Background
Currently, the Global Navigation Satellite System (GNSS) mainly includes GPS in the united states, GLONASS in russia, beidou in china, and GALILEO in europe. The signals of the Global Navigation Satellite System (GNSS) are generally interfered when being received. Taking the GPS in the united states as an example, if the interference power exceeds the GPS signal reception power by 24dB, the C/a code receiver of the commercial GPS cannot keep track of the signal. Experiments show that the jammer with the power of 1W can enable a C/A code receiver within 85 kilometers to be incapable of working.
Many contexts require high interference rejection of GNSS receivers. In a differential network base station of the GNSS, a receiver needs to have anti-interference performance; the GNSS time of the communication network base station and the mobile phone network base station is provided by a receiver with better anti-interference performance; ocean fleet activities are global, and it cannot be guaranteed that the ocean fleet does not work in unfriendly or even malicious environments. Near an aviation airport where airplanes approach each other, the management and scheduling of airports requires GNSS signals to be of higher accuracy and not threatened by interference. The GNSS receiver is required to have certain anti-interference capability in city control networks, geodetic surveying, precision engineering surveying, crustal motion monitoring, resource exploration and the like and special application related to safety departments.
Aiming at navigation signal interference, the technical system is divided into compression interference, deception interference and distributed three-dimensional interference. The pressing type interference is to press the signal at the front end of the GNSS receiver by transmitting an interference signal with a certain level through a transmitter, so that the receiver cannot receive satellite signals; the deceptive jamming is that a transmitter transmits a fake signal which has the same parameters as the GNSS signal but different information codes, so that a receiver receives wrong positioning information; the distributed three-dimensional interference is the omnibearing three-dimensional interference which is carried out by applying a plurality of various interference machines on the ground and in the air. The specific interference pattern is various, such as wideband gaussian noise, continuous wave, swept continuous wave, pulsed continuous wave, am continuous wave, narrowband/wideband fm signal, etc.
The development of the anti-jamming technology of the GNSS receiver mainly includes the following three technologies: 1. time domain and frequency domain filtering technology: this technique is implemented at a digital intermediate frequency. It uses DFT (discrete fourier transform) technique in the processing of digital intermediate frequency signals. Taking simple frequency domain amplitude processing as an example, if there is no radio frequency interference in the signal, the thermal noise power spectrum is fairly uniform in the frequency domain; if there is a narrow-band interference in the signal, it will have abnormal spectral lines in the frequency domain, and such abnormal spectral lines are adaptively filtered out before the DFT inversion. Mitre corporation developed a GPS narrowband interference frequency domain removal chip. 2. The spatial filtering technology comprises the following steps: an adaptive array nulling technique is applied to a GNSS receiver. The self-adaptive antenna array comprises a plurality of antenna array elements, each array element is connected with a microwave network, the microwave network is connected with a processor, the processor processes signals from the microwave network and then feeds back and adjusts the microwave network, the gain and the phase of each array element are controlled to change, and a zero point facing to an interference direction is generated in an antenna directional diagram, so that interference is counteracted. The number of the nulls is determined by the number of the antennas, and generally M array elements can control M-1 null points. Under an ideal condition, the adaptive antenna can improve the anti-interference performance of the GNSS receiver by 40-50 dB. 3. Space-time adaptive (STAP) techniques: compared with the simple time domain and frequency domain technology, the spatial domain filtering technology has obvious advantages, simple realization and small calculated amount, but has two defects: if the number of array units is M, the maximum number of nulls which can be generated by the array is M-1, and the maximum number of interferences which can be eliminated by the array is the same. This statement does not exclude the possibility that a null could eliminate both interferers, since the aperture of the array is of the order of the carrier wavelength, and the actual angular separation of the two interferers, although relatively large, is still small compared to the beam width. In practical application, the number of array elements is limited in consideration of constraints such as cost, size and power, so that the anti-interference performance of the adaptive array is reduced. Secondly, if the angle interval of one interference source to a certain GNSS satellite is very close, the signal of the GNSS satellite is attenuated to be unusable aiming at the null of the interference space. In practice, although the angular separation of the interference from the satellite is already relatively large, it is still relatively small compared to the beamwidth of the array. In order to eliminate the influence of different error sources, signals of as many GNSS satellites as possible are required. Aiming at the defects, a space-time adaptive (STAP) technology is provided on the basis of a space domain filtering technology, time domain delay is added on the space domain filtering, a space-time two-dimensional filter is constructed by utilizing a space domain array and the time domain delay, each stage of delay forms FIR filtering when each array element channel is seen in isolation, and interference can be removed in the time domain; from the same time delay node, different array elements form self-adaptive filtering of a space domain, a space interference source can be distinguished, and space domain null is formed to inhibit space domain interference; the STAP can be considered that the space-time domain filtering is popularized to the space-time domain or the time domain filtering is popularized to the space-time two-dimensional filtering, and the STAP has the capability of eliminating interference on a space-frequency two-dimensional plane. On the premise of not increasing array elements, the method greatly increases the degree of freedom of the array and qualitatively improves the anti-interference capability of narrow-band interference. However, the method has the disadvantages of large operation amount and difficult realization.
The Digital Beam Forming (DBF) technology in navigation signal processing is also a new technology which is developed currently, and the basic idea of the technology is to utilize the direction function product theorem of an antenna array, control the direction function of the antenna array by weighting on antenna array elements, compensate the echo phase difference caused by the propagation path difference caused by different sensors in spatial positions, realize in-phase superposition, and achieve the purpose of controlling an antenna array directional diagram to dynamically generate high-gain narrow beams in the useful signal direction. To form multiple beams, a plurality of beamformers may be used that direct steering vectors in different directions.
Disclosure of Invention
The invention provides a satellite navigation anti-interference system and a method based on digital beam forming and space-time zero-setting cascade connection, which cascade digital beam forming and space-time zero-setting processing and improve the anti-interference performance of the whole system.
The invention can also form a plurality of independent beams to carry out anti-interference processing, the working state of the multi-beam realizes flexible scanning of the beams and quick change of the waveform through continuous change of the weight, and each beam can deal with a plurality of interference directions by combining self-adaptive interference zero setting.
The technical scheme of the invention is as follows:
an anti-interference system based on digital beam forming and space-time zeroing cascade connection comprises an antenna array, a radio frequency module and an anti-interference processing module which are sequentially connected, and is characterized in that the anti-interference processing module comprises a digital beam forming module and a space-time zeroing processing module; the digital beam forming module is used for weighting and synthesizing each antenna array element signal into a reference beam signal pointing to the satellite according to a steering vector obtained by the satellite pointing control information, so that the reference beam signal forms the maximum gain in the satellite direction, and meanwhile, an orthogonal vector signal orthogonal to the steering vector is obtained, and an interference signal in the reference beam signal is predicted; the output of the digital beam forming module is cascaded with the space-time zero-setting processing module, and the space-time zero-setting processing module is used for receiving the reference beam signal output by the digital beam forming module and the orthogonal vector signal orthogonal to the guide vector, performing adaptive zero-setting anti-interference processing on the interference signal in the reference beam signal, eliminating the interference signal in the reference beam signal, and outputting a useful satellite signal.
The digital beam forming module comprises a guiding vector calculation module and an orthogonal vector calculation module, wherein the guiding vector calculation module calculates satellite position information according to the output of the inertial navigation system and ephemeris forecast, and simultaneously calculates the pitch angle and the azimuth angle of the satellite in a view field to obtain an N-dimensional guiding vector pointing to the satellite, so that the SINR of the designated satellite in the view field is maximized; the orthogonal vector calculation module extracts one dimension of the N-dimensional input signal as a reference beam signal, and simultaneously multiplies the N-dimensional input signal by an N-1-dimensional orthogonal matrix of the N-dimensional steering vector to obtain an N-1-dimensional orthogonal vector signal orthogonal to the steering vector.
The space-time zero-setting processing module comprises an adaptive zero-setting anti-interference weight calculation module, and the adaptive zero-setting anti-interference weight calculation module calculates an adaptive zero-setting weight through a random gradient LMS adaptive algorithm on an input N-1 dimensional orthogonal vector signal under a minimum mean square error criterion.
A joint anti-interference system comprising a plurality of independent digital beam forming and space-time zero-setting based cascades comprises an antenna array, a radio frequency module and an anti-interference processing module which are connected in sequence, and is characterized in that the anti-interference system comprises a plurality of independent anti-interference processing subsystems, and each anti-interference processing subsystem comprises the radio frequency module, the digital beam forming module and the space-time zero-setting processing module which are connected in sequence respectively; the digital beam forming module is used for weighting and synthesizing each antenna array element signal into a reference beam signal pointing to the satellite according to a steering vector provided by the satellite pointing control information, so that the reference beam signal forms maximum gain in the satellite direction, an orthogonal vector signal orthogonal to the steering vector is obtained at the same time, and an interference signal in the reference beam signal is predicted; the output of the digital beam forming module is cascaded with the space-time zero-setting processing module, and the space-time zero-setting processing module is used for receiving the reference beam signal output by the digital beam forming module and the orthogonal vector signal orthogonal to the guide vector, performing adaptive zero-setting anti-interference processing on the interference signal in the reference beam signal, eliminating the interference signal in the reference beam signal, and outputting a useful satellite signal.
An anti-interference method based on digital beam forming and space-time zero-setting cascade is characterized in that two technologies of digital beam forming and space-time zero-setting anti-interference are combined, through the digital beam forming technology, according to a steering vector obtained by satellite directional control information, each antenna array element signal is weighted and synthesized into a reference beam signal pointing to a satellite, so that the reference beam signal forms maximum gain in the satellite direction, an orthogonal vector signal orthogonal to the steering vector is obtained at the same time, and an interference signal in the reference beam signal is predicted; and then, the interference signals in the predicted reference beam signals are subjected to adaptive zero-adjusting anti-interference processing by a space-time zero-adjusting anti-interference technology, so that the interference signals in the reference beam signals are eliminated, and useful satellite signals are output.
The steering vector obtained according to the pointing control information of the satellite refers to: according to the output of the inertial navigation system and ephemeris forecast, calculating satellite position information, and simultaneously calculating a pitch angle and an azimuth angle of the satellite in the view field to obtain a guide vector pointing to the satellite, so that the SINR of the designated satellite in the view field is maximized; the steering vector is obtained by the following method:
assuming that the array element positions are written in vector form as
The signal direction vector is
If the modulus of the signal direction vector is 1, the spherical coordinate of the signal direction vector is
Wherein,
when the angle between the signal direction vector and the positive direction of the z axis is theta, the x axis rotates anticlockwise to reach the angle rotated by the projection of the signal direction vector on the xoy plane when viewed from the positive direction of the z axis, and then the rectangular coordinate of the signal direction vector is expressed as follows:
the wave path difference between the arrival of the signal at each array element and the origin is expressed as
Dot multiplication of an array element vector and a signal vector; the expression from which the steering vector is derived is:
the weighting and synthesizing of the antenna array element signals into a reference beam signal pointing to a satellite means that: extracting one dimension of the N-dimensional input signal as a reference beam signal according to a steering vector pointing to the satellite, wherein the one-dimension reference beam signal is obtained by the following method:
if the system receives the N-dimensional array signal:
X(θ)=(x0,x1,…,xN-1)
the maximum composite signal after digital beamforming is:
d=X(θ)AH(θ)
wherein A (theta) is a guide vector, AH(θ) represents the conjugate transpose of steering vector a (θ); d is the one-dimensional reference beam signal, and d includes all signals of a certain satellite;
the orthogonal vector signal orthogonal to the steering vector acquired at the same time is the remaining N-1 dimensional signal:
Y=X(θ)P⊥(AH(θ))=X(θ)(I-AH(θ)A(θ)/(A(θ)AH(θ))) representing said N-1 orthogonal vector signals; wherein, P⊥Represents an orthogonal projection, and I represents an identity matrix; y does not contain any navigation signals.
The space-time zero-setting anti-interference processing comprises adaptive zero-setting anti-interference weight calculation, wherein the adaptive zero-setting anti-interference weight calculation refers to that an N-1-dimensional orthogonal vector signal is sent to an adaptive filter for processing, and the adaptive filter comprises two parts: the filtering process: calculating a response of the linear filter output to the input signal; an estimation error is then generated by comparing the output result field expected response. The self-adaptive process comprises the following steps: automatically adjusting filter parameters according to the estimate; these two processes work together to form a feedback loop; in the filtering process, the expected response d (n) is processed together with the input vector x (n), and given an input, the filter produces an outputAn estimate as an expected response d (n); defining an estimation error e (n) as the difference between the expected response and the actual filter output, adding the estimation error e (n) and an input vector x (n) to the adaptive control part, wherein a feedback loop surrounding the weight vector is closed loop, and the closed loop algorithm is the following according to a stochastic gradient LMS adaptive algorithm proposed by Widrow and Hoff:
in the above formula, mu is the step size factor, w (n) is the real-time weight, d*Representing a conjugate operation.
The steering vector includes corrections and compensations for the amplitude and phase of the antenna, X (θ) — (X) for the received array signal0,x1,…,xN-1) After compensation correction, it is expressed as:
Xcorrect(θ)=(x0,x1/M1(θ,φ),…,xN-1/MN-1(θ,φ))
the above formula uniformly represents the correction and compensation of amplitude and phase; wherein M isiAnd (theta, phi) represents the amplitude and phase difference between each antenna element and the reference element obtained by measurement.
The antenna array comprises a non-uniform array or an nonlinear array, and the phase difference of the non-uniform array or the nonlinear array caused by the wave path difference is as follows: if the unit vector for a certain incoming wave traveling direction in space is
I.e. the connecting line direction of the carrier and the satellite, and the wave path difference is the vector and the displacement vector r from each array element to the reference array element
iThe inner product of (2) is then the compensation parameter of each array element is
Way I:
and a path Q:
where λ is the wavelength.
The invention has the technical effects that:
the invention designs an anti-interference system based on digital beam forming and space-time zero-tuning cascade aiming at high requirements of anti-interference performance of a GNSS (Global Navigation Satellite System) receiver in many contexts, and the anti-interference system comprises an antenna array, a radio frequency module and an anti-interference processing module, can replace an antenna and a radio frequency part of a traditional GNSS receiver to be connected with a baseband part of a Navigation receiver, and can also replace the antenna part to be directly connected with a general GNSS Navigation receiver. The system cascades the digital beam forming and the space-time zero-setting treatment, combines the digital beam forming and the space-time zero-setting anti-interference, and improves the anti-interference performance of the whole system because the array gain of the digital beam forming is approximately equal to the number of array elements and the anti-interference capacity of N array elements can be improved by about N times; the invention also designs a combined anti-interference system comprising a plurality of independent anti-interference subsystems based on digital beam forming and space-time zero-setting cascade connection, which comprises a plurality of independent anti-interference subsystems based on digital beam forming and space-time zero-setting cascade connection, can simultaneously form a plurality of independent beams for anti-interference processing, each subsystem can carry out independent pointing control and self-adaptive anti-interference processing aiming at a certain satellite and can coordinate to work, the working state of the multi-beam realizes flexible scanning and rapid agility of the waveform of the beam through continuous weight value conversion, and meanwhile, the combined anti-interference zero-setting method can deal with a plurality of interference directions, has stronger protection capability on satellite signals in a complex interference scene, stronger satellite pointing capability and stronger robustness, and has correction capability on amplitude and phase inconsistency among receiving channels.
Detailed Description
The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
As shown in fig. 1, which is a schematic structural diagram of an anti-interference system according to an embodiment of the present invention, the anti-interference system includes three main components, namely an antenna array 3 (note: the number of array elements 31 is variable, and a quaternary array is shown in the schematic diagram), a multi-channel down-conversion radio frequency module 4, and an anti-interference processing module 5; the output of the anti-jamming system is connected to the GNSS satellite navigation receiver baseband section 6. The multi-channel down-conversion radio frequency module 4 is responsible for completing the amplification and frequency conversion work of signals, converting the frequency of an input GNSS carrier frequency signal into a low intermediate frequency for digital signal processing, and amplifying the signals to a level suitable for sampling; the output digital intermediate frequency signal enters an anti-interference processing module 5 for anti-interference processing, and the output signal processed by the anti-interference processing module 5 is subjected to baseband processing and navigation positioning calculation. Another general connection method is shown in fig. 2, the rf section 4 includes a multi-channel down-conversion rf module 41 and an up-conversion rf module 42, which is responsible for re-converting the processed if signal to a carrier frequency to ensure the versatility of the interface with the GNSS receiver. Therefore, the anti-interference system designed by the invention can replace the connection between the antenna and the radio frequency part of the traditional GNSS receiver and the baseband part of the navigation receiver, and can also replace the direct connection between the antenna part and the general GNSS navigation receiver.
The anti-interference processing module 5 of the present invention cascades the digital beam forming module and the space-time zeroing processing module, as shown in fig. 3, a schematic diagram of the cascade relation between the digital beam forming module and the space-time zeroing processing module in the anti-interference processing module 5 of the anti-interference system of the present invention is shown. The anti-interference processing module comprises a digital beam forming module and a space-time zero-setting processing module; the digital beam forming module is used for weighting and synthesizing each antenna array element signal into a reference beam signal pointing to the satellite according to a steering vector obtained by the satellite pointing control information, so that the reference beam signal forms the maximum gain in the satellite direction, and meanwhile, an orthogonal vector signal orthogonal to the steering vector is obtained, and an interference signal in the reference beam signal is predicted; the output of the digital beam forming module is cascaded with the space-time zero-setting processing module, and the space-time zero-setting processing module is used for receiving the reference beam signal output by the digital beam forming module and the orthogonal vector signal orthogonal to the guide vector, performing adaptive zero-setting anti-interference processing on the interference signal in the reference beam signal, eliminating the interference signal in the reference beam signal, and outputting a useful satellite signal. The digital beam forming module comprises a guiding vector calculation module and an orthogonal vector calculation module, wherein the guiding vector calculation module calculates satellite position information according to the output of the inertial navigation system and ephemeris forecast, and simultaneously calculates the pitch angle and the azimuth angle of the satellite in a view field to obtain an N-dimensional guiding vector pointing to the satellite, so that the SINR of the designated satellite in the view field is maximized; the orthogonal vector calculation module extracts one dimension of the N-dimensional input signal as a reference beam signal, and simultaneously multiplies the N-dimensional input signal by an N-1-dimensional orthogonal matrix of the N-dimensional steering vector to obtain an N-1-dimensional orthogonal vector signal orthogonal to the steering vector. The space-time zero-setting processing module comprises an adaptive zero-setting anti-interference weight calculation module, and the adaptive zero-setting anti-interference weight calculation module calculates an adaptive zero-setting weight through a random gradient LMS adaptive algorithm on the input N-1 dimensional orthogonal vector signal under the minimum mean square error criterion.
As shown in fig. 4, an anti-interference system including multiple independent digital beam forming and space-time zeroing cascade connections includes an antenna array, a radio frequency module, and an anti-interference processing module, which are connected in sequence, where the anti-interference system includes multiple independent anti-interference processing subsystems, and each anti-interference processing subsystem includes a radio frequency module, a digital beam forming module, and a space-time zeroing processing module, which are connected in sequence; each subsystem can perform independent pointing control and self-adaptive anti-interference processing on a certain satellite, and can work coordinately to ensure that the system can effectively cover satellite signals. Fig. 4 is a schematic diagram showing a cascade relationship of a plurality of independent anti-interference subsystems of the anti-interference system of the present invention. Each anti-interference processing subsystem respectively comprises a multi-channel down-conversion radio frequency module DDC, a digital beam forming module DBF-1 and a space-time zero-setting processing module STAP-1, a multi-channel down-conversion radio frequency module DDC, a digital beam forming module DBF-2 and a space-time zero-setting processing module STAP-2 which are connected in sequence, a multi-channel down-conversion radio frequency module DDC, a digital beam forming module DBF-N and a space-time zero-setting processing module STAP-N, and intermediate frequency signals IF _1, IF _2, IF. The digital beam forming module is used for weighting and synthesizing each antenna array element signal into a reference beam signal pointing to the satellite according to a steering vector provided by the satellite pointing control information, so that the reference beam signal forms the maximum gain in the satellite direction, and meanwhile, an orthogonal vector signal orthogonal to the steering vector is obtained, and an interference signal in the reference beam signal is predicted; the output of the digital beam forming module is cascaded with the space-time zero-setting processing module, and the space-time zero-setting processing module is used for receiving the reference beam signal output by the digital beam forming module and the orthogonal vector signal orthogonal to the guide vector, performing adaptive zero-setting anti-interference processing on the interference signal in the reference beam signal, eliminating the interference signal in the reference beam signal, and outputting a useful satellite signal.
An anti-interference method based on digital beam and space-time zero-setting cascade is characterized in that two technologies of digital beam forming and space-time zero-setting anti-interference are combined, through the digital beam forming technology, according to a steering vector obtained by satellite directional control information, antenna array element signals are weighted and synthesized into a reference beam signal pointing to a satellite, so that the reference beam signal forms maximum gain in the satellite direction, an orthogonal vector signal orthogonal to the steering vector is obtained at the same time, and an interference signal in the reference beam signal is predicted; and then, performing further anti-interference processing on the interference signal in the predicted reference beam signal by a space-time zero-setting anti-interference technology, eliminating the interference signal in the reference beam signal and outputting a useful satellite signal.
The difference between digital beam forming and traditional adaptive nulling is that a phased array technology is adopted on a reference signal, a guide vector (namely an array flow pattern) of a satellite is utilized to enable the reference signal beam to point to the satellite, and due to practical physical limitations such as array scale, the reference beam also has side lobes in a non-satellite direction, and interference signals can enter from the side lobes, so that useful satellite signals and interference signals exist in the reference signal, and at the moment, the output of beam forming is cascaded with traditional adaptive nulling anti-interference so as to further improve the anti-interference performance. As shown in fig. 3, the N-dimensional input signal outputs a one-dimensional reference signal according to a steering vector provided by the pointing control information, and at the same time, the N-1-dimensional orthogonal projection can be obtained by the N-1-dimensional orthogonal matrix of the N-dimensional input signal and the steering vector, because of the orthogonality with the steering vector, the N-1-dimensional output signal does not include a satellite signal but includes an interference signal, so that cancellation processing of an interference signal in a reference beam can be performed by an adaptive nulling algorithm, and the output after the interference cancellation is a satellite signal in the beam pointing direction.
The N-dimensional steering vector obtained from the pointing control information is: generally, according to Inertial Navigation System (INS) output and ephemeris forecast, satellite position information is calculated, and a pitch angle and an azimuth angle of a satellite in a field of view are calculated, so that a steering vector pointing to the satellite can be obtained, and a beam forming module maximizes SINR of a specified satellite in the field of view according to the steering vector.
Digital beamforming requires the synthesis of the best signal in a certain direction, and this N-dimensional vector is the steering vector. In array signal processing, array flow patterns are generally represented by N-dimensional vectors
The task of resisting disturbance is to find and
and an orthogonal N-dimensional weight value vector W is combined with a pointing control module in the navigation processor to calculate satellite coordinates by using ephemeris and time, and the satellite coordinates are converted into a steering vector and output to a beam forming module and a preset zero anti-interference weight value calculation module.
Assuming that the array element positions are written in vector form as
The signal direction vector is
The signal direction vector has a modulus of 1, and the spherical coordinate is
The included angle between the signal direction vector and the positive direction of the z axis is shown, and theta is the angle rotated by the x axis anticlockwise when the signal direction vector is viewed from the positive direction of the z axis to the angle rotated by the projection of the signal direction vector on the xoy plane. The rectangular coordinates of the signal bearing vector can be expressed as:
the difference in the path length of the signal between the arrival of each array element and the origin can be expressed as
Dot multiplication of an array element vector and a signal vector; the expression from which the steering vector is derived is:
according to the output of an Inertial Navigation System (INS) and ephemeris forecast, satellite position information is calculated, the pitch angle and the azimuth angle of the satellite in the view field are calculated at the same time, and the pitch angle and the azimuth angle are substituted into the formula, so that a guide vector pointing to the satellite can be obtained, and the SINR of the specified satellite in the view field is maximized.
According to the above expression of the steering vector, if the one-dimensional linear array is taken as an example for analysis, the steering vector can be obtained as follows:
the system receives the N-dimensional array signals as follows:
X(θ)=(x0,x1,…,xN-1)
the maximum composite signal after digital beamforming is:
d=X(θ)AH(θ)
wherein A isH(θ) represents the conjugate transpose of the steering vector a (θ). d is the one-dimensional reference signal, and according to the beam forming principle, d includes all signals of a certain satellite.
The remaining N-1 dimensional signals are: y ═ X (θ) P⊥(AH(θ))=X(θ)(I-AH(θ)A(θ)/(A(θ)AH(θ))) in any of the N-1 dimensions, where P is⊥For orthogonal projection, I represents an identity matrix, Y represents the orthogonal vector signal of the N-1 dimension, and since it is orthogonal projection, Y does not contain any navigation signal, d is interference with any component that can be predicted by Y,interference in d is achieved by eliminating the predicted portion of Y in d.
The space-time zero-setting anti-interference processing comprises adaptive zero-setting anti-interference weight calculation and adaptive zero-setting anti-interference weight calculation, and a space-time adaptive two-dimensional processing STAP technology is generally adopted. Space-time adaptive (STAP) technology utilizes a space domain array and time domain delay to construct a space-time two-dimensional filter. Each array element channel is isolated, each stage of time delay forms FIR filtering, and interference can be removed in a time domain; from the same time delay node, different array elements form self-adaptive filtering of a space domain, a space interference source can be distinguished, and space domain null suppression space domain interference is formed. The STAP can be considered that the space-time domain filtering is popularized to the space-time domain or the time domain filtering is popularized to the space-time two-dimensional filtering, and the STAP has the capability of eliminating interference on a space-frequency two-dimensional plane.
The number of array elements is set as M, and the number of time delay units is set as N. The time delay of each time delay unit is Ts. The weight w is an MN × 1-dimensional weight vector, and the two-dimensional filter structure is shown in fig. 7.
For the constrained optimization problem, the value obtained by calculating the weight vector through an LMS (Least-Mean-Square) algorithm under the minimum Mean Square error criterion represents an estimation, and when the iteration number tends to be infinite, the estimated expected value approaches a wiener solution for the generalized stationary process. The spatial nulling technique obtained based on the principle adopts a self-adaptive algorithm to correct the weighted value of the antenna, so that the gain and the phase of each array element are changed, zero points facing to the direction of an interference source are generated in a directional diagram of the antenna array, and the strong interference level can be effectively reduced to a low noise level, thereby achieving the purpose of resisting interference.
The LMS algorithm is a member of the family of stochastic gradient algorithms. The algorithm uses deterministic gradients in the stochastic input wiener filter recursive computation. One significant feature of the LMS algorithm is its simplicity. In addition, the method does not need to calculate related correlation functions and matrix inversion operation, and is easy to implement.
In general, the LMS includes two basic processes:
■ Filtering Process: firstly, calculating the response of the output of the linear filter to an input signal; an estimation error is then generated by comparing the output result field expected response.
■ adaptive procedure: the filter parameters are automatically adjusted according to the estimate.
The space-time zero-setting anti-interference processing is to send an N-1-dimensional orthogonal vector signal into an adaptive filter to perform adaptive zero-setting anti-interference weight calculation, and the adaptive filter comprises two parts: the filtering process: calculating a response of the linear filter output to the input signal; an estimation error is then generated by comparing the output result field expected response. The self-adaptive process comprises the following steps: automatically adjusting filter parameters according to the estimate; these two processes work together to form a feedback loop, as shown in fig. 8.
During the filtering process, the expected response d (n) participates in the processing together with the input vector y (n). In this case, given an input, the filter produces an output
As an estimate of the expected response d (n). Thus, the estimation error e (n) can be defined as the difference between the desired response and the actual filter output. The estimation error e (n) and the input vector y (n) are both applied to the adaptive control part, so that the feedback loop around the weight vectors is closed loop. According to the random gradient LMS adaptive algorithm proposed by Widrow and Hoff, the closed-loop algorithm is as follows:
in the above formula, mu is the step size factor, w (n) is the real-time weight, d*Representing a conjugate operation.
For a DBF system, calibration and compensation of the antenna amplitude and phase are key technologies for implementation. Since neither the actual rf nor the antenna is ideal, the rf may introduce a fixed phase shift, and the antenna elements have different gains and phases in different directions, an actual antenna gain pattern may be as shown in fig. 5. Besides the pointing information, the main control part of the system needs to provide compensation coefficients caused by channel and radio frequency imperfections and apply the compensation coefficients to the compensation correction of the N-dimensional steering vectors.
In order to correct the effect of antenna inconsistency, it is necessary to measure the amplitude and phase of the antenna, and to measure the gain and phase difference M between each antenna element and the reference elementi(theta, phi). The specific measurement method can be completed by a darkroom method or a specially designed receiver algorithm in an open place. For a received array signal X (theta) — (X)0,x1,…,xN-1) After compensation correction, it is expressed as:
Xcorrect(θ)=(x0,x1/M1(θ,φ),…,xN-1/MN-1(θ,φ))
the above equation represents the correction and compensation of amplitude and phase.
Because the array cannot be a linear array or even a uniform array due to the limitation of the conditions of a carrier and the like, the phase calculation of the array flow pattern is not necessarily a multiple relation, and the phase difference caused by the wave path difference needs to be obtained.
As shown in FIG. 6, the unit vector of the traveling direction of a certain incoming wave in space is
(i.e. the direction of the connection between the carrier and the satellite), the wave path difference is the vector and the displacement vector r from each array element to the reference array element
iInner product of, compensation parameters of each array element are
Way I:
and a path Q:
where λ is the wavelength.
It should be noted that the above-mentioned embodiments enable a person skilled in the art to more fully understand the invention, without restricting it in any way. Therefore, although the present invention has been described in detail in this specification and examples, it will be understood by those skilled in the art that the present invention may be modified and equivalents may be substituted; all technical solutions and modifications thereof which do not depart from the spirit and scope of the present invention are intended to be covered by the scope of the present invention.