阅读说明：本技术 用于在gnss接收器中执行联合信道和时间估计的方法和装置 (Method and apparatus for performing joint channel and time estimation in a GNSS receiver ) 是由 J·A·G·莫利纳 于 2018-05-24 设计创作，主要内容包括：提供了一种用于与包括多个卫星的全球导航卫星系统(GNSS)一起使用的接收器。每个卫星传输包含扩频码的相应导航信号。GNSS保持被编码到导航信号中的时间参考。接收器包括：接收器时钟；以及至少一个天线,所述天线用于在多个相应信道上接收多个信号,每个信道由传输卫星和在所述信道的相对端处的接收天线限定。接收器进一步包括至少一个相关器,所述相关器用于计算(i)在多个信道上接收的信号与(ii)由接收器提供的导航信号的参考版本之间的互相关函数。接收器被配置成使用计算的互相关函数来执行以下各项的联合估计：(i)接收器时钟相对于由GNSS保持的时间参考的时钟偏差,以及(ii)包括作为时间延迟的函数的多个信道的组合贡献的复合信道。(A receiver for use with a Global Navigation Satellite System (GNSS) including a plurality of satellites is provided. Each satellite transmits a respective navigation signal containing a spreading code. The GNSS maintains a time reference encoded into the navigation signals. The receiver includes: a receiver clock; and at least one antenna for receiving a plurality of signals on a plurality of respective channels, each channel being defined by a transmission satellite and a receiving antenna at an opposite end of the channel. The receiver further includes at least one correlator for calculating a cross-correlation function between (i) signals received on the plurality of channels and (ii) a reference version of a navigation signal provided by the receiver. The receiver is configured to use the calculated cross-correlation function to perform a joint estimation of: (i) a clock bias of the receiver clock relative to a time reference maintained by the GNSS, and (ii) a composite channel comprising a combined contribution of the plurality of channels as a function of time delay.)

1. A receiver for use with a Global Navigation Satellite System (GNSS) comprising a plurality of satellites, wherein each satellite transmits a respective navigation signal containing a spreading code, and wherein the GNSS maintains a time reference encoded into the navigation signal, the receiver comprising:

a receiver clock;

at least one antenna for receiving a plurality of signals on a plurality of respective channels, each channel being defined by a transmission satellite and a receiving antenna at an opposite end of the channel; and

at least one correlator for calculating a cross-correlation function between (i) a signal received on a plurality of channels and (ii) a reference version of a navigation signal provided by a receiver;

wherein the receiver is configured to perform a joint estimation using the calculated cross-correlation function of: (i) a clock bias of the receiver clock relative to a time reference maintained by the GNSS, and (ii) a composite channel comprising a combined contribution of the plurality of channels as a function of time delay.

2. The receiver of claim 1, wherein the signals received on the respective channels include contributions from one or more of the navigation signals.

3. The receiver of claim 2, wherein the receiver is configured to detect a given navigation signal included in the signal received on a given channel based on a calculated cross-correlation function between the signal received on the given channel and a reference version of the given navigation signal.

4. The receiver of claim 3, wherein the receiver is further configured to determine a code phase alignment between the received given navigation signal and the provided reference version of the given navigation signal using a calculated cross-correlation function between the received signal on the given channel and the reference version of the given navigation signal.

5. The receiver of claim 4, wherein code phase alignment depends on: (i) a travel time of a given navigation signal from a satellite to a receiver, and (ii) a clock bias between a receiver clock and a time reference maintained by the GNSS.

6. The receiver of claim 4 or 5, wherein the calculated cross-correlation function used to perform the joint estimation is centered around the determined code phase alignment for each satellite.

7. The receiver of any of claims 2 to 6, wherein the contribution to the one or more navigation signals comprises a plurality of non-line-of-sight components.

8. The receiver of any preceding claim, wherein the arrival of the reference line-of-sight component is time delayed.

9. A receiver as claimed in any preceding claim, wherein the clock bias of the receiver clock is estimated as an offset from the approximate clock bias.

10. The receiver of claim 9, wherein the approximate clock bias is equal to an initially estimated clock bias having a typical accuracy (1- σ) in the range of 0.1-1 microseconds, or in the range of 0.2-0.8 microseconds, or in the range of 0.25-0.6 microseconds.

11. The receiver of claim 9, wherein the receiver is configured to estimate the approximate clock bias by determining a maximum of a composite cross-correlation function for the composite channel using the calculated cross-correlation function prior to performing the joint estimation.

12. The receiver of claim 11, wherein the complex cross-correlation function is defined by equation (1).

13. The receiver of any preceding claim, wherein performing joint estimation of clock bias comprises searching for a predefined range of possible clock bias values.

14. A receiver according to any preceding claim, wherein the composite channel is specified by a set of coefficients, each coefficient corresponding to a particular time delay and representing an aggregate level of power transmission through the composite channel for that time delay.

15. The receiver of claim 14, wherein the receiver comprises at least one probability distribution function for each coefficient in a set of coefficients.

16. The receiver of claim 15, wherein the provided probability distribution function is based on an estimate of the composite channel obtained using the received signal from a previous epoch.

17. The receiver of claim 16, wherein the receiver is configured to update the probability distribution function based on the estimated composite channel.

18. The receiver of any of claims 15 to 17, wherein the receiver performs joint estimation using a monte carlo search based on trial values of the composite channel generated from the probability density function to find the trial value and clock bias that best fits the calculated cross-correlation function.

19. The receiver of any preceding claim, wherein performing joint estimation is defined by equation (3).

20. The receiver according to any of claims 1 to 18, wherein the receiver comprises a plurality of antennas, and the joint estimation exploits spatial correlation between signals received by different antennas.

21. The receiver of claim 20, wherein performing joint estimation is defined by equation (4).

22. A receiver as claimed in any preceding claim, wherein the estimated clock bias has an accuracy of less than 100 nanoseconds, or less than 80 nanoseconds, or in the range of 25 to 75 nanoseconds, or of the order of 50 nanoseconds.

23. A method of operating a receiver for use with a Global Navigation Satellite System (GNSS) comprising a plurality of satellites, wherein each satellite transmits a respective navigation signal containing a spreading code, and wherein the GNSS maintains a time reference encoded into the navigation signal, the receiver comprising a receiver clock and at least one antenna for receiving a plurality of signals on a plurality of respective channels, each channel being defined by a transmitting satellite and a receiving antenna at an opposite end of the channel, the method comprising:

calculating a cross-correlation function between (i) a signal received on a plurality of channels and (ii) a reference version of a navigation signal provided by a receiver;

performing a joint estimation using the calculated cross-correlation function of: (i) a clock bias of the receiver clock relative to a time reference maintained by the GNSS, and (ii) a composite channel comprising a combined contribution of the plurality of channels as a function of time delay.

24. The method of claim 23, wherein the receiver is located in an indoor environment.

Technical Field

The present invention relates to Global Navigation Satellite Systems (GNSS) and in particular to performing joint channel and time estimation in a GNSS receiver for use in indoor conditions.

Background

Global Navigation Satellite Systems (GNSS) are becoming increasingly important in a wide range of applications including in-vehicle navigation support, location-based services for smart phones, and the like. Example implementations of GNSS are described in WO 2006/063613 and WO 2007/101454. Such a GNSS comprises a constellation of typically twenty to thirty satellites, each of which transmits a navigation signal incorporating a known spreading code (a predetermined pseudo-random noise sequence of bits or chips) unique to that satellite. Similar to a conventional CDMA (code division multiple access) system, the receiver cross-correlates the received signals with spreading codes for various satellites to acquire navigation signals for respective satellites that are currently in view (above the horizon). The navigation signals for the satellites also encode the position of the satellites and the time of transmission to a high degree of accuracy.

For position determination, the receiver locates the maximum of the cross-correlation function between the acquired navigation signals of a given satellite and a version of the spreading code for that satellite that is stored or internally generated within the receiver. The maximum of the cross-correlation function corresponds to the code phase alignment (zero offset) between the received spreading code and the inner spreading code and depends on (i) the signal travel time from the satellite to the receiver, which depends on the spatial position of the satellite and the spatial position of the receiver, and (ii) any timing (clock) offset (offset) between the satellite and the receiver. By acquiring code phase information from multiple satellites, it is possible to account for the (usually) unknown spatial position and clock bias of the receiver.

The accuracy of any such position determination depends on how accurately the code phase alignment can be determined based on the cross-correlation function. Most conventional GNSS navigation signals exhibit a single maximum in their autocorrelation function, which follows a known triangular shape near zero code phase offset. In addition, the new GNSS signals modulate the transmitted navigation signals with a Binary Offset Carrier (BOC). This has the effect of narrowing the main peak of the autocorrelation function to support more accurate estimation of the position, albeit at the expense of the appearance of new correlation peaks in the autocorrelation function, thereby introducing ambiguity (ambiguity) in the code phase estimation.

A typical approach for tracking code phase is to compute the cross-correlation function for three different delays, referred to as early, immediate (Prompt) and late, where immediate represents the expected delay (e.g., as determined from signal acquisition). The code phase alignment can then be determined by fitting the expected correlation peak to the correlation values for the three different delays. Additional details regarding BOC modulation and fitting of correlation peaks are provided in EP 3104195.

As mentioned above, obtaining code phase alignment for signals from multiple satellites allows determination of unknown spatial position and clock bias of the receiver. While GNSS signals are primarily associated with previous aspects (position determination-hence reference for navigation), they are also significant in later aspects (interest). Thus, determining the clock bias of the receiver allows obtaining a very accurate time at the receiver with reference to a very accurate time maintained throughout the GNSS.

GNSS signals are therefore used for time synchronization purposes in many different applications including the telecommunications, financial and energy sectors [1 ]. In general, the utilization of GNSS signals is typically limited to outdoor applications (where there is no GNSS coverage problem), while the use of GNSS signals under indoor conditions is more challenging [2] - [4 ]. For example, in an indoor environment, the receiver may or may not receive a direct line of sight (LOS) signal from a satellite, and may also receive a significant multipath component, i.e., a non-line of sight (NLOS) component. Because the NLOS component has a system delay relative to the LOS component due to the extra travel distance, this may introduce potential bias in the delay (and hence timing) estimation.

In the special case of telecommunication applications, GNSS synchronization is employed in macro cells by telecommunication service providers, while other synchronization techniques, such as wire-based solutions, like precise time protocols, are being considered for indoor small cells to achieve timing accuracy in the sub-microsecond range (depending on the specific network topology and other requirements).

Accurate indoor time synchronization is expected to be important for future 5G telecommunication networks, where indoor small cells may be used to fill (post) indoor areas to provide 5G services. For the transmission of such new 5G signals, strict synchronization requirements between different cells will be required, in the order of 50 nanoseconds or less [5 ]. Furthermore, accurate time synchronization may also potentially be used to provide accurate positioning based on ranging measurements derived from 5G signals.

While GNSS signals can still be used for positioning and timing applications under mild (mil) indoor conditions, this is at the cost of degraded performance due to deep fading and the effects of strong multipath from non line-of-sight (NLOS) components. In the specific case of timing applications, mass-market timing receivers (see, e.g., [6]]) Even in the case of tracking only one GNSS signal (if the receiver's position isA prioriKnown) and can also operate with a desired accuracy of the order of 500 nanoseconds (1-sigma) [6]. However, the accuracy is still significantly lower than the level required for the proposed 5G service described above.

Recently, new commercial timing services based on the Low Earth Orbit (LEO) constellation of iridium have become available to provide accurate timing (and positioning) in indoor conditions [7], [8 ]. For this service, a timing accuracy of 100 nanoseconds (1- σ) is claimed by the operator, which, although still not matching, is closer to the above-mentioned strict requirements for telecommunications applications.

Disclosure of Invention

The invention is defined in the appended claims.

A receiver for use with a Global Navigation Satellite System (GNSS) including a plurality of satellites is provided. Each satellite transmits a respective navigation signal containing a spreading code. The GNSS maintains a time reference encoded into the navigation signals. The receiver includes: a receiver clock; and at least one antenna for receiving a plurality of signals on a plurality of respective channels, each channel being defined by a transmission satellite and a receiving antenna at an opposite end of the channel. The receiver further includes at least one correlator for calculating a cross-correlation function between (i) signals received on the plurality of channels and (ii) a reference version of a navigation signal provided by the receiver. The receiver is configured to use the calculated cross-correlation function to perform a joint estimation of: (i) a clock bias of the receiver clock relative to a time reference maintained by the GNSS, and (ii) a composite channel comprising a combined contribution of the plurality of channels as a function of time delay.

Such a receiver supports the use of GNSS signals for fine time synchronization in indoor conditions, thereby helping to mitigate deep fading and NLOS multipath conditions in this type of scenario. The methods described herein seek to provide robust and accurate time synchronization under indoor conditions without the need to provide any additional infrastructure (e.g., a wired network and/or other devices other than a single GNSS receiver).

Drawings

Various implementations of the present invention will now be described in detail, by way of example only, with reference to the following drawings:

FIG. 1 is a schematic diagram of a receiver for receiving GNSS navigation signals.

Fig. 2 is a plot showing an example of the distortion introduced into the correlation function by the indoor propagation channel for a given antenna and satellite.

Fig. 3 is a plot showing an example of distortion introduced into the correlation function by the indoor propagation channel for different antennas and a given satellite.

Fig. 4 shows an example of a power delay curve (profile) for a simulated indoor condition.

Fig. 4A and 4B show subsets of the data from fig. 4, each showing a power delay curve for a single respective satellite.

Detailed Description

FIG. 1 is a high-level schematic block diagram of a receiver 701 suitable for use with a GNSS for receiving navigation signals from respective satellites. The receiver 701 may be provided as a stand-alone unit or may be incorporated into some larger device, such as a mobile (cellular) telephone, a cellular radio access node (e.g., a 5G small cell), a computer, an automobile or other form of vehicle, a hospital bed, an aircraft or ship, a shipping container, and so forth. In operation, receiver 701 includes at least one antenna 715 for receiving GNSS signals. The antenna 715 is linked to an RF conditioning, down-conversion (down-conversion) and a/D conversion unit 720, which unit 720 in turn passes the baseband complex envelope of the received signal to a channel acquisition and tracking unit 725.

The receiver 701 also includes a code generator 712, which code generator 712 can generate (or store) a spreading code (pseudo-random noise sequence) for a satellite constellation. The channel acquisition and tracking unit 725 is provided with correlated spreading codes by the code generator 712 so that the incoming channel(s) can be acquired from the baseband complex envelope of the received signal. The channel acquisition unit 725 supports sequential or parallel operation of the channel acquisition unit as appropriate (i.e., depending on the number of receiver channels) with an appropriate number of spreading codes. In the illustrated embodiment, the channel acquisition and tracking unit 725 includes R correlators 728.

The receiver 701 is configured to perform a joint estimation of (i) the clock bias and (ii) the composite channel, as described in more detail below. This processing may be integrated into the channel acquisition unit 725 or may be performed in a separate unit within the receiver 701 (not shown in fig. 1), for example in a processor, Digital Signal Processor (DSP), or any other suitable component.

Some navigation systems may include local elements (sometimes referred to as pseudolites). These are local positioning systems, e.g. at airports, which may supplement positioning signals from satellite navigation systems to provide additional position information. It will be appreciated that the receiver 701 as described herein is intended to receive and process any suitable navigation signal (whether from a satellite, a pseudolite, or from any other relevant source) and perform the joint estimation process accordingly.

Fig. 2 is a plot showing an example of the distortion introduced into the correlation function by the indoor propagation channel for a given antenna and satellite (as obtained by means of the channel models in [9], [10 ]). A reference correlation function (dashed line) is included for comparison and has peaks aligned with zero. The observed correlation function (solid line) is distorted and slightly delayed (shifted to the right) due to the presence of the (delayed) NLOS component, which also participates in the correlation function and thus produces a bias in the result (in contrast, the reference correlation function can be considered as the correlation function that would be produced by receiving only the LOS component). Note in particular that the maximum of the distortion correlation function for the indoor channel in fig. 2 is biased (delayed) with respect to the reference correlation function, which will then introduce bias into any position and/or timing estimate.

Fig. 3 is a plot showing an example of the distortion introduced into the correlation function by the indoor propagation channel for different antennas and a given satellite (again as obtained by means of the channel models in [9], [10 ]). Note that the correlation function is normalized so that the antenna with the highest correlation has a unity value. (different antennas typically have different positions and orientations and therefore experience slightly different channels from each other). It can be seen that the antenna with the strongest correlation value also has the best shape in terms of a well-defined peak of the correlation function. In contrast, the correlation functions for other antennas are typically somewhat distorted, with lower and less well-defined peaks (due to distortion from NLOS components and at least partial LOSs of LOS components). Therefore, any position and/or timing estimate based directly on the correlation function of these distortions is susceptible to bias or error.

The methods described herein help overcome the bias/distortion such as illustrated in fig. 2 and 3. We consider for exampleA lower receiver having an array of N antennas (where N ≧ 1), and receiving M navigation signals (where M ≧ 1, all with equivalent correlation properties and transmitted within the same frequency band) from a given GNSS constellation (or any other satellite-based navigation system). The signals received from the M satellites are considered to be affected by the indoor propagation channel. Multiple Input Multiple Output (MIMO) systems based on this definition, and including MIMO systemsM×NA composite channel of combined contributions of the individual propagation channels provides a method for jointly estimating the composite channel and time synchronization (between the receiver clock and the time reference of the GNSS) based on the received satellite navigation signals.

The following pieces of information are considered to be knownA priori：

a) A baseline (phase offset) between each of the antennas and the phase center of the receiver.

b) The location of the receiver;

c) coarse clock skew for a receiver；

With regard to (a), this is a property of the receiver itself and can be determined using known measurement equipment. With respect to (b) and (c), these may be estimated using conventional techniques such as those implemented in existing GNSS receivers. For example, the receiver may first perform conventional acquisition of the navigation signal to obtain this position and timing (clock bias) information before performing the additional processing described herein. Alternatively (or additionally), the location and/or timing information may be known from other sources-e.g. the receiver may be located in a building whose location is already known, e.g. from mapping information, also the timing information may be available from an external source, such as a telecommunications network or the like.

A composite channel is defined as an aggregate (aggregate) contribution of M × N propagation channels, corresponding to N received versions of M satellite signals (one for each antenna), as a function of time delay relative to a LOS signal (even though such LOS signal may not be visible to the receiver). The composite channel gives a measure of the power delay curve of the resulting aggregated MIMO channel-i.e. the total aggregated power contributed by all received signals as a function of time delay. Based on the estimate of the composite channel, an estimate of the clock bias of the receiver may also be derived-i.e., a joint estimation process is performed.

In general, each of the coefficients of the composite channel (one coefficient per time delay, each coefficient reflecting the level of power transfer through the channel for that time delay) will follow a Probability Density Function (PDF), which isA prioriUnknown and highly dependent on the propagation conditions of the receiver environment. As described below, the estimation of the composite channel is performed as a multi-hypothesis estimation problem, where a different PDF hypothesis is evaluated for each coefficient of the composite channel.

Estimation of composite channel based on composite cross-correlation functionComplex cross correlation functionDepending on the clock skew. Thus, at different discrete clock offset valuesThe following estimation of the complex cross-correlation function is meaningful. In practice, the complex cross-correlation function also depends on other parameters, such as clock driftAnd the phase center position of the receiver relative to the N antennas (among others).

Thus, it is possible to provideWherein the state vector of the receiverAnd sub-index 0 refers to the phase center of the array of antennas (i.e., the phase center position of the receiver). State vector for receiverIs based on a complex cross-correlation function of [11]Is derived as:

(equation 1)

Wherein the following definitions apply:

a)equivalent to state vector of each antenna. Attention is paid toWhereinIs the phase center and the second of the receiverjVector between the antennas. In fact:，

b) is for an integration period corresponding to consideration in composite channel estimationjOf an antennaKSnapshot of one sample (in sampling period)）（x _j Thus being represented at the corresponding antennajThe received signal at (a),

c)is directed to the state vectorFor the firstjVector of basis functions for each satellite of an antennaThe components of the composition are as follows,

d) to the firstiA satellite andjbasis function of antennaAt the moment of timetIs defined asWhereinIs composed ofiA complex baseband model of the signal transmitted by each satellite.

(appendix A below provides a derivation of equation (1)).

As noted above, the coarse clock bias may be estimated using any suitable prior art, for example, as implemented in current receivers. To help reduce the computational complexity of the joint composite channel and time estimation process, the clock bias error (typically on the order of hundreds of ns) typically obtained by prior art (state-of-the-art) under indoor conditions may be further reduced to help constrain the search space in the joint estimation problem. For example, an intermediate clock bias estimation may be performed to estimate the phase by using a complex cross-correlation functionTo improve (refine) the initial coarse estimate. Then, the intermediate clock bias estimationCan be derived as:

(equation 2)

WhereinαIs dependent on the coarse clock bias estimate at the current epoch (epoch)/integration periodA threshold value of quality of.

The received signal at each antenna may be viewed as the sum of a number of separate components, each component corresponding to the transmitted navigation signal but subject to variations in relative delay and amplitude. In this respect, we can consider the line of sight (LOS) as having zero delay and all other components as having positive delay. In some cases, the LOS component may not be visible.

To formalize this, we define a composite channel vectorContaining information specific to interestDCoefficients of a composite channel of discrete time delays (uniformly distributed covering extended channel delays) such thatIs the second of the composite channel vectormA coefficient; coefficients corresponding to LOS contributionh ₁The actual clock bias of the receiver, i.e. the time reference relative to the GNSS, is marked (the actual clock bias will be estimated by the procedure described herein). Thus, in practice, for a given componenth _m The value of m specifies the time delay (relative to the LOS component), andh _m is determined to have a time delaymThe amplitude (or other measure of the strength) of the component(s).

Coefficients of components of a composite channelIs considered to follow an arbitrary set of PDFsWhereinTo aim atlAn assumption and the firstmPDFs taken by the individual coefficients, and wherein the total is for the PDFsN _l Each hypothesis is followed by each coefficient. We will base on the intermediate clock biasAnd (positive or negative) integer offsetFine clock offset estimator (to be estimated)Is defined asWhereinIs the time delay resolution of the composite channel vector. Based on this, for the secondlH and of one hypothesisβMay be defined as a solution to the constrained optimization problem, according to equation (3):

(equation 3)

Wherein the following definitions apply:

a.is in PDF for the channelTo (1) alA composite channel vector estimated under an assumption wherein，

b.Is at the firstlThe integer offset estimated under each hypothesis,

c.based on intermediate clock skewTo constrain the error to integer offsetsβEstimated value (total ofThe estimated value),

d.is an estimated complex cross-correlation vector for a MIMO system, which is inQPair composite cross correlation function at individual clock offset valuesUniformly sampled (with time delay resolution)) By mean of intermediate clock offset estimationIs centered such that，

e.Is comprised of clock deviationA vector of samples of the reference correlation function (for the GNSS signals under consideration) centered.

Note that it is thisβIs finally meaningful to define the time at the receiver with reference to the time in the GNSS.

In the high-level term (term), equation (3) defines (i) the observed signal value (according to (d) above byA sum of squared differences between representations) and (ii) a predicted value for the hypothesis, the predicted value determined based on: (a) the above original navigation signal (denoted by s according to (e)), (b) according to (a) above, according to the predicted channelA pattern of hypothesized reflections for the signal received at each antenna, and a hypothesized time offset of the receiver clock, the time offset being determined according to (b) aboveAnd (4) showing.Andthe better the match to the received data, the smaller the sum of squared differences.

Optimization is affected by two conditions. First, the method is defined toA limited range of values (±) -ofW) This reduces the search space with respect to the intermediate clock bias. Note that this is primarily a computational problem-with more available computational resources, it can be addressed toExploring a wider search range, in some cases, an arrival may use a coarse clock bias to perform the search without first deriving an intermediate clock biasTo the extent of (c). Secondly, the first step is to carry out the first,is constrained by a probability density function consisting ofAs discussed in more detail below. Consider thatAs a form of regularization, which ensures that the difference is differentIs physically reasonable (rather than allowing any arbitrary valuesA value of (d); this also helps to reduce the risk of over-fitting.

The optimization problem proposed in equation (3) can be solved, for example, using the monte carlo method, whereby the coefficients of the composite channel vector h are treated as coming from the respective arbitrary PDFExtracted random variables (with evaluation for each coefficient in each hypothesisIndividual monte carlo runs). In practice, the actual distribution of coefficients depends strongly on the indoor channel propagation conditions. Thus, a plurality of PDF hypotheses for the coefficients of the composite channel are estimated, and following this method, the estimated composite channel vectorIs thatN _l An estimated channel vectorOne of (i) the estimated composite cross-correlation and (ii) a reference model based on the estimated composite channel vector.

As set forth below, algorithm 1 summarizes the consideration of N for each of the coefficients of the composite channel_lPDF hypothesis sum N_cThe individual monte carlo runs are based on the composite channel and fine clock bias estimates of equation (3) (i.e.,and）。

as set forth below, forN=1 (single antenna used by receiver) andN _l in the specific case of =1 (taking into account the single assumption for each coefficient in the estimation process), algorithm 1A summarizes the composite channel and fine clock bias estimation based on equation (3) such thatIs directed tomThe PDF of the individual coefficients is,is the state vector of the antenna of the receiver, and x₁Collected by the antenna of the receiver (gather)KA snapshot of individual samples.

In general, PDFMay be arranged in any desired distribution. One option is to define a set of default PDFs that represent different types of environments from light-indoor (light-indoor) to deep-indoor (deep-indoor) conditions, as derived from theoretical models, ray-tracing models, and/or from real-world data for those environments — e.g., using a database that represents indoor conditions. For a static receiver, another possibility is to estimate the coefficients based on previously available epochsTo estimate for a composite channelDPDF of coefficientsI.e. the PDF is constructed or improved based on the coefficients estimated so far. For a sufficiently long period of time (e.g. several days) such that all possible geometries between GNSS satellites and receiver antennas are encountered, the result isThe propagation conditions of the receiver should be characterized for the indoor conditions in which the receiver is located. In this case, it is preferable that the air conditioner,can be viewedIs a signature of the environment surrounding the receiver. Such estimation may be performed in a sequential (iterative) manner for each epoch, updating the estimated PDF of the previous epochTo provide a new composite channel vector estimate for the current epoch。

In an alternative approach, h and h are defined by exploiting the spatial correlation of NLOS multipath components between antennasβIs estimated. In other words, NLOS components "seen" by different antennas (located in slightly different locations) may have some degree of cross-correlation. The degree of this correlation depends on the distance between the antennas and the indoor propagation conditions.

Joint estimators can be defined as solutions to constrained optimization problems

(equation 4)

Wherein the following definitions apply:

a)whereinIs associated with a spatial covariance matrix of multipath error modeling errorsjA characteristic value estimated as，

b)Is a vector containing multipath error modeling errors with estimates for each antennaIs defined as:

(equation 5), wherein，

c)Is directed tojEstimated composite cross-correlation vector of each antenna atQPair composite cross correlation function at individual clock offset valuesUniformly sampled (with time delay resolution)) By mean of intermediate clock offset estimationIs centered such that，

d)Is estimated as:

(equation 6).

Algorithm 2 summarizes each consideration in the coefficients for the composite channelN _l A PDF hypothesis sumN _c Alternate composite channel and fine clock bias based on equation (4) at Monte Carlo run timeAnd (6) estimating.

Note that the inputs and outputs in algorithm 2 are the same as for algorithm 1. Thus, algorithm 2 may be considered an alternative method for performing the same overall analysis as performed in algorithm 1. Further, those skilled in the art will appreciate that these methods are provided by way of example, and that other methods may be employed depending on the circumstances of any given implementation. For example, rather than using the Monte Carlo method, definitions can be made across overlaysAndperforms a systematic search over a predetermined grid of spaces of possible (expected) values. Furthermore, rather than performing a monte carlo or grid search, some implementations might use an optimization (feedback) scheme that seeks to converge on some minimum value for equation (3) (similar to hill-climbing routines (hill-bounding routine), seeking only a minimum value rather than a maximum value). In all cases, it will be understood that finding the minimum configuration will generally mean finding an approximate minimum (rather than an absolute minimum), for example, due to limitations in the analysis performed (such as considering only the absolute minimum in equation (3))Is/are as followsN _W Value).

By considering the results from [9]]、[10]The method described herein has been simulated via a later correlation-level semi-analytical method using an implementation of algorithm 1, with a receiver having a linear array of four antennas, each with a 10 cm spacing between them. In the simulation, the simulation was carried out,N _l =3 (i.e. three assumptions are considered about the PDF of each of the composite channel coefficients) And is andN _c = 10. The spatial correlation between the signals received for each antenna is taken into account in the model. For reference, a timing solution obtained based on a received signal from one of the antennas of the array and a conventional Weighted Least Squares (WLS) based solution are used. In both cases, a single integration period is considered.

Fig. 4 shows an example of a power delay curve (channel model based on [9 ]) for simulated indoor conditions with more than 20 dB of attenuation and significant NLOS components for all satellites. In particular, fig. 4 shows simulation results for a set of 11 satellites in view (SV 1, SV2, … SV 11), where the results for each satellite are plotted with symbols having a particular shape/color for the respective satellite. Fig. 4A and 4B show subsets of data from fig. 4, where fig. 4A shows results for one satellite (denoted SV 2) and fig. 4B shows results for another satellite (denoted SV 8).

For a given satellite, each symbol with a particular shape/color represents a signal component from that satellite. It can be seen that the number of signal components produced by a single navigation signal (i.e. from one satellite) is relatively high-typically more than 10 or 20, typically in the range of 30-120, e.g. 50-100, but may still be higher. Each signal component is plotted based on:

i) x-axis — delay for an ideal LOS component in nanoseconds with respect to that component. By way of scaling, a delay of 10 nanoseconds corresponds to an additional travel distance of about 3 meters.

ii) y-axis — normalized power (dB), where here normalization indicates that the total received power (across all components) for each satellite is set to a consistent level (i.e., the same for all satellites).

Note that there is considerable variation in the signal components received from different satellites, for example, fig. 4B for satellite SV8 shows significantly more components with higher delays (> 100 ns) than fig. 4A for satellite SV 2. Furthermore, it will be appreciated that the desired accuracy in the delay estimate (of around 50 ns) is significantly less than the delay of some NLOS components (even for satellite SV 2).

The simulation results are summarized in table 1, which shows the timing errors (RMS and 95 percentile) for the WLS-based reference method and the methods described herein, with algorithm 1 being used in particular (note that additional calibration and satellite system related errors are not considered in the simulation). In particular, these figures relate to and determineThe associated timing error is either a function of the timing error,is the clock bias of the receiver compared to the time reference maintained by the GNSS. The results obtained show a significant improvement in the time estimation of clock bias performed with the method described herein compared to conventional time estimation methods. This improvement is achieved by the herein described method of better exploiting NLOS signals (mitigating the bias introduced in the clock bias estimation) and by exploiting the signal diversity introduced by an array with multiple antennas (although the herein described method may be used even if only a single antenna is available, such as by exploiting algorithm 1A as presented above).

TABLE 1 summary of the results

Although the method described herein has a higher computational complexity than conventional timing architectures, it is well suited for parallel implementations in order to reduce the computational burden. Furthermore, with relatively slow decorrelation (typically in the order of several seconds) of the indoor channels for the static receiver, the computation of the composite channel can be performed over a longer period of time.

The methods described herein allow satellite navigation signals, including but not limited to current GNSS, to help achieve timing accuracies below about 50 ns (1- σ) under indoor conditions. Such timing accuracy may help to support the use of current GNSS signals in indoor timing applications (without having to rely on other commercial services), such as the proposed 5G network operations mentioned above. In the methods described herein, a joint composite channel and time estimation method is provided to address low carrier noise (C/No) conditions and the high impact of NLOS multipath components typically experienced in such indoor environments. As described above, the method has been evaluated via software simulations considering the actual indoor propagation channel, and shows a significant improvement in the resulting time synchronization performance over conventional time estimation methods.

Although the method described herein is primarily intended to allow accurate determination of the timing offset of the receiver clock for the case of indoor reception of GNSS signals, it may also be used more generally for determining multipath components of the received signal (such as shown in fig. 4). This method may be used to help remove bias (distortion) from the cross-correlation function, thereby providing a more accurate estimate of the position of the receiver (instead of or as well as the clock offset) in the indoor environment. Furthermore, the method may be used to obtain information about the indoor environment. For example, if the observed multipath components change, this may be indicative of some change in conditions within the environment (such as a door opening or closing), and may therefore be used, for example, in some forms of monitoring systems.

Although the methods described herein are primarily intended for use with existing or future GNSS's, such as GPS and Galileo, the term GNSS should be understood to include any system that provides suitable satellite (or pseudolite) signals for performing the methods described herein, even though such signals are not necessarily provided for navigation purposes (or at least are not their primary purpose). For example, satellite signals used therewith may be primarily intended for time synchronization and/or for telecommunications (rather than for position location).

The method described herein provides a receiver for use with a Global Navigation Satellite System (GNSS) comprising a plurality of satellites. Each satellite transmits a respective navigation signal that includes a spreading code. The GNSS maintains a time reference that is encoded into the navigation signal. The receiver includes a receiver clock and at least one antenna for receiving a plurality of signals over a plurality of respective channels, each channel defined by a transmission satellite and a receiving antenna at opposite ends of the channel. The receiver further comprises at least one correlator for calculating a cross-correlation function between (i) a signal received over the plurality of channels and (ii) a reference version of the navigation signal provided by the receiver. The receiver is configured to use the calculated cross-correlation function to perform a joint estimation of: (i) a clock bias of a receiver clock relative to a time reference maintained by the GNSS and (ii) a composite channel comprising a combined contribution of the plurality of channels as a function of time delay.

The signals received over the respective channels typically include contributions from a plurality of navigation signals; these contributions are typically below the noise level, but can be discerned by cross-correlating with known codes for different navigation signals. Each cross-correlation function is typically calculated based on (i) the corresponding received signal (for a given channel) and (ii) the corresponding navigation signal (for a given satellite). The receiver will typically detect only a subset of the navigation signals-the remainder may correspond to, for example, sub-horizon satellites. It is also possible that (in some cases, the receiver may have almanac (almanac) information available and know in advance which satellites are currently below the horizon) that only a given navigation signal may be detected on a subset of the channels, e.g., due to strong attenuation or fading on one or more other channels. If no navigation signal is located (i.e. no correlation peak is found) for a given cross-correlation function for the corresponding channel and satellite, the cross-correlation function can be subtracted (counted) from the joint estimate.

In general, a receiver is configured to determine code phase alignment between a received navigation signal and a reference version of a given navigation signal using a calculated cross-correlation function between the received signal on a given channel and the reference version of the navigation signal. The code phase alignment depends on: (i) a travel time of a given navigation signal from a satellite to a receiver, and (ii) a clock bias between a receiver clock and a time reference maintained by the GNSS. For receivers of known position, the propagation time can be determined, thereby exposing the clock bias to the measurement.

The receiver described herein is well suited for use in indoor conditions, which typically suffer from (i) fading or sometimes complete absence of a direct (line-of-sight) component, and (ii) the typical presence of multiple non-line-of-sight components. Joint estimation utilizes information received on all channels, including information from multipath components, to help achieve higher clock bias accuracy than can typically be achieved with the prior art.

In some implementations, the clock bias of the receiver clock is estimated as an offset from the approximate clock bias. The approximate clock bias can be derived from conventional techniques and has an uncertainty greater than the desired accuracy of the clock bias provided herein. For example, the approximate clock bias may have a typical accuracy (1- σ) in the range of 0.1-1 microsecond, or in the range of 0.2-0.8 microsecond, or in the range of 0.25-0.6 microsecond. In some implementations, the receiver may be configured to estimate the approximate clock bias by using the calculated cross-correlation function to determine a maximum of a composite cross-correlation function for the composite channel prior to performing the joint estimation. (this latter approach will generally provide a more accurate preliminary estimate for the approximate clock bias than conventional techniques, and thus reduce the search space for joint estimation, as described below).

The composite channel may represent a multiple-input multiple-output (MIMO) channel, where the multiple inputs correspond to various satellites and the multiple outputs correspond to different antennas. However, in some cases, the receiver may have only a single antenna, in which case there is only a single output. Note that in the case where there are multiple antennas (i.e., MIMO channels), the joint estimation can exploit spatial correlation between signals received by different antennas to enhance the accuracy of the joint estimation.

The joint estimation of the clock bias and the composite channel can be considered as an optimization problem, i.e. finding the clock bias and the composite channel that will most closely reproduce the calculated cross-correlation function. The optimization may be performed using any suitable search technique, such as a grid search, an iterative technique (such as hill climbing), and so forth. For clock skew, the search may be performed within a window centered (at least approximately) on the overall code phase alignment of the calculated cross-correlation function, or by using any other suitable method. As noted above, having an a priori known estimate for the clock bias reduces the search space, which may be smaller the more accurate the a priori known estimate.

The composite channel represents a more complex parameter space. For example, a composite channel may be specified by a set of coefficients, each coefficient corresponding to a particular time delay and representing a level of aggregate power transmission through the composite channel for that time delay (e.g., as part of the total power received in the channel). In some implementations, the receiver includes at least one probability distribution function for each coefficient in the set of coefficients, and the probability distribution function is used to generate a trial value (trial value) for the coefficient. In some cases, the search space may include the use of multiple different probability distribution functions for the coefficients. There are various ways to define or provide such probability distribution functions. For example, the probability distribution function may be determined from simulations of typical indoor propagation channels, such as shown in fig. 4, 4A, and 4B. Another possibility is that the probability distribution function provided is based on an estimate of the composite channel obtained using the received signal from the previous epoch (if available). For example, the first epoch measured may use a probability distribution function obtained from the simulation, whereas the subsequent epoch may use a probability distribution function determined from the earlier epoch. In some cases, the receiver may be configured to update the existing probability distribution function after each new joint estimate of the composite channel becomes available, thus allowing continuous and uninterrupted (iterative) enhancement of the probability distribution function used for the coefficients.

Also provided herein is a method of operating a receiver for use with a Global Navigation Satellite System (GNSS) comprising a plurality of satellites, wherein each satellite transmits a respective navigation signal comprising a spreading code, and wherein the GNSS maintains a time reference that is encoded into the navigation signal. The receiver includes a receiver clock and at least one antenna for receiving a plurality of signals over a plurality of respective channels, each channel defined by a transmission satellite and a receiving antenna at opposite ends of the channel. The method includes calculating a cross-correlation function between (i) a signal received over a plurality of channels and (ii) a reference version of a navigation signal provided by a receiver; and performing, using the calculated cross-correlation function, a joint estimation of (i) a clock bias of the receiver clock relative to a time reference maintained by the GNSS and (ii) a composite channel comprising a combined contribution of the plurality of channels as a function of time delay. The method is particularly suitable for use in indoor environments which are typically affected by fading of multipath (non-line-of-sight) components and/or direct (line-of-sight) components. This approach may generally benefit from the same functionality and enhancements discussed above.

The methods described herein may be implemented as some form of receiver, whether provided as a standalone receiver or incorporated into some other type of device, such as a mobile phone, a cellular radio access node (e.g., a 5G small cell), a tablet, a vehicle GPS (car, airplane, boat, bicycle, etc.), a camera, or any other suitable location monitoring/tracking system. A receiver may implement such a method at least in part by executing program instructions on one or more processors (such as a DSP, GPU, etc.), and/or a receiver may implement such a method at least in part by utilizing dedicated hardware-e.g., one or more chips specially designed to support the method, e.g., by providing a plurality of hardware correlators, etc. A computer program may also be provided to implement the methods described herein. Such computer programs typically comprise computer program instructions that may be provided on a non-transitory computer readable storage medium, such as a compact disc (CD or DVD), hard disk, or flash memory. The computer program may be loaded into the computer memory from such a storage medium or may be downloaded into the computer memory via a network, such as the internet. An apparatus (e.g. a receiver) loaded with a computer program may comprise one or more processors for executing the computer program, the computer program comprising instructions for causing the apparatus to carry out a method such as that described above.

***

In conclusion, various implementations have been described herein, but these have been provided by way of example only, and many variations and modifications to such implementations will be apparent to those skilled in the art and fall within the scope of the invention as defined by the appended claims and their equivalents.

Reference to the literature

[1] "GNSS marker Report", 5 th 2017 (ISSN 2443-.

[2] G, Seco-Granado, J.A. Lopez-Salcedo, D.jimenez-Banos, G.Lopez-Risueno, "Challengen Indonger Global Navigation Satellite Systems: unknown places in resources in Signal Processing", IEEE Signal Processing journal, Vol.29, No. 2, p.108, 131. 2012.

[3] G, Hein and a. Teuber, "GNSS inoors, lighting the facing, Part 3", Inside GNSS, pages 45-53, month 7/8 of 2008.

[4] D, Rubin, T, Young, "Femtocells: Bridging removable Location and Timing inoors", InsideGNSS, autumn 2008.

[5]“ LTE; Evolved Universal Terrestrial Radio Access (E-UTRA); Base Station (BS) radio transmission and reception (3GPP TS 36.104 version 13.7.0 Release 13)”，ETSI TS 136 104 v13.7.0（2017-04）。

[6] "NEO/LEA-M8T u-blob M8 Current GNSS timing modules, Data Sheet", available at www.u-blob.com, 6 months and 21 days 2016.

[7] Cordaro et al, "An Alternative Source of Timing and Location using the Low Earth Orbit Iridium Satellite Conference," Joint Navigation Conference, "Howland, Redunn, 2017.

[8] D, Lawrence et al, "Innovation: Navigation from LEO", GPS World, 6 months 2017.

[9] T. Jost et al, "A Wireless band satellite-to-inductor Channel Model for Navigation Applications", IEEE Transactions on Antennas and Propagation, Vol.62, No. 10, p.2014, p.5307-.

[10] Report ITU-R P.2145-2, "Model Parameters for the Physical-Statistical Wideband Model in Recommendation ITU-R P.681", 9 months 2017.

[11] J.A. Garcia-Molina and J.A. Fernandez-Rubio, "Exploiting Spatial Diversity in Low-cost SDR Platforms" the MIMO-GNSS propach ", 6th International Coloqualum on Scientific and Fundamental Aspects of GNSS/Galileo (6 th International Association for the Scientific and Fundamental Aspects of GNSS/Galileo), Banana Paeonia, 2017 for 10 months.

Appendix A: derivation of the composite cross-correlation function of equation (1)

We define the signal vector that collects all signal snapshots observed through the N antennas as:

，（i）

whereinVector of motionAll complex amplitudes for the MN path in the MIMO-GNSS system are collected,collecting noise of all antennas (among them)) And based on a single antenna basis functionDefining a multi-antenna basis function matrix as:

。（ii）

we define a cost functionSo that the state vector of the receiverBy minimizing the MIMO Maximum Likelihood Estimation (MLE)To find:

。 （iii）

to find an efficient implementation of the MLE of the state vector of the receiver, we define the cross-correlation，And. Based on the previous definition, and taking into account the least squares estimate of the complex magnitude vector c is:

， （iv）

it can be easily seen that the cost function can be rewritten as:

（v）

thus, the MLE of the state vector of the receiver can be redefined as:

。 （vi）

to apply this MLE of the receiver's state vector to a real-time receiver, it is convenient to re-express the maximization problem in (reformulate) (vi) based on the cross-correlation for each antenna of the receiver:andresult in

（vii）

Wherein the cost function is referred to as to be related to the state vectorThe maximized "composite cross-correlation function", which can be rewritten as:

（viii）

(corresponding to equation 1 above).