Layered beam training method and apparatus based on channel map assistance

Abstract

This invention relates to the field of large-scale multiple-input multiple-output (MIMO) communication technology, and particularly to a hierarchical beamforming training method and apparatus based on channel map assistance. The method includes: constructing a channel map guided by beamforming codebooks, wherein the channel map records the equivalent channel gain corresponding to each grid point within a preset area and each beamforming codeword; acquiring partial location information of the user equipment, including multiple mutually exclusive spatial sub-regions where the user is located and the prior probabilities of each sub-region; calculating the beam potential energy of all beamforming codewords based on the channel map and partial location information; pruning the complete binary search tree based on the beam potential energy of each beamforming codeword; and traversing the simplified search tree according to a hierarchical beamforming training strategy to obtain near-optimal beamforming codewords. This solves the problems of unknown precise user location before beamforming training, unclear coordination mechanism between channel map and observation information, and difficulty in balancing training overhead and computational complexity.

Description

Technical Field

[0001] This invention relates to the field of large-scale multiple-input multiple-output communication technology, and in particular to a hierarchical beam training method and apparatus based on channel map assistance. Background Technology

[0002] In 6G communication scenarios such as millimeter wave and terahertz, massive MIMO (Multiple-Input Multiple-Output) technology achieves highly directional transmission by configuring a large number of antenna arrays, significantly improving communication capacity and spectral efficiency. Beamforming, as a key means of acquiring channel state information, is a prerequisite for user equipment access and reliable communication. However, traditional beamforming methods face two major challenges: first, in massive antenna array scenarios, traversing beams for searching requires exponentially increasing pilot overhead, leading to excessively high training costs and time consumption; second, the precise location of user equipment is usually unknown before beamforming, and the effectiveness of location-based channel prior information heavily depends on location accuracy, with location uncertainty significantly reducing beamforming performance.

[0003] Channel maps, as an emerging carrier of prior channel information, can provide location-related spatial / channel information without consuming additional wireless resources, offering a new approach to solving the aforementioned problems. While channel maps have been applied to scenarios such as environment-aware hybrid beamforming, their application in beam training still faces significant limitations: most research focuses on scalar path loss maps, making them difficult to adapt to multi-DOF, multi-antenna systems; existing channel map-assisted beam training schemes lack interpretability and policy control, and have not effectively addressed the coordination mechanism between location uncertainty and channel map / observation information; furthermore, the balance between training overhead and computational complexity is insufficient, making it difficult to meet practical deployment requirements.

[0004] Therefore, there is an urgent need for a beam training technology that can integrate partial location information with channel maps to achieve low overhead, high accuracy and low complexity in scenarios with unknown locations, so as to promote the practical application of channel maps in large-scale MIMO beam training. Summary of the Invention

[0005] This invention provides a hierarchical beam training method and apparatus based on channel map assistance to solve problems such as unknown user location before beam training, unclear coordination mechanism between channel map and observation information, and difficulty in balancing training overhead and computational complexity.

[0006] A first aspect of this invention provides a channel map-assisted hierarchical beam training method, comprising the following steps: Construct a beamforming codebook-guided channel map, wherein the channel map records the equivalent channel gain corresponding to each grid point and each beamforming codeword within a preset area; Obtain partial location information of the user device, wherein the partial location information includes multiple mutually exclusive spatial sub-regions where the user is located and the prior probability of each sub-region; Calculate the beam potential energy of all beam codes based on the channel map and the partial location information; The pre-constructed complete binary search tree is pruned according to the beam potential energy of each beam codeword to obtain a simplified search tree; The simplified search tree is traversed according to the hierarchical beam training strategy to obtain near-optimal beam codes.

[0007] Optionally, the construction of the beamforming codebook-guided channel map includes: The preset communication area is quantized into a set of grid points according to preset horizontal and vertical spacing to generate an orthogonal beamforming codebook based on discrete Fourier transform. The equivalent channel gain of each grid point and each beamcode in the orthogonal beamforming codebook is calculated using ray tracing technology to construct a channel map adapted to multiple-input multiple-output communication systems.

[0008] Optionally, calculating the beam potential of all beamcodes based on the channel map and the partial location information includes: The grid-level codeword weights of the grid points in each sub-region are determined based on the equivalent channel gain and the prior probability of each sub-region. A gain threshold and indicator function are introduced to filter the effective weights in the equivalent channel gain; The beam potential energy of the bottom-level beam codeword is calculated based on the effective weights and the grid-level codeword weights of each sub-region grid point. Based on a hierarchical recursive approach, the beam potential energy of each beam codeword is calculated according to the beam potential energy of the lowest-level beam codeword.

[0009] Optionally, the step of pruning the pre-constructed complete binary search tree according to the beam potential energy of each beam codeword to obtain a simplified search tree includes: The complete binary search tree is traversed using a preset potential energy screening threshold, and compared with the beam potential energy of each beam codeword. If the beam potential energy of any beam codeword is less than the preset potential energy screening threshold, its corresponding node and subtree are removed. If the beam potential energy of any beam codeword is greater than or equal to the preset potential energy screening threshold, its corresponding node and subtree are retained to obtain the simplified search tree. The complete binary search tree is composed of the hierarchical beam training levels and corresponding codewords.

[0010] Optionally, the hierarchical beam training strategy is a training strategy based on maximizing beam potential reward or a low-complexity two-layer look-ahead training strategy.

[0011] Optionally, traversing the simplified search tree according to the hierarchical beam training strategy to obtain near-optimal beam codes includes: Based on the hierarchical beam training strategy, the root node of the simplified search tree is initialized to the top-level codeword corresponding to omnidirectional transmission, the initial value of the search iteration counter is -1, and the initial value of the initial search layer is 0. Extract the two-layer lookahead subtrees corresponding to the current root node to identify the subtree topology as any one of full tree structure, asymmetric binary structure, and single-chain structure, and adaptively select the near-optimal beamcode of the subtree topology.

[0012] Optionally, the adaptive selection of near-optimal beamcodes for the subtree topology includes: If the subtree topology is a full tree structure, the search level in the next time step is the current search level plus 1; If the subtree topology is an asymmetric binary structure, calculate the layering and traversal search overhead after beam potential weighting. If the layering search overhead is small, the next search layer is the current search layer plus 1; otherwise, it is the current search layer plus 2. If the subtree topology is a single-chain structure, the next search layer is the current search layer plus 2.

[0013] A second aspect of the present invention provides a channel map-assisted hierarchical beam training device, comprising: A base station module, which is configured with a multi-antenna array, is used to transmit beam signals and receive feedback commands; The user equipment module is used to receive the beam signal sent by the base station module and feed back the optimal beam codeword index. A channel map construction module is used to construct a beamforming codebook-guided channel map, wherein the channel map records the equivalent channel gain corresponding to each grid point and each beamforming codeword within a preset area. The user location acquisition module is used to acquire partial location information of the user device, wherein the partial location information includes multiple mutually exclusive spatial sub-regions where the user is located and the prior probability of each sub-region; The potential energy calculation module is used to calculate the beam potential energy of all beam codes based on the channel map and the partial location information. The pruning module is used to prune the pre-built complete binary search tree according to the beam potential energy of each beam codeword to obtain a simplified search tree; The traversal module is used to traverse the simplified search tree according to the hierarchical beam training strategy to obtain near-optimal beam codes.

[0014] A third aspect of the present invention provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the channel map-assisted hierarchical beam training method as described in the above embodiments.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described channel map-assisted hierarchical beam training method.

[0016] The hierarchical beam training method and apparatus based on channel map assistance proposed in this invention can integrate user location information with the prior value of the channel map without the need for high-precision positioning. This reduces the invalid search space, significantly reduces training overhead and computational complexity, and improves the beam training accuracy and communication spectrum efficiency of large-scale MIMO systems. It is applicable to 6G communication scenarios such as millimeter wave and terahertz.

[0017] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating a channel map-assisted hierarchical beam training method according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the execution of a channel map-assisted hierarchical beam training method according to an embodiment of the present invention. Figure 3 This is a schematic diagram of a multi-layer channel map model for multi-beam codebooks provided according to an embodiment of the present invention; Figure 4 This is a simulation diagram illustrating the comparison of the improvement trend of beam training gain with the beam training process (i.e., under different training overheads) according to an embodiment of the present invention. Figure 5 This is a simulation diagram showing the performance comparison of the final transmission rate under different beam training methods with different signal-to-noise ratios according to an embodiment of the present invention. Figure 6 A block diagram of a channel map-assisted hierarchical beam training device according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention.

[0019] Explanation of reference numerals in the attached figures: 60-A hierarchical beam training device based on channel map assistance; 601-Base station module; 602-User equipment module; 603-Channel map construction module; 604-User location acquisition module; 605-Potential energy calculation module; 606-Pruning module; 607-Traversal module; 701-Memory; 702-Processor; 703-Communication interface. Detailed Implementation

[0020] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0021] The following description, with reference to the accompanying drawings, illustrates a hierarchical beam training method and apparatus based on a channel map. As mentioned in the background section, in current 6G communication scenarios such as millimeter wave and terahertz, beam training is a crucial step in acquiring channel state information and a prerequisite for reliable user equipment access. However, as the scale of antenna arrays continues to expand, traditional beam training schemes face severe challenges: on the one hand, traversing beam search requires exponentially increasing pilot overhead, resulting in excessively high training latency and difficulty in adapting to high real-time communication requirements; on the other hand, the precise location of user equipment is usually unknown before beam training, and existing channel map-based auxiliary schemes mostly rely on high-precision location information. Location uncertainty will significantly reduce beam training performance, and the coordination mechanism between channel map and real-time observation information is unclear. At the same time, the balance design between training overhead and computational complexity is insufficient, making it difficult to meet the actual deployment requirements. This invention provides a hierarchical beam training method based on channel map assistance. In this method, based on the channel map guided by the beamforming codebook, the user's partial location information and probability distribution are first fused to define the beam potential and establish a simplified search tree; then, a low-complexity two-layer look-ahead hierarchical beam training scheme is proposed, and the search level is dynamically adjusted by analyzing the local tree structure to balance performance and complexity. This solves the problems of exponential growth in pilot overhead, high training latency, and susceptibility of beam training performance to user location uncertainty in traditional beam training schemes under large-scale MIMO communication scenarios. It also addresses the issues of unclear coordination mechanism between channel map and observation information, and difficulty in balancing computational complexity and training accuracy.

[0022] Specifically, Figure 1This is a schematic flowchart of a channel map-assisted hierarchical beam training method provided in an embodiment of the present invention.

[0023] like Figure 1 As shown, the channel map-assisted hierarchical beam training method includes the following steps: In step S101, a beamforming codebook-guided channel map is constructed, wherein the channel map records the equivalent channel gain corresponding to each grid point in a preset area and each beamforming codeword.

[0024] In some embodiments, constructing a beamforming codebook-guided channel map includes: The preset communication area is quantized into a set of grid points according to preset horizontal and vertical spacing to generate an orthogonal beamforming codebook based on discrete Fourier transform. The equivalent channel gain of each grid point and each beamcode in the orthogonal beamforming codebook is calculated using ray tracing technology to construct a channel map adapted to multiple-input multiple-output communication systems.

[0025] In actual implementation, the spatial range of the preset communication area is set as follows: (Unit: m), along the horizontal direction at intervals (Unit: m), vertical direction with spacing (Unit: m) Perform uniform quantization to obtain a set of grid points. .in, , These represent the number of grid points in the horizontal and vertical directions, respectively. This indicates the rounding up operation. For the first Line number The geographic coordinates of the grid points satisfy , ( (The starting coordinates of the lower left corner of the region).

[0026] Based on grid point set Set base station configuration The antennas employ a uniform linear array (ULA) layout with a specific antenna spacing. ( The wavelength of electromagnetic waves, At the speed of light, (for carrier frequency) to generate 3D Discrete Fourier Transform (DFT) Matrix The first of the matrix The column vector is the first one. Beamforming code The codebook set is The code satisfies orthogonality. ,in, This indicates the conjugate transpose. Kronecker function ( hour Otherwise, it is 0).

[0027] The equivalent channel gain of each grid point and each beamcode is calculated using ray tracing technology to construct a channel map architecture model adapted to multi-degree-of-freedom multi-antenna systems. The channel map architecture model stores the three-dimensional mapping relationship on the base station side to support fast lookup of equivalent channel gain by grid points and codewords.

[0028] In step S102, partial location information of the user equipment is obtained, wherein the partial location information includes multiple mutually exclusive spatial sub-regions where the user is located and the prior probability of each sub-region.

[0029] In actual implementation, based on the geographical characteristics of the communication area (such as building boundaries and road distribution), the potential user area is divided into... Non-overlapping spatial sub-regions ,satisfy ( Each sub-region Defined by geographic coordinate boundaries, containing Grid points, i.e. ,in, sub-region The first Grid points. Statistical modeling methods are used to calculate sub-regions. Prior probability .

[0030] In step S103, the beam potential of all beam codes is calculated based on the channel map and partial location information.

[0031] In some embodiments, the beam potential of all beamcodes is calculated based on the channel map and partial location information, including: The grid-level codeword weights of grid points in each sub-region are determined based on the equivalent channel gain and the prior probability of each sub-region. Introduce a gain threshold and indicator function to filter the effective weights in the equivalent channel gain; The beam potential energy of the bottom-level beam codeword is calculated based on the effective weights and the grid-level codeword weights of each sub-region grid point. Based on a hierarchical recursive approach, the beam potential energy of each beam codeword is calculated according to the beam potential energy of the lowest-level beam codeword.

[0032] In actual implementation, for each sub-region Grid points within sub-regions and beamcode The definition of grid-level codeword weights is a comprehensive representation of the probability of a user equipment at a grid point and the communication gain of the corresponding beam. ,in, The spatial distribution probability of grid points. This represents the equivalent channel gain in the channel map.

[0033] Simultaneously, the beam potential of the bottom-level narrow-beam codeword is calculated, assuming the total number of layers in the layered beam training is... The lowest-level beamcode is the narrow beamcode, corresponding to the complete codebook. Introducing a gain threshold (Based on the sparsity characteristics of the channel, the maximum equivalent channel gain is usually taken.) times, ), through indicator functions Effective weights for selecting high-gain grid points ( like (otherwise it is 0), thus the lowest level beamcode The potential energy formula is:

[0034] in, ; Secondly, based on the hierarchical recursive approach, considering the recursive calculation of the upper-layer beam potential, for the th... Layered beamcode Its potential energy is obtained by summing the potentials of the two sub-beam codewords in the next layer, to ensure the accuracy and efficiency of the potential energy calculation. The recursive formula is as follows:

[0035] in, , The first Layer , The power weight of each codeword.

[0036] In step S104, the pre-constructed complete binary search tree is pruned according to the beam potential energy of each beam codeword to obtain a simplified search tree.

[0037] In some embodiments, the pre-constructed full binary search tree is pruned according to the beam potential energy of each beam codeword to obtain a simplified search tree, including: The complete binary search tree is traversed using a preset potential energy screening threshold, and the beam potential energy of each beam codeword is compared with that of the beam codeword. If the beam potential energy of any beam codeword is less than the preset potential energy screening threshold, its corresponding node and subtree are removed. If the beam potential energy of any beam codeword is greater than or equal to the preset potential energy screening threshold, its corresponding node and subtree are retained to obtain a simplified search tree. The complete binary search tree consists of the hierarchical beam training levels and corresponding codewords.

[0038] In actual implementation, for the complete binary search tree of beamcodes composed of the hierarchical beam training levels and corresponding codewords, a potential energy screening threshold is set. Traverse the complete binary search tree and remove all trees whose potential energy satisfies... The nodes and their corresponding subtrees retain potential energy. The nodes are reconstructed according to their original hierarchical relationships to form a simplified binary search tree. The number of nodes in this tree is much smaller than that in a complete tree, which significantly reduces the search range for subsequent training.

[0039] In step S105, the simplified search tree is traversed according to the hierarchical beam training strategy to obtain the near-optimal beam codeword.

[0040] In some embodiments, the hierarchical beam training strategy is a training strategy based on maximizing beam potential reward or a low-complexity two-layer look-ahead training strategy.

[0041] In some embodiments, traversing the simplified search tree according to a hierarchical beam training strategy to obtain near-optimal beam codes includes: Based on the hierarchical beam training strategy, the root node of the simplified search tree is initialized to the top-level codeword corresponding to omnidirectional transmission, the initial value of the search iteration counter is -1, and the initial value of the initial search layer is 0. Extract the two-layer lookahead subtrees corresponding to the current root node to identify the subtree topology as a full tree structure, an asymmetric binary structure, or a single-chain structure, and adaptively select the near-optimal beamcode of the subtree topology.

[0042] In actual implementation, based on the hierarchical beam training strategy, the root node of the simplified search tree is initialized as the top-level codeword corresponding to omnidirectional transmission. Iteration counter Initial search layer . No. In the round of iteration, starting with the current root node Based on this, two layers of look-ahead subtrees are extracted, and the subtree topology is identified as a full tree structure, an asymmetric binary structure, or a single-chain structure. The near-optimal beamcode of the subtree topology is then adaptively selected.

[0043] If the subtree topology is a full tree structure, the search level at the next time step is the current search level plus 1; if the subtree topology is an asymmetric binary structure, calculate the layering and traversal search overhead after beam potential weighting. If the layering search overhead is small, the next search level is the current search level plus 1, otherwise it is the current search level plus 2; if the subtree topology is a single-chain structure, the next search level is the current search level plus 2.

[0044] Furthermore, after receiving the beamcode from the base station, the user equipment feeds back the index of the codeword with the highest received power. This codeword is then updated as the new root node, and the simplified search tree is also updated. (Iteration counter) Repeat the subtree identification, search layer decision, and feedback update steps until... Output near-optimal beamcode.

[0045] It should be noted that the base station in this embodiment of the invention is configured with a multi-antenna array for transmitting beam signals and receiving feedback instructions, while the user equipment is a single-antenna device, and some of its location information is derived from historical movement trajectories, spatial constraints, or sensor data.

[0046] It is understandable that the beam training optimization problem is a multi-objective optimization problem coupling channel prior information and real-time observation, making it difficult to provide a closed-form solution analytically that allows the system to simultaneously achieve the lowest training overhead and the highest beam matching accuracy. This embodiment of the invention uses channel map prior modeling, beam potential energy quantitative assessment, simplified search tree pruning, and hierarchical training strategy iterative search to jointly complete the selection and determination of near-optimal beam codes. It should be noted that this embodiment of the invention does not have strict requirements on the type of base station antenna array (such as uniform linear array, uniform planar array) or the moving speed of user equipment. When the number of base station antennas increases or the number of user accesses increases, the levels and number of nodes of the simplified search tree are adjusted accordingly, but the pruning mechanism based on beam potential energy and the hierarchical training strategy can always achieve efficient beam training through dimensionality reduction.

[0047] Finally, as Figure 4 and 5 As shown, the embodiments of the present invention can achieve better beam training results than related technologies under the same simulation conditions. The simulation parameters are set as follows: number of base station antennas... (Uniform linear array deployment); Preset communication area range ; Mesh quantization spacing Number of user equipment sub-regions The prior probabilities of the sub-regions are respectively , , carrier frequency Signal-to-noise ratio Beam potential energy screening threshold .like Figure 4 As shown, under different search step conditions, compared with the traditional hierarchical search method without channel map assistance and the beam recommendation method based on deep learning, the channel map-assisted hierarchical beam training scheme proposed in this embodiment of the invention can achieve higher training gain and lower training overhead, with training overhead reduced by about 40% and training gain increased by about 35%. Meanwhile, as... Figure 5 As shown, under different signal-to-noise ratio constraints, this scheme achieves superior communication spectral efficiency regardless of the number of users accessing the network. Specifically, on the one hand, compared with training without channel map assistance, the spectral efficiency of channel map-assisted training can be significantly improved by at least 1.2 bps / Hz; on the other hand, compared with the beam selection method that directly calls map gain, the two-layer look-ahead strategy proposed in this embodiment of the invention can further improve the spectral efficiency by about 0.8 bps / Hz. Therefore, it can be concluded that, based on the relevant methods and apparatus of this application, the base station can significantly save pilot resources and computing resources without requiring high-precision positioning of user equipment, achieving low-overhead, high-precision beam training and ensuring reliable connection in 6G communication scenarios.

[0048] The hierarchical beam training method based on channel map assistance proposed in this invention, without requiring high-precision positioning of user equipment or strict adaptation to the multi-degree-of-freedom characteristics of multi-antenna systems, constructs a beamforming codebook-guided channel map to mine location-related channel prior information, calculates beam potential energy by fusing partial location probability distributions and prunes the search tree, and combines two layers of look-ahead local subtree topology recognition and dynamic search hierarchy decision-making strategies to achieve efficient beam traversal in a simplified search space. This maximizes the improvement of beam training accuracy and efficiency, ensures reliable communication in large-scale MIMO systems, and effectively reduces pilot overhead and computational complexity. It is particularly suitable for 6G full-spectrum communication scenarios such as millimeter wave and terahertz, and can adaptively adjust the training strategy for different channel conditions and user mobility states, exhibiting excellent location robustness. It can also effectively solve the technical problems of high overhead, large latency, poor location adaptability, and insufficient collaborative application of channel maps in traditional beam training schemes in large-scale MIMO communication scenarios. This is of great significance for promoting the engineering implementation of channel map technology and the practical development of large-scale MIMO systems.

[0049] Next, referring to the accompanying drawings, a channel map-assisted hierarchical beam training device based on an embodiment of the present invention is described.

[0050] Figure 6 This is a block diagram of a channel map-assisted hierarchical beam training device according to an embodiment of the present invention.

[0051] like Figure 6As shown, the channel map-assisted hierarchical beam training device 60 includes: a base station module 601, a user equipment module 602, a channel map construction module 603, a user location acquisition module 604, a potential energy calculation module 605, a pruning module 606, and a traversal module 607.

[0052] The system includes a base station module 601 configured with a multi-antenna array for transmitting beam signals and receiving feedback commands. A user equipment module 602 receives the beam signals transmitted by the base station module and provides feedback on the optimal beam codeword index. A channel map construction module 603 constructs a beamforming codebook-guided channel map, where the channel map records the equivalent channel gain corresponding to each grid point within a preset area and each beamforming codeword. A user location acquisition module 604 acquires partial location information of the user equipment, including multiple mutually exclusive spatial sub-regions where the user is located and the prior probabilities of each sub-region. A potential energy calculation module 605 calculates the beam potential energy of all beam codeswords based on the channel map and partial location information. A pruning module 606 prunes the pre-constructed complete binary search tree based on the beam potential energy of each beam codeword to obtain a simplified search tree. A traversal module 607 traverses the simplified search tree according to a hierarchical beam training strategy to obtain near-optimal beam codeswords.

[0053] In some embodiments, the channel map construction module 603 includes: The generation unit is used to quantize the preset communication area into a set of grid points according to the preset horizontal and vertical spacing, so as to generate an orthogonal beamforming codebook based on discrete Fourier transform. The building unit is used to calculate the equivalent channel gain of each grid point and each beamcode in the orthogonal beamforming codebook using ray tracing technology, so as to build a channel map adapted to multiple input multiple output communication systems.

[0054] In some embodiments, the potential energy calculation module 605 includes: The determination unit is used to determine the grid-level codeword weights of grid points in each sub-region based on the equivalent channel gain and the prior probability of each sub-region. A filtering unit is used to introduce a gain threshold and an indicator function to filter the effective weights in the equivalent channel gain; The first calculation unit is used to calculate the beam potential energy of the bottom-level beam codeword based on the effective weight and the grid-level codeword weight of each sub-region grid point; The second calculation unit is used to calculate the beam potential energy of each beam codeword based on the beam potential energy of the lowest-level beam codeword in a hierarchical recursive manner.

[0055] In some embodiments, the trimming module 606 includes: The complete binary search tree is traversed using a preset potential energy screening threshold, and the beam potential energy of each beam codeword is compared with that of the beam codeword. If the beam potential energy of any beam codeword is less than the preset potential energy screening threshold, its corresponding node and subtree are removed. If the beam potential energy of any beam codeword is greater than or equal to the preset potential energy screening threshold, its corresponding node and subtree are retained to obtain a simplified search tree. The complete binary search tree consists of the hierarchical beam training levels and corresponding codewords.

[0056] In some embodiments, the hierarchical beam training strategy is a training strategy based on maximizing beam potential reward or a low-complexity two-layer look-ahead training strategy.

[0057] In some embodiments, the traversal module 607 includes: The search iteration unit is used to initialize the root node of the simplified search tree as the top-level codeword corresponding to omnidirectional transmission based on the hierarchical beam training strategy. The initial value of the search iteration counter is -1, and the initial value of the initial search layer is 0. The identification and selection unit is used to extract the two-layer look-ahead subtrees corresponding to the current root node, so as to identify the subtree topology as any one of the full tree structure, asymmetric binary structure and single-chain structure, and adaptively select the near-optimal beamcode of the subtree topology.

[0058] In some embodiments, the identification and selection unit includes: If the subtree topology is a full tree structure, the search level in the next time step is the current search level plus 1; If the subtree topology is an asymmetric binary structure, calculate the layering and traversal search overhead after beam potential weighting. If the layering search overhead is small, the next search layer is the current search layer plus 1; otherwise, it is the current search layer plus 2. If the subtree topology is a single-linked structure, the next search level is the current search level plus 2.

[0059] It should be noted that the foregoing explanation of the embodiment of the channel map-assisted hierarchical beam training method also applies to the channel map-assisted hierarchical beam training device of this embodiment, and will not be repeated here.

[0060] The hierarchical beam training device based on channel map assistance proposed in this embodiment of the invention, without requiring high-precision positioning of user equipment, reduces pilot overhead and computational cost in beam training as much as possible through the coordinated operation of storing and updating multi-beam channel maps on the base station side, beam potential energy calculation, search tree construction and training strategy execution, combined with the signal reception and feedback mechanism on the user equipment side, thereby ensuring high-spectral-efficiency transmission and high-precision beam matching of the communication system.

[0061] Figure 7This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention.

[0062] The electronic device may include a memory 701, a processor 702, and a computer program stored in the memory 701 and executable on the processor 702. When the processor 702 executes the program, it implements the channel map-assisted hierarchical beam training method provided in the above embodiments.

[0063] Furthermore, electronic devices also include: Communication interface 703 is used for communication between memory 701 and processor 702.

[0064] The memory 701 is used to store computer programs that can run on the processor 702.

[0065] The memory 701 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0066] If the memory 701, processor 702, and communication interface 703 are implemented independently, then the communication interface 703, memory 701, and processor 702 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0067] Optionally, in a specific implementation, if the memory 701, processor 702, and communication interface 703 are integrated on a single chip, then the memory 701, processor 702, and communication interface 703 can communicate with each other through an internal interface.

[0068] The processor 702 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0069] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described channel map-assisted hierarchical beam training method.

[0070] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0071] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0072] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0073] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0074] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0075] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0076] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0077] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A hierarchical beam training method based on channel map assistance, characterized in that, For use in multiple-input multiple-output communication systems, the following steps are included: Construct a beamforming codebook-guided channel map, wherein the channel map records the equivalent channel gain corresponding to each grid point and each beamforming codeword within a preset area; Obtain partial location information of the user device, wherein the partial location information includes multiple mutually exclusive spatial sub-regions where the user is located and the prior probability of each sub-region; Calculate the beam potential energy of all beam codes based on the channel map and the partial location information; The pre-constructed complete binary search tree is pruned according to the beam potential energy of each beam codeword to obtain a simplified search tree; The simplified search tree is traversed according to the hierarchical beam training strategy to obtain near-optimal beam codes.

2. The hierarchical beam training method based on channel map assistance according to claim 1, characterized in that, The construction of the beamforming codebook-guided channel map includes: The preset communication area is quantized into a set of grid points according to preset horizontal and vertical spacing to generate an orthogonal beamforming codebook based on discrete Fourier transform. The equivalent channel gain of each grid point and each beamcode in the orthogonal beamforming codebook is calculated using ray tracing technology to construct a channel map adapted to multiple-input multiple-output communication systems.

3. The hierarchical beam training method based on channel map assistance according to claim 1, characterized in that, The step of calculating the beam potential energy of all beamcodes based on the channel map and the partial location information includes: The grid-level codeword weights of the grid points in each sub-region are determined based on the equivalent channel gain and the prior probability of each sub-region. A gain threshold and indicator function are introduced to filter the effective weights in the equivalent channel gain; The beam potential energy of the bottom-level beam codeword is calculated based on the effective weights and the grid-level codeword weights of each sub-region grid point. Based on a hierarchical recursive approach, the beam potential energy of each beam codeword is calculated according to the beam potential energy of the lowest-level beam codeword.

4. The hierarchical beam training method based on channel map assistance according to claim 1, characterized in that, The step of pruning the pre-constructed complete binary search tree according to the beam potential energy of each beam codeword to obtain a simplified search tree includes: The complete binary search tree is traversed using a preset potential energy screening threshold, and compared with the beam potential energy of each beam codeword. If the beam potential energy of any beam codeword is less than the preset potential energy screening threshold, its corresponding node and subtree are removed. If the beam potential energy of any beam codeword is greater than or equal to the preset potential energy screening threshold, its corresponding node and subtree are retained to obtain the simplified search tree. The complete binary search tree is composed of the hierarchical beam training levels and corresponding codewords.

5. The hierarchical beam training method based on channel map assistance according to claim 1, characterized in that, The hierarchical beam training strategy is either a training strategy based on maximizing beam potential energy reward or a low-complexity two-layer look-ahead training strategy.

6. The hierarchical beam training method based on channel map assistance according to claim 1, characterized in that, The step of traversing the simplified search tree according to the hierarchical beam training strategy to obtain near-optimal beam codes includes: Based on the hierarchical beam training strategy, the root node of the simplified search tree is initialized to the top-level codeword corresponding to omnidirectional transmission, the initial value of the search iteration counter is -1, and the initial value of the initial search layer is 0. Extract the two-layer lookahead subtrees corresponding to the current root node to identify the subtree topology as any one of full tree structure, asymmetric binary structure, and single-chain structure, and adaptively select the near-optimal beamcode of the subtree topology.

7. The hierarchical beam training method based on channel map assistance according to claim 6, characterized in that, The adaptive selection of near-optimal beamcodes for the subtree topology includes: If the subtree topology is a full tree structure, the search level in the next time step is the current search level plus 1; If the subtree topology is an asymmetric binary structure, calculate the layering and traversal search overhead after beam potential weighting. If the layering search overhead is small, the next search layer is the current search layer plus 1; otherwise, it is the current search layer plus 2. If the subtree topology is a single-chain structure, the next search layer is the current search layer plus 2.

8. A hierarchical beam training device based on channel map assistance, characterized in that, include: A base station module, which is configured with a multi-antenna array, is used to transmit beam signals and receive feedback commands; The user equipment module is used to receive the beam signal sent by the base station module and feed back the optimal beam codeword index. A channel map construction module is used to construct a beamforming codebook-guided channel map, wherein the channel map records the equivalent channel gain corresponding to each grid point and each beamforming codeword within a preset area. The user location acquisition module is used to acquire partial location information of the user device, wherein the partial location information includes multiple mutually exclusive spatial sub-regions where the user is located and the prior probability of each sub-region; The potential energy calculation module is used to calculate the beam potential energy of all beam codes based on the channel map and the partial location information. The pruning module is used to prune the pre-built complete binary search tree according to the beam potential energy of each beam codeword to obtain a simplified search tree; The traversal module is used to traverse the simplified search tree according to the hierarchical beam training strategy to obtain near-optimal beam codes.

9. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, the processor executing the program to implement the channel map-assisted hierarchical beam training method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the channel map-assisted hierarchical beam training method as described in any one of claims 1-7.

Images