A multi-dimensional beamforming method based on FPGA

Abstract

The application discloses a kind of based on FPGA multidimensional beam forming method, comprising: receiving the spatial azimuth angle of incoming wave signal, elevation angle parameter;Receive array antenna element position information;Receive signal frequency information;Array antenna is modeled, constructs the spatial normalization vector of incoming wave signal;Based on the spatial azimuth angle, elevation angle and signal wavelength information, calculate the phase offset value of signal emission source to element and coordinate origin;The phase offset value is corrected;According to the phase offset value after correction, the phase of incoming wave signal is changed, and the beam pointing on signal direction of arrival or required emission direction is realized in cooperation with the phase shift characteristics of antenna itself.The application actually realizes beam pointing, and actual result precision meets demand;Meanwhile, compared with the traditional beam forming implementation method, the application method obtains the resource advantage of serial computation on the basis of retaining parallel computation timing stability.

Description

Technical Field

[0001] This invention relates to the field of communication signal processing technology, and more specifically to a multidimensional beamforming method based on FPGA. Background Technology

[0002] Phased array radar, also known as electronically scanned radar, has significant advantages over traditional mechanical radar, such as beam agility, low intercept rate, and high reliability. The beam control system is the key component of a phased array radar system. It changes the phase wave forward tilt angle by controlling the phase of each radiating element of the array antenna, and ultimately controls the beam pointing direction, thereby increasing the system's probability of acquiring useful signals while suppressing interference signals.

[0003] Beamforming is based on array antennas. As the number of antenna array elements increases, the computational load rises sharply, placing higher demands on the on-chip resources of the FPGA and limiting the options for engineering implementation. Summary of the Invention

[0004] The purpose of this invention is to provide a multidimensional beamforming method based on FPGA to solve the problem of balancing the calculation speed and resource consumption of traditional beam pointing.

[0005] To achieve the above objectives, the present invention employs the following technical solution:

[0006] A multidimensional beamforming method based on FPGA includes:

[0007] The spatial azimuth and elevation angle parameters of the received incoming signal; the position information of the receiving array antenna elements; and the frequency information of the received signal.

[0008] Model the array antenna and construct the spatial normalized vector of the incoming signal; based on the spatial azimuth angle, elevation angle and signal wavelength information, calculate the phase offset value between the signal source and the array element and the coordinate origin;

[0009] The phase offset value is corrected;

[0010] The phase of the incoming signal is changed based on the corrected phase offset value, and the phase shift characteristics of the antenna itself are used to achieve beam pointing in the direction of the incoming signal or the desired transmission direction.

[0011] Furthermore, the step of modeling the array antenna and constructing the spatially normalized vector of the incoming signal includes:

[0012] Choose any point on the array antenna as the origin of the coordinate system, where s k (n) represents the incident elevation angle and azimuth angle, respectively θ. k and The k-th incoming wave signal has a spatially normalized vector relative to the coordinate system as follows:

[0013]

[0014] Further, the step of calculating the phase offset value between the signal transmitting source and the array element and the coordinate origin based on the spatial azimuth angle, elevation angle, and signal wavelength information includes:

[0015] Let the spatial coordinates of array element m be:

[0016] m = [x m ,y m ,z m ]

[0017] The distance difference between the signal source and array element m and the origin is expressed as:

[0018]

[0019] Corresponding phase offset value:

[0020]

[0021] In the formula, λ is the signal wavelength.

[0022] Furthermore, let the element spacing be d, the incident angle of the incoming signal s(n) be θ, the beam pointing output signal be y(n), the array antenna weight matrix be w, and the first element be taken as the reference element;

[0023] Based on the above modeling process, the array steering vector can be obtained as follows:

[0024] a(θ) = [1, e -jφ ,…,e -j(M-1)φ ] T

[0025] Where j is the imaginary unit, e is the base of the natural logarithm, φ = 2πdsinθ / λ, and M is the number of array elements;

[0026] Write the weights of each element tap in vector form:

[0027] w = [w1, w2, ..., w M ] T

[0028] The final output of the array is:

[0029] y(n)=w H a(θ)s(n)

[0030] The superscript H indicates conjugate transpose;

[0031] When w = a(θ), that is, after pre-shifting the incoming wave signal s(n) by a phase shift equal to the phase offset value (M-1)φ, it can be guaranteed that there is the maximum gain in the transmission or reception direction θ of the incoming wave signal, that is, the beam pointing is achieved.

[0032] Furthermore, calculating the phase shift requires calculating trigonometric functions, which are then iteratively calculated using planar rotation:

[0033] Vector (X) i ,Y i Planar rotation can be represented by a matrix as follows:

[0034]

[0035] Among them, (X) j ,Y j ) represents the rotated vector, and θ is the rotation angle; when θ is rotated by θ... n After execution in steps with a step size, it is represented as:

[0036]

[0037] Among them, (X) n+1 ,Y n+1 () is the vector after the (n+1)th rotation;

[0038] To suit the computing characteristics of FPGA, tanθ n Expressing it as a power of 2 simplifies the calculation:

[0039]

[0040] By setting a value for n, trigonometric function values ​​with the required accuracy can be obtained through iteration.

[0041] Furthermore, the calculation of the trigonometric functions is implemented in a trigonometric function module; a serial-parallel combination approach is adopted, using a trigonometric function module during calculation, but not directly outputting the results but temporarily storing the data, while suppressing the enable of new angle input, and opening the enable after all array element phase shift calculations are completed, and outputting the data in parallel.

[0042] Further, the phase offset value is corrected, including:

[0043] An iterative method is used to constrain the phase between (-π, π], where N is the number of iterations. The specific calculation is as follows:

[0044]

[0045] The computation maintains a structure that combines serial and parallel operations while suppressing new inputs.

[0046] Furthermore, the phase shift characteristics of the antenna itself are used to achieve beam pointing of the signal in the upward or desired transmission direction, i.e., y(n) = w H a(θ)s(n).

[0047] Compared with the prior art, the present invention has the following technical features:

[0048] This invention addresses the shortcomings of traditional beam pointing parameter calculation, such as large computational load in fully parallel calculation and unstable results in fully serial calculation. It adopts a combination of serial and parallel methods to calculate beam pointing parameters, absorbing the advantages of both parallel and serial calculations. It achieves the resource advantages of serial calculation while retaining the timing stability of parallel calculation. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the process of the present invention;

[0050] Figure 2 A schematic diagram of the array antenna model;

[0051] Figure 3 A schematic diagram for modeling the beam pointing of a linear array;

[0052] Figure 4 A schematic diagram of matrix modeling for planar rotation;

[0053] Figure 5 A schematic diagram of the MATLAB beam pointing pattern simulation results;

[0054] Figure 6 The error curve of the PFGA implementation result relative to the MATLAB result. Detailed Implementation

[0055] This invention addresses the problems of high computational complexity in fully parallel calculations and unstable results in fully serial calculations in traditional beam pointing parameter calculations. It proposes a multidimensional beamforming method based on FPGA (see appendix). Figure 1 The present invention specifically includes the following steps:

[0056] Step 1: Receive the spatial azimuth and elevation angle parameters of the incoming wave signal; receive the position information of the array antenna elements; receive the signal frequency (wavelength) information.

[0057] Step 2: Model the array antenna and construct the spatial normalized vector of the incoming signal; based on the spatial azimuth angle, elevation angle and signal wavelength information, calculate the phase offset value between the signal source and the array element and the origin.

[0058] Specifically, to allocate the phase of the array antenna and obtain the desired beam pointing result, the array antenna is modeled as follows: Figure 2As shown, taking an arbitrary array spatial form of a general M-element array as an example, any point on the array antenna is selected as the origin of the coordinate system, where s k (n) represents the incident elevation angle and azimuth angle, respectively θ. k and The k-th incoming wave signal has a spatially normalized vector relative to the coordinate system as follows:

[0059]

[0060] If we let the spatial coordinates of array element m be:

[0061] m = [x m ,y m ,z m ]

[0062] The distance difference between the signal source and array element m and the origin can be expressed as:

[0063]

[0064] The corresponding phase offset value can be written as:

[0065]

[0066] In the formula, λ is the signal wavelength.

[0067] To simplify the process, we will now use an M-element uniform linear array as an example to illustrate the impact of phase offset on beam pointing results. For example... Figure 3 As shown, let the element spacing be d, the incident angle of the incoming wave signal s(n) be θ, the beam pointing output signal be y(n), the array antenna weight matrix be w, and the first element be the reference element.

[0068] Based on the above modeling process, the array steering vector can be obtained as follows:

[0069] a(θ) = [1, e -jφ ,…,e -j(M-1)φ ] T

[0070] Where j is the imaginary unit and e is the base of the natural logarithm;

[0071] φ=2πdsinθ / λ

[0072] Write the weights of each element tap in vector form:

[0073] w = [w1, w2, ..., w M ] T

[0074] The final output of the array is:

[0075] y(n)=wH a(θ)s(n)

[0076] The superscript H indicates conjugate transpose.

[0077] It can be seen that, given a fixed signal incident angle θ, whether a signal can pass through the spatial filter depends entirely on the weights w. Choosing different weights can result in zero output from the array for the incident signal in the θ direction, or it can amplify the amplitude of the incident signal in that direction by a factor of two before outputting it. In other words, by determining the weight coefficients w through certain optimization criteria, it is entirely possible to amplify signals in some directions or block signals in others.

[0078] In particular, when w = a(θ), that is, after the incoming wave signal s(n) is pre-shifted by a phase shift equal to the phase offset value (M-1)φ, the maximum gain in the transmission or reception direction θ of the incoming wave signal can be guaranteed, that is, the beam pointing can be achieved.

[0079] Calculating phase offset requires calculating trigonometric functions. To reduce the use of on-chip resources on the FPGA, this invention utilizes a trigonometric function calculation module that employs planar rotation to iteratively calculate the trigonometric functions.

[0080] Figure 3 The middle vector (X) i ,Y i Planar rotation can be represented by a matrix as follows:

[0081]

[0082] Among them, (X) j ,Y j ) represents the rotated vector, and θ is the rotation angle; when θ is rotated by θ... n After execution in steps with a step size, it is represented as:

[0083]

[0084] To suit the computing characteristics of FPGA, tanθ n Expressing it as a power of 2 simplifies the calculation:

[0085]

[0086] When a suitable value of n is set, trigonometric function values ​​with the required accuracy can eventually be obtained through iteration.

[0087] In step 2, when calculating the phase offset value, multiple phase data need to be output simultaneously based on the number of arrays. Traditional fully parallel algorithms use multiple trigonometric function calculation modules to output data in parallel, which can ensure the correctness of the corresponding output timing when the input angle changes rapidly. However, the required resources will increase sharply when the number of arrays increases. Traditional serial algorithms use only one trigonometric function calculation module to serially output the phase data of each array element. When the input angle changes rapidly, it will cause the output data to become chaotic.

[0088] This invention employs a combined serial-parallel approach. During computation, a trigonometric function module is used, but the results are not directly output; instead, the data is temporarily stored. Simultaneously, enabling new angle inputs is suppressed. Once the phase shift calculations for all array elements are complete, the enable is activated, and the data is output in parallel. This ensures computational stability and reduces resource utilization.

[0089] Step 3: Correct the phase offset value.

[0090] An iterative method is used to constrain the phase between (-π, π]. The number of iterations N is set according to actual needs, and the specific calculation is as follows:

[0091]

[0092] The computation still maintains the suppression of new inputs, combining serial and parallel structures.

[0093] Step 4: Change the phase of the output data according to the beam direction.

[0094] The phase of the incoming signal is changed based on the corrected phase offset value, that is, y(n) = w H s(n), in conjunction with the phase shift characteristic a(θ) of the antenna itself, achieves beam pointing upwards or in the desired transmission direction, i.e., y(n) = w H a(θ)s(n).

[0095] Experimental verification:

[0096] Figure 5 The results of the MATLAB beam pointing pattern simulation are as follows. Figure 6 The error curves for the PFGA implementation results compared to the MATLAB results are shown. It is evident that the present invention does indeed achieve beam pointing, and the actual accuracy meets the requirements. Furthermore, compared to traditional beamforming implementation methods, the present invention achieves the resource advantages of serial computation while retaining the timing stability of parallel computing.

[0097] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A multidimensional beamforming method based on FPGA, characterized in that, include: The spatial azimuth and elevation parameters of the received incoming wave signal; Receive array antenna element position information; Receive signal frequency information; Model the array antenna and construct the spatial normalized vector of the incoming signal; based on the spatial azimuth angle, elevation angle and signal wavelength information, calculate the phase offset value between the signal source and the array element and the coordinate origin; The phase offset value is corrected; The phase of the incoming signal is changed according to the corrected phase offset value, and the phase shift characteristics of the antenna itself are combined to achieve beam pointing of the signal in the upward or desired transmission direction. When calculating the phase shift value, trigonometric functions are used, and the calculation of the trigonometric functions is implemented in a trigonometric function module. A serial-parallel approach is adopted. During the calculation, a trigonometric function module is used, but the result is not directly output. Instead, the data is temporarily stored. At the same time, the enable of new angle input is suppressed. When the phase shift calculation of all array elements is completed, the enable is turned on, and the data is output in parallel.

2. The FPGA-based multidimensional beamforming method according to claim 1, characterized in that, The process of modeling the array antenna and constructing the spatially normalized vector of the incoming signal includes: Choose any point on the array antenna as the origin of the coordinate system, where The incident elevation angle and azimuth angle are respectively and The k-th incoming wave signal has a spatially normalized vector relative to the coordinate system as follows: 。 3. The FPGA-based multidimensional beamforming method according to claim 1, characterized in that, The calculation of the phase offset value between the signal source and the array element and the origin of the coordinate system based on the spatial azimuth angle, elevation angle, and signal wavelength information includes: Let the spatial coordinates of array element m be: The distance difference between the signal source and array element m and the origin is expressed as: Corresponding phase offset value: In the formula The wavelength is the signal wavelength.

4. The FPGA-based multidimensional beamforming method according to claim 1, characterized in that, The spacing between array elements is d , incoming wave signal s ( n The angle of incidence is The beam pointing output signal is The array antenna weight matrix is And the first array element is used as the reference array element; Based on the above modeling process, the array steering vector can be obtained as follows: in, j The imaginary unit, e The base of the natural logarithm, M represents the number of array elements; Write the weights of each element tap in vector form: The final output of the array is: The superscript H indicates conjugate transpose; when That is, for the incoming wave signal Pre-set the size equal to the phase offset value After phase shift, the direction of transmission or reception of the incoming wave signal can be guaranteed. The maximum gain is achieved at the top, which is also the point where beam pointing is realized.

5. The FPGA-based multidimensional beamforming method according to claim 1, characterized in that, Calculating the phase shift requires calculating trigonometric functions, which are then calculated iteratively using planar rotation: vector Planar rotation can be represented by a matrix as follows: in, This represents the rotated vector. For rotation angle; when by After execution in steps with a step size, it is represented as: in, This is the vector after the (n+1)th rotation; To suit the computing characteristics of FPGA, Expressing it as a power of 2 simplifies the calculation: By preset By iterating over the value of , trigonometric function values ​​with the required accuracy can be obtained.

6. The FPGA-based multidimensional beamforming method according to claim 1, characterized in that, Correcting the phase offset value includes: The phase is constrained using an iterative method. Between these numbers, N represents the number of iterations, calculated as follows: The computation maintains a structure that combines serial and parallel operations while suppressing new inputs.

7. The FPGA-based multidimensional beamforming method according to claim 1, characterized in that, The phase shift characteristics of the antenna itself are used to achieve beam pointing of the signal upwards or in the desired transmission direction, i.e. .

Images