Apparatus for estimating a direction of arrival and corresponding method

Abstract

An apparatus for estimating a direction of arrival includes an antenna, a beamforming network, and an evaluator. The antenna is configured to receive signals, is circularly polarized, and includes a plurality of different radiation patterns. The beamforming network is configured to provide based on signals received by the antenna decomposed signals that are received by associated radiation patterns of the plurality of radiation patterns. The evaluator is configured to estimate the direction of arrival based on the decomposed signals and based on information describing signal receiving characteristics of the antenna. The invention also refers to a corresponding method.

Claims

1. Apparatus for estimating a direction of arrival, wherein the apparatus comprises an antenna, a beamforming network, and an evaluator, wherein the antenna is configured to receive signals, wherein the antenna is circularly polarized, wherein the antenna comprises a plurality of different radiation patterns, wherein the beamforming network is configured to provide based on signals received by the antenna decomposed complex signals that are received by associated radiation patterns of the plurality of radiation patterns, wherein the evaluator is configured to use the amplitude and phase response of the different radiation patterns to estimate the direction of arrival and an inclination of the signal source relative to the antenna based on the decomposed complex signals and based on information describing signal receiving characteristics of the antenna, and wherein the information describing signal receiving characteristics of the antenna comprises a set of steering vectors depending on angles describing the position of a signal source emitting the signals received by the antenna relative to the antenna and depending on an angle describing an inclination of a polarization of the signal source relative to the antenna.

2. Apparatus of claim 1, wherein the antenna is either right hand circularly polarized or left hand circularly polarized.

3. Apparatus of claim 1, wherein the antenna comprises a plurality of antenna elements, wherein the antenna elements of the plurality of antenna elements comprise different radiation patterns, and wherein the beamforming network is configured to provide the decomposed complex signals so that the decomposed complex signals are received with radiation patterns of individual antenna elements or with a combination of radiation patterns of at least two antenna elements.

4. Apparatus of claim 3, wherein the antenna elements of the plurality of antenna elements are all either right hand circularly polarized or left hand circularly polarized.

5. Apparatus of claim 3, wherein the antenna elements of the plurality of antenna elements are located in a plane.

6. Apparatus of claim 1, wherein the evaluator is configured to estimate an inclination of a signal source emitting the signals received by the antenna.

7. Apparatus of claim 1, wherein the information describing signal receiving characteristics of the antenna refers to a co-elevation and an azimuth describing the position of the signal source relative to the antenna.

8. Apparatus of claim 1, wherein the information describing signal receiving characteristics of the antenna refers to an inclination of the signal source relative to the antenna.

9. Apparatus of claim 1, wherein the set of steering vectors is based on dividing a range of possible inclinations of the signal source into partition intervals.

10. Apparatus of claim 9, wherein a width of the partition intervals is a measure for a resolution concerning an estimation of the inclination of the signal source.

11. Apparatus of claim 9, wherein the partition intervals are set based on a projection similarity measure, wherein for chosen reference values of the angles describing the position of a signal source relative to the antenna a reference inclination is chosen as a reference steering vector, wherein steering vectors belonging to same reference values of the angles are projected on the reference steering vector, and wherein in case an acquired projection value lies within a given value interval, then the steering vectors are considered as identical.

12. Apparatus of claim 11, wherein steering vectors belonging to same reference values of the angles are projected on the reference steering vector and normalized for different inclination values.

13. Apparatus of claim 11, wherein the given value interval is given by a lower threshold and a value close to one.

14. Method for estimating a direction of arrival, comprising: receiving signals emitted by a signal source with a circularly polarized antenna comprising a plurality of different radiation patterns, providing based on the received signals decomposed complex signals that are received by associated radiation patterns of the plurality of radiation patterns, and using the amplitude and phase response of the different radiation patterns to estimate the direction of arrival and an inclination of the signal source relative to the antenna based on the decomposed complex signals and based on information describing signal receiving characteristics of the antenna, and wherein the information describing signal receiving characteristics of the antenna comprises a set of steering vectors depending on angles describing the position of a signal source emitting the signals received by the antenna relative to the antenna and depending on an angle describing an inclination of a polarization of the signal source relative to the antenna.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Embodiments of the present invention will be detailed subsequently referring to the appended drawings, in which:

(2) FIG. 1 shows a comparison between phase differences of lobes 2, 3 and 4 with respect to lobe 1 of a multibeam antenna,

(3) FIG. 2 illustrates an embodiment of the apparatus with a 22 mulitbeam antenna structure and the direction finding scenario including the signal source and its coordinates relation,

(4) FIG. 3 shows example axial ratio values of the respective lobes L.sub.1 to L.sub.5 on the XZ plane,

(5) FIG. 4 shows the results of a projection of A(0, ) on the reference steering vector A(0, 0) for 90<=<=90,

(6) FIG. 5 shows the results of a projection of A(20, ) on the reference steering vector A(0, 0) for 90<=<=90,

(7) FIG. 6 shows the results of a projection of A(20, ) on A(0, .sub.ref 0) for .sub.ref belonging to the set {78, 55, 32, 9, 14, 37, 60, 83),

(8) FIGS. 7A-C shows three spectra for distinct source inclinations: a) =0, b) =45, and c) =90 at =5,

(9) FIGS. 8A-C shows three spectra for distinct source inclinations: a) =0, b) =45, and c) =90 at =50,

(10) FIG. 9 shows a different embodiment of the apparatus and

(11) FIG. 10 shows another embodiment of the apparatus.

DETAILED DESCRIPTION OF THE INVENTION

(12) FIG. 1 shows a comparison between phase differences of lobes 2, 3 and 4 with respect to lobe 1 of a multibeam antenna which is similar to the antenna given by [1]. Shown is on the x-axis the observation angle in and on the y-axis the phase differences in . These phase differences are given for the lobes L.sub.2, L.sub.3, and L.sub.4 with respect to the first lobe L.sub.1 of the four beams of the multibeam antenna taught by [1] at an azimuth =0 and to an inclination =0. For each lobe two curves are shown: RP: source inclination =0 and MS: source inclination =90, respectively.

(13) An embodiment of the apparatus 1 is shown in FIG. 2. In the apparatus 1, a multibeam antenna (MBA) 2 is utilized having diverse radiation patterns. The antenna elements 5 support only a single circular polarization: RHCP (in the shown embodiment) or LHCP. If the amplitude and phase response of the patterns are used in a subspace-based method, for instance, a direction of arrival (DoA) estimation of the source signal stemming from the signal source 10 is possible along with an estimation of the source leaning with inclination angle of the signal source 10 relative to the antenna 2.

(14) FIG. 2 shows the geometry by example of a 2-by-2 elements array providing here five beams, based on the antenna provided by [1] with 4 antenna elements 5: four beams are given by the radiation patterns of the four antenna elements and one beam is the result of the combination of the radiation patterns of the antenna elements. That all antenna elements 5 are right handed circularly polarized is indicated by the white arrows.

(15) The array aperture is in the XY plane. The signal source 10 appears at arbitrary co-elevation and azimuth (here: =0) and is arbitrarily oriented with respect to the receive array denoted by the inclination angle . A beam-forming network (BFN) 3 provides the excitation vectors that may be used for forming the beam patterns, which can be selected via an RF switch (see FIG. 10). The resulting decomposed signals are processed by the evaluator 4.

(16) The discussion in the following is based on the single source scenario depicted in FIG. 2 for =0. The method is, however, generally applicable to every multibeam antenna and multiple sources.

(17) The signal model reads

(18) $\begin{array}{cc}\underset{\underset{x}{}}{\left[\begin{array}{c}{x}_1\\ x_2\\ x_3\\ x_4\\ x_5\end{array}\right]}=\underset{\underset{A}{}}{\left[\begin{array}{c}{a}_1\left({}_1,{}_1\right)\\ a_2\left({}_1,{}_1\right)\\ a_3\left({}_1,{}_1\right)\\ a_4\left({}_1,{}_1\right)\\ a_5\left({}_1,{}_1\right)\end{array}\right]}{s}_1+\underset{\underset{N}{}}{\left[\begin{array}{c}{n}_1\\ n_2\\ n_3\\ n_4\\ n_5\end{array}\right]},& \left(1\right)\end{array}$
where x is the vector of complex received signals, A is the complex responses of the lobes including the round trip phase difference of the signal, N is the vector of complex white Gaussian noise having a variance of .sup.2 and s.sub.1 is the complex source signal.

(19) A covariance matrix is then formed. It is given by
R=xx.sup.H=(As.sub.1+N)(As.sub.1+N).sup.H=As.sub.1s.sub.1.sup.HA.sup.H+.sup.2I(2)
due to the fact that noise is uncorrelated with the signal.

(20) Equivalently, there is
R.sup.2I=s.sub.1.sup.2AA.sup.H(3)

(21) The rest of the algorithm depends on the selection of the subspace-based method, which makes use of the covariance matrix defined above, so the following results are legitimate for any subspace-based estimation such as MUSIC, ESPRIT, Maximum Likelihood, root-MUSIC, CAPON among others [7].

(22) For the estimation, the steering vectors A(, ) for each distinct and may be used. This is for the discussed case =0. Hence, in a general form, the steering vectors A(, , ) may be used.

(23) The two orthogonal source polarization A(, 0) and A(, 90) are measured for 90<<90 per each beam output. Using these two steering vectors, the rest follows from

(24) $\begin{array}{cc}{A}_i\left(,\right)=\left[\begin{array}{cc}\cos \left(\right)& \sin \left(\right)\end{array}\right]\left[\begin{array}{c}{A}_i\left(,0^{}\right)\\ A_i\left(,{90}^{}\right)\end{array}\right]& \left(4\right)\end{array}$

(25) for the each beam i (in our example i=1, 2, 3, 4, 5). Then, steering vectors for an arbitrary source inclination is obtained.

(26) The fact is, as long as the steering vector covariance matrix AA.sup.H is considerably different for each value of , the polarization sensitivity is achieved.

(27) Assuming all the beams are perfectly RHCP for all range, then as the linearly polarized source leaning changes, the beam output signals are shifted with exactly the same phase, providing no change on the steering vector covariance matrix. As a result, the polarization sensitivity is not achieved.

(28) However, different levels of achieved RHCP provides the sensitivity. This is where the multibeam antenna MBA 2 comes into action. Having an MBA yields an axial ratio pattern as exemplarily portrayed in FIG. 3 in which the axial ratios AR in dB for the five beams of the antenna 2 in FIG. 2 is shown. The AR is given depending on the observation angle in .

(29) Each beam or each sub-set of beamsif some beams repeat because of symmetryhas its unique axial ratio characteristics with respect to observation direction. This is peculiar to multibeam antennas and results in polarization sensitivity. Thus, a multibeam antenna as used in the apparatus provides a well resolved direction of arrival estimation characteristics and source inclination estimation.

(30) Before the DoA estimation is done, the steering vector set matching the source inclination is needed. The set is in one embodiment obtained by measurements and is in a different embodiment obtained by measurements and by calculations based on these measurements.

(31) In one embodiment, the range of possible source inclination is divided into partitions. The partitioning intervals are arranged by a projection similarity measure. For a certain value =.sub.ref, a reference .sub.ref is selected, forming the reference steering vector. Then, all the steering vectors having the same .sub.ref value is projected on the reference steering vector and normalized for different values:

(32) $\begin{array}{cc}\mathrm{proj}\left({}_{\mathrm{ref}},\right)=\sqrt{\frac{{A\left({}_{\mathrm{ref}},\right)}^{H}A\left({}_{\mathrm{ref}},{}_{\mathrm{ref}}\right)}{.Math.A\left({}_{\mathrm{ref}},\right).Math..Math.A\left({}_{\mathrm{ref}},{}_{\mathrm{ref}}\right).Math.}}.& \left(5\right)\end{array}$

(33) This formula is given for =0. The general formula is accordingly given for proj(.sub.ref, .sub.ref, ).

(34) If the projection value is almost 1, the steering vectors can be considered identical. In practice, a lower boundary, i.e., a threshold, is selected for differentiating between steering vectors. Above the boundary, steering vectors are considered identical, below steering vectors are different. The threshold is selected with respect to axial ratio beamwidth, signal demodulation and noise; it should be close to 1, for example, 0.998.

(35) Towards normal direction of the MBA, the projection is close to 1 independent of as shown in FIG. 4 for .sub.ref=0 because of the absence of polarization sensitivity (compare FIG. 3). In FIGS. 4 to 6 the norm of projection is shown in dependency on the inclination angle in .

(36) As the co-elevation values get larger, there appears an immediate polarization sensitivity (axial ratio diversity) for some partitions. In FIG. 5, the situation is demonstrated by taking .sub.ref=20 and .sub.ref=0 for which, there is a polarization sensitivity having a 30 resolution. FIG. 5 also shows that the resolution is about +17 and the partition interval will have accordingly a width of about 34. The resolution is here given by the values for the norm of projection greater than 0.998, i.e. close enough to the value of the reference inclination.

(37) Assuming, for the time being, the resolution is identical for all .sub.ref, the range of inclination angles can be divided into equidistant partitions, as shown in FIG. 6. For =20 eight partitions are expected to cover the area above the threshold line. The resolution interval boundaries are indicated by two circles.

(38) In the example, the following steering vectors were used: A(20, 78), A(20, 55), A(20, 32), A(20, 9), A(20, 14), A(20, 37), A(20, 60), and A(20, 83).

(39) The selection is made in one embodiment regarding to this similarity projection method for the other instances of and . For the interval having no polarization sensitivity, only a single steering vector, i.e., a single partition, is used for forming the spectrum. This fact decreases the number of comparisons that may be used.

(40) After determining the steering reference vectors that may be used and after partitioning the p spectrum for each (and ), the estimation can be initiated with any subspace based method. The number of partitions, which directly corresponds to the polarization sensitivity of the MBA, increases as increases.

(41) In the following table, a sample partitioning for the given MBA of the embodiment of the apparatus of FIG. 2 is demonstrated, which is formed using equidistant approach with some overlapping. Here, for the sake of simplicity, the azimuth is still given by =0.

(42) TABLE-US-00001 Values Number of Partitions [2, 2] 1 [7, 3] [3, 7] 2 [12, 8] [8, 12] 4 [17, 13] [13, 17] 6 [22, 18] [18, 22] 8 [27, 23] [23, 27] 8 [32, 28] [28, 32] 10 [37, 33] [33, 37] 10 [42, 38] [38, 42] 10 [47, 43] [43, 47] 10 [52, 48] [48, 52] 12 [57, 53] [53, 57] 12 [62, 58] [58, 62] 14 [67, 63] [63, 67] 16 [72, 68] [68, 72] 18 [77, 73] [73, 77] 22 [82, 78] [78, 82] 26 [87, 83] [83, 87] 34

(43) The table gives example numbers of steering vector partitions for the given range of the co-elevation (here: the azimuth is set to 0).

(44) As seen in FIG. 6, for some steering vectors there appears an overlapping above the threshold line as a consequence of the equidistant partitioning. The resolution at the respective steering vectors could therefore be doubled in the worst case, as the estimator favors one of the intervals in the spectrum while yielding a coarse estimation of the source polarization. By having nonequidistant partitioning, there exists no overlapping as well, hence the MBA achieves a nonoverlapping and a bounded resolution capability. In short, the method yields a guide on how to select the reference partitioning steering vector set, having the constraint of covering all the field above the threshold line.

(45) Using the MUSIC algorithm for the verification, some test results are presented to show the capability of multibeam antenna to resolve the direction of arrival of signals together with the increasing sensitivity to source inclination as the co-elevation increases.

(46) FIGS. 7A-C demonstrates the output of the partitioned spectrum with the source being placed at 5 and for three different inclinations: =0, 45, and 90.

(47) FIGS. 8A-C demonstrates the result for the same source inclination set when the source is placed at co-elevation =50.

(48) In both figures, FIGS. 7A-C and FIGS. 8A-C, it can be seen that the range of possible inclinations is divided into different partitions. Hence, the spectrum obtained by processing the decomposed signals with regard to their amplitudes and phases and using the set of steering vectors describing the spatial distribution of the signal receiving sensitivity of the different radiation patterns and describing effects of an inclination of the signal source is analyzed based on the different partitions with regard to the inclination of the signal source. The further processing is done by any subspace-based method according to the state of art, e.g. MUSIC.

(49) The results supports that the MBA achieves the polarization sensitivity together with its axial ratio pattern. There appears a very good accuracy on direction of arrival estimation, together with the partitioned source inclination estimation. An aspect here is that, the MBA suppresses the high axial ratio angles of the beams with its smaller amplitude responses at those angles, which acts like a weighting, yielding a polarization sensitivity together with a more robust direction of arrival estimation, thus using only RHCP beams without any secondary polarization pattern.

(50) An embodiment of an apparatus 1 for DoA and polarisation estimation is portrayed in FIG. 9.

(51) The Antenna Array 2 with different antenna elements 5 connects to an beam forming network (BFN) 3e.g. realized by a Butler matrixwhich decomposes the received antenna signals into decomposed signals. The estimation is performed in the Signal Processing device or evaluator 4. The steering vector set of the respective antenna 2 obtained from measurement or simulation is provided via a storage unit 6. The evaluator 4 provides estimations for the direction of arrival and the inclination of the polarization of the signal source. A controller 7 controls the apparatus 1. The separation of the BFN and the Signal Processing is logical. The BFN can be also part of the Signal Processing unit, representing a digital signal decomposition.

(52) Another embodiment of the apparatus 1 for DoA and polarization estimation is portrayed in FIG. 10.

(53) The Antenna Array 2 with antenna elements 5 (having all the same circular polarization) connects to an BFN 3, which decomposes the antenna signals into decomposed signals. In contrast to the embodiment in FIG. 9, only one decomposed signal is selected at an instant of time via a switch unit 8. Supposing an analogue implementation of the BFN 3 and the switch unit 8, this allows for reducing the number of signal branches to one. The selection of the signal is controlled by the controller 7. It ensures a time-synchronous switching. The estimation is performed in the evaluator 4. The steering vector set of the respective antenna 2 is provided by the storage unit 6. The separation of the beam forming network 3, the switch unit 8, and the evaluator 4 is logical. The BFN 3 and the switch unit 8 are in a different embodiment part of the evaluator 4, representing a digital signal decomposition and selection.

(54) The provided apparatus and method provide at least the following benefits:

(55) Reduced costs for implementation as single polarization involves only a single signal branch for each element, halving the effort compared to dual polarized solutions.

(56) Because of the diverse patterns, the DoA and the polarization (inclination, orientation) of the source signal can be concurrently estimated using an array with one polarization.

(57) Applying partitioning of the range of inclination angles reduces the signal processing effort for estimation, allowing for usage of cheaper processing platforms such as microcontrollers (e.g. ARM family) or digital signal processors.

(58) Possible application areas are:

(59) Direction and orientation estimation of radio transmitters/transponders such as RFID transponders, wireless sensors, mobile devices (e.g. mobile phones, laptops, tablet computers), vehicles, aircrafts.

(60) Direction finding for military purpose and security (radio reconnaissance).

(61) Sensor applications: determination of inclination/positioning/orientation

(62) Although some aspects have been described in the context of an apparatus, it is clear that these aspects also represent a description of the corresponding method. Analogously, aspects described in the context of a method step also represent a description of a corresponding block or item or feature of a corresponding apparatus.

(63) Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperating with a programmable computer system, such that one of the methods described herein is performed.

(64) Generally, embodiments of the present invention can be implemented as a computer program product with a program code, the program code being operative for performing one of the methods when the computer program product runs on a computer. The program code may for example be stored on a machine readable carrier.

(65) Other embodiments comprise the computer program for performing one of the methods described herein, stored on a machine readable carrier.

(66) In other words, an embodiment of the inventive method is, therefore, a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.

(67) A further embodiment of the inventive methods is, therefore, a data carrier (or a digital storage medium, or a computer-readable medium) comprising, recorded thereon, the computer program for performing one of the methods described herein. The data carrier, the digital storage medium or the recorded medium are typically tangible and/or non-transitory.

(68) A further embodiment of the inventive method is, therefore, a data stream or a sequence of signals representing the computer program for performing one of the methods described herein.

(69) The data stream or the sequence of signals may for example be configured to be transferred via a data communication connection, for example via the Internet.

(70) A further embodiment comprises a processing means, for example a computer, or a programmable logic device, configured to or adapted to perform one of the methods described herein.

(71) A further embodiment comprises a computer having installed thereon the computer program for performing one of the methods described herein.

(72) A further embodiment according to the invention comprises an apparatus or a system configured to transfer (for example, electronically or optically) a computer program for performing one of the methods described herein to a receiver. The receiver may, for example, be a computer, a mobile device, a memory device or the like. The apparatus or system may, for example, comprise a file server for transferring the computer program to the receiver.

(73) In some embodiments, a programmable logic device (for example a field programmable gate array) may be used to perform some or all of the functionalities of the methods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor in order to perform one of the methods described herein. Generally, the methods are advantageously performed by any hardware apparatus.

(74) The apparatus described herein may be implemented using a hardware apparatus, or using a computer, or using a combination of a hardware apparatus and a computer.

(75) The methods described herein may be performed using a hardware apparatus, or using a computer, or using a combination of a hardware apparatus and a computer.

(76) While this invention has been described in terms of several embodiments, there are alterations, permutations, and equivalents which fall within the scope of this invention. It should also be noted that there are many alternative ways of implementing the methods and compositions of the present invention. It is therefore intended that the following appended claims be interpreted as including all such alterations, permutations and equivalents as fall within the true spirit and scope of the present invention.

REFERENCES

(77) [1] L. Weisgerber and A. E. Popugaev, Multibeam antenna array for RFID applications, in 2013 European Microwave Conference (EuMC), Nuremberg, Germany, Oct. 10-16, 2013, pp. 84-87. [2] K. A. Gotsis, K. Siakavara, and J. N. Sahalos, On the direction of arrival (doa) estimation for a switched-beam antenna system using neural networks, IEEE Transactions on Antennas and Propagation, vol. 57, no. 5, pp. 1399-1411, May 2009. [3] W. X. Sheng, J. Zhou, D. G. Fang, and Y. C. Gu, Super-resolution DOA estimation in switch beam smart antenna, in Proc. 5th Int. Symp. Antennas, Propagation and EM Theory ISAPE 2000, August 2000, pp. 603-606. [4] L. Weisgerber and A. E. Popugaev, Antenna concepts for DoA estimation in traffic scenarios, in Proc. IEEE-APS Topical Conf. Antennas and Propagation in Wireless Communications (APWC), August 2014, pp. 423-426. [5] R. Schmidt, Multiple emitter location and signal parameter estimation, IEEE Transactions on Antennas and Propagation, vol. 34, no. 3, pp. 276-280, March 1986. [6] Y. Han, Q. Fang, F. Yan, M. Jin, and X. Qiao, Joint DOA and polarization estimation for unequal power sources based on reconstructed noise subspace, Journal of Systems Engineering and Electronics, vol. 27, no. 3, pp. 501-513, June 2016. [7] J. Foutz, A. Spanias, and M. K. Banavar, Narrowband Direction of Arrival Estimation for Antenna Arrays, ser. Synthesis Lectures on Antennas #8. Morgan & Claypool, 2008.