Wireless communication physical isolation method and system based on 6g spatial multiplexing

Abstract

The application discloses a wireless communication physical isolation method and system based on 6G space multiplexing, and relates to the technical field of wireless communication. The method comprises the following steps: dynamically segmenting a target area into physical isolation microzones according to user distribution density; deploying a multi-cavity compound eye bionic antenna unit; taking the unit as a communication neuron node to construct a Mesh network; driving the antenna unit to collect multidimensional signals and perform space spectrum calculation to generate an interference thermodynamic map; uploading the thermodynamic map to the Mesh network, and completing air interface resource orthogonal distribution based on node cooperation to output a resource scheme; and dynamically adjusting the microzones in fine granularity according to user density changes in the communication process. The technical problems of low spectrum efficiency and low throughput of a wireless communication system caused by same-frequency interference in a high-density user scenario are solved, and the technical effects of significantly reducing interference through physical isolation and space division multiplexing, dynamically optimizing resources, and thus improving system capacity and communication quality are achieved.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and more specifically to a method and system for physical isolation of wireless communication based on 6G spatial multiplexing. Background Technology

[0002] With the large-scale commercialization of 5G networks and the advancement of 6G technology research and development, the demand for wireless communication services is experiencing explosive growth. Especially in ultra-high user density scenarios such as concerts, sporting events, and large-scale exhibitions, the dense access of terminal devices and the simultaneous initiation of high-speed data services by a massive number of users, such as high-definition video streaming, image sharing, and real-time social networking, pose serious challenges to traditional wireless networks. Although existing communication systems have deployed a large number of AAU devices, utilized full spectrum resources, and optimized scheduling strategies, they still struggle to effectively address co-channel interference and uplink noise rise caused by user aggregation. This leads to a sharp decline in network throughput efficiency, and in severe cases, even communication paralysis, impacting user experience and causing significant economic losses. Current mainstream solutions mainly rely on macro base station overlay, cell splitting, and beamforming technologies, but their essence remains based on traditional cellular architecture, resulting in limited airspace resource utilization and persistent inter-user interference. Summary of the Invention

[0003] This application provides a method and system for physical isolation of wireless communication based on 6G spatial multiplexing, which is used to address the technical problem of low spectral efficiency and throughput of wireless communication systems caused by co-channel interference in high-density user scenarios. It achieves the technical effect of significantly reducing interference and realizing dynamic resource optimization through physical isolation and spatial multiplexing, thereby improving system capacity and communication quality.

[0004] In view of the above problems, this application provides a method and system for physical isolation of wireless communication based on 6G spatial multiplexing.

[0005] The first aspect of this application provides a wireless communication physical isolation method based on 6G spatial multiplexing. The method includes: dynamically dividing a target area into N physically isolated micro-cells based on real-time user distribution density obtained through user equipment location sensing; deploying N multi-cavity compound-eye bionic antenna units on top of each of the N physically isolated micro-cells; using the N multi-cavity compound-eye bionic antenna units as N communication neuron nodes, connecting them according to a preset topology via low-latency direct links to construct a mesh network; and driving the N multi-cavity compound-eye bionic antenna units to acquire multi-dimensional signals. The data is processed to perform spatial spectrum calculations, resulting in N interference heatmaps. These N interference heatmaps are then uploaded to the Mesh network. Based on the collaborative efforts of the N communication neuron nodes, orthogonal allocation of air interface resources is performed, outputting N orthogonal resource allocation schemes. Each orthogonal resource allocation scheme includes a spatial beam pointing angle, frequency subcarriers, and a code-domain pilot sequence. During the communication service process for the N physically isolated micro-cells using these N orthogonal resource allocation schemes, fine-grained dynamic adjustments are made to the N physically isolated micro-cells based on the dynamic changes in the real-time user distribution density.

[0006] A second aspect of this application provides a wireless communication physical isolation system based on 6G spatial multiplexing. The system includes: a region segmentation module: dynamically segmenting a target area into N physically isolated micro-regions based on real-time user distribution density obtained through user equipment location sensing; an antenna deployment module: deploying N multi-cavity compound-eye bionic antenna elements on top of each of the N physically isolated micro-regions; a network construction module: using the N multi-cavity compound-eye bionic antenna elements as N communication neuron nodes, connecting them according to a preset topology via low-latency direct links to construct a mesh network; and a spatial spectrum calculation module: driving the N multi-cavity compound-eye bionic antenna elements. Multi-dimensional signal data is collected, spatial spectrum calculation is performed, and N interference heatmaps are obtained. The resource matching module uploads the N interference heatmaps to the Mesh network, and performs orthogonal allocation of air interface resources based on the collaboration of the N communication neuron nodes, outputting N orthogonal resource allocation schemes. Each orthogonal resource allocation scheme includes a spatial beam pointing angle, frequency subcarriers, and code-domain pilot sequences. The isolation micro-cell adjustment module, during the communication service process of the N physically isolated micro-cells using the N orthogonal resource allocation schemes, performs fine-grained dynamic adjustments to the N physically isolated micro-cells based on the dynamic changes in the real-time user distribution density.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] Based on the real-time user distribution density obtained through user equipment location sensing, the target area is dynamically divided into N physically isolated micro-zones. N multi-cavity compound-eye bionic antenna units are deployed on top of each of the N physically isolated micro-zones. These N multi-cavity compound-eye bionic antenna units serve as N communication neuron nodes, connected via low-latency direct links according to a preset topology to construct a mesh network. The N multi-cavity compound-eye bionic antenna units are driven to collect multi-dimensional signal data, perform spatial spectrum calculations, and obtain N interference heatmaps. The N interference heatmaps are uploaded to the mesh network, and based on the collaborative orthogonal allocation of air interface resources by the N communication neuron nodes, N orthogonal resource allocation schemes are output. Each orthogonal resource allocation scheme includes a spatial beam pointing angle, frequency subcarriers, and code-domain pilot sequences. During the communication service process of the N physically isolated micro-zones using the N orthogonal resource allocation schemes, fine-grained dynamic adjustments are made to the N physically isolated micro-zones according to the dynamic changes in the real-time user distribution density. The system achieves the technical effect of significantly reducing interference and optimizing dynamic resources through physical isolation and spatial multiplexing, thereby improving system capacity and communication quality. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic diagram of the wireless communication physical isolation method based on 6G spatial multiplexing provided in this application embodiment;

[0011] Figure 2 A schematic diagram of a wireless communication physical isolation system structure based on 6G spatial multiplexing provided in an embodiment of this application.

[0012] Figure labeling: Region segmentation module 11, antenna deployment module 12, network construction module 13, spatial spectrum calculation module 14, resource matching module 15, isolation micro-area adjustment module 16. Detailed Implementation

[0013] This application provides a method and system for physical isolation of wireless communication based on 6G spatial multiplexing. It addresses the technical problem of low spectral efficiency and throughput of wireless communication systems in high-density user scenarios due to co-channel interference. The method achieves the technical effect of significantly reducing interference and realizing dynamic resource optimization through physical isolation and spatial multiplexing, thereby improving system capacity and communication quality.

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0015] Example 1, as Figure 1 As shown, this application provides a method for physical isolation of wireless communication based on 6G spatial multiplexing, the method comprising:

[0016] Based on the real-time user distribution density obtained from user equipment location sensing, the target area is dynamically divided into N physically isolated micro-zones.

[0017] In one embodiment, the target area is first defined as the coverage space in a three-dimensional coordinate system, and its corresponding reference grid scale, density threshold, physical isolation constraint parameter set, segmentation update cycle, and density statistics time window are set. Then, a millimeter-wave antenna array or equivalent positioning sensing unit pre-deployed in the target area transmits multi-beam scanning signals to the area at the segmentation update cycle. Through multipath separation algorithms and joint calculations, the three-dimensional coordinate set of each user equipment in the three-dimensional coordinate system is obtained. Next, based on the user equipment's three-dimensional coordinate set and the preset reference grid scale, the real-time user distribution density is calculated. Then, based on this density threshold and the corresponding isolation constraints, the target area's regional grid is dynamically segmented to form N physically isolated micro-areas. The micro-area boundary and topology information, which can be directly used for subsequent antenna deployment mapping, interference spectrum generation, and orthogonal allocation of air interface resources, is output, achieving the effect of improving spatial reuse efficiency and overall system capacity while ensuring the physical isolation of the micro-areas.

[0018] In one possible implementation, the target area is dynamically divided into N physically isolated micro-regions based on the real-time user distribution density obtained through user equipment location sensing. The method includes:

[0019] A millimeter-wave antenna array pre-deployed in the target area transmits a multi-beam scanning signal to the target area to receive multipath signals reflected by user equipment, and the original signal dataset is obtained by summarizing the signals. After calculating the signal time-of-flight difference and carrier phase offset of the original signal dataset, the direct path signal features are extracted by a multipath separation algorithm, and the direct path feature parameter set is output. The angle of arrival and time difference of arrival are jointly calculated on the direct path feature parameter set, and the three-dimensional coordinate set of user equipment is output. After generating the real-time user distribution density based on the three-dimensional coordinate set of user equipment, the target area is divided into the N physically isolated micro-regions based on the density threshold and physical isolation constraints.

[0020] Optionally, active multi-beam scanning is first performed using a pre-deployed millimeter-wave antenna array in the target area. This array consists of M independently controllable elements and operates in the millimeter-wave frequency band, such as 24 GHz to 100 GHz. The system generates narrow-beam scanning signals at different azimuth and elevation angles according to a segmentation update cycle to spatially cover the target area. Each scan generates a reference signal sequence with a known structure and records the transmission timestamp t. tx User equipment generates reflected and scattered signals under the illumination of the scanning beam. The multipath echoes are received by the millimeter-wave antenna array, and the receiving end records the reception timestamp t. rx The received signal amplitude A, phase φ, and frequency offset information are used to form the original signal dataset S. raw ={A k ,φ k ,t rxk ,f k}, where k represents the multipath component number. Subsequently, the time-of-flight difference and carrier phase offset are calculated on the original signal dataset; that is, by calculating ΔT... k =t rxk t tx And by combining this with the known speed of light c, the estimated path distance d is obtained. k =c·ΔT k Simultaneously, the phase offset Δφ is calculated based on the carrier phase change. k This is used to compensate for frequency offset and for fine-grained distance estimation. To distinguish between direct and reflected path signals, the system employs a multipath separation algorithm for parameter estimation. For example, parameter decomposition methods based on MUSIC, ESPRIT, or sparse representation are used to extract the component with the highest energy, shortest propagation path, and stable angle as the direct path signal, and output the corresponding direct path feature parameter set F. los ={d los ,AoA los, φ los ,P los Next, the angle of arrival (AoA) and time difference of arrival (TDoA) are jointly calculated for the direct path characteristic parameter set. During this process, the incident angle AoA is calculated based on the phase difference of the array received signals. los If the system contains multiple synchronous array units, triangulation is performed using the time difference of arrival (TDoA) between different arrays, combined with distance estimation (d). los The position (x, y, z) of the user equipment in the three-dimensional coordinate system is solved using geometric positioning equations. By filtering and fusing multiple measurement results, a stable three-dimensional coordinate set P={p i (x i ,y i ,z iThen, the target area is divided into multiple three-dimensional mesh units g according to a preset reference grid scale. j And count the number of users n in each grid cell. j Combined with the mesh volume V j Calculate density ρ j =n j / V j This forms the real-time user distribution density matrix {ρ j To avoid instantaneous fluctuations, a moving average or exponentially weighted smoothing process can be performed within the time window to obtain a stable real-time user distribution density. Finally, density-driven dynamic aggregation and splitting are performed based on density thresholds and physical isolation constraints, resulting in N physically isolated micro-regions that satisfy both density conditions and physical isolation constraints. Each physically isolated micro-region contains clearly defined spatial boundaries, center coordinates, and adjacency information. Through the above steps, high-precision user spatial distribution acquisition based on millimeter-wave positioning and sensing is achieved, and dynamic micro-region partitioning based on density-driven and isolation constraints is completed, providing a precise spatial foundation for subsequent spatial reuse and orthogonal resource allocation.

[0021] In one possible implementation, after generating a real-time user distribution density based on the user equipment's three-dimensional coordinate set, the target region is divided into the N physically isolated micro-regions based on a density threshold and physical isolation constraints. The method includes:

[0022] The target region is divided into multiple regional grids based on a preset reference grid scale; multiple grid-level user densities are obtained by projecting the three-dimensional coordinate set of the user equipment onto the multiple regional grids, which constitute the real-time user distribution density; the multiple regional grids are subjected to density-driven dynamic aggregation and splitting based on density thresholds and physical isolation constraints to obtain the N physically isolated micro-regions.

[0023] Optionally, a reference grid size, such as 0.5m to 3m, is preset based on the spatial scale and positioning accuracy of the application scenario. In a three-dimensional coordinate system, the target area is divided equidistantly along the x, y, and z directions, forming multiple regular three-dimensional mesh units g. j Each grid cell has a defined spatial boundary range [x j1 ,x j2 ]、[y j1 ,y j2 ]、[z j1 ,z j2 and volume V j For planar scenes, such as halls or plazas, a two-dimensional mesh can be used, discretizing only the xy-plane. Then, for each user device, the three-dimensional coordinates p... i (x i ,y i ,z iThe grid cell g to which it belongs is determined based on its coordinate range. j And include it in the user count n of that grid cell. j After the statistics are completed, the user density ρ is calculated for each grid cell. j In a two-dimensional scene, n j / A j This yields the grid-level user density. To improve stability, the user density can be smoothed using a moving average or exponential weighting within a preset time window to obtain the real-time user distribution density. Then, a density threshold ρ is set. th When ρ j ≥ρ th When ρ is high, the corresponding grid cell is marked as a high-density grid; when ρ is high, the corresponding grid cell is marked as a high-density grid. j <ρ th The time markers are designated as low-density grids. High-density grids will form a candidate hotspot set H. Next, density-driven dynamic aggregation is performed on the high-density grids; that is, according to spatial connectivity rules, such as the six-neighbor or twenty-six-neighbor connectivity criterion, adjacent high-density grids are clustered and merged to form several candidate aggregation blocks. For each candidate aggregation block, its spatial centroid, coverage area, and equivalent size parameters are calculated, and then based on preset physical isolation constraint parameters, such as the minimum center-to-center distance d between adjacent micro-regions... min Maximum permissible coverage scale L max Predicted beam sidelobe interference threshold I th Then, the candidate aggregation blocks are evaluated. When the size of a certain aggregation block exceeds L... max When the density gradient or density valley location is less than d, the region is split into multiple sub-blocks using k-means clustering or density peak-based region segmentation algorithms. min If necessary, the boundary grids are reassigned or buffer grid bands are introduced until the isolation constraints are met. Simultaneously, low-density grid regions can be merged into adjacent micro-regions based on adjacency relationships, or retained as transitional regions to maintain regional continuity. Finally, after iterative aggregation and splitting, N physically isolated micro-regions satisfying both density conditions and physical isolation constraints are output. Each micro-region is defined by its contained grid set, spatial boundary contour, center coordinates, and adjacency relationships. Through the above-described process of gridding statistics, density discrimination, connectivity aggregation, and constraint verification splitting, adaptive micro-region construction based on real-time user distribution is achieved. This ensures that the region division reflects both user density characteristics and meets physical isolation requirements, providing a structured spatial unit foundation for subsequent spatial beam control and orthogonal resource allocation.

[0024] N multi-cavity compound eye bionic antenna units are deployed on top of the N physically isolated micro-regions respectively.

[0025] In one embodiment, after dividing the target area into N physically isolated micro-regions, N multi-cavity compound eye bionic antenna units are deployed at the top of each of the N physically isolated micro-regions according to the principle of one unit per micro-region, to form directional coverage and physical isolation for each micro-region. Typically, the installation point is located at a preset height at the top of the micro-region, such as the ceiling, truss, or light fixture in an indoor scene, or the pole or cantilever support in an outdoor scene. It is ensured that the main radiation direction of the antenna unit can cover the authorized user activity area within the micro-region without being blocked. The multi-cavity compound eye bionic antenna unit consists of several independently controllable miniature radiators, each of which is arranged in multiple metal cavities. These metal cavities are connected by a metal cavity waveguide structure for electromagnetic energy guidance and isolation. On the one hand, the metal cavities provide electromagnetic shielding boundaries for each radiator, suppressing undesirable coupling between radiators; on the other hand, the metal cavity waveguide constrains the energy propagation path, so that the electromagnetic field mainly propagates within the direction specified by the cavity and waveguide, structurally weakening non-main direction radiation, thereby achieving sidelobe attenuation. Through the above deployment process, a controllable, highly directional coverage unit is formed above each physically isolated micro-region. By leveraging the sidelobe attenuation capability provided by the metal cavity waveguide structure, energy leakage across micro-regions is reduced from the source, providing a feasible hardware foundation and engineering implementation path for physical isolation communication under subsequent space reuse.

[0026] The N multi-cavity compound eye bionic antenna units are used as N communication neuron nodes and connected according to a preset topology through low-latency direct links to construct a Mesh network.

[0027] In one embodiment, after deploying N multi-cavity compound-eye bionic antenna units on top of N physically isolated micro-cells, each multi-cavity compound-eye bionic antenna unit is defined as a communication neuron node and interconnected via low-latency direct links according to a preset topology. This preset topology is determined based on the regional adjacency relationships and spatial distribution characteristics of the N physically isolated micro-cells, typically including ring topologies and star topologies. Next, based on this preset topology, the underlying transmission architecture of the N communication neuron nodes is constructed, and a corresponding layered protocol stack is loaded onto the constructed underlying transmission architecture, thereby building a mesh network for collaborative sensing, map sharing, and resource collaborative allocation. After construction, the mesh network undergoes consistency verification and operational initialization; that is, any node or a designated coordinating node broadcasts network parameters, such as synchronization clock, frame structure, and service channel mapping, and each node completes time alignment and sends back confirmation. When the overall network connectivity and latency indicators meet the requirements, the mesh network enters the working state. Through the above process, the N antenna elements not only act as independent beam generation entities, but also form a low-latency collaborative network as neuron nodes with computing and communication capabilities. This provides an implementable distributed infrastructure for the rapid sharing of subsequent interference heatmaps, three-dimensional resource conflict detection, and orthogonal resource allocation iteration.

[0028] In one possible implementation, a mesh network is constructed by connecting low-latency direct links according to a preset topology, the method comprising:

[0029] The network topology is configured based on the regional adjacency and spatial distribution characteristics of the N physically isolated micro-regions, wherein a ring topology is deployed in high-user-density areas and a star topology is deployed in boundary areas. Based on the network topology, dual-channel physical links are configured between the N communication neuron nodes to complete the construction of the underlying transmission architecture. A millimeter-wave wireless channel is enabled when the node spacing is less than or equal to a preset distance threshold, and a fiber optic wired channel is enabled when the distance exceeds the preset threshold. The N communication neuron nodes are equipped with N virtual channels for transmitting interference heatmaps. A layered protocol stack is loaded into the underlying transmission architecture to complete the construction of the Mesh network. The layered protocol stack includes a transport layer time-sensitive network protocol, a network layer distributed hash table addressing protocol, and a link layer binding protocol.

[0030] Optionally, based on the regional adjacency relationships and spatial distribution characteristics of the N physically isolated micro-regions generated from the aforementioned micro-region division results, for the set of micro-regions located in high-user-density areas, a ring topology or multi-ring nested topology is adopted as the current network topology. This involves connecting the communication neuron nodes in this set end-to-end in spatial order to form a closed-loop structure, and adding diagonal connections as needed to create redundant paths, thereby improving link reliability and data routing capabilities. For micro-region nodes located in low-density areas, a star topology is adopted as the current network topology. This involves selecting the node with the best geometric center or link conditions as the convergence node, and establishing direct connections between the remaining boundary nodes and this convergence node, thereby reducing the number of links and simplifying edge management. This method generates a topology connection table, clearly defining the set of adjacent nodes for each communication neuron node. After determining the network topology, the system calculates the spatial distance between each pair of adjacent nodes in the topology connection table. When the calculated spatial distance is less than or equal to a preset distance threshold, a millimeter-wave wireless direct connection channel is activated. Point-to-point connection is achieved through narrow beam alignment, link training, and power control, and link latency and bit error rate are tested. When the spatial distance is greater than the preset distance threshold, a fiber optic or high-speed Ethernet wired channel is activated to complete the physical connection, and port binding and link authentication are performed. Each physical link establishes a primary and backup dual-channel structure. The wireless and wired links can be configured in primary / backup mode or load-sharing mode to meet the dual requirements of low latency and high reliability. Based on this, logical virtual channels are divided within each communication neuron node, with at least one dedicated virtual channel configured for transmitting interference heatmap data. The virtual channels are isolated from the physical links through VLAN identification or logical port mapping to ensure that the heatmap data has independent bandwidth and priority. Subsequently, a link layer binding protocol is loaded at the link layer of the underlying transmission architecture to manage the binding of multiple physical interfaces, monitor link status, and enable automatic switching. A distributed hash table (DHT) addressing protocol is loaded at the network layer, allowing each communication neuron node to perform decentralized addressing and route lookup based on its node ID, automatically constructing multiple paths and maintaining a routing table. A time-sensitive network (TSN) protocol is loaded at the transport layer, setting deterministic scheduling time slots and bandwidth reservation mechanisms for high-priority data streams such as interference heatmaps and resource allocation control information, ensuring that end-to-end transmission latency meets preset thresholds. After protocol loading, each node establishes a unified clock reference and network operating parameters through handshake and synchronization processes, generating a complete Mesh routing table and link status table. Finally, through topology configuration, dual-channel link establishment, and layered protocol loading, a Mesh network structure with low latency, high reliability, and distributed addressing capabilities is formed. This enables N communication neuron nodes to achieve real-time sharing and collaborative computation of interference heatmaps without central control, providing stable underlying network support for subsequent orthogonal allocation of air interface resources and conflict resolution.

[0031] The N multi-cavity compound eye bionic antenna units are driven to collect multi-dimensional signal data, perform spatial spectrum calculation, and obtain N interference heat maps.

[0032] In one embodiment, after N multi-cavity compound eye bionic antenna elements are deployed and connected to the Mesh network, the system performs unified clock synchronization and data acquisition task distribution to each antenna element, driving it to acquire multi-dimensional signal data in its corresponding physically isolated micro-area, including user signal strength in each direction, multipath signal phase difference, subcarrier noise floor, etc. To eliminate the spectrum offset caused by the difference in acquisition time of different nodes, each antenna element records a local timestamp and performs clock drift compensation before and after acquisition, so that the observations within the same acquisition period can be aligned and fused. Subsequently, for possible signal leakage, the corresponding intensity detection is performed according to the probe antenna, and then the interference direction distribution and intensity matrix are estimated by the spatial spectrum estimation algorithm on the supplemented multi-dimensional signal data, thereby generating N interference heat maps containing the interference direction distribution and interference intensity matrix, with each interference heat map corresponding to a multi-cavity compound eye bionic antenna element. Through the above process, the system can plot the direction and intensity distribution of interference sources in each micro-region in real time at node-level resolution, providing directly usable input data for subsequent graph aggregation, resource conflict detection and orthogonal allocation of air interface resources in the Mesh network, thereby enabling engineering-based support for physical isolation optimization under spatial reuse.

[0033] In one possible implementation, the N multi-cavity compound eye bionic antenna elements are driven to acquire multi-dimensional signal data, perform spatial spectrum calculations, and obtain N interference heat maps. The method includes:

[0034] The user signal strength, multipath signal phase difference, and subcarrier noise floor are collected within the first physical isolation micro-region by several independently controllable miniature radiators in the first multi-cavity compound eye bionic antenna unit. Distributed signal leakage intensity detection of adjacent physical isolation micro-regions is performed by probe antennas deployed at the boundary of the first physical isolation micro-region. Based on the user signal strength, multipath signal phase difference, and subcarrier noise floor, a spatial spectrum estimation algorithm is executed to generate a first interference heat map containing the interference direction distribution and interference intensity matrix.

[0035] Optionally, measurement tasks are first issued to several independently controllable miniature radiators within the first multi-cavity compound-eye bionic antenna unit during each acquisition cycle. Each miniature radiator receives the uplink pilot and reference signals of user equipment within the first physically isolated micro-area according to a preset time-division polling or parallel reception method. At the baseband side, the user signal strength, multipath signal phase difference, and subcarrier noise floor within the first physically isolated micro-area are calculated. The user signal strength index can be obtained by statistically analyzing RSRP / RSSI or equivalent received power by user or by angular sector. The multipath signal phase difference can be obtained by extracting the phase and calculating the phase difference from the complex channel estimates of the same user in different receiving channels, different arrival paths, or adjacent time slots, used to characterize the multipath structure and interference superposition characteristics. The subcarrier noise floor can be obtained by estimating the noise power spectral density on subcarriers that do not carry effective signals or are unloaded by the pilot, and can be averaged by subband. To ensure data comparability, timestamps are recorded during acquisition, and automatic gain control and amplitude-phase calibration are performed. After removing outliers, the final multi-dimensional signal data for the first physically isolated micro-area is formed. Subsequently, probe antennas are deployed at the boundary of the first physically isolated micro-region. These probe antennas are arranged at preset intervals of P probe points along the boundary of the first micro-region, for example, one every 5m to 20m. Each probe antenna faces the direction of the adjacent micro-region and measures the leakage components of the transmitted signals or common reference signals of the adjacent micro-regions within the same acquisition period as the antenna element. The leakage intensity of each probe point, as well as the corresponding azimuth angle and boundary position index, are output. To reduce the impact of instantaneous fading, the leakage intensity can be short-time averaged to form a boundary leakage vector, which is used to reflect the spatial distribution of cross-micro-region leakage interference. Then, the multi-channel received signals of the first antenna element are constructed into an array observation vector, and the spatial covariance matrix is ​​calculated. Then, the spatial covariance matrix is ​​combined with the subcarrier noise basis to perform noise whitening or noise subspace separation. The multipath phase difference characteristics are used to enhance and adjust the suspected multipath or interference components, thereby improving the resolution capability of the interference incident direction. Then, spatial spectrum estimation is performed on a preset azimuth and elevation angle scanning grid to calculate the angular domain power spectrum. The interference direction distribution of the first micro-region is obtained by fusing the spectrum at each frequency point using a weighted sum or by taking the maximum value. Further, the interference direction distribution and the frequency domain interference power estimates are combined into an interference intensity matrix, where each matrix element represents the interference energy intensity at the corresponding frequency point. Finally, the interference direction distribution and the interference intensity matrix are encapsulated into a first interference heatmap, with necessary structured fields added, including angular grid resolution, frequency domain resolution, peak interference direction set, leakage intensity markers for adjacent micro-region directions, and metadata such as timestamps, node IDs, and micro-region IDs. Through the above process of multi-dimensional acquisition by micro-radiators, boundary probe leakage detection, and spatial spectrum estimation fusion, a first interference heatmap capable of characterizing the direction and intensity distribution of interference sources can be generated, providing direct input for subsequent cross-node spectrum aggregation and orthogonal resource allocation.

[0036] In one possible implementation, the method further includes:

[0037] Using the N interference thermal maps as the basis for beam control, the N multi-cavity compound eye bionic antenna elements are controlled by a beamforming algorithm to generate highly directional main lobe beams toward the N physically isolated micro-regions, and spatial nulls are formed in the direction of adjacent micro-regions.

[0038] Optionally, the system uses the N interference heatmaps generated by each communication neuron node as the direct basis for beam control. A beamforming algorithm is used to jointly control the amplitude and phase of each miniature radiator in the N multi-cavity compound eye bionic antenna elements, resulting in high directivity main lobe coverage within each physically isolated micro-region, while simultaneously creating spatial nulls in adjacent micro-regions to suppress leakage interference. Specifically, for the nth physically isolated micro-region, the target main lobe pointing angle pair (θ) is calculated based on the three-dimensional coordinates of the authorized user within that micro-region. n ,φ n And by combining the micro-region boundaries and adjacency relationships, the set of directions Ω of adjacent micro-regions is obtained. n ={(θ {n,1} ,φ {n,1} ),…,(θ {n,m} ,φ {n,m} Simultaneously, the set of interference peak directions and their intensity weights are extracted from the corresponding nth interference heatmap to distinguish between strong interference directions that must be suppressed and secondary directions that can be weakened, and a null depth threshold (e.g., ≥20dB suppression) and main lobe gain are set. Subsequently, the nth multi-cavity compound eye bionic antenna element is considered as an array composed of Q independent controllable micro radiators, and an array steering vector a is established based on the array geometry, frequency point, and cavity waveguide structure parameters. n(θ,φ)Then, beam weight vectors are constructed based on this, where each weight includes amplitude and phase control values. Beam design constraints are divided into main lobe constraints, null constraints, interference weighting constraints, and power and hardware constraints. The main lobe constraint requires the antenna array to form the most concentrated main radiation beam in the target coverage direction of the physically isolated micro-area, maximizing the radiation gain in that direction, or at least ensuring it is not lower than a preset main lobe gain threshold, to guarantee that the communication quality and signal strength of licensed users within the micro-area meet service requirements. The null constraint requires the formation of significant energy suppression regions, i.e., spatial nulls, in other physically isolated micro-area directions adjacent to the micro-area, ensuring that the radiation power in these directions is lower than a preset null threshold. The trap depth threshold effectively reduces cross-micro-area signal leakage and co-channel interference; interference weighting constraint applies a higher level of suppression priority to high-intensity interference directions identified in the interference heatmap, prioritizing the reduction of radiation energy from strong interference directions during beam optimization, thus achieving key suppression of critical interference sources; power and hardware constraints ensure that the total transmit power limit of antenna elements, the maximum output capability of a single radiating element, and phase adjustment accuracy are met during beam optimization, ensuring that the calculated beam control parameters can be realistically implemented under existing RF and phased array hardware conditions.

[0039] Subsequently, the system employs linearly constrained minimum variance, MVDR, or zero-force beamforming algorithms for solution. For example, the objective function is to minimize the output power in adjacent and interfering directions, while simultaneously satisfying main lobe gain constraints and null constraints; or, in multi-user scenarios, weighted minimum mean square error is used to jointly optimize the multi-user beams. After obtaining the weight vector, it is converted into amplitude control words and phase control words for each micro-radiator, and written into the phased array network or controllable feed network through the RF control interface of the antenna element, thereby forming the corresponding highly directional main lobe and spatial nulls at the physical level. If multiple licensed users need to be covered, a multi-beam superposition method can be used, i.e., a sub-beam weight vector is generated for each user direction, and the vectors are superimposed and normalized according to the power allocation coefficient to obtain the final beam weight vector, while maintaining the null constraints for adjacent micro-area directions. Then, after the weights are distributed, the antenna element verifies the beam effect using backhaul CSI, leakage measurements from the boundary probe antenna, and SINR statistics of users in the micro-area. If the leakage power in adjacent micro-regions is still higher than the threshold or the main lobe coverage is insufficient, a secondary optimization iteration is triggered. This involves adjusting constraint parameters, such as increasing null weights, changing the main lobe direction, or reselecting the service user set, until the main lobe coverage and adjacent direction nulls are met. The entire process is repeated according to a preset update cycle or triggered by a significant change in the interference thermal map, achieving real-time beam control based on the map-driven mechanism. Through this process, the system uses N interference thermal maps as input to achieve feasible precise main lobe focusing and adjacent direction spatial null beam generation, thereby effectively suppressing cross-micro-region leakage interference under the same frequency condition, strengthening physical isolation, and improving spatial reuse efficiency.

[0040] In one possible implementation, the center-to-center distance between adjacent physically isolated micro-regions in the N physically isolated micro-regions is greater than the beam sidelobe interference threshold, and the micro radiator achieves sidelobe attenuation through a metal cavity waveguide structure.

[0041] Optionally, to ensure the physical isolation effect between micro-cells at the structural level, a dual control mechanism of center-to-center spacing constraint and antenna structure sidelobe suppression is introduced when dividing the physically isolated micro-cells. Specifically, firstly, in the spatial planning stage of N physically isolated micro-cells, the Euclidean distance between the geometric center coordinates of any pair of adjacent physically isolated micro-cells (i,j) is calculated. The system pre-determines the minimum safe center-to-center spacing threshold D based on the radiation pattern parameters of the antenna elements used, the sidelobe level SLL, and the maximum allowable neighboring cell leakage interference power. min This threshold can be determined by the following principle: when the main lobe points towards the local microregion and the side lobes face the neighboring region, at a distance D... min After the sidelobe field strength at a certain point is attenuated, its interference power does not exceed the preset interference value. Only when d ij ≥D minWhen two micro-regions meet the beamside lobe interference safety distance constraint, if they do not, the center spacing is increased during the micro-region division stage by boundary back-off, buffer band introduction, or micro-region re-splitting until the threshold condition is met. This ensures sufficient physical isolation margin between adjacent micro-regions on a spatial scale. At the antenna structure level, sidelobes are engineered and suppressed using a metal cavity waveguide structure. Each multi-cavity compound eye bionic antenna element consists of several miniature radiators, each encapsulated in an independent metal cavity. Electromagnetic energy coupling and directional transmission are achieved between the cavities through the metal waveguide structure. The metal cavity acts as a shield and absorber for electromagnetic waves in undesirable directions, while the waveguide structure constrains the main radiation direction, preferentially radiating electromagnetic energy along the main lobe direction and reducing lateral and rearward leakage. By optimizing the cavity size, waveguide opening angle, and cavity spacing, the sidelobe level relative to the main lobe gain can be reduced to a preset attenuation level. By achieving a synergistic design that satisfies the sidelobe safety threshold through the center-to-center spacing of micro-regions and realizes sidelobe attenuation through the metal cavity waveguide structure, the interference distance between micro-regions is increased on the macroscopic spatial scale, while the energy leakage in non-primary directions is reduced at the microscopic antenna radiation level. This effectively suppresses sidelobe interference between adjacent physically isolated micro-regions and improves the physical isolation reliability and system stability under spatial reuse conditions.

[0042] The N interference heatmaps are uploaded to the Mesh network. Based on the collaboration of the N communication neuron nodes, air interface resources are orthogonally allocated, and N orthogonal resource allocation schemes are output. The orthogonal resource allocation schemes include spatial beam pointing angle, frequency subcarrier, and code domain pilot sequence.

[0043] In one embodiment, after each communication neuron node generates an interference heatmap locally, the system uploads and shares N interference heatmaps to the Mesh network. Each node then collaboratively completes orthogonal allocation of three-dimensional air interface resources across the spatial, frequency, and code domains. This resolves conflicts and iterates resource mapping, resulting in N orthogonal resource allocation schemes. Each scheme includes a spatial beam pointing angle, frequency subcarriers, and a code pilot sequence. The spatial beam pointing angle refers to the angle parameter corresponding to the main lobe energy being concentrated in a specific spatial direction using beamforming technology of the array antenna; it typically includes azimuth and elevation angles. The frequency subcarriers are multiple narrowband orthogonal subcarriers divided from the entire bandwidth. Each subcarrier can independently carry data and is a resource allocation unit in the frequency dimension. The code pilot sequence is a known signal sequence used for channel estimation, synchronization, and demodulation reference. Different users or different cells can use different pilot sequences for differentiation. Finally, the system forms a resource mapping table from N orthogonal resource allocation schemes and distributes it to the corresponding multi-cavity compound eye bionic antenna unit and baseband processing unit through the Mesh network. The antenna unit generates the main lobe beam according to the spatial beam pointing angle, and the baseband scheduler performs frequency domain scheduling according to the frequency domain subcarriers and inserts code domain pilot sequences at the physical layer to complete link estimation and synchronization. This achieves orthogonalization of spatial, frequency, and code domain resources across micro-cells under the same frequency condition, reduces neighboring cell interference, enhances physical isolation effect, and improves overall spectrum efficiency and system capacity.

[0044] In one possible implementation, the N interference heatmaps are uploaded to the Mesh network, and air interface resources are orthogonally allocated based on the collaboration of the N communication neuron nodes, outputting N orthogonal resource allocation schemes. The method includes:

[0045] The N communication neuron nodes upload the N interference heatmaps to the Mesh network through the N virtual channels, perform heatmap aggregation, and generate a global interference matrix. Based on the global interference matrix, perform pairwise adjacent micro-region three-dimensional resource conflict detection to obtain N conflict identifier sets, wherein the three-dimensional resource conflict detection covers spatial domain conflict detection, frequency domain conflict detection, and code domain conflict detection. Iterate the conflict resolution and resource mapping of the N conflict identifier sets until the N orthogonal resource allocation schemes are output.

[0046] Optionally, each nth communication neuron node uploads the interference heatmap generated by its node to the Mesh network through a pre-configured nth virtual channel. These virtual channels have independent priorities and bandwidth guarantees and are used to carry the heatmap data. After receiving the interference heatmaps from all nodes, each node performs a heatmap aggregation operation, that is, extracts the interference intensity of each micro-region pointing to other micro-regions as weighted boundary values ​​to construct a micro-region-micro-region interference relationship matrix G, where the matrix elements G... (i,j)This represents the leakage interference intensity of micro-region i in the direction of micro-region j. After network-wide synchronous confirmation, a global interference matrix G is generated. full This serves as the unified input for subsequent 3D resource conflict detection. Subsequently, based on the global interference matrix G... full Based on the current initial or previous round of resource allocation results, three-dimensional resource conflict detection is performed on each pair of adjacent microcells (i,j). In spatial conflict detection, the difference in the current main beam pointing angle between the two microcells is calculated; if it is a three-dimensional angle, the combined angle difference of azimuth and elevation is taken. When the main beam pointing angle difference is less than a preset angle threshold of 10 degrees, a spatial conflict is determined to exist, and this is combined with G... full(i,j) Determine the conflict intensity level, for example, G full(i,j) ≥G high Determined as a high-intensity conflict; G mid ≤G full(i,j) <G high Determined as a medium-intensity conflict; below G mid The collision is classified as low-intensity. In frequency domain collision detection, the minimum frequency interval Δf between the allocated subcarriers of the two micro-cells is calculated. ij If Δf ij If the frequency overlap is less than the preset protection bandwidth Δf, a frequency domain conflict is determined to exist. The conflict intensity can be determined based on the frequency overlap ratio or the corresponding interference matrix weight G. full(i,j) A hierarchical classification is performed. In code domain collision detection, the pilot sequence p allocated to the two micro-regions is calculated. i With p j The normalized cross-correlation value C between them ij If C ijA cross-correlation value greater than 0.2 indicates a code domain conflict. A higher cross-correlation value indicates a higher conflict intensity level; for example, 0.2–0.4 indicates a low-level conflict, 0.4–0.7 indicates a medium-level conflict, and greater than 0.7 indicates a high-level conflict. Next, for each microcell n, the detected conflict information between it and its neighboring microcells is aggregated to generate a conflict identifier set. Each conflict identifier entry includes the conflicting neighbor cell ID, conflict type (spatial, frequency, code domain), conflict intensity level, and corresponding key parameters, ultimately forming N conflict identifier sets. Then, the system sorts the conflict identifier sets of each microcell according to a preset priority, such as spatial domain first, then frequency domain, and finally code domain. Then, by performing beam pointing angle adjustment, subcarrier reallocation, and pilot sequence reconstruction on the microcell groups corresponding to each type of conflict, the corresponding spatial domain resolution scheme, frequency domain resolution scheme, and code domain resolution scheme are obtained. Finally, by restoring and backtracking these conflict resolution schemes, and then mapping the conflict resolution results to obtain the spatial beam pointing angle, frequency subcarrier, and code pilot sequence, N orthogonal resource allocation schemes are formed, thus completing the three-dimensional orthogonal resource configuration. Through the above process, a quantifiable three-dimensional conflict detection and hierarchical resolution mechanism based on the global interference matrix is ​​realized, ensuring that spatial, frequency, and code domain resources maintain an engineering-executable orthogonal relationship between adjacent micro-cells, thereby guaranteeing the physical isolation effect and system stability under spatial reuse conditions.

[0047] In one possible implementation, conflict resolution and resource mapping are iterated on the N conflict identifier sets until the N orthogonal resource allocation schemes are output, the method comprising:

[0048] Based on a preset conflict resolution priority, the N conflict identifier sets are decomposed to obtain spatial conflict micro-region groups, frequency conflict micro-region groups, and code conflict micro-region groups. Beam pointing angle adjustment, subcarrier reallocation, and pilot sequence reconstruction are performed on the spatial conflict micro-region groups, frequency conflict micro-region groups, and code conflict micro-region groups, respectively, to obtain spatial resolution schemes, frequency resolution schemes, and code resolution schemes. After restoring the spatial resolution schemes, frequency resolution schemes, and code resolution schemes, conflict backtracking verification is performed, and conflict resolution iteration is triggered based on the verification results until N conflict-free resource allocation schemes are obtained. Resource mapping is performed on the N conflict-free resource allocation schemes to obtain the N orthogonal resource allocation schemes.

[0049] Optionally, a preset conflict resolution priority is first obtained, typically in the order of spatial domain > frequency domain > code domain. Then, conflicts are categorized according to the rule of recording only the highest priority conflict type for each pair of adjacent microcells. Specifically, for each pair of adjacent microcells (i,j), its conflict identifier entry is read. If multiple conflict types exist simultaneously, only the highest priority conflict type is retained as the unique classification result for that adjacent pair. For example, when both spatial and code domain conflicts exist, (i,j) is only assigned to the spatial conflict group and not recorded again in the code domain conflict group. Similarly, when both frequency and code domain conflicts exist, it is only assigned to the frequency domain conflict group. After traversal, three grouping results are obtained: spatial conflict microcell group GS, frequency conflict microcell group GF, and code domain conflict microcell group GC. Each group contains several pairs of adjacent microcells and their conflict strength level information. Subsequently, for each microcell group, the system executes the corresponding resolution process to generate the corresponding resolution scheme. During the generation of the spatial domain resolution scheme, for each pair of adjacent micro-cells (i,j) in the spatial domain conflict micro-cell group GS, the system reads the current beam pointing angle and angle difference, and sets the adjustment step size according to the conflict intensity level. Generally, stronger conflicts take a larger step size. Then, the beam is adjusted using the principle of minimum disturbance while ensuring coverage. For example, based on the global interference matrix, the side with higher interference contribution that needs to be adjusted first is determined, and its beam pointing angle is deflected away from the neighboring cell, so that the updated angle difference is ≥10°. If a single deflection is still insufficient, fine-tuning is continued or the system switches to a feasible angle in the set of alternative beam pointing angles. For multiple conflict edges involving the same micro-cell, a weighted merging strategy is used to calculate the comprehensive adjustment amount of the micro-cell at once, avoiding repeated swings within the same round. This yields the spatial domain resolution scheme PlanS, which includes the updated beam pointing angle value of each relevant micro-cell. During the frequency domain resolution scheme generation process, for each pair of adjacent microcells (i,j) in the frequency domain conflict microcell group GF, the minimum frequency interval of their current subcarrier set is checked. If the constraint that the minimum frequency interval ≥ the guard bandwidth is not met, subcarrier migration is performed on one side. That is, the edge subcarriers on the conflicting side are preferentially moved to the candidate set that are not occupied by adjacent microcells and whose interval with them meets the guard bandwidth. If there are no available idle subcarriers, the bandwidth of the two microcells is compressed or time slotted replacement is used according to the service priority. That is, some subcarriers are transferred to different time units until the guard bandwidth constraint is met. This yields the frequency domain resolution scheme PlanF, which includes the subcarrier set update of each relevant microcell. During the code domain resolution scheme generation process, for each pair of adjacent microcells (i,j) in the code domain conflict microcell group GC, the current pilot sequence is read and the normalized cross-correlation value is calculated. If the normalized cross-correlation value > 0.2, the pilot sequence is reallocated to the conflicting side. The sequence selection follows the principle of low cross-correlation priority, that is, candidate sequences with cross-correlation with the pilots of neighboring micro-regions not greater than 0.2 are selected from the preset pilot sequence pool, and sequences that meet the threshold with all their neighbors are given priority.If there are not enough candidates, the sequence is reconstructed, for example by changing the cyclic shift, changing the orthogonal overlay code, or switching to another sequence family, until the cross-correlation threshold constraint is met, thus obtaining the code domain resolution scheme PlanC, which includes pilot updates for each relevant micro-region.

[0050] Afterwards, the system merges PlanS, PlanF, and PlanC according to priority to restore the candidate resource scheme for this round. Based on this scheme, it re-executes three-dimensional conflict detection. During backtracking verification, it no longer uses the coarse rule of only recording the highest priority conflict, but instead performs a complete detection of all three types of conflicts for each pair of adjacent micro-cells to identify new conflicts that were missed due to coarse grouping or introduced by adjustments. It verifies and outputs a new set of conflict identifiers and counts whether any conflict entries still exist. Then, if backtracking verification finds that conflicts still exist, the new set of conflict identifiers is regrouped again according to the rules of "spatial domain > frequency domain > code domain" and "only the highest priority conflict type is retained for a pair of adjacent cells" to enter the next round of resolution. If it is found that some code domain conflicts have been automatically eliminated after spatial domain resolution, these conflicts will not enter subsequent code domain resolution, thereby reducing the number of sequence reconstructions. The iterative process of "coarse grouping + backtracking verification" continues until all adjacent micro-cell pairs meet the following conditions under complete detection: angle difference ≥ 10°, subcarrier spacing ≥ Δf, cross-correlation ≤ 0.2, or the preset maximum number of iterations is reached. This outputs the approximate solution with the minimum current conflict and marks the residual conflict level. Finally, after backtracking verification confirms no conflict, the system writes the final resource triplet for each micro-cell into the resource mapping table, including the beam number of the b-antenna element, the set of subcarrier indices that the scheduler can directly configure, the physical layer pilot sequence ID, and its generation parameters. The mapping table is distributed to the corresponding multi-cavity compound eye bionic antenna element and baseband scheduling unit via the Mesh network, enabling each micro-cell to perform beamforming, frequency domain scheduling, and pilot insertion according to its orthogonal scheme. This achieves feasible three-dimensional orthogonal resource configuration and physically isolated communication under the condition of spatial reuse at the same frequency.

[0051] During the communication service of the N physically isolated micro-cells using the N orthogonal resource allocation schemes, the N physically isolated micro-cells are dynamically adjusted in a fine-grained manner according to the dynamic changes in the real-time user distribution density.

[0052] In one embodiment, after communication services are provided in N physically isolated micro-cells according to corresponding orthogonal resource allocation schemes, the system does not maintain the micro-cell structure and resource configuration unchanged. Instead, based on the continuous updates of real-time user distribution density, it executes a fine-grained dynamic adjustment mechanism to achieve closed-loop optimization where the network follows the users. Specifically, the system continuously performs user location sensing and density statistics at a preset update cycle, generates a new real-time user distribution density matrix, and compares it with the density distribution of the previous cycle. When any trigger condition is detected, the dynamic adjustment process is initiated. These trigger conditions typically include density growth in a micro-cell exceeding a corresponding threshold, density in a micro-cell remaining below a corresponding threshold, hotspot center migration distance exceeding a spatial threshold, or micro-cell load rate exceeding a preset load limit. After the adjustment is triggered, the system performs fine-grained reconstruction according to the principle of local priority and gradual adjustment. In this process, firstly at the structural level, the affected micro-regions are re-partitioned at the grid level. Only the boundary grids of hotspot migration areas or areas with abnormal loads are reassigned or split / merged, without re-partitioning the entire region, thereby reducing the disturbance range. For example, when a bimodal density structure appears within a micro-region, it is split into two sub-micro-regions according to the density valley value. When two adjacent micro-regions have low densities and insufficient loads, a merging operation can be performed. Subsequently, at the resource level, the adjusted micro-regions are re-matched with resource triples. If the micro-region boundary changes but the adjacency relationship remains unchanged, the original beam pointing angle can be reused first, with only the main lobe direction fine-tuned. If the adjacency relationship changes or a new micro-region is generated, local three-dimensional resource conflict detection is triggered. Only the spatial, frequency, and code domain orthogonal allocation is re-performed within the set of affected micro-regions, without redistributing all micro-regions in the entire network, thereby reducing overhead. Subsequently, at the beam control level, the target pointing angle is recalculated based on the new micro-cell boundary and peak user density. While maintaining an angle difference of ≥10° with neighboring cells, the beam is slightly adjusted to ensure the main lobe consistently covers densely populated user areas. Simultaneously, the null direction is reassessed based on the updated interference heatmap to determine if correction is needed. Finally, to avoid communication interruptions, a dual-configuration transition mechanism is employed when resource or micro-cell structure adjustments occur. This means that while the old resource configuration remains valid, a new resource mapping table is first sent to the control channel, and a smooth handover is completed within a preset handover time slot. For ongoing service sessions, short-term parallel resource allocation or buffering mechanisms ensure uninterrupted operation. Through this closed-loop mechanism of continuous sensing, local reconstruction, resource remapping, and smooth handover, fine-grained dynamic adjustments to N physically isolated micro-cells are achieved. This allows the micro-cell structure and resource configuration to match changes in user distribution in real time, maintaining orthogonality across the spatial, frequency, and code domains while improving spectrum utilization efficiency, reducing interference risks, and enhancing system capacity and service stability.

[0053] Example 2, based on the same inventive concept as the 6G spatial multiplexing-based wireless communication physical isolation method in the aforementioned examples, such as... Figure 2 As shown, this application provides a wireless communication physical isolation system based on 6G spatial multiplexing. The system and method embodiments in this application are based on the same inventive concept. The system includes: a region segmentation module 11: dynamically segmenting the target area into N physically isolated micro-regions based on the real-time user distribution density obtained through user equipment positioning and sensing; an antenna deployment module 12: deploying N multi-cavity compound eye bionic antenna units on top of each of the N physically isolated micro-regions; a network construction module 13: using the N multi-cavity compound eye bionic antenna units as N communication neuron nodes, connecting them according to a preset topology via low-latency direct links to construct a Mesh network; and a spatial spectrum calculation module 14: driving the N multi-cavity compound eye bionic antenna units to collect multi-dimensional signal data and perform spatial spectrum calculation. Inter-spectral calculation yields N interference heatmaps; Resource matching module 15: Uploads the N interference heatmaps to the Mesh network, performs orthogonal allocation of air interface resources based on the collaboration of the N communication neuron nodes, and outputs N orthogonal resource allocation schemes, wherein the orthogonal resource allocation schemes include spatial beam pointing angle, frequency subcarriers, and code domain pilot sequences; Isolation micro-cell adjustment module 16: During the communication service process of the N physically isolated micro-cells using the N orthogonal resource allocation schemes, performs fine-grained dynamic adjustment of the N physically isolated micro-cells according to the dynamic changes in the real-time user distribution density.

[0054] Furthermore, the region segmentation module 11 also includes:

[0055] A millimeter-wave antenna array pre-deployed in the target area transmits a multi-beam scanning signal to the target area to receive multipath signals reflected by user equipment, and the original signal dataset is obtained by summarizing the signals. After calculating the signal time-of-flight difference and carrier phase offset of the original signal dataset, the direct path signal features are extracted by a multipath separation algorithm, and the direct path feature parameter set is output. The angle of arrival and time difference of arrival are jointly calculated on the direct path feature parameter set, and the three-dimensional coordinate set of user equipment is output. After generating the real-time user distribution density based on the three-dimensional coordinate set of user equipment, the target area is divided into the N physically isolated micro-regions based on the density threshold and physical isolation constraints.

[0056] Furthermore, the region segmentation module 11 also includes:

[0057] The target region is divided into multiple regional grids based on a preset reference grid scale; multiple grid-level user densities are obtained by projecting the three-dimensional coordinate set of the user equipment onto the multiple regional grids, which constitute the real-time user distribution density; the multiple regional grids are subjected to density-driven dynamic aggregation and splitting based on density thresholds and physical isolation constraints to obtain the N physically isolated micro-regions.

[0058] Furthermore, the network construction module 13 also includes:

[0059] The network topology is configured based on the regional adjacency and spatial distribution characteristics of the N physically isolated micro-regions, wherein a ring topology is deployed in high-user-density areas and a star topology is deployed in boundary areas. Based on the network topology, dual-channel physical links are configured between the N communication neuron nodes to complete the construction of the underlying transmission architecture. A millimeter-wave wireless channel is enabled when the node spacing is less than or equal to a preset distance threshold, and a fiber optic wired channel is enabled when the distance exceeds the preset threshold. The N communication neuron nodes are equipped with N virtual channels for transmitting interference heatmaps. A layered protocol stack is loaded into the underlying transmission architecture to complete the construction of the Mesh network. The layered protocol stack includes a transport layer time-sensitive network protocol, a network layer distributed hash table addressing protocol, and a link layer binding protocol.

[0060] Furthermore, the spatial spectrum calculation module 14 also includes:

[0061] The user signal strength, multipath signal phase difference, and subcarrier noise floor are collected within the first physical isolation micro-region by several independently controllable miniature radiators in the first multi-cavity compound eye bionic antenna unit. Distributed signal leakage intensity detection of adjacent physical isolation micro-regions is performed by probe antennas deployed at the boundary of the first physical isolation micro-region. Based on the user signal strength, multipath signal phase difference, and subcarrier noise floor, a spatial spectrum estimation algorithm is executed to generate a first interference heat map containing the interference direction distribution and interference intensity matrix.

[0062] Furthermore, the spatial spectrum calculation module 14 also includes:

[0063] Using the N interference thermal maps as the basis for beam control, the N multi-cavity compound eye bionic antenna elements are controlled by a beamforming algorithm to generate highly directional main lobe beams toward the N physically isolated micro-regions, and spatial nulls are formed in the direction of adjacent micro-regions.

[0064] Furthermore, the spatial spectrum calculation module 14 also includes:

[0065] The center-to-center distance between adjacent physical isolation micro-regions in the N physical isolation micro-regions is greater than the beam sidelobe interference threshold, and the micro radiator achieves sidelobe attenuation through a metal cavity waveguide structure.

[0066] Furthermore, the resource matching module 15 also includes:

[0067] The N communication neuron nodes upload the N interference heatmaps to the Mesh network through the N virtual channels, perform heatmap aggregation, and generate a global interference matrix. Based on the global interference matrix, perform pairwise adjacent micro-region three-dimensional resource conflict detection to obtain N conflict identifier sets, wherein the three-dimensional resource conflict detection covers spatial domain conflict detection, frequency domain conflict detection, and code domain conflict detection. Iterate the conflict resolution and resource mapping of the N conflict identifier sets until the N orthogonal resource allocation schemes are output.

[0068] Furthermore, the resource matching module 15 also includes:

[0069] Based on a preset conflict resolution priority, the N conflict identifier sets are decomposed to obtain spatial conflict micro-region groups, frequency conflict micro-region groups, and code conflict micro-region groups. Beam pointing angle adjustment, subcarrier reallocation, and pilot sequence reconstruction are performed on the spatial conflict micro-region groups, frequency conflict micro-region groups, and code conflict micro-region groups, respectively, to obtain spatial resolution schemes, frequency resolution schemes, and code resolution schemes. After restoring the spatial resolution schemes, frequency resolution schemes, and code resolution schemes, conflict backtracking verification is performed, and conflict resolution iteration is triggered based on the verification results until N conflict-free resource allocation schemes are obtained. Resource mapping is performed on the N conflict-free resource allocation schemes to obtain the N orthogonal resource allocation schemes.

[0070] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0071] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0072] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A method for physical isolation in wireless communication based on 6G spatial multiplexing, characterized in that, The method includes: Based on the real-time user distribution density obtained through user equipment location sensing, the target area is dynamically divided into N physically isolated micro-zones; N multi-cavity compound eye bionic antenna elements are deployed on the top of the N physically isolated micro-regions respectively; The N multi-cavity compound eye bionic antenna units are used as N communication neuron nodes and connected according to a preset topology through low-latency direct connection links to construct a Mesh network; The N multi-cavity compound eye bionic antenna units are driven to collect multi-dimensional signal data, perform spatial spectrum calculation, and obtain N interference heat maps; Upload the N interference heatmaps to the Mesh network, and perform orthogonal allocation of air interface resources based on the collaboration of the N communication neuron nodes, outputting N orthogonal resource allocation schemes, wherein the orthogonal resource allocation schemes include spatial beam pointing angle, frequency subcarrier and code domain pilot sequence; During the communication service of the N physically isolated micro-cells using the N orthogonal resource allocation schemes, the N physically isolated micro-cells are dynamically adjusted in a fine-grained manner according to the dynamic changes in the real-time user distribution density.

2. The wireless communication physical isolation method based on 6G spatial multiplexing as described in claim 1, characterized in that, Referring to the real-time user distribution density obtained through user equipment location sensing, the target area is dynamically divided into N physically isolated micro-regions. The method includes: By using a millimeter-wave antenna array pre-deployed in the target area, a multi-beam scanning signal is transmitted to the target area to receive multipath signals reflected by user equipment and summarize them to obtain the original signal dataset. After calculating the signal time-of-flight difference and carrier phase offset of the original signal dataset, the direct path signal features are extracted by the multipath separation algorithm, and the direct path feature parameter set is output. The arrival angle and arrival time difference are jointly calculated on the direct trajectory characteristic parameter set to output the user equipment three-dimensional coordinate set; After generating the real-time user distribution density based on the user equipment's three-dimensional coordinate set, the target region is divided into the N physically isolated micro-regions based on the density threshold and physical isolation constraints.

3. The wireless communication physical isolation method based on 6G spatial multiplexing as described in claim 2, characterized in that, After generating the real-time user distribution density based on the user equipment's three-dimensional coordinate set, the target region is segmented into the N physically isolated micro-regions based on a density threshold and physical isolation constraints. The method includes: The target region is divided into multiple regional grids based on a preset reference grid scale; By projecting the three-dimensional coordinate set of the user equipment onto the multiple regional grids, multiple grid-level user densities are obtained, which constitute the real-time user distribution density; Based on density thresholds and physical isolation constraints, the multiple regional meshes are subjected to density-driven dynamic aggregation and splitting to obtain the N physically isolated micro-regions.

4. The wireless communication physical isolation method based on 6G spatial multiplexing as described in claim 1, characterized in that, The method involves using the N multi-cavity compound eye bionic antenna elements as N communication neuron nodes, connecting them according to a preset topology via low-latency direct links to construct a mesh network. The network topology is configured based on the regional adjacency relationship and spatial distribution characteristics of the N physically isolated micro-regions, wherein a ring topology is deployed in high user density areas and a star topology is deployed in boundary areas. Based on the network topology, a dual-channel physical link is configured between the N communication neuron nodes to complete the construction of the underlying transmission architecture. When the node spacing is less than or equal to a preset distance threshold, a millimeter-wave wireless channel is enabled; when the spacing is greater than the preset distance threshold, a fiber optic wired channel is enabled. The N communication neuron nodes are deployed with N virtual channels for transmitting interference heat maps. A layered protocol stack is loaded into the underlying transmission architecture to complete the construction of the Mesh network. The layered protocol stack includes a transport layer time-sensitive network protocol, a network layer distributed hash table addressing protocol, and a link layer binding protocol.

5. The wireless communication physical isolation method based on 6G spatial multiplexing as described in claim 1, characterized in that, The method involves driving the N multi-cavity compound eye bionic antenna elements to acquire multi-dimensional signal data, performing spatial spectrum calculations, and obtaining N interference heat maps. The user signal strength, multipath signal phase difference, and subcarrier noise floor are collected in the first physically isolated micro-region by several independently controllable miniature radiators in the first multi-cavity compound eye bionic antenna unit. Distributed signal leakage intensity detection of adjacent physically isolated micro-regions is performed using probe antennas deployed at the boundary of the first physically isolated micro-region. Based on the user signal strength, multipath signal phase difference, and subcarrier noise basis, a spatial spectrum estimation algorithm is executed to generate a first interference heat map containing the interference direction distribution and interference intensity matrix.

6. The wireless communication physical isolation method based on 6G spatial multiplexing as described in claim 1, characterized in that, The method further includes: Using the N interference thermal maps as the basis for beam control, the N multi-cavity compound eye bionic antenna elements are controlled by a beamforming algorithm to generate highly directional main lobe beams toward the N physically isolated micro-regions, and spatial nulls are formed in the direction of adjacent micro-regions.

7. The wireless communication physical isolation method based on 6G spatial multiplexing as described in claim 4, characterized in that, Upload the N interference heatmaps to the Mesh network, perform orthogonal allocation of air interface resources based on the collaboration of the N communication neuron nodes, and output N orthogonal resource allocation schemes. The method includes: The N communication neuron nodes upload the N interference heatmaps to the Mesh network through the N virtual channels, perform heatmap aggregation, and generate a global interference matrix; Based on the global interference matrix, three-dimensional resource conflict detection is performed on each pair of adjacent micro-regions to obtain N conflict identifier sets, wherein the three-dimensional resource conflict detection covers spatial domain conflict detection, frequency domain conflict detection and code domain conflict detection. The conflict resolution and resource mapping are iterated on the N conflict identifier sets until the N orthogonal resource allocation schemes are output.

8. The wireless communication physical isolation method based on 6G spatial multiplexing as described in claim 7, characterized in that, The method involves iterative conflict resolution and resource mapping on the N conflict identifier sets until the N orthogonal resource allocation schemes are output. Based on the preset conflict resolution priority, the N conflict identifier sets are decomposed to obtain spatial domain conflict micro-region groups, frequency domain conflict micro-region groups and code domain conflict micro-region groups. Beam pointing angle adjustment, subcarrier reallocation, and pilot sequence reconstruction are performed on the spatial conflict micro-cell group, frequency conflict micro-cell group, and code conflict micro-cell group, respectively, to obtain the spatial resolution scheme, frequency resolution scheme, and code resolution scheme; After restoring the spatial domain resolution scheme, frequency domain resolution scheme, and code domain resolution scheme, conflict backtracking verification is performed, and conflict resolution iteration is triggered based on the verification results until N conflict-free resource allocation schemes are obtained. Resource mapping is performed on the N conflict-free resource allocation schemes to obtain the N orthogonal resource allocation schemes.

9. The wireless communication physical isolation method based on 6G spatial multiplexing as described in claim 5, characterized in that, The center-to-center distance between adjacent physical isolation micro-regions in the N physical isolation micro-regions is greater than the beam sidelobe interference threshold, and the micro radiator achieves sidelobe attenuation through a metal cavity waveguide structure.

10. A wireless communication physical isolation system based on 6G spatial multiplexing, characterized in that, The system is used to execute the wireless communication physical isolation method based on 6G spatial multiplexing as described in any one of claims 1 to 9, including: Region segmentation module: Based on the real-time user distribution density obtained from user equipment positioning and sensing, the target area is dynamically segmented into N physically isolated micro-regions; Antenna deployment module: N multi-cavity compound eye bionic antenna units are deployed on top of the N physically isolated micro-regions respectively; Network construction module: The N multi-cavity compound eye bionic antenna units are used as N communication neuron nodes and connected according to a preset topology through low-latency direct links to construct a Mesh network; Spatial spectrum calculation module: drives the N multi-cavity compound eye bionic antenna units to collect multi-dimensional signal data, performs spatial spectrum calculation, and obtains N interference heat maps; Resource matching module: Uploads the N interference heatmaps to the Mesh network, performs orthogonal allocation of air interface resources based on the collaboration of the N communication neuron nodes, and outputs N orthogonal resource allocation schemes, wherein the orthogonal resource allocation schemes include spatial beam pointing angle, frequency subcarrier and code domain pilot sequence; Isolation micro-cell adjustment module: During the communication service of the N physically isolated micro-cells using the N orthogonal resource allocation schemes, the module performs fine-grained dynamic adjustment of the N physically isolated micro-cells based on the dynamic changes in the real-time user distribution density.

Images