Beam weight generation method, device and equipment

Abstract

The embodiment of the invention provides a beam weight generation method, device and equipment, and relates to the technical field of communication, and the method comprises the steps: respectively setting signal snapshots and array manifold vectors of an expected angle and an unexpected angle of a to-be-generated beam; wherein the expected angle is an angle corresponding to a main lobe of a to-be-generated beam, and the unexpected angle comprises an angle corresponding to a side lobe of the to-be-generated beam; generating a cost function according to the set signal snapshot and the array manifold vector; the variable of the cost function is the weight of a to-be-generated beam, and the dependent variable is the sum of a first preset parameter and a second preset parameter; and performing iterative operation on the cost function on the basis of a preset local particle swarm optimization algorithm based on a von Noiemann topological structure so as to obtain a target weight of the to-be-generated beam. According to the embodiment of the invention, the technical problem that the beam directivity and the main lobe power are difficult to consider in the existing weight generation mode of the multi-peak beam can be solved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a beam weight generation method, device and equipment. Background Art

[0002] Since the frequency band of the millimeter wave communication system is high and the path loss is large, beamforming technology is usually used to enable the user terminal to obtain a better signal to interference and noise ratio (SINR).

[0003] In practical applications, the millimeter wave communication system may have a situation where there are many access users but a small single-user business volume. In this case, the use of multi-peak beams is a better solution, and its advantages are reflected in two aspects: First, when the downlink channel state information reference signal (CSI-RS) is scanned by beam, the use of multi-peak beam scanning can increase the angle of each scan, thereby reducing the scanning cycle and reducing the user terminal delay; second, based on the beam scanning results, when transmitting data to the user terminal, the use of multi-peak beams can serve more users at the same time, which can also reduce the user terminal delay.

[0004] However, when the number of antennas is small, it is difficult for the existing multi-peak beam weight generation method to take into account both the beam directivity and the main lobe power, resulting in a low SINR of the user terminal. Summary of the invention

[0005] The embodiments of the present application provide a beam weight generation method, device and equipment, which can solve the technical problem that the existing multi-peak beam weight generation method is difficult to take into account both beam directivity and main lobe power.

[0006] In a first aspect, an embodiment of the present application provides a method for generating beam weights, the method comprising:

[0007] The signal snapshot and array manifold vector of the desired angle of the beam to be generated, and the signal snapshot and array manifold vector of the undesired angle are respectively set; wherein the desired angle is the angle corresponding to the main lobe of the beam to be generated, and the undesired angle includes the angle corresponding to the side lobe of the beam to be generated;

[0008] Generate a cost function according to the signal snapshot and array manifold vector of the desired angle and the signal snapshot and array manifold vector of the undesired angle; the variable of the cost function is the weight of the beam to be generated, and the dependent variable is the sum of a first preset parameter and a second preset parameter, the first preset parameter is the difference between the power of the main lobe and the power of the signal snapshot of the desired angle, and the second preset parameter is the difference between the power of the side lobe and the power of the signal snapshot of the undesired angle;

[0009] Based on a preset local particle swarm optimization algorithm based on a von Neumann topology structure, the cost function is iteratively calculated to obtain a target weight of the beam to be generated.

[0010] In some embodiments, it also includes:

[0011] Determining the first preset parameter according to the signal snapshot and array manifold vector of the expected angle and the conjugate transposed matrix of the weight;

[0012] The second preset parameter is determined according to the signal snapshot and array manifold vector of the non-desired angle and the conjugate transposed matrix of the weight.

[0013] In some embodiments, the local particle swarm optimization algorithm based on the preset von Neumann topology structure performs iterative calculations on the cost function to obtain the target weight of the beam to be generated, including:

[0014] Initializing a preset number of particles, and arranging the particles at each node in the von Neumann topology structure; wherein the dimension of the particle is 2N, N is the number of array elements corresponding to the beam to be generated, N dimensions of the 2N dimensions of the particle represent the amplitude of the particle, and the other N dimensions represent the phase of the particle, and N is a positive integer;

[0015] Traversing each of the particles, and performing the following operations on the currently traversed target particle: updating the speed of the target particle; updating the position of the target particle based on the updated speed of the target particle; converting the updated position of the target particle into the weight of the beam to be generated and inputting it into the cost function to obtain the cost function value corresponding to the target particle;

[0016] After traversing each particle, or when the cost function meets the preset convergence condition or number of iterations, the final position of the particle corresponding to the minimum cost function value obtained during the calculation process is determined, and the weight converted from the final position is determined as the target weight.

[0017] In some embodiments, updating the speed of the target particle includes:

[0018] Traverse each dimension of the target particle, and based on the preset update parameters, update the speed of the target particle currently traversed to the dth dimension:

[0019] The update parameters include at least one of the following parameters: current time t, update time interval T, the d-th dimension velocity of the target particle at time tT, the d-th dimension position coordinates of the target particle, the d-th dimension of the historical optimal solution of the target particle, and the d-th dimension of the historical optimal solution of the target particle in the neighborhood; d is a positive integer, and d≤2N.

[0020] In some embodiments, updating the speed of the target particle further includes:

[0021] If the target particle has an updated velocity v in any dimension i The modulus is greater than the preset speed threshold V max , then according to the speed threshold V max Adjust the v i .

[0022] In some embodiments, updating the position of the target particle based on the updated speed of the target particle includes:

[0023] The position coordinate of the target particle i in the dth dimension is updated using the following formula:

[0024] x i,d =x i,d (tT)+v i,d (t) × T;

[0025] Among them, x i,d represents the updated position of the target particle i in the dth dimension, x i,d (tT) represents the position of the target particle i in the dth dimension at time tT, v i,d (t) represents the updated velocity of the target particle i in the dth dimension at the current time t.

[0026] In some embodiments, updating the position of the target particle based on the updated speed of the target particle further includes:

[0027] If the updated position coordinate x of the target particle i in the dth dimension i,d The corresponding normalized parameter value is greater than the preset first parameter threshold, then the updated position coordinate x of the target particle i in the dth dimension is adjusted. i,d , so that the updated position coordinate x of the target particle i in the dth dimension i,d The normalized parameter value is equal to the first parameter threshold;

[0028] If the updated position coordinate x of the target particle i in the dth dimensioni,d The corresponding normalized parameter value is less than the preset second parameter threshold, then the updated position coordinate x of the target particle i in the dth dimension is adjusted. i,d , so that the updated position coordinate x of the target particle i in the dth dimension i,d The normalized parameter value of is equal to the second parameter threshold; wherein the second parameter threshold is less than the first parameter threshold.

[0029] In a second aspect, an embodiment of the present application provides a beam weight generation device, the device comprising:

[0030] A setting module, used to set the signal snapshot and array manifold vector of the desired angle of the beam to be generated, and the signal snapshot and array manifold vector of the undesired angle respectively; wherein the desired angle is the angle corresponding to the main lobe of the beam to be generated, and the undesired angle includes the angle corresponding to the side lobe of the beam to be generated;

[0031] A processing module, used to generate a cost function according to the signal snapshot and array manifold vector of the desired angle and the signal snapshot and array manifold vector of the undesired angle; the variable of the cost function is the weight of the beam to be generated, and the dependent variable is the sum of a first preset parameter and a second preset parameter, the first preset parameter is the difference between the power of the main lobe and the power of the signal snapshot of the desired angle, and the second preset parameter is the difference between the power of the side lobe and the power of the signal snapshot of the undesired angle;

[0032] The operation module is used to iteratively operate the cost function based on a preset local particle swarm optimization algorithm based on the von Neumann topology structure to obtain the target weight of the beam to be generated.

[0033] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory communicatively connected to the processor;

[0034] The memory stores computer-executable instructions;

[0035] The processor is used to execute the computer-executable instructions stored in the memory to implement the beam weight generation method provided in the first aspect.

[0036] In a fourth aspect, a computer-readable storage medium is provided in an embodiment of the present application, wherein the computer-readable storage medium stores computer-executable instructions. When a processor executes the computer-executable instructions, the beam weight generation method provided in the first aspect is implemented.

[0037] The beam weight generation method, device and equipment provided in the embodiments of the present application iterate the cost function through a local particle swarm optimization algorithm based on the von Neumann topology structure to obtain the target weight of the beam to be generated. When the number of antennas is small, the sidelobe power can be minimized while ensuring the main lobe power is constant. This can improve the directionality of the multi-peak beam, reduce interference to other users in the cell, and improve the SINR on the user side. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the following is a brief introduction to the drawings required for use in the embodiments or the related technical descriptions. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 A schematic diagram of a flow chart of a beam weight generation method provided in an embodiment of the present application;

[0040] Figure 2 A schematic diagram of a von Neumann topology structure provided in an embodiment of the present application;

[0041] Figure 3 Another schematic diagram of a flow chart of a beam weight generation method provided in an embodiment of the present application;

[0042] Figure 4 A schematic diagram of a program module of a beam weight generating device provided in an embodiment of the present application;

[0043] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0045] The terms "first", "second", etc. in the specification and claims of this application and the above drawings are used to distinguish similar or similar objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise noted. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, for example, they can be implemented in an order other than those given in the diagrams or descriptions of the embodiments of this application.

[0046] The term "module" used in the embodiments of the present application refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic, or a combination of hardware and software code that can perform the functions associated with the element.

[0047] The following is an explanation of some of the terms involved in the embodiments of the present application:

[0048] Millimeter wave communication system: Millimeter wave (mmWave) is an electromagnetic wave between microwave and light wave. Usually, the millimeter wave frequency band is 30GHz to 300GHz, and the corresponding wavelength is 1mm to 10mm. Millimeter wave communication system refers to a communication system that uses millimeter wave as a carrier for transmitting information.

[0049] SINR: Signal to Interference and Noise Ratio, refers to the ratio of the strength of the received useful signal to the strength of the received interference signal (noise and interference); it can be simply understood as "signal-to-noise ratio".

[0050] CSI-RS: Channel State Information Reference Signal, used to measure wireless channel quality and perform operations such as beamforming.

[0051] PSO: Particle Swarm Optimization is a population-based random optimization technology that originates from the study of bird flock predation behavior. It seeks the optimal solution through collaboration and information sharing among individuals in the group.

[0052] Von Neumann topology: A topology that is often used to define linear logic for multidimensional arrays.

[0053] In some embodiments, since the frequency band of the millimeter wave communication system is relatively high and the path loss is relatively large, beamforming technology is usually used to generate a multi-peak beam so that the user side can obtain a relatively good SINR.

[0054] In some embodiments, the process of generating a multi-peak beam may include the following steps:

[0055] Step 1: The network device sends a downlink CSI-RS signal to the user equipment.

[0056] Step 2: The user equipment feeds back the original signal to the baseband chip.

[0057] Step 3: The baseband chip obtains multiple user angles based on the above original signal processing.

[0058] Step 4: The baseband chip generates multi-peak beam weights based on the above multiple user angles.

[0059] Step 5: The baseband chip transmits the multi-peak beam weights to the beamforming chip.

[0060] Step 6: The beamforming chip modifies the amplitude and phase of the original signal based on the multi-peak beam weights to form a multi-peak beam.

[0061] In practical applications, there are usually two important factors that affect the SINR on the user side. One is the directivity of the beam. The better the beam directivity, the less interference from other beams, which can improve the SINR on the user side. The other is that based on the high path loss characteristics of millimeter waves, providing a larger array gain in the main lobe direction of the beam also helps to improve the SINR on the user side. However, how to generate a multi-peak beam that takes into account both beam directivity and main lobe power is a technical problem that needs to be solved urgently.

[0062] In some implementations, multiple discrete Fourier transform single-basis beam patterns can be used to combine into a multi-peak beam pattern with multiple directions. Specifically, two discrete Fourier transform beam weights can be superimposed, and the modulus of the largest element of the weight modulus can be used as a reference to normalize the power of the entire panel weight according to this modulus. However, in this way, since the maximum power of the array element is constant, even if the sidelobe power can be reduced after superposition, the mainlobe power will also decrease, and it is impossible to effectively ensure that the SINR on the user side is maintained at a high level.

[0063] In other embodiments, the antenna panel can be split into multiple sub-panels, and the mutual interference between the sub-panels is not considered, and a discrete Fourier transform beam is used on each sub-panel to form a multi-peak beam. When the number of antenna elements is large, this method can achieve good results, but when the number of antenna elements is small, the mutual interference between the sub-panels cannot be ignored, resulting in interference lobes, and there may be large interference at the target angle. Therefore, although the main lobe power generated by this method is higher, when the number of antenna elements is small, the SINR on the user side cannot be guaranteed to be at a high level.

[0064] In order to solve the above technical problems, a beam weight generation method is provided in an embodiment of the present application. By calling a local particle swarm optimization algorithm based on a von Neumann topology structure, the generated cost function is iteratively calculated to obtain the target weight of the beam to be generated. When the number of antennas is small, the side lobe power can be minimized on the basis of ensuring a certain main lobe power, thereby improving the directivity of the multi-peak beam, reducing interference to other users in the cell, and improving the SINR on the user side. Please refer to the following embodiments of the present application for details.

[0065] Reference Figure 1 , Figure 1 The following is a flow chart of a beam weight generation method provided in an embodiment of the present application. The beam weight generation method includes:

[0066] S101. Set the signal snapshot and array manifold vector of the desired angle of the beam to be generated, and the signal snapshot and array manifold vector of the undesired angle respectively; wherein the desired angle is the angle corresponding to the main lobe of the beam to be generated, and the undesired angle includes the angle corresponding to the side lobe of the beam to be generated.

[0067] Among them, for the signal snapshot at the expected angle, it refers to the sampling and recording of the signal from the expected direction, and the expected angle usually refers to the direction of the signal you want to receive; for the signal snapshot at the unexpecting angle, it refers to the sampling and recording of the signal from the unexpecting direction, and the unexpecting angle usually refers to directions other than the expected direction.

[0068] It should be noted that the selection of specific desired angles and undesired angles may vary depending on the application scenario and requirements. In practical applications, it is necessary to select appropriate desired angles and undesired angles according to the nature of the signal and processing requirements.

[0069] In beamforming, the array manifold vector (AMV) refers to the distribution of the signals of each antenna unit in the array antenna in space. For the array manifold vector at the desired angle, it indicates that the array antenna receives and amplifies the signal from the desired direction; for the array manifold vector at the undesired angle, it indicates that the array antenna receives and amplifies the signal from the undesired direction. At the desired angle, the amplitude of the array manifold vector is large, while at the undesired angle, the amplitude of the array manifold vector is small.

[0070] The calculation of the array manifold vector is usually based on the signal amplitude and phase information of each antenna unit in the array antenna. In practical applications, the array manifold vector can be obtained by sampling and processing the received signal, and its amplitude and phase information can be used to achieve beamforming and directional signal reception.

[0071] In beamforming, the main lobe is the direction with the highest radiation intensity in the beam, that is, the direction in which the antenna radiation energy is most concentrated. In the direction of the main lobe, the radiation intensity is the largest, and the angle corresponding to the main lobe can be regarded as the above-mentioned desired angle in this application. Side lobes are lobes other than the main lobe in the beam, and their intensity gradually decreases on both sides of the main lobe. The appearance of side lobes is due to the coupling between certain units of the antenna array or the discontinuity of the antenna structure. Although the radiation intensity of the side lobe is smaller than that of the main lobe, in some applications, the characteristics of the side lobe are also very important. For example, in radar and sonar systems, the radiation of the side lobe may have an adverse effect on the system performance, so measures need to be taken to reduce the intensity of the side lobe. In this application, the angle corresponding to the side lobe can be regarded as the above-mentioned undesired angle.

[0072] S102 , generating a cost function according to the signal snapshots and array manifold vectors at the desired angle and the signal snapshots and array manifold vectors at the undesired angle.

[0073] Among them, the variable of the above-mentioned cost function is the weight of the beam to be generated, and the dependent variable is the sum of a first preset parameter and a second preset parameter, the first preset parameter is the difference between the power of the main lobe and the power of the signal snapshot at the desired angle, and the second preset parameter is the difference between the power of the side lobe and the power of the signal snapshot at the non-desired angle.

[0074] The above cost function can also be called a loss function. In practical applications, it can be used as a learning criterion to link with the optimization problem, that is, to solve and evaluate the model by minimizing the cost function. In beamforming, the above cost function can be used to optimize the pointing of the beam to achieve the desired signal enhancement effect.

[0075] In the embodiment of the present application, the optimization goal of the above cost function is to minimize the side lobe power while ensuring that the main lobe power is constant.

[0076] S103, based on a preset local particle swarm optimization algorithm based on the von Neumann topology structure, iteratively calculate the above cost function to obtain a target weight of the beam to be generated.

[0077] It is understandable that in the traditional particle swarm optimization algorithm, the connection between particles is random, which will lead to a slow convergence speed of the algorithm and it is easy to fall into the local optimal solution. Alternatively, for the global particle swarm optimization algorithm (using star topology), each particle will be connected to every other particle, so although the convergence speed is faster, it is still easy to fall into the local optimal solution.

[0078] In an embodiment of the present application, a local particle swarm optimization algorithm based on the von Neumann topology structure is provided. The algorithm adopts the von Neumann topology structure, that is, the particles are arranged into a specific structure according to certain rules, and each particle only exchanges information with adjacent particles, which can effectively avoid the problem of local optimal solution.

[0079] For example, refer to Figure 2 , Figure 2 A schematic diagram of the structure of a von Neumann topology structure provided in an embodiment of the present application.

[0080] exist Figure 2 The relationship between a particle P and its neighboring particles D is shown in FIG.

[0081] from Figure 2 It can be seen that in the above von Neumann topology, each particle only exchanges information with adjacent particles, and does not exchange information with all particles.

[0082] In an embodiment of the present application, the local particle swarm optimization algorithm based on the von Neumann topology structure is used to iteratively calculate the cost function to find the weight that can make the function value of the cost function reach the minimum value. The weight can be used as the target weight of the beam to be generated.

[0083] In some implementations, after the target weight of the beam to be generated is obtained, a corresponding beam may be formed according to the target weight, wherein the beam may be a multi-peak beam.

[0084] The beam weight generation method provided in the embodiment of the present application obtains the target weight of the beam to be generated by calling the local particle swarm optimization algorithm based on the von Neumann topology structure, iteratively calculating the cost function, and minimizing the sidelobe power while ensuring the main lobe power when the number of antennas is small. This can improve the directionality of the multi-peak beam, reduce interference to other users in the cell, and improve the SINR on the user side.

[0085] Based on the contents described in the above embodiments, in some embodiments of the present application, the goal of beam design can be converted into a mathematical description, and the relevant concepts involved in the PSO process are introduced. Among them, the optimization goal of beam design is to minimize the sidelobe power while ensuring that the main lobe power is constant.

[0086] In some embodiments, the first preset parameter can be determined based on the signal snapshot and array manifold vector at the expected angle, and the conjugate transposed matrix of the weights; the second preset parameter can be determined based on the signal snapshot and array manifold vector at the non-expected angle, and the conjugate transposed matrix of the weights.

[0087] In some implementations, based on the above optimization objective, and the signal snapshots and array manifold vectors of the desired angles of the beam to be generated, and the signal snapshots and array manifold vectors of the undesired angles, the following cost function may be generated:

[0088] J(w)=(p1-w H p s ) 2 +α∫(w H p n -p2) 2 ;

[0089] Where w is the weight of the beam to be generated, J(w) is the cost function value, p1 is the signal snapshot at the desired angle, and p s is the array manifold vector of the desired angle, w H is the conjugate transposed matrix of w, β is the preset adjustment factor, p n is the array manifold vector of the undesired angle, p2 is the signal snapshot of the undesired angle, ∫ represents the integral operation, and the value range of the integral operation is the angle at which the above side lobe is higher than the specified side lobe envelope.

[0090] In some embodiments, the beam weight w is a vector.

[0091] The sidelobe envelope of a beam refers to the situation in which the amplitudes of the side lobes on both sides of the main lobe of the beam gradually decrease during beam forming.

[0092] In the embodiment of the present application, the target weight of the beam to be generated can be determined by finding a weight that can make the cost function value of the above cost function reach the minimum value.

[0093] Optionally, the constraints of the above cost function may be:

[0094] subject to‖w‖ ∞ =1

[0095] Among them, among them ‖w‖ ∞ represents the infinity norm of w.

[0096] In some implementations, a preset local particle swarm optimization algorithm based on a von Neumann topology structure may be called to iteratively calculate the above cost function to obtain a target weight of the beam to be generated.

[0097] Reference Figure 3 , Figure 3 Another flowchart of a beam weight generation method provided in an embodiment of the present application is shown below. In some embodiments, the beam weight generation method includes:

[0098] S301, initializing particles.

[0099] In some implementations, a preset number of particles may be initialized, and the initialized particles may be deployed at each node in the von Neumann topology structure.

[0100] Optionally, assuming that the number of antenna array elements is N, the dimension D of each of the above particles can be set to 2N, where N dimensions represent the amplitude of the particle, and another N dimensions represent the phase of the particle, and N is a positive integer.

[0101] In some embodiments, the value ranges of all parameters of each dimension of the above-mentioned particles can be normalized to [0,1], wherein the actual amplitude value range corresponding to the dimension representing the amplitude is [0,1], and the actual amplitude value range corresponding to the dimension representing the phase is [0,2π].

[0102] S302: Update the speed of the traversed target particles.

[0103] In some implementations, each particle may be traversed, and the speed of the currently traversed target particle may be updated.

[0104] In some embodiments, the speed of the target particle in the dth dimension currently traversed can be updated based on preset update parameters; the update parameters include at least one of the following parameters: current time t, update time interval T, the speed of the target particle in the dth dimension at time tT, the position coordinates of the target particle in the dth dimension, the dth dimension of the historical optimal solution of the target particle, and the dth dimension of the historical optimal solution of the target particle in the neighborhood; wherein d is a positive integer, and d≤2N.

[0105] Optionally, each dimension of the target particle may be traversed, and the following operations may be performed on the dimension d currently traversed by the target particle:

[0106] Use the following formula to update the velocity of the d-th dimension of the target particle i to v i,d (t):

[0107] v i,d (t) = β1 × v i,d (tT)+β2×rand×(p i best,d -x i,d )+β3×rand×

[0108] (p i local best,d -x i,d );

[0109] Among them, β1, β2, and β3 are all preset scaling factors, t represents the current time, and v i,d(tT) represents the velocity of the target particle i in the dth dimension at time tT, x i,d represents the position coordinate of the d-th dimension of the target particle i, rand is a random variable that is regenerated each time it is used, and p i best,d represents the dth dimension of the historical optimal solution of target particle i, p i local best,d represents the d-th dimension of the historical optimal solution of the target particle i in the neighborhood, d is a positive integer, and d≤2N.

[0110] In the particle swarm optimization algorithm, each particle has a historical optimal solution, which records the optimal position that the particle has ever reached. Specifically, each particle will update its speed and position based on its own experience and group experience; each particle will compare its current position with the position of the historical optimal solution. If the current position is better, it will update its historical optimal solution position. The continuous updating of the historical optimal solution can help particles find better solutions and gradually approach the global optimal solution of the problem during the iteration process. At the same time, self-cognition ability can also help particles better understand their own status and behavior performance, so as to better adjust their behavior strategies and improve the optimization effect.

[0111] In the particle swarm optimization algorithm, each particle interacts and cooperates with its neighboring particles to jointly find the optimal solution to the problem. In this process, each particle pays attention to the behavior and experience of the particles in its neighborhood and uses this information to adjust its own behavior strategy. Specifically, each particle updates its speed and position based on the experience and behavior of the particles in its neighborhood. The historical optimal solution that the particles in the neighborhood have reached is one of the very important reference information in the particle update process. Each particle compares its current position with the position of the historical optimal solution in the neighborhood. If the current position is better, it will update its historical optimal solution position. The continuous update of the historical optimal solution in the neighborhood can also help particles find better solutions and gradually approach the global optimal solution of the problem during the iteration process. At the same time, by paying attention to the behavior and experience of particles in the neighborhood, particles can better understand the status and behavior of other particles, so as to better adjust their own behavior strategies and improve the optimization effect.

[0112] In some embodiments, if the updated velocity v of the target particle i in any dimension is i The modulus is greater than the preset speed threshold V max , then the speed of the target particle i is limited.

[0113] In some embodiments, if the target particle has an updated velocity v in any dimension i The modulus is greater than the preset speed threshold Vmax , then according to the speed threshold V max Adjust v i .

[0114] Optionally, if the updated velocity v of the target particle i in any dimension i The modulus is greater than the preset speed threshold V max , then v i Adjust to v i ':

[0115] v i '=V max *v i / |v i |.

[0116] In some embodiments, v i Can be a vector, |v i | indicates v i The model, V max By multiplying v i / |v i |, you can put v i Adjust to v i ' while maintaining the original direction.

[0117] S303: Update the position of the traversed target particle.

[0118] In some implementations, each particle may be traversed, and after the speed of the currently traversed target particle is updated, the position of the target particle may be updated based on the updated speed of the target particle.

[0119] Optionally, the position coordinate of the d-th dimension of the target particle i can be updated using the following formula:

[0120] x i,d =x i,d (tT)+v i,d (t) × T;

[0121] Among them, x i,d represents the updated position of the target particle i in the dth dimension, x i,d (tT) represents the position of the target particle i in the dth dimension at time tT, v i,d (t) represents the updated velocity of the target particle i in the dth dimension at the current time t.

[0122] In some embodiments, if the updated position coordinate x of the target particle i in the dth dimension is i,dIf the corresponding normalized parameter value is greater than a preset first parameter threshold, or less than a preset second parameter threshold, the position of the target particle i in the dth dimension is restricted, wherein the second parameter threshold is less than the first parameter threshold.

[0123] Optional, if the updated position coordinate x of the target particle i in the dth dimension i,d If the corresponding normalized parameter value is greater than the preset first parameter threshold, the updated position coordinate x of the target particle i in the dth dimension is adjusted. i,d , so that the updated position coordinate x of the target particle i in the dth dimension i,d The normalized parameter value of is equal to the first parameter threshold; if the updated position coordinate x of the target particle i in the dth dimension i,d If the corresponding normalized parameter value is less than the preset second parameter threshold, the updated position coordinate x of the target particle i in the dth dimension is adjusted. i,d , so that the updated position coordinate x of the target particle i in the dth dimension i,d The normalized parameter value of is equal to the second parameter threshold.

[0124] In some implementations, the first parameter threshold is 1, and the second parameter threshold is 0.

[0125] S304: Use the cost function to score the traversed target particles.

[0126] In some implementations, the updated position of the target particle may be converted into the weight of the beam to be generated and then input into the cost function to obtain the cost function value corresponding to the target particle, and the cost function value may be used as the score of the target particle i.

[0127] For example, the position coordinate x of the target particle i in the D dimension can be i After converting to w, it is substituted into the cost function, from which the cost function value corresponding to the target particle i can be calculated.

[0128] S305: Determine whether a preset convergence condition or number of iterations is met.

[0129] In some implementations, after scoring the target particles, it may be determined whether a preset convergence condition or number of iterations is satisfied. If not, S306 is executed; if yes, S307 is executed.

[0130] The number of iterations can be set based on experience without fixed criteria, and the more iterations, the better the effect.

[0131] The convergence condition may be that the change in the cost function is less than a preset threshold value, and the threshold value may be an empirical value or obtained through a method such as cross-validation.

[0132] S306: Determine whether all particles have been traversed.

[0133] In some implementations, when the preset convergence condition or the number of iterations is not met, it may be determined whether all particles have been traversed. If so, S307 is executed; if not, the process returns to S302.

[0134] In some implementations, when it is determined that not all particles have been traversed, the traversal may continue to the next particle, and the operations in S302 to S304 may be performed on the next traversed particle.

[0135] S306. Output the particle corresponding to the minimum cost function value.

[0136] In some embodiments, after determining that a preset convergence condition or number of iterations is met, or after traversing each particle, the particle corresponding to the minimum cost function value obtained during the calculation process is output, and the weight obtained by converting the final position of the particle can be determined as the target weight of the beam to be generated.

[0137] The beam weight generation method provided in the embodiment of the present application can significantly improve the main lobe power, compared with the method of combining multiple discrete Fourier transform single-basis beam patterns into a multi-peak beam pattern with multiple directions, so that the user side can obtain a better SINR; compared with the method of splitting the antenna panel into multiple sub-panels, the multi-peak beam generated by the local particle swarm optimization algorithm based on the von Neumann topology structure has a higher main lobe power and lower side lobe power at the target angle, and can also obtain a better SINR on the user side.

[0138] Based on the contents described in the above embodiments, the present application also provides a beam weight generation device, referring to Figure 4 , Figure 4 A schematic diagram of a program module of a beam weight generating device provided in an embodiment of the present application. In some embodiments, the beam weight generating device 40 includes:

[0139] The setting module 401 is used to respectively set the signal snapshot and array manifold vector of the desired angle of the beam to be generated, and the signal snapshot and array manifold vector of the undesired angle; wherein the desired angle is the angle corresponding to the main lobe of the beam to be generated, and the undesired angle includes the angle corresponding to the side lobe of the beam to be generated.

[0140] The processing module 402 is used to generate a cost function based on the signal snapshot and array manifold vector at the desired angle and the signal snapshot and array manifold vector at the undesired angle; the variable of the cost function is the weight of the beam to be generated, and the dependent variable is the sum of a first preset parameter and a second preset parameter, the first preset parameter is the difference between the power of the main lobe and the power of the signal snapshot at the desired angle, and the second preset parameter is the difference between the power of the side lobe and the power of the signal snapshot at the undesired angle.

[0141] The operation module 403 is used to iteratively operate the cost function based on a preset local particle swarm optimization algorithm based on the von Neumann topology structure to obtain the target weight of the beam to be generated.

[0142] The beam weight generation device provided in the embodiment of the present application, by calling the local particle swarm optimization algorithm based on the von Neumann topology structure, iteratively calculates the cost function to obtain the target weight of the beam to be generated. When the number of antennas is small, the sidelobe power can be minimized while ensuring the main lobe power is constant. This can improve the directionality of the multi-peak beam, reduce interference to other users in the cell, and improve the SINR on the user side.

[0143] In some embodiments, the setting module 401 is further used to:

[0144] Determining the first preset parameter according to the signal snapshot and array manifold vector of the expected angle and the conjugate transposed matrix of the weight;

[0145] The second preset parameter is determined according to the signal snapshot and array manifold vector of the non-desired angle and the conjugate transposed matrix of the weight.

[0146] In some embodiments, the operation module 403 is used to:

[0147] Initializing a preset number of particles, and arranging the particles at each node in the von Neumann topology structure; wherein the dimension of the particle is 2N, N is the number of array elements corresponding to the beam to be generated, N dimensions of the 2N dimensions of the particle represent the amplitude of the particle, and the other N dimensions represent the phase of the particle, and N is a positive integer;

[0148] Traversing each of the particles, and performing the following operations on the currently traversed target particle: updating the speed of the target particle; updating the position of the target particle based on the updated speed of the target particle; converting the updated position of the target particle into the weight of the beam to be generated and inputting it into the cost function to obtain the cost function value corresponding to the target particle;

[0149] After traversing each particle, or when the cost function meets the preset convergence condition or number of iterations, the final position of the particle corresponding to the minimum cost function value obtained during the calculation process is determined, and the weight converted from the final position is determined as the target weight.

[0150] In some embodiments, the operation module 403 is used to:

[0151] Traversing each dimension of the target particle, and updating the speed of the target particle in the dth dimension currently traversed based on a preset update parameter;

[0152] The update parameters include at least one of the following parameters: current time t, update time interval T, the d-th dimension velocity of the target particle at time tT, the d-th dimension position coordinates of the target particle, the d-th dimension of the historical optimal solution of the target particle, and the d-th dimension of the historical optimal solution of the target particle in the neighborhood; d is a positive integer, and d≤2N.

[0153] In some embodiments, the operation module 403 is used to:

[0154] If the target particle has an updated velocity v in any dimension i The modulus is greater than the preset speed threshold V max , then according to the speed threshold V max Adjust the v i .

[0155] In some embodiments, the operation module 403 is used to:

[0156] The position coordinate of the target particle i in the dth dimension is updated using the following formula:

[0157] x i,d =x i,d (tT)+v i,d (t) × T;

[0158] Among them, x i,d represents the updated position of the target particle i in the dth dimension, x i,d (tT) represents the position of the target particle i in the dth dimension at time tT, v i,d (t) represents the updated velocity of the target particle i in the dth dimension at the current time t.

[0159] In some embodiments, the operation module 403 is used to:

[0160] If the updated position coordinate x of the target particle i in the dth dimension i,d The corresponding normalized parameter value is greater than the preset first parameter threshold, then the updated position coordinate x of the target particle i in the dth dimension is adjusted.i,d , so that the updated position coordinate x of the target particle i in the dth dimension i,d The normalized parameter value is equal to the first parameter threshold;

[0161] If the updated position coordinate x of the target particle i in the dth dimension i,d The corresponding normalized parameter value is less than the preset second parameter threshold, then the updated position coordinate x of the target particle i in the dth dimension is adjusted. i,d , so that the updated position coordinate x of the target particle i in the dth dimension i,d The normalized parameter value of is equal to the second parameter threshold; wherein the second parameter threshold is less than the first parameter threshold.

[0162] It can be understood that the implementation principle and method of the above-mentioned beam weight generation 40 are the same as the implementation principle and method of the beam weight generation method described in the above-mentioned embodiment. Therefore, the description of each embodiment in the above-mentioned beam weight generation method can be referred to and will not be repeated here.

[0163] Furthermore, based on the contents described in the above embodiments, an electronic device is also provided in an embodiment of the present application, which includes at least one processor and a memory; wherein the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory to implement the various steps in the beam weight generation method described in the above embodiments.

[0164] In order to better understand the embodiments of the present application, refer to Figure 5 , Figure 5 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application.

[0165] like Figure 5 As shown, the electronic device 50 of this embodiment includes: a processor 501 and a memory 502; wherein:

[0166] Memory 502, used to store computer-executable instructions;

[0167] The processor 501 is configured to execute the computer-executable instructions stored in the memory to implement the various steps in the beam weight generation method described in the above embodiment.

[0168] Optionally, the memory 502 may be independent or integrated with the processor 501 .

[0169] When the memory 502 is independently provided, the device further includes a bus 503 for connecting the memory 502 and the processor 501 .

[0170] An embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable instructions. When a processor executes the computer-executable instructions, each step of the beam weight generation method described in the above embodiment is implemented. For details, please refer to the relevant description in the above method embodiment.

[0171] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0172] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0173] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The above-mentioned module-composed unit can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0174] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The steps of the method disclosed in the present application may be directly implemented as being executed by a hardware processor, or may be implemented by a combination of hardware and software modules in the processor.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A beam weight generation method, characterized in that: The method comprises: The signal snapshot and array manifold vector of the desired angle of the beam to be generated, and the signal snapshot and array manifold vector of the undesired angle are respectively set; wherein the desired angle is the angle corresponding to the main lobe of the beam to be generated, and the undesired angle includes the angle corresponding to the side lobe of the beam to be generated; Generate a cost function according to the signal snapshot and array manifold vector of the desired angle and the signal snapshot and array manifold vector of the undesired angle; the variable of the cost function is the weight of the beam to be generated, and the dependent variable is the sum of a first preset parameter and a second preset parameter, the first preset parameter is the difference between the power of the main lobe and the power of the signal snapshot of the desired angle, and the second preset parameter is the difference between the power of the side lobe and the power of the signal snapshot of the undesired angle; Based on a preset local particle swarm optimization algorithm based on a von Neumann topology structure, the cost function is iteratively calculated to obtain a target weight of the beam to be generated.

2. The method according to claim 1, characterized in that Also includes: Determining the first preset parameter according to the signal snapshot and array manifold vector of the expected angle and the conjugate transposed matrix of the weight; The second preset parameter is determined according to the signal snapshot and array manifold vector of the non-desired angle and the conjugate transposed matrix of the weight.

3. The method according to claim 1 or 2, characterized in that: The preset local particle swarm optimization algorithm based on the von Neumann topology structure is used to iterate the cost function to obtain the target weight of the beam to be generated, including: Initializing a preset number of particles, and arranging the particles at each node in the von Neumann topology structure; wherein the dimension of the particle is 2N, N is the number of array elements corresponding to the beam to be generated, N dimensions of the 2N dimensions of the particle represent the amplitude of the particle, and the other N dimensions represent the phase of the particle, and N is a positive integer; Traversing each of the particles, and performing the following operations on the currently traversed target particle: updating the speed of the target particle; updating the position of the target particle based on the updated speed of the target particle; converting the updated position of the target particle into the weight of the beam to be generated and inputting it into the cost function, and calculating the cost function value corresponding to the target particle; After traversing each particle, or when the cost function meets the preset convergence condition or number of iterations, the final position of the particle corresponding to the minimum cost function value obtained during the calculation process is determined, and the weight converted from the final position is determined as the target weight.

4. The method according to claim 3, characterized in that The updating of the speed of the target particle comprises: Traversing each dimension of the target particle, and updating the speed of the target particle in the dth dimension currently traversed based on a preset update parameter; The update parameters include at least one of the following parameters: current time t, update time interval T, the d-th dimension velocity of the target particle at time tT, the d-th dimension position coordinates of the target particle, the d-th dimension of the historical optimal solution of the target particle, and the d-th dimension of the historical optimal solution of the target particle in the neighborhood; d is a positive integer, and d≤2N.

5. The method according to claim 4, characterized in that The updating of the speed of the target particle further includes: If the target particle has an updated velocity v in any dimension i The modulus is greater than the preset speed threshold V max , then according to the speed threshold V max Adjust the v i .

6. The method according to claim 4, characterized in that The updating of the position of the target particle based on the updated speed of the target particle comprises: The position coordinate of the target particle i in the dth dimension is updated using the following formula: x i,d =x i,d (t-T)+v i,d (t)×T; Among them, x i,d represents the updated position of the target particle i in the dth dimension, x i,d (tT) represents the position of the target particle i in the dth dimension at time tT, v i,d (t) represents the updated velocity of the target particle i in the dth dimension at the current time t.

7. The method according to claim 6, characterized in that The updating of the position of the target particle based on the updated speed of the target particle further includes: If the updated position coordinate x of the target particle i in the dth dimension i,d The corresponding normalized parameter value is greater than the preset first parameter threshold, then the updated position coordinate x of the target particle i in the dth dimension is adjusted. i,d , so that the updated position coordinate x of the target particle i in the dth dimension i,d The normalized parameter value is equal to the first parameter threshold; If the updated position coordinate x of the target particle i in the dth dimension i,d The corresponding normalized parameter value is less than the preset second parameter threshold, then the updated position coordinate x of the target particle i in the dth dimension is adjusted. i,d , so that the updated position coordinate x of the target particle i in the dth dimension i,d The normalized parameter value of is equal to the second parameter threshold; wherein the second parameter threshold is less than the first parameter threshold.

8. A beam weight generating device, characterized in that: The device comprises: A setting module, used to set the signal snapshot and array manifold vector of the desired angle of the beam to be generated, and the signal snapshot and array manifold vector of the undesired angle respectively; wherein the desired angle is the angle corresponding to the main lobe of the beam to be generated, and the undesired angle includes the angle corresponding to the side lobe of the beam to be generated; A processing module, used to generate a cost function according to the signal snapshot and array manifold vector of the desired angle and the signal snapshot and array manifold vector of the undesired angle; the variable of the cost function is the weight of the beam to be generated, and the dependent variable is the sum of a first preset parameter and a second preset parameter, the first preset parameter is the difference between the power of the main lobe and the power of the signal snapshot of the desired angle, and the second preset parameter is the difference between the power of the side lobe and the power of the signal snapshot of the undesired angle; The operation module is used to iteratively operate the cost function based on a preset local particle swarm optimization algorithm based on the von Neumann topology structure to obtain the target weight of the beam to be generated.

9. An electronic device, characterized in that: include: A processor and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor is used to execute the computer-executable instructions stored in the memory to implement the beam weight generation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions. When the processor executes the computer-executable instructions, the beam weight generation method according to any one of claims 1 to 7 is implemented.

Images