Interference Mitigation for Received GPS and Radio Communication Signals

Abstract

System and method for concurrent STAP (space-time adaptive processing) filtering of interference signals (e.g., jamming signals) from signals received at a broadband antenna array via satellite and terrestrial broadcasts and communication systems operating in the 950 MHz to 1.65 GHz L-band, including but not limited to Link 16 terrestrial communications, Iridium satellite communications, and the Global Positioning System (GPS) and other Global Navigation Satellite Systems (GNSS). While the GPS/GNSS signals are protected against jamming using Nulling of the interfering signals, the Iridium and other communication systems' signals are protected using Beamforming. Use of a common broadband antenna and common front-end signal processing of the received signals enables an integrated system for use on a wide range of platforms, including without limitation size-, weight-, and power-constrained platforms such as drones, unmanned aerial vehicles (UAVs), and helicopters.

Claims

1. An antijam system for suppressing interference signals received with Iridium and GPS and other GNSS signals comprising: An antenna array having a plurality of antenna elements with signal reception coverage over the 950 MHz to 1.65 GHz frequency band; a RF front end for outputting downconverted and digitized antenna array received signals; means for digitally downconverting the digitized signals to baseband, suppressing the interference signals by filtering the baseband signals, and outputting filtered Iridium and GPS and other GNSS signals; and a navigation system for determining antenna array orientation.

2. The system of claim 1, wherein signal reception over the 950 MHz to 1.65 GHz frequency band includes reception of satellite and terrestrial radio communication signals.

3. The system of claim 2, wherein the satellite and terrestrial radio communication signals include Iridium and GPS and other GNSS signals.

4. The system of claim 1, wherein the antenna array is a broadband antenna array selected from the group consisting of multi-element arrays of patch-elements, polarimetric elements, and loop elements.

5. The system of claim 1, wherein the means comprise a general purpose processor and a hardware device.

6. The system of claim 5, wherein the hardware device is a logic device.

7. The system of claim 6, wherein the logic device is a Field Programmable Gate Array (FPGA).

8. The system of claim 6, wherein the logic device is an Application Specific Integrated Circuit (ASIC).

9. The system of claim 1, wherein the means comprise a processor programmed to receive antenna array orientation information from the navigation system and digitally downconvert the digitized signals to baseband, suppress the interference signals by filtering the baseband signals, and output filtered Iridium and GPS and other GNSS signals.

10. The system of claim 3, wherein the interference signals are suppressed from the Iridium signals in a first filter.

11. The system of claim 10, wherein the first filter is a first STAP filter.

12. The system of claim 10, wherein the interference signals are suppressed from the Iridium signals in the first STAP filter by beamforming.

13. The system of claim 3, wherein the interference signals are suppressed from the GPS and other GNSS signals in a second filter.

14. The system of claim 13, wherein the second filter is a second STAP filter.

15. The system of claim 14, wherein the interference signals are suppressed from the GPS and other GNSS signals in the second STAP filter by nulling.

16. The system of claim 10, wherein the filtered Iridium signals are digitally up-converted for RF conversion and up-conversion.

17. The system of claim 13, wherein the filtered GPS and other GNSS signals are output to a GPS and other GNSS receiver.

18. The system of claim 1, wherein the navigation system is an inertial navigation system.

19. The system of claim 18, wherein the inertial navigation system comprises an inertial measurement unit.

20. The system of claim 19, wherein the inertial measurement unit comprises at least one gyroscope.

21. The system of claim 1, wherein the navigation system includes at least one magnetic field sensor.

22. The system of claim 5, wherein the general purpose processor is a general purpose computer.

23. A method for suppressing interference signals received with Iridium and GPS and other GNSS signals comprising the steps of: receiving Iridium signals, GPS and other GNSS signals, and interference signals at an antenna array; downconverting the received signals; digitizing the downconverted signals; downconverting the digitized signals to baseband; determining position and orientation of the antenna array; suppressing the interference signals in the baseband signals using satellite almanac data and the antenna array position and orientation, from: the Iridium signals in a first filter, and the GPS and other GNSS signals in a second filter; up-converting, and RF-converting and up-converting the filtered Iridium signals; and outputting separate filtered RF up-converted Iridium and digital GPS and other GNSS signals.

24. The method of claim 23, wherein the antenna array is a broadband antenna array.

25. The method of claim 23, wherein the downconverting and digitizing of the antenna array received signals is performed in an RF front end.

26. The method of claim 23, wherein the digital downconverting of the digitized signals to baseband is performed using digital downconverters.

27. The method of claim 23, wherein the first filter is a STAP filter.

28. The method of claim 23, wherein the second filter is a STAP filter.

29. The method of claim 23, wherein the steps of digital downconverting, suppressing of the interference signals, and digital up-converting of the filtered Iridium signals are performed in a hardware device.

30. The method of claim 29, wherein the hardware device is a logic device.

31. The method of claim 30, wherein the logic device is a Field Programmable Gate Array (FPGA).

32. The method of claim 30, wherein the logic device is an Application Specific Integrated Circuit (ASIC).

33. The method of claim 23, wherein the steps of digital downconverting, suppressing of the interference signals, and digital up-converting of the filtered Iridium signals are performed in a software device.

34. The method of claim 33, wherein the software device is a processor programmed to perform the steps of digital downconverting, suppressing of the interference signals, and digital up-converting of the filtered Iridium signals.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The instant specification is better understood using the drawings below. The drawings, however, are primarily for illustration and must not to be construed as limiting.

[0020] FIG. 1 is a block diagram of the main components of the integrated Iridium/GPS anti-jamming (AJ) system.

[0021] FIG. 2 is a block diagram of Beamforming showing an example division of signal processing tasks between hardware (HW) and software (SW).

[0022] FIG. 3 is a block diagram of additional detail of the design of the Beamformer.

[0023] FIG. 4 is a block diagram of signal processing steps used to obtain the Beamformer weights in the Fourier domain.

[0024] FIG. 5 is a table of Beamforming protection levels for Iridium SVs 64, and 47 from a test conducted on 11 Oct. 2016.

[0025] FIG. 6 is a table of Beamforming protection levels for Iridium SVs 46, 20, 49, 11, and 3 from a test conducted on 17 Oct. 2016.

[0026] FIG. 7 is a table of Beamforming protection levels for Iridium SVs 31, 32, 91, 58, and 96 from a test conducted on 2 Nov. 2016.

DETAILED DESCRIPTION OF THE INVENTION

[0027] FIG. 1 is an overview of one embodiment of the system depicting an eight element sCRPA 11 broadband receive antenna array integrated with an Iridium TX transceive patch antenna element 12. (The transceive patch element 12 supports Iridium user uplink data transmissions to the Iridium SV.) The specific types of patch element and array, and number of antenna elements are illustrative and not limiting. The antenna may be composed of multi-element arrays of patch-elements, and/or polarimetric elements, such as bent monopole elements and/or loop elements. Some, not all, classes of polarimetric-element arrays require mode-forming. For example, mode-forming is required with antenna systems based on small CRPA (Controlled Reception Pattern Antenna) arrays, whereas the larger ADAP (Advanced Digital Antenna Production) antenna array does not require mode-forming for effective antijam processing of the received signals. All such antenna array based implementations requiring antijam processing with and without mode-forming are within the scope of this disclosure and the claims herein.

[0028] Signal outputs from the antenna array are downconverted and digitized into complex baseband samples, which are optionally available for digital modeforming (e.g., Butler array), if desired, prior to the signal processing for suppressing the GPS-bands jamming and Iridium-band jamming 13. The depicted embodiment shows an Iridium unprotected bypass mode of operation 14, whereby, if desired by the user, the Iridium radio bypasses its AJ protection. A circulator 15 enables passage of both Iridium uplink and downlink signals.

[0029] The GPS/Iridium AJ system 13 implements Nulling for the protection of the GPS L1 and/or L2 band transmissions and Beamforming for the protection of the Iridium user downlink transmission. Both Nulling and Beamforming algorithms run concurrently. Algorithms may be implemented in software (SW) or in a combination of hardware (HW), such as Field Programmable Gate Array (FPGA) or Application Specific Integrated Circuit (ASIC) technology, and software running on a general purpose processor (GPP). An illustrative split between HW and SW functionality appears in FIG. 2. The most compute-intensive aspects of the algorithms (e.g., fast-Fourier transform (FFT), matrix inversion/factorization) are preferably implemented in HW (but may also be implemented in SW), while less demanding aspects are implemented in SW (e.g., determination of beam direction, HW control). The particular division of work between HW and SW described herein is understood to be illustrative and not limiting.

[0030] The Nulling solution for the protection of GPS involves computing sample covariance matrices from the complex-valued, baseband (and, optionally, mode-formed) data output from the antenna array. The inverse of the sample covariance matrix is multiplied by a pseudo-steering vector to obtain the STAP-filter's complex-valued weight vector. The pseudo-steering vector is a vector with a unity-valued element that normally corresponds to the branch center-tap of the antenna array reference element (or the reference omni-mode for mode-formed arrays). The Nulling solution is well-known art with already known performance characteristics, so further details are not provided. See, e.g., Understanding GPS: Principles and Applications, Editors: E. D. Kaplan and C. J. Hegarty, Artech House, 2006.

[0031] The Beamforming solution is closely examined in FIG. 2 with an illustrative but not limiting embodiment. Digital down converters (DDCs) 21 convert the real-valued outputs of analog-to-digital (A/D) converters (not shown) to complex baseband form. The example here shows without limitation that the eight DDC converter-bank outputs (which correspond, respectively, to the eight signals received from the eight antenna elements) are, optionally, input to a digital mode-forming block 22, if mode forming is implemented, to select the five most efficient modes for the example sCRPA array depicted in FIG. 1. (The choice of five modes is illustrative, and not limiting.) Alternatively, the mode-forming step may be avoided and further STAP filtering performed directly using the complex baseband signals. The complex-valued time-domain samples (from mode forming, if implemented, or otherwise) are also input to the aforementioned Nulling (not shown). However, for use in Beamforming the samples are preferably buffered (FIFO 23) before being STAP-filtered. The FIFOs permit matching the complex baseband samples that are to be filtered to their appropriate filter weights. The weights themselves are obtained by processing K-sample buffered samples 24 (e.g., K=8192 complex samples/mode).

[0032] In this latter processing, described in further detail below, the time-domain K-samples are Fourier transformed (e.g., via N.sub.F-bin fast Fourier transform (FFT)), and the transformed data is formed into cross-spectral density matrices (CSDMs) 24, which may also be referred to as Fourier domain sample covariance matrices (per Owsley). The CSDMs are employed in the computation of Fourier domain STAP filter weights 25 as discussed below. Inverse FFTs are employed to convert the Fourier domain weights into the time-domain 25 for use in the STAP filter 26. The STAP filter employs N.sub.F taps per finite impulse response (FIR) filter branch (FIG. 3) where one filter branch is allocated to each antenna array element or mode from the antenna array output. (For illustration only. in FIG. 3, L=5 for a 5-mode system.) The complex-valued samples are digitally up-converted (See e.g., DUC 27), which is the inverse processing to that of the DDC, followed, if and as required, by digital-to-analog (D/A) conversion 28 and RF up-conversion (not shown), and onward transmission to the Iridium radio.

[0033] FIG. 3 illustrates the FIR filter signal processing 31 within the Beamformer 26, where N.sub.F-tap filtering is applied to each of the L modes or antenna element signals (in FIG. 2 we illustrate for L=5 modes as an example). The term w.sub.k,n in FIG. 3 denotes the tap n complex-valued weight applied to the antenna element k or mode-k branch. The antenna element k or mode-k antenna array response at the frequency ω in the Iridium space vehicle (SV) signal direction (θ, φ) (where θ=azimuth, φ=elevation in the platform's local Cartesian frame), may be represented as A.sub.k(θ, φ, ω). We call (θ, φ) the Iridium signal direction-of-arrival (DOA).

[0034] Thus, FIG. 3 also indicates that beamforming requires antenna pattern data (inclusive of RF electronics effects) for obtaining the correct steering vector to the Iridium SV. The direction-of-arrival (DOA) depends on the line-of-sight (LOS) to the Iridium SV from the platform hosting the antijam system (AJ system), inertial navigation system (INS), and GPS and Iridium receivers. The LOS is determined from both platform orientation ((INS)-derived information) and platform position (GPS-derived information) in space. For satellite communication, Iridium SV almanac information is also needed to determine which Iridium SV is in view, as LOS also requires position information about the Iridium SV.

[0035] The Beamforming algorithm employed herein is Capon's algorithm (See, e.g., Capon, and Owsley). This algorithm minimizes the STAP filter output power to achieve nulling of the interference, while at the same time the gain response of the STAP filter is constrained to unity to prevent nulling of the Iridium user's downlink signal.

[0036] FIG. 4 depicts the key signal processing steps involved in obtaining the Beamformer weights. The DDC 21 sample sequences are denoted x.sub.k (nT) where T is the DDC sampling interval and k=1,2, . . . , L−1, L for an antenna array of L antenna elements or modes, (e.g., L=5). In FIG. 4, sample blocks of length K complex samples per mode or antenna element are sub-divided into non-overlapping segments each of length N.sub.F samples, and thus K=MN.sub.F, for which M is the number of segments. Each N.sub.F length segment is transformed using N.sub.F-bin DFT/FFT operations to obtain a K-sample transformed data (Fourier coefficients) set per mode. Relatively long sample blocks (large K) are needed to obtain good covariance estimates. It has been found in experimentation with a prototype system that K may range from 2048 to 8192 samples. Prototyping also shows that N.sub.F=16 yields acceptable performance.

[0037] The Fourier bin m segment (or block) v FFT coefficients for each antenna element or mode are accumulated into a length L complex-valued column vector denoted X.sub.m,v=[X.sub.l,v(ω.sub.m) X.sub.2,v(ω.sub.m) . . . X.sub.L,v(ω)].sup.T in FIG. 4. The Fourier bin m cross-spectral density matrix (CSDM) (also called Fourier domain sample covariance matrix) is obtained via the outer products summations:

[00001]$\begin{array}{cc}{R}_m=\frac{1}{M}\underset{v-0}{\overset{M1}{.Math.}}{\underset{¯}{X}}_{m,v}{\underset{¯}{X}}_{m,v}^H\in C^{L\times L}& \left(1\right)\end{array}$

These CSDMs are applicable for the observation time k, and the computed beamformer STAP weights (see below) apply to these same samples as well.

[0038] Following Owsley, the following quadratic programming problem (QPP), when solved using the steering vector a.sub.m,p (defined below), yields the Fourier domain STAP weights, Ŵ.sub.m:

[00002]$\begin{array}{cc}{\underset{¯}{\hat{W}}}_m=\underset{{\underset{¯}{W}}_m}{\mathrm{argmin}}{\underset{\_}{W}}_m^H{R}_m{\underset{¯}{W}}_m\mathrm{subject}\mathrm{to}{\underset{\_}{W}}_m^H{\underset{\_}{a}}_{m,p}=1& \left(2a\right)\end{array}$

for m=0,1, . . . , N.sub.F−1 in general, or only some relevant subset of these indices such as those covering the Iridium user downlink band.

[0039] The steering vector is

a.sub.m,p=[A.sub.1(θ.sub.p, φ.sub.p, ω.sub.m) A.sub.2(θ.sub.p,φ.sub.p, ω.sub.m) . . . A.sub.L(θ.sub.p,φ.sub.p, ω.sub.m)].sup.T ∈C.sup.L   (2b)

This is from pre-stored antenna pattern and includes all RF electronics effects at frequency ω.sub.m (baseband equivalent, normalized frequency scale) in the pth signal direction (θ.sub.p,φ.sub.p) (i.e., azimuth, elevation) over all of the available antenna array modes k=1,2, . . . L. Thus, A.sub.k(θ, φ, ω) represents the mode (or antenna element, if there is no mode forming) k antenna pattern.

[0040] The linear equality constraint in (2a) constrains the STAP filter gain to be unity in the desired DOA, and across the band for which the constraints are applied. Implicitly, a linear phase response is also imposed over that band. However, the Iridium transmissions are very narrowband and so the linear phase response characteristic of the solution from solving (2a) is not critical.

[0041] From the method of Lagrange multipliers, the solution to the constrained QPP in (2a) is as follows, which is the well-known solution to the Capon beamforming weights design problem:

[00003]$\begin{array}{cc}{\underset{¯}{\hat{W}}}_m=\frac{1}{{\underset{¯}{a}}_{m,p}^H{R}_m^1{\underset{¯}{a}}_{m,p}}{R}_m^1{\underset{¯}{a}}_{m,p}=\left[\begin{array}{c}{\hat{W}}_1\left({\omega }_m\right)\\ {\hat{W}}_2\left({\omega }_m\right)\\ \mathrm{.Math.}\\ {\hat{W}}_L\left({\omega }_m\right)\end{array}\right]\in C^L& \left(3\right)\end{array}$

[0042] The computation (3) is done for bin m covering the 10.5 MHz Iridium user band at complex baseband. The Fourier domain weights for a bin that is not in this band are simply set to zero. From Equation (3) we see that a given L-element solution vector contains the Fourier domain weights for all the modes or antenna elements at that bin index. Thus, an implementation must search through all solution vectors to extract all of the Fourier domain weights for any given mode or antenna element of interest.

[0043] The Fourier domain weights from (3) are transformed into the time-domain using the inverse FFT so that filtering may be realized using STAP.

[0044] The Beamformer design described herein has been implemented as a prototype and tested a number of times against live in-sky Iridium satellite signals with broadband (BB) injected jamming signals. The Beamforming anti-jam (AJ) test procedure used was as follows: [0045] a. Align 0-deg azimuth (AZ) angle of the antenna array to the true North direction. [0046] b. Use software to predict a tracked Iridium satellite and ensure it is a rising satellite. [0047] c. Without jamming, let the Iridium phone track the satellite by going to “registered” state. [0048] d. Turn on the broadband (BB) jammer to jam out the Iridium phone. The Iridium phone should go to “searching for network” state after the jammer is turned on. The jamming power at this time is the OMNI break point. Because we don't have enough time to dwell in the OMNI mode to search for the OMNI break point during AJ tests, this OMNI break point needs to be determined before the AJ tests. [0049] e. Switch from OMNI mode to Beamforming ADAPT mode and wait for the Iridium phone to recover to the “registered” state. (The ADAPT mode pertains to Iridium handset operation in the presence of jammer signals. [0050] f. Check if the Iridium phone remains in “registered” state for 15 seconds. If so, increase the jamming power by 5 dB. If the Iridium phone remains in “registered” state for another 15 seconds, then increase the jamming power by another 5dB. Repeat the process until the Iridium phone goes to “searching for network” state; the jamming power at this time is at the ADAPT break point. [0051] g. Record the jamming power at OMNI break point and ADAPT break point, and the difference is the AJ protection of Beamforming.

[0052] FIG. 5, FIG. 6, and FIG. 7 provide tables of protection levels achieved based on this test procedure in the case of one BB jammer where each test was performed on a different day. Protection levels ranging from 25 to 40 dB were observed over a range of conditions. The OMNI and ADAPT break points are given as attenuation levels in decibels for the respective jammer power.

[0053] The scope of the invention is to be limited only by the claims, and not by the drawings or description herein. The words “including”, “comprising”, “having,” “with” and other like words used herein are to be interpreted broadly and comprehensively. Moreover, any embodiments disclosed in the subject application must not be taken as the only possible embodiments. Other embodiments will occur to those skilled in the art and are within the scope of the claims herein.