CN120416863B - Beam control optimization method and system for millimeter wave network - Google Patents
Beam control optimization method and system for millimeter wave networkInfo
- Publication number
- CN120416863B CN120416863B CN202510913118.1A CN202510913118A CN120416863B CN 120416863 B CN120416863 B CN 120416863B CN 202510913118 A CN202510913118 A CN 202510913118A CN 120416863 B CN120416863 B CN 120416863B
- Authority
- CN
- China
- Prior art keywords
- phase
- optimization
- channel capacity
- reflection
- vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/24—Cell structures
- H04W16/28—Cell structures using beam steering
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Radio Transmission System (AREA)
Abstract
The invention discloses a beam control optimization method and a system for a millimeter wave network, which relate to the technical field of beam control optimization and comprise the following steps of adopting a DTW algorithm dynamic time warping algorithm to match multi-hop path delay characteristics between a base station and user equipment, screening an optimal reflection path sequence, obtaining path topology, constructing a secondary unconstrained binary optimization model, obtaining discrete phase distribution through quantum annealing solution, calculating quantization errors of the discrete phases, adopting an interpolation algorithm frequency domain to compensate, obtaining a phase compensation vector, loading the compensation vector to programmable super-surface hardware, forming a reflection link, constructing a phase error covariance matrix, carrying out incremental correction on weight parameters of an optimization model based on the matrix, and finally, combining correction weight and the compensation vector to construct an objective function, so that channel capacity optimization is realized, the problems of difficult multi-hop path selection, insufficient phase error compensation and high beam optimization complexity in millimeter wave communication are solved, and channel capacity and communication quality are improved.
Description
Technical Field
The present invention relates to the field of beam control optimization technology, and more particularly, to a beam control optimization method and system for a millimeter wave network.
Background
In the development of the current fifth generation (5G) and future sixth generation (6G) mobile communication systems, millimeter wave communication technology has become a research hotspot due to its extremely high bandwidth and transmission rate. However, millimeter wave signals face serious problems of path loss, weak penetrating capacity, easy shielding and the like in the transmission process, particularly in complex urban environments, the situation that a direct path is blocked is very common, and the communication stability and the service coverage range of the millimeter wave signals are greatly limited.
To overcome these physical limitations, beam steering techniques based on reflection paths and programmable subsurface have been increasingly developed in recent years. The programmable super surface is a novel artificial structure, and the phase, amplitude and direction of the incident electromagnetic wave can be dynamically adjusted, so that the defect of a direct channel is overcome by constructing a multi-hop reflection link. However, how to select an optimal path from a plurality of reflection paths, how to construct an optimal beam forming strategy, and how to cope with discrete errors and regulation deviation introduced by actual hardware are still critical issues to be solved.
In the prior art, heuristic search or a traditional optimization algorithm is mostly adopted to configure the super-surface state, but the defects of low solving efficiency, insufficient optimality and difficulty in adapting to dynamic channel change are commonly existed. Meanwhile, most methods neglect the influence of discrete hardware characteristics on the overall beam control precision, so that the performance in actual deployment is reduced, and the potential of the super surface in millimeter wave communication cannot be fully exerted.
The present invention proposes a solution to the above-mentioned problems.
Disclosure of Invention
In order to overcome the defects in the prior art, the embodiment of the invention provides a beam control optimization method and a system for a millimeter wave network, which realize high-precision beam control and channel capacity optimization based on programmable super-surface in the millimeter wave network by combining dynamic time regular path matching, quantum annealing optimization model solving and frequency domain error compensation so as to solve the problems of difficult selection of multi-hop reflection paths, insufficient phase dispersion error compensation and high complexity of beam optimization in millimeter wave communication, thereby causing difficult improvement of channel capacity and communication quality.
In order to achieve the above purpose, the present invention provides the following technical solutions:
A beam control optimization method for millimeter wave network includes the steps of matching multi-hop path delay characteristics between a base station and user equipment by means of a DTW algorithm, screening an optimal reflection path sequence, obtaining a topological structure of the optimal reflection path sequence, constructing a secondary unconstrained binary optimization model, carrying out quantum annealing solution to obtain discrete phase distribution, calculating quantization error characteristic values of the discrete phase distribution, carrying out frequency domain compensation on the quantization error characteristic values by means of an interpolation algorithm to obtain a phase compensation vector, loading the phase compensation vector to a programmable super-surface hardware unit to obtain a first reflection link, constructing a phase error covariance matrix, carrying out incremental correction on weight parameters of the secondary unconstrained binary optimization model based on the phase error covariance matrix, constructing an objective function based on the corrected weight parameters and the phase compensation vector, and carrying out channel capacity reconstruction by means of a gradient projection method to obtain optimal channel capacity.
In a preferred embodiment, the method adopts a DTW algorithm to match multi-hop path delay characteristics between a base station and user equipment, and screens an optimal reflection path sequence, specifically, time delay sequence data of a plurality of candidate reflection paths between the base station and the user equipment are collected, a path delay characteristic library is constructed, the time delay sequence of each candidate reflection path is aligned by adopting the DTW algorithm based on a preset reference time delay template, the minimum accumulated path distance is calculated, and the plurality of candidate reflection paths are screened according to the minimum accumulated path distance, so that the optimal reflection path sequence is obtained.
In a preferred embodiment, the topology structure of the optimal reflection path sequence is obtained, a secondary unconstrained binary optimization model is constructed and quantum annealing solution is carried out to obtain discrete phase distribution, specifically, an activation state variable and a phase adjustment variable of a programmable ultra-surface hardware unit are determined according to the topology structure of the optimal reflection path sequence, the activation state variable and the phase adjustment variable are used as input, a secondary unconstrained binary optimization model is constructed based on preset activation state constraint conditions and phase discretization constraint conditions and maximum signal intensity is used as an optimization target, the secondary unconstrained binary optimization model is mapped into an element Xin Moxing of a quantum annealing machine, and the discrete phase distribution is obtained by adopting a quantum annealing algorithm to solve.
In a preferred embodiment, the method comprises the steps of calculating a quantization error characteristic value of discrete phase distribution, carrying out frequency domain compensation on the quantization error characteristic value by adopting an interpolation algorithm to obtain a phase compensation vector, specifically, calculating a difference value between the discrete phase distribution and a preset continuous phase distribution, calculating the quantization error value of each programmable ultra-surface hardware unit based on the difference value, carrying out Fourier transform on the quantization error value, extracting frequency domain characteristic components to obtain the quantization error characteristic vector, carrying out spectrum analysis on the error characteristic vector in a frequency domain, adopting an adaptive fluctuation detection method to identify a missing frequency band, carrying out dynamic division on the frequency band of the quantization error characteristic vector by adopting a fuzzy clustering algorithm to obtain a plurality of first frequency bands, calculating the average value of the quantization error characteristic vector of each first frequency band and the ratio of standard deviation, dynamically adjusting the node interval density of spline interpolation for three times according to the ratio, carrying out adaptive interpolation compensation on the missing frequency band according to the adjusted node interval density, and inversely transforming the frequency domain compensation result after interpolation to the time domain to generate the phase compensation vector.
In a preferred embodiment, the phase compensation vector is loaded to the programmable subsurface hardware unit to obtain a first reflection link, and a phase error covariance matrix is constructed, specifically, the phase response parameters of the programmable subsurface hardware unit are adjusted according to the phase compensation vector to obtain a first reflection link, a phase error sample set is constructed by measuring the deviation value of the actual phase response and the preset first phase response through the first reflection link, and the phase error covariance matrix is obtained by calculating the phase error covariance value among each programmable subsurface hardware unit based on the phase error sample set.
In a preferred embodiment, the incremental correction is performed on the weight parameters of the secondary unconstrained binary optimization model based on the phase error covariance matrix, specifically, the phase error covariance matrix is used as a regularization term to be input into the secondary unconstrained binary optimization model, and the weight parameters of the secondary unconstrained binary optimization model are iteratively updated by adopting a gradient descent method of a self-adaptive step length until the error converges, so as to obtain corrected weight parameters.
In a preferred embodiment, the objective function is constructed based on the corrected weight parameter and the phase compensation vector, and channel capacity reconstruction is performed through a gradient projection method to obtain an optimal channel capacity, specifically, a multi-objective optimization function is constructed by taking the corrected weight parameter and the phase compensation vector as inputs, iterative optimization is performed on the multi-objective optimization function through the gradient projection method, the maximum channel capacity meeting preset power constraint is calculated, a first channel capacity value is initialized, the gradient direction of the multi-objective optimization function under the current parameter is calculated, the gradient direction is projected to a feasible solution space, the first channel capacity value is dynamically updated, and iterative iteration is repeated until the increment of the first channel capacity value is smaller than a preset first threshold value to obtain the optimal channel capacity.
The beam control optimization method and the system for the millimeter wave network have the technical effects and advantages that:
1. The invention realizes accurate modeling and beam forming of complex channel states in a high-frequency communication environment by introducing multi-level and multi-stage joint designs such as a dynamic time warping algorithm, multi-hop reflection path modeling, quantum annealing solving, frequency domain compensation, gradient projection optimization and the like. Compared with the traditional method, the method can not only effectively match the optimal reflection path between the base station and the user and remarkably improve the signal coverage and stability, but also solve the problem of large-scale combination optimization in phase control by constructing a secondary unconstrained binary optimization model and combining quantum annealing solution. On the basis, the discrete phase quantization error is compensated by using a frequency domain interpolation algorithm, so that the phase control precision is obviously improved. The system further builds a phase error covariance matrix through actual measurement, and performs incremental correction on the weight parameters of the optimized model based on the phase error covariance matrix, so that the robustness of the model to hardware uncertainty is enhanced. Finally, the channel capacity reconstruction under the multi-objective constraint is realized by means of a gradient projection method, and the aim of improving the overall communication performance of the system is fulfilled. The invention ensures high calculation efficiency, simultaneously effectively combines precision, self-adaptability and engineering realization, is particularly suitable for the next generation millimeter wave communication system under ultra-dense deployment and dynamic change scenes, and has remarkable practical value and industrial application prospect.
Drawings
Fig. 1 is a flow chart of the beam control optimization method for millimeter wave networks according to the present invention.
Fig. 2 is a schematic structural diagram of a beam steering optimization system for a millimeter wave network according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Embodiment 1, fig. 1 shows a beam control optimization method for a millimeter wave network according to the present invention, which includes the following steps:
S1, matching multi-hop path delay characteristics between a base station and user equipment by adopting a DTW algorithm, and screening an optimal reflection path sequence;
in this example, a DTW algorithm is used to match the multi-hop path delay characteristics between the base station and the user equipment, and an optimal reflection path sequence is screened, which specifically includes:
Acquiring time delay sequence data of a plurality of candidate reflection paths between a base station and user equipment, and constructing a path time delay feature library;
Based on a preset reference time delay template, aligning time delay sequences of each candidate reflection path by adopting a DTW algorithm, and calculating the minimum accumulated path distance;
And screening the plurality of candidate reflection paths according to the minimum accumulated path distance to obtain an optimal reflection path sequence.
Illustratively, in a millimeter wave communication environment, there are multiple reflection paths between a Base Station (BS) and a User Equipment (UE). The system collects the time delay sequence on each candidate reflection path through a ranging module or a time of arrival (ToA) estimation technique. For example, there are three reflection paths:
The time delay sequences on the three candidate reflection paths of the path A are respectively 2.1ns, 2.3ns and 2.5ns, the time delay sequences on the four candidate reflection paths of the path B are respectively 1.9ns, 2.0ns, 2.4ns and 2.6ns, the time delay sequences on the three candidate reflection paths of the path C are respectively 2.0ns, 2.4ns and 2.7ns, and the time delay sequence data are constructed into a path time delay feature library. In addition, the system sets a set of reference delay templates T representing ideal path characteristics, which represent propagation delay characteristics under ideal reflection topology, specifically 2.0ns, 2.4ns and 2.7ns, and the templates can be obtained according to historical environmental data, manual setting or model deduction modes. And finally, dynamically aligning the delay sequence of the path A, B, C with a reference template by using a DTW algorithm, and calculating the minimum accumulated path distance. The results are DTW (a, T) =0.1, DTW (B, T) =0.2, DTW (C, T) =0.4, respectively, with path a having the smallest distance ordered according to the smallest DTW distance, and thus determined as the optimal reflection path. Optionally, a threshold value can be set or the first N paths are selected as the optimal path sequence, and the finally output optimal reflection path sequence is the path A.
S2, obtaining a topological structure of an optimal reflection path sequence, constructing a secondary unconstrained binary optimization model, and carrying out quantum annealing solution to obtain discrete phase distribution;
in the example, the topological structure of the optimal reflection path sequence is obtained, a secondary unconstrained binary optimization model is constructed, and quantum annealing solution is carried out to obtain discrete phase distribution, specifically:
determining an activation state variable and a phase adjustment variable of the programmable subsurface hardware unit according to the topological structure of the optimal reflection path sequence;
Taking an activation state variable and a phase adjustment variable as inputs, and constructing a secondary unconstrained binary optimization model by taking maximum signal strength as an optimization target based on preset activation state constraint conditions and phase discretization constraint conditions;
Mapping the model into the Xin Moxing of the quantum annealing machine, and solving by adopting a quantum annealing algorithm to obtain discrete phase distribution.
It should be noted that the topology may be represented as a directed graph, where each reflection point corresponds to a programmable subsurface hardware unit. Each cell is provided with two variables, wherein an active state variable indicates whether the cell is active or not and a phase adjustment variable indicates the adjusted discrete phase. The construction process of the secondary unconstrained binary optimization model aims at realizing signal enhancement in millimeter wave communication, and firstly, the state of each programmable ultra-surface hardware unit, namely whether to participate in reflection and a specific phase adjustment value thereof, is determined according to a selected optimal reflection path sequence. To adapt the discrete phase settings actually available to the hardware, each phase adjustment variable is discretized into a fixed set of angles and encoded by a set of mutually exclusive binary variables. On the basis, the model uses the maximum signal intensity of the receiving end as an optimization target, and converts the coupling relation and the phase difference influence among all reflection units into a quadratic objective function related to binary variables. The objective function does not contain any constraint condition, can be converted into a standard secondary unconstrained binary optimization model form through reasonable construction, and is suitable for efficient solution of discrete optimization methods such as quantum annealing and the like. The model can simultaneously consider the interaction among a plurality of units and the discrete regulation and control behaviors thereof, thereby providing an optimal strategy in a feasible solution space for the subsequent optimization process.
Further, the specific steps of the quantum annealing solution are as follows:
the initial state is set as a quantum superposition state;
gradually evolving the system in the simulated annealing path, wherein the Hamiltonian quantity evolves from the initial Hamiltonian quantity to a target Hamiltonian quantity;
The lowest energy state (i.e. the bit string of the optimal solution) is obtained after the final measurement;
the process finds the bit configuration that maximizes the objective function by minimizing the hamiltonian.
S3, calculating quantization error characteristic values of discrete phase distribution, and carrying out frequency domain compensation on the quantization error characteristic values by adopting an interpolation algorithm to obtain a phase compensation vector;
in this example, the quantization error eigenvalue of the discrete phase distribution is calculated, and the interpolation algorithm is adopted to perform frequency domain compensation on the quantization error eigenvalue, so as to obtain a phase compensation vector, which specifically includes:
Calculating the difference value of the discrete phase distribution and the preset continuous phase distribution, and calculating the quantization error value of each programmable subsurface hardware unit based on the difference value;
performing Fourier transform on the quantized error value, and extracting frequency domain feature components to obtain quantized error feature vectors;
carrying out spectrum analysis on the error feature vector in the frequency domain, and identifying a missing frequency band by adopting a self-adaptive fluctuation detection method;
dynamically dividing the frequency bands of the quantization error feature vectors by adopting a fuzzy clustering algorithm to obtain a plurality of first frequency bands;
Calculating the ratio of the quantization error feature vector mean value and the standard deviation of each first frequency band, and dynamically adjusting the node interval density of cubic spline interpolation according to the ratio;
and carrying out self-adaptive interpolation compensation on the missing frequency band according to the adjusted node interval density, and inversely transforming the frequency domain compensation result after interpolation to a time domain to generate a phase compensation vector.
In millimeter wave networks, in order to reduce the error caused between the discrete phase distribution and the ideal continuous phase distribution, it is necessary to perform frequency domain compensation processing on the phase quantization error. The method comprises the steps of firstly calculating the difference between the actual discrete phase value and the expected continuous value of each super-surface unit, and carrying out Fourier transformation on the error component vectors to obtain a frequency domain representation reflecting error frequency characteristics, namely error characteristic vectors. And then, identifying the missing frequency bands in the frequency domain signal by a self-adaptive fluctuation detection method, and dividing the frequency spectrum by fuzzy clustering to obtain a plurality of error frequency bands. And then, calculating the ratio of the mean value and the standard deviation of the error vector in each frequency band to reflect the stability or the fluctuation of the error distribution of the frequency band, dynamically adjusting the node interval density of cubic spline interpolation according to the stability or the fluctuation, and carrying out interpolation compensation on the missing frequency band. And finally, restoring the frequency domain compensation result to the time domain through inverse Fourier transformation to generate a phase compensation vector which can be used for phase control, thereby improving the beam precision and the overall performance of the system.
S4, loading a phase compensation vector to the programmable subsurface hardware unit to obtain a first reflection link, and constructing a phase error covariance matrix;
In this example, the phase compensation vector is loaded to the programmable subsurface hardware unit to obtain the first reflection link, and a phase error covariance matrix is constructed, specifically:
adjusting phase response parameters of the programmable ultra-surface hardware unit according to the phase compensation vector to obtain a first reflection link;
measuring a deviation value of an actual phase response and a preset first phase response through a first reflection link, and constructing a phase error sample set;
and calculating a phase error covariance value among each programmable subsurface hardware unit based on the phase error sample set to obtain a phase error covariance matrix.
It should be noted that after the phase compensation vector is obtained, the vector is loaded into the programmable subsurface hardware unit for adjusting the phase response of each unit, thereby forming the actual first reflection link. In order to evaluate the accuracy of the phase control of this link, the system compares the phase response of the actual reflected link with the theoretical preset response, calculates the phase deviation of each of the subsurface units, and composes the deviation data into a sample set. Based on the sample set, the phase error covariance between any two units is calculated by a statistical method, so that a complete phase error covariance matrix is constructed. The matrix not only reflects the spatial correlation distribution of errors, but also provides a data basis for accurate correction of weight parameters in a follow-up optimization model, and is beneficial to further improving the accuracy and robustness of beam control.
S5, performing incremental correction on weight parameters of the secondary unconstrained binary optimization model based on the phase error covariance matrix;
In this example, the incremental correction is performed on the weight parameters of the secondary unconstrained binary optimization model based on the phase error covariance matrix, specifically:
Inputting the phase error covariance matrix as a regularization term into a secondary unconstrained binary optimization model;
And iteratively updating the weight parameters of the secondary unconstrained binary optimization model by adopting a gradient descent method of the self-adaptive step length until the error converges to obtain corrected weight parameters.
It should be noted that after phase compensation is completed and the phase compensation is loaded onto the programmable super surface, the system obtains the actual phase response deviation of each unit through measurement of the first reflection link, and constructs a phase error sample set covering all units. After further statistical analysis, a phase error covariance matrix reflecting the error correlation between units is calculated, and the matrix is used as external feedback information of the current model.
In the original quadratic unconstrained binary optimization model, the coupling relationships between the individual subsurface elements are represented as a weight matrix to describe their joint impact. And the original weight setting can not accurately reflect the error propagation effect due to the fact that the actual hardware has discreteness and response offset. Therefore, a new model structure with regularization term is adopted, a phase error covariance matrix is used as a regularization factor to be embedded into an optimized model, an original objective function is corrected, the model is enabled to consider actual error association while optimizing a target, and robustness to noise and interference is enhanced.
In order to update the weight parameters efficiently, an adaptive step-size-based gradient descent method is adopted for iterative updating. The specific flow is as follows:
Initializing a weight matrix as parameters obtained after the last quantum annealing optimization;
Adding the covariance matrix as a regularization penalty term into an objective function to form a new loss function with constraint;
calculating the gradient direction of the current loss function to each weight;
Adjusting the weight value according to the gradient direction, introducing dynamic step control, and automatically adjusting the iteration amplitude of each step according to the descending speed of the loss function;
And continuing iteration until the decreasing amplitude of the loss function is smaller than a preset convergence threshold value, indicating that the model fully absorbs the error statistical characteristics, and outputting the finally corrected weight parameters.
S6, constructing an objective function based on the corrected weight parameters and the phase compensation vector, and reconstructing the channel capacity by a gradient projection method to obtain the optimal channel capacity.
In this example, an objective function is constructed based on the corrected weight parameter and the phase compensation vector, and channel capacity reconstruction is performed by a gradient projection method to obtain an optimal channel capacity, which specifically includes:
taking the corrected weight parameters and the phase compensation vector as inputs to construct a multi-objective optimization function;
performing iterative optimization on the multi-objective optimization function by using a gradient projection method, and calculating the maximum channel capacity meeting the preset power constraint;
initializing a first capacity value of a channel, and calculating the gradient direction of a multi-objective optimization function under the current parameters;
projecting the gradient direction to a feasible solution space, and dynamically updating a first capacity value of a channel;
and repeating the iteration until the increment of the first capacity value of the channel is smaller than a preset first threshold value, and obtaining the optimal channel capacity.
It should be noted that, the reconstruction process is based on two types of input, namely, an optimized model weight parameter obtained by incremental correction in the previous stage, which already contains the adaptation to the hardware error characteristic, and a phase compensation vector generated based on frequency domain compensation, which is used for reflecting the actual regulation and control capability of the current programmable ultra-surface hardware unit. The method aims at improving the system channel capacity, a multi-objective optimization function is constructed, and the following factors are comprehensively considered:
the method comprises the steps of maximizing beam gain, minimizing signal interference, smoothing phase jump of adjacent units, and meeting the condition that the transmitting power does not exceed a preset constraint value. The form of the objective function gives consideration to transmission performance and engineering constraint, and the whole is a non-convex optimization problem, but the structure has gradient conductivity.
Furthermore, for solving the objective function efficiently, a gradient projection method is adopted for iterative optimization, and the main steps are as follows:
Initializing a channel capacity value, and setting the channel capacity value as the theoretical capacity of the current system under the original configuration;
calculating the gradient direction of the multi-objective optimization function under the current phase compensation and weight combination;
projecting the gradient direction into a feasible solution space meeting the power and the physical adjustable range;
Updating the channel capacity estimation value according to the projection result, and recording the optimization step length;
if the increment between the current capacity value and the previous round is smaller than a set threshold value or the change of the optimization function tends to be stable, stopping iteration;
And finally outputting the phase configuration and the model weight corresponding to the maximum channel capacity for controlling the actual beam pointing. By the gradient projection optimization method, the system can continuously approach the optimal value of the channel capacity under the condition of multiple constraints, and can effectively avoid sinking into a local optimal solution. Meanwhile, the method fully utilizes the complementary advantages of the pre-optimization model and the frequency domain error compensation result, realizes the integrated processing from error correction to capacity enhancement, and greatly improves the transmission efficiency and the robustness of the millimeter wave communication link
Embodiment 2, fig. 2 shows a beam control optimization system for a millimeter wave network, which comprises a path matching module, an optimization solving module, a phase compensation module, a hardware control module, a parameter correction module and a capacity reconstruction module:
The path matching module is used for matching the multi-hop path delay characteristics between the base station and the user equipment by adopting a DTW algorithm and screening an optimal reflection path sequence;
the optimization solving module is used for obtaining the topological structure of the optimal reflection path sequence, constructing a secondary unconstrained binary optimization model and carrying out quantum annealing solving to obtain discrete phase distribution;
determining an activation state variable and a phase adjustment variable of the programmable subsurface hardware unit according to the topological structure of the optimal reflection path sequence;
Taking an activation state variable and a phase adjustment variable as inputs, and constructing a secondary unconstrained binary optimization model by taking maximum signal strength as an optimization target based on preset activation state constraint conditions and phase discretization constraint conditions;
Mapping the secondary unconstrained binary optimization model into an element Xin Moxing of a quantum annealing machine, and solving by adopting a quantum annealing algorithm to obtain discrete phase distribution;
the phase compensation module is used for calculating the quantization error characteristic value of discrete phase distribution, and carrying out frequency domain compensation on the quantization error characteristic value by adopting an interpolation algorithm to obtain a phase compensation vector;
Calculating the difference value of the discrete phase distribution and the preset continuous phase distribution, and calculating the quantization error value of each programmable subsurface hardware unit based on the difference value;
performing Fourier transform on the quantized error value, and extracting frequency domain feature components to obtain quantized error feature vectors;
carrying out spectrum analysis on the error feature vector in the frequency domain, and identifying a missing frequency band by adopting a self-adaptive fluctuation detection method;
dynamically dividing the frequency bands of the quantization error feature vectors by adopting a fuzzy clustering algorithm to obtain a plurality of first frequency bands;
Calculating the ratio of the quantization error feature vector mean value and the standard deviation of each first frequency band, and dynamically adjusting the node interval density of cubic spline interpolation according to the ratio;
performing adaptive interpolation compensation on the missing frequency band according to the adjusted node interval density, and inversely transforming the frequency domain compensation result after interpolation to a time domain to generate a phase compensation vector;
The hardware control module is used for loading the phase compensation vector to the programmable subsurface hardware unit to obtain a first reflection link and constructing a phase error covariance matrix;
The parameter correction module is used for carrying out incremental correction on the weight parameters of the secondary unconstrained binary optimization model based on the phase error covariance matrix;
And Rong Liangchong the construction module is used for constructing an objective function based on the corrected weight parameters and the phase compensation vector, and reconstructing the channel capacity by a gradient projection method to obtain the optimal channel capacity.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product.
Those of ordinary skill in the art will appreciate that the various illustrative modules and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist alone physically, or two or more modules may be integrated into one module.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Finally, the foregoing description of the preferred embodiment of the invention is provided for the purpose of illustration only, and is not intended to limit the invention to the particular form disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.
Claims (6)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510913118.1A CN120416863B (en) | 2025-07-03 | 2025-07-03 | Beam control optimization method and system for millimeter wave network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510913118.1A CN120416863B (en) | 2025-07-03 | 2025-07-03 | Beam control optimization method and system for millimeter wave network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120416863A CN120416863A (en) | 2025-08-01 |
| CN120416863B true CN120416863B (en) | 2025-09-30 |
Family
ID=96506745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510913118.1A Active CN120416863B (en) | 2025-07-03 | 2025-07-03 | Beam control optimization method and system for millimeter wave network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120416863B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113973305A (en) * | 2021-10-26 | 2022-01-25 | 西安电子科技大学 | Intelligent reflecting surface position and beam joint optimization method carried on unmanned aerial vehicle |
| CN119135219A (en) * | 2024-09-13 | 2024-12-13 | 浙江大学 | A distributed RIS fast beamforming method based on quantum annealing |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024246710A1 (en) * | 2023-05-26 | 2024-12-05 | Latys Intelligence Inc. | Method of and system for pre-deployment and post-deployment optimization of reconfigurable intelligent surfaces |
| US20250005856A1 (en) * | 2023-06-28 | 2025-01-02 | Microsoft Technology Licensing, Llc | Metasurface design and placement |
-
2025
- 2025-07-03 CN CN202510913118.1A patent/CN120416863B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113973305A (en) * | 2021-10-26 | 2022-01-25 | 西安电子科技大学 | Intelligent reflecting surface position and beam joint optimization method carried on unmanned aerial vehicle |
| CN119135219A (en) * | 2024-09-13 | 2024-12-13 | 浙江大学 | A distributed RIS fast beamforming method based on quantum annealing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120416863A (en) | 2025-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113128119B (en) | Filter Reverse Design and Optimization Method Based on Deep Learning | |
| CN113392961B (en) | Method for extracting mesoscale eddy track stable sequence and predicting cyclic neural network | |
| CN111062512B (en) | Wind power prediction method | |
| CN109299430A (en) | Short-term wind speed prediction method based on two-stage decomposition and extreme learning machine | |
| CN112512069B (en) | Network intelligent optimization method and device based on channel beam pattern | |
| CN110601777A (en) | Method for estimating satellite-ground downlink co-channel interference under low-orbit mobile satellite constellation | |
| Mom et al. | Application of artificial neural network for path loss prediction in urban macrocellular environment | |
| CN115243382B (en) | Fair resource allocation method for cellular network users based on RIS-assisted edge computing | |
| CN119109527B (en) | Antenna polarization simulation method based on phase and amplitude calibration | |
| CN119047226B (en) | FPGA-based complex electromagnetic environment simulation method | |
| CN109560849B (en) | A Robust Adaptive Algorithm for Beamforming Based on Second-Order Cone Planning | |
| CN117354652B (en) | High-speed multimode optical module performance optimization method based on variation self-encoder | |
| CN114826462A (en) | Beam domain channel amplification method for large-scale MIMO statistical port selection | |
| CN114169251A (en) | An ultra-short-term wind power prediction method | |
| CN111105005A (en) | A wind power prediction method | |
| CN109753763B (en) | Wind-solar combined output probability modeling method | |
| CN118118069B (en) | Robust self-adaptive beam forming method based on deep expansion network | |
| CN120416863B (en) | Beam control optimization method and system for millimeter wave network | |
| CN118869023A (en) | A space-time joint extrapolation method for 6G near-field non-stationary channels based on deep learning | |
| CN113076699B (en) | Antenna optimization method based on Bayesian optimization of multi-output Gaussian process | |
| Nouir et al. | Comparison of neural network learning algorithms for prediction enhancement of a planning tool | |
| CN114117917A (en) | A Multi-objective Optimization Modeling Method for Ship Magnetic Dipole Array | |
| WO2024113313A1 (en) | Generating digital predistortion models using performance metrics | |
| CN113592565A (en) | Method for predicting importance of nodes of power system | |
| CN119363652B (en) | Time delay optimization method for ship fusion communication |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |