still shows a significant peak (shown at 60) and numerous secondary peaks. However, when motion compensation is used only along direction D1, the sensitivity decreases significantly at the angles where secondary peaks exist. In particular, the sensitivity of the solid line is significantly suppressed at the angle corresponding to the first peak (51) in the dashed line pattern. In some cases, the sensitivity pattern of the solid line exhibits zeros corresponding to the peaks in the dashed line gain pattern (e.g., as shown at 53). Therefore, when correlation is performed using the weighted phasor sequence vector W, the risk of the receiver "locking" to an unwanted signal direction is significantly reduced.

[0163] It is noteworthy that the peak of the gain pattern of the weighted phasor sequence vector W has a slightly smaller amplitude and is offset by a small amount of δθ in angle compared to the dashed phasor sequence. However, this does not adversely affect the receiver's ability to lock onto the desired SL signal X1, especially due to the significant suppression of sidelobe gain along the direction of the unwanted signal. Therefore, it will be understood that the motion-compensated correlation process using the weighted phasor sequence vector W significantly improves the correlation in traditionally difficult environments such as multipath environments or in the presence of strong unwanted signals such as those from spoofers.

[0164] Now let's go back Figure 3 Step S104 of the flowchart outlined therein determines a first direction (direction of signal enhancement) and a second direction (direction of signal suppression), and considers an exemplary method for determining said directions. The main steps of this method are as follows: Figure 6 The flowchart is shown, and it is executed by the direction determination unit 6.

[0165] Figure 6The flowchart illustrates a “brute-force” search of the sky to generate candidate directions for receiving unwanted signals (such as NSL signals). It should be noted that the desired NSL direction is generally known due to knowledge of the satellite's position within the positioning constellation (e.g., through broadcast almanacs and ephemeris data stored on receiver 100). At step S104a, the direction determination unit 6 generates motion-compensated phasors representing the receiver's motion in multiple candidate directions, selected by choosing appropriate values ​​for elevation angle α and azimuth angle β to provide full sky coverage for all possible directions where signals can be received. Figure 7 Candidate directions 70 are shown at specific angles α and β. Motion-compensated phasors are generated based on the receiver's measured or assumed movement along the candidate directions.

[0166] For each candidate direction, a motion-compensated correlation signal is calculated by correlating the local signal and the received signal and applying the motion-compensated phasor sequence for that candidate direction (step S104b). As previously described, the motion-compensated phasor sequence can be applied to at least one of the local signal, the received signal, and the correlation result.

[0167] At step S104c, the direction determination unit 6 analyzes the motion-compensated correlation signal to determine whether a candidate direction is a possible direction for receiving a signal. This is typically based on an analysis of the signal-to-noise ratio (SNR) of the motion-compensated correlation signal for that candidate direction. For example, if a high SNR is determined, the candidate direction is a possible direction for receiving a signal. If the candidate direction is not a SL direction to a trusted source, then such a signal may be an unwanted NSL signal (e.g., a reflected signal).

[0168] If the motion-compensated correlation signal has a low signal-to-noise ratio (e.g., SNR of about 1 or less), it can be inferred that no reflected signal was received along the candidate direction.

[0169] At step S104d, motion-compensated phasors for determining candidate directions that may have received unwanted (e.g., reflected NSL) signals can be stored in local memory 16. These stored phasor sequences and associated directions can be reused in appropriate subsequent time periods, assuming that the receiver's motion and the first and second directions remain substantially constant throughout the time period, so as to generate a weighted phasor sequence vector W in step S107. This advantageously optimizes processing and power resource usage without requiring the phasor sequence to be regenerated in S106.

[0170] Besides identifying the direction of potentially reflected NSL signals, this "brute-force" search can also be used to identify potential spoofing sources, such as cheaters. The aforementioned techniques can be used to distinguish between trusted signals and spoofed signals sent by a trusted remote location source based on knowledge of the signal's direction of arrival and its orientation towards a trusted remote source.

[0171] This brute force search of the sky is a computationally intensive process, which in one embodiment can be determined by using a terrain (e.g. three dimensional (3D)) model (e.g. a 3D city model) of the environment around the receiver 100 to determine the desired directions and the undesired directions. Such a model can be stored in the local memory 16 and provided to the direction determination unit 6 as required.

[0172] In embodiments using a 3D model, the map can be used to aim candidate directions where a reflected signal can be received, rather than the “brute force” approach of essentially scanning all elevation and azimuth angles as described above. For example, motion compensation can be provided along a straight line direction to a trusted source and an analysis of the total signal strength from that trusted source and the signal to noise ratio of the motion compensated related signal can be performed. Based on the analysis, it can be inferred whether a reflected signal is received. For example, if the signal to noise ratio of the motion compensated related signal is relatively low, but the receiver 100 receives a high signal strength from that satellite, it can be inferred that a reflected signal dominates the received signal from that satellite. The 3D model of the environment can then be used to determine candidate directions where a reflected signal can be received (e.g. reflected from a building) at the receiver.

[0173] If a signal component is received in a direction in the brute force scan of the sky where no reflection can occur based on the 3D model, such a 3D model can be used to determine a spoofing source, e.g. a deceiver.

[0174] Further details on using a 3D model to generate candidate directions can be found in published document WO2019 / 058119.

[0175] In some embodiments, a 3D model of the receiver environment can be used to pre- predict where a reflection is likely to be received and the determination of the first direction and the second direction in step S104 can be performed based on the pre- prediction without using the method of Figure 6

[0176] In the examples outlined so far, the first direction in which the signal is to be enhanced is the SL signal to a trusted source. This is expected to be the case in most situations. However, the invention can be used to provide preferential gain in any desired direction, e.g. if the objective is to locate a deceiver source, there can be use cases where the desired first direction is the SL direction to a deceiver. In this case, the weighting phasor sequence W will be used to actively suppress the SL or NSL satellite signal.

[0177] ​The present application is particularly suitable for use in a receiver 100 using a single antenna, wherein the method performed by the receiver software is used to spatially distinguish the received signal. However, it is envisaged that in other embodiments, the receiver can comprise two or more antennas, which can be used in conjunction with the method as described above to further aid in distinguishing the direction of the received signal.

Claims

1. A method performed in a positioning system, comprising: (a) receiving signal data at a receiver from one or more remote sources; (b) measuring or assuming movement of the receiver, wherein assuming movement of the receiver is based on a pattern of receiver movement in a prior epoch; (c) determining a first direction in which signal received at the receiver is expected to be enhanced; (d) determining a second direction in which signal received at the receiver is expected to be suppressed; (e) obtaining first and second sequences of phasors respectively indicative of measured or assumed movement of the receiver in the first and second directions, the first and second sequences of phasors each comprising one or more phasors, the phasors comprising an amplitude and an angle; (f) generating a third sequence of phasors based on a weighted combination of the first and second sequences of phasors in accordance with the determined first and second directions, wherein the third sequence of phasors is generated using an estimation process based on an estimate of an expected gain of signal received in the first direction and an expected gain of signal received in the second direction; (g) providing a local signal; and (h) providing a correlation signal using the third sequence of phasors, wherein providing the correlation signal comprises correlating the local signal with the received signal data and combining the third sequence of phasors with a result of the correlation such that signal received in the second direction is suppressed relative to signal received in the first direction.

2. The method of claim 1, wherein, The first direction is a straight line direction from the remote source to the receiver.

3. The method of claim 1 or 2, wherein, The second direction is a non-straight line direction from the remote source to the receiver, or a straight line direction from an untrusted remote source to the receiver.

4. The method of claim 1 or 2, wherein, The first and second sequences of phasors are derived from measured or assumed movement of the receiver over time.

5. The method of claim 1 or 2, wherein, The estimation process is a least squares fitting process.

6. The method of claim 1 or 2, wherein, The third sequence of phasors W is generated by: W = S + Z where S is a matrix representing the first and second sequence of phasors, S + is the pseudo-inverse of S, and Z is a matrix representing the desired gains of the first and second directions.

7. The method of claim 1 or 2, wherein, The first direction is determined based on known or estimated positions of the remote source.

8. The method of claim 1 or 2, wherein, The determination of the second direction is based on an analysis of the received signal data.

9. The method of claim 8, wherein, The analysis comprises: generating respective sequences of phasors for one or more respective directions based on measured or assumed movement of the receiver in the respective one or more directions; for each direction, providing a direction correlation signal using the respective sequence of phasors, wherein providing the direction correlation signal comprises correlating the local signal with the received signal data and correlating at least one of the local signal, the received signal data and a result of the correlation with the respective sequence of phasors, and determining the second direction based on an analysis of the respective one or more direction correlation signals.

10. The method of claim 9, wherein, A sequence of phasors is generated for a straight line direction between the receiver and a remote source, and wherein; The analysis further comprises determining whether the received signal comprises a component received in a direction different to the straight line direction, wherein the determination is based on a signal strength of the received signal data from the remote source and a signal to noise ratio of the direction correlation signal.

11. The method of claim 9, wherein, generating a respective sequence of phasors for a plurality of directions distributed over all possible directions in which a signal can be received at the receiver, and wherein the analysis further comprises determining whether a signal is received along each of the plurality of directions based at least on a signal-to-noise ratio of the respective direction-dependent signal.

12. The method of any one of claims 1-2 and 9-11, wherein, The determination of the second direction is based on a terrain model of an environment in which the receiver is located, wherein the terrain model is used to predict the presence of a reflected signal received at the receiver.

13. The method of any one of claims 1-2 and 9-11, wherein, The determination of the first direction and the determination of the second direction are based on prior knowledge of an environment in which the receiver is located.

14. The method of any one of claims 9-11, further comprising storing a sequence of phasors in addressable memory, the sequence of phasors corresponding to a measurement or assumed movement of the receiver in the determined second direction.

15. The method of any one of claims 1-2 and 9-11, further comprising storing at least one of the determined first direction and the determined second direction and / or at least one sequence of phasors obtained in step (e) in addressable memory.

16. The method of any one of claims 1-2 and 9-11, further comprising storing the generated third sequence of phasors in addressable memory.

17. The method of any one of claims 1-2 and 9-11, further comprising: providing a local frequency or phase reference using a local oscillator; determining a shift between the local frequency or phase reference and a received frequency or received phase of a first reference signal received from at least one of the remote sources, the first reference signal having a known or predictable frequency or phase; and wherein at least one of the first and second sequences of phasors obtained in step (e) is indicative of the determined shift.

18. The method of any one of claims 1-2 and 9-11, further comprising: providing a local frequency or phase reference using a local oscillator; determining a shift between the local frequency or phase reference and a received frequency or received phase of a first reference signal received from at least one of the remote sources, the first reference signal having a known or predictable frequency or phase; and using the shift to provide the local signal.

19. The method of claim 18, wherein, The method comprises determining a sequence of shifts between the local frequency or phase reference and the received frequency or received phase of the first reference signal over time, and using the sequence of shifts to provide the local signal.

20. The method of any one of claims 1-2, 9-11 and 19, further comprising determining a position of the receiver based on the correlation signals.

21. The method of any one of claims 1-2, 9-11, and 19, wherein, The receiver is a GNSS receiver, and at least one of the remote sources comprises at least one GNSS satellite.

22. A computer program product comprising executable instructions which, when executed by a processor in a positioning system, cause the processor to perform the steps of any one of the preceding claims.

23. A positioning system, comprising: a receiver configured to receive signal data from one or more remote sources; a motion module configured to provide a measured or assumed motion of the receiver, wherein the assumed movement of the receiver is based on a pattern of receiver movement in a previous epoch; a direction determination unit configured to determine a first direction and a second direction, wherein a signal received at the receiver along the first direction is expected to be enhanced and a signal received at the receiver along the second direction is expected to be suppressed; a local signal generator configured to provide a local signal; a phasor generation unit configured to: (i) obtain a first phasor sequence and a second phasor sequence indicative of a measured or assumed movement of the receiver in the first direction and the second direction, respectively, the first and second phasor sequences each comprising one or more phasors, the phasors comprising an amplitude and an angle, and (ii) generate a third phasor sequence based on a weighted combination of the first and second phasor sequences in accordance with the determined first and second directions, wherein the third phasor sequence is generated using an estimation process based on an expected gain of a signal received along the first direction and an expected gain of a signal received along the second direction; and a correlation unit configured to provide a correlation signal using the third phasor sequence, wherein providing the correlation signal comprises correlating the local signal with received signal data and combining the third phasor sequence with a result of the correlation such that a signal received along the second direction is suppressed relative to a signal received along the first direction.

24. The positioning system of claim 23, further comprising a positioning unit configured to determine a position of the receiver based on the correlation signal.

25. The positioning system of claim 23 or 24, wherein, The motion module comprises at least one inertial sensor.

26. The positioning system of claim 23 or 24, further comprising an addressable memory configured to store at least one of the first and second directions determined by the direction determination unit and / or at least one of the phasor sequences generated by the phasor generation unit.

27. The positioning system of claim 23 or 24, wherein, The receiver exactly comprises one antenna for receiving the signal data.

28. The positioning system of claim 23 or 24, further comprising: a local oscillator configured to provide a local frequency or phase reference; and a local oscillator offset determination unit configured to determine an offset between the local frequency or phase reference and a received frequency or received phase of a first reference signal received from at least one of the remote sources, the first reference signal having a known or predictable frequency or phase, and wherein at least one of the first and second phasor sequences obtained by the phasor generation unit is indicative of the determined offset.

29. The positioning system of claim 23 or 24, further comprising: a local oscillator configured to provide a local frequency or phase reference; and a local oscillator offset determination unit configured to determine an offset between the local frequency or phase reference and a received frequency or received phase of a first reference signal received from at least one of the remote sources, the first reference signal having a known or predictable frequency or phase, and wherein at least one of the first and second phasor sequences obtained by the phasor generation unit is indicative of the determined offset. a local oscillator offset determination unit configured to determine an offset between the local frequency or phase reference and a received frequency or received phase of a first reference signal received from at least one of the remote sources, the first reference signal having a known or predictable frequency or phase, and wherein the local signal generator is configured to provide the local signal using the local frequency or phase reference from the local oscillator and the determined offset.

30. The positioning system of claim 29, wherein, the local oscillator offset determination unit is configured to calculate a sequence of offsets between the local frequency or phase reference and the received frequency or received phase of the first reference signal as a function of time.

31. The positioning system of any one of claims 23, 24, and 30, wherein, the positioning system is provided on a single positioning device.

32. The positioning system of any one of claims 23, 24, and 30, wherein, the positioning system is configured as a distributed system.

33. The positioning system of any one of claims 23, 24, and 30, wherein, the estimation process is a least squares fitting process.

34. The positioning system of any one of claims 23, 24, and 30, wherein, the third sequence of phasors W is generated by: W = S + Z where S is a matrix representing the first and second sequence of phasors, S + is the pseudo-inverse of S, and Z is a matrix representing the desired gains of the first and second directions.

35. The positioning system of any one of claims 23, 24, and 30, wherein, the receiver is a GNSS receiver and at least one of the remote sources comprises at least one GNSS satellite.

Images