Ultra-large-scale MIMO channel modeling methods, devices, media and products

Abstract

This application provides a method, apparatus, medium, and product for ultra-large-scale MIMO channel modeling, relating to the field of wireless communication technology. The method includes: obtaining the cluster power of each cluster according to a preset channel model; determining the spatially non-stationary SnS state of each cluster according to the cluster power or a preset statistical distribution model; determining a first power attenuation factor for different clusters at the base station based on the SnS state of each cluster; determining a second power attenuation factor for each element of the user equipment; and generating channel coefficients based on the first and second power attenuation factors. The solution of this application provides a simple, physically meaningful, and efficient channel modeling method, effectively overcoming the shortcomings of existing technologies and providing reliable theoretical support for the design and optimization of ultra-large-scale MIMO systems.

Description

Technical Field

[0001] This application relates to the field of wireless communication technology, specifically to a method, apparatus, medium, and product for ultra-large-scale MIMO channel modeling. Background Technology

[0002] With the rapid development of mobile communication technology, research on sixth-generation mobile communication systems (6G) has gradually unfolded. 6G systems aim to provide higher spectral efficiency, greater system capacity, and more reliable communication services to meet the needs of emerging applications such as immersive communication, holographic communication, and the Internet of Things. Extremely Large-Scale Multi-Input Multi-Output (XL-MIMO), as one of the key candidate technologies for 6G, significantly increases the number of antenna elements (typically hundreds to thousands), thereby improving the system's spectral efficiency, capacity, and reliability.

[0003] However, the introduction of XL-MIMO also brings new challenges. Because XL-MIMO systems are equipped with large-scale antenna arrays, the propagation environment observed by antenna elements at different spatial locations varies significantly, leading to spatial non-stationarity (SnS) in the channel. This non-stationarity manifests in two main ways: first, objects of finite size (such as people or trees) may not completely block the entire antenna array, causing the propagation paths of some elements to be obstructed; second, finite-sized scatterers may no longer function as complete scattering sources for the entire array in a large-aperture array, resulting in the power of the scattered signal being concentrated in a portion of the array. This spatial non-stationarity places higher demands on channel modeling for XL-MIMO systems.

[0004] In existing channel modeling techniques, there are two main approaches for modeling spatial nonstationarity: the Birth and Death Process (BD) and the Visible Range (VR). The BD method simulates the appearance and disappearance of clusters along the array axis by defining the cluster generation and death rates on each antenna element. However, it has the following drawbacks: 1) The cluster "birth rate" and "death rate" are difficult to obtain from measured or simulated data; 2) The model parameters have a weak correlation with the real physical environment, making it difficult to intuitively reflect the physical phenomena in the channel; 3) When a cluster's "birth" or "death" occurs, the cluster power exhibits abrupt changes, lacking a smooth transition. The Visible Range method simulates cluster visibility by defining a region on the antenna array, but it also has the following problems: 1) The size, center point, and shape of the Visible Range depend on the propagation environment and the distribution of scatterers, making it difficult to accurately obtain the parameters; 2) At the boundary of the Visible Range, the cluster power exhibits abrupt changes, lacking a reasonable characterization of the transition region. Summary of the Invention

[0005] At least one embodiment of this application provides a method, apparatus, medium, and product for modeling ultra-large-scale MIMO channels. It addresses the problems in the prior art modeling methods mentioned above, such as the difficulty in accurately describing the spatial non-stationarity of XL-MIMO channels, which leads to difficulties in obtaining model parameters, unclear physical meaning, and sudden changes in cluster power.

[0006] To solve the above-mentioned technical problems, this application is implemented as follows:

[0007] In a first aspect, embodiments of this application provide a method for modeling ultra-large-scale MIMO channels, including:

[0008] Based on the preset channel model, obtain the cluster power of each cluster;

[0009] Based on the power of each cluster or a preset statistical distribution model, determine the spatial non-stationary SnS state of each cluster;

[0010] Based on the SnS state of each cluster, determine the first power attenuation factor for different clusters at the base station.

[0011] Determine the second power attenuation factor for each element of the user equipment;

[0012] Channel coefficients are generated based on the first power attenuation factor and the second power attenuation factor.

[0013] Optionally, based on the SnS state of each cluster, a first power attenuation factor for different clusters at the base station is determined, including:

[0014] When the SnS state indicates that it is in a non-stationary state, the visibility probability and visibility area of ​​the corresponding SnS state cluster are calculated, and the first power attenuation factor of the SnS state cluster is determined based on the visibility probability and the visibility area.

[0015] When the SnS state cluster represents a stable state, the first power attenuation factor of all array elements of the corresponding non-SnS state cluster is determined to be 1.

[0016] Optionally, based on the power of each cluster, the spatially non-stationary SnS state of each cluster is determined, including:

[0017] Based on the power of each cluster, the spatial nonstationary probability of each cluster is calculated using at least one of the following methods: multi-user joint analysis to obtain the probability distribution, single-user independent analysis to obtain the probability distribution, and cluster normalized power correlation modeling.

[0018] Based on the spatial non-stationary probability of each cluster, a binarization decision is made for each cluster to determine the spatial non-stationary SnS state of each cluster.

[0019] Optionally, the visibility probability and visible region of the corresponding SnS state cluster are calculated, including:

[0020] Determine a first ratio between the number of visible array elements in the SnS state cluster and the total number of array elements in the VLS; establish a functional relationship between the visibility probability and the cluster power based on ray tracing simulation or channel measurement data; calculate the visibility probability of the SnS state cluster according to the functional relationship; or, calculate the visibility probability of the SnS state cluster according to the preset statistical distribution model.

[0021] Based on the visibility probability, the visible area of ​​the SnS state cluster is calculated using at least one of the following methods: center-radial method, corner-triggered method, and near-field preferred method.

[0022] Optionally, determining the first power attenuation factor of the SnS state cluster based on the visibility probability and the visible area includes:

[0023] If the SnS state cluster is within the visible area, the first power attenuation factor of the SnS state cluster is determined to be 1 according to a preset spatial attenuation function;

[0024] If the SnS state cluster is outside the visible area, a first power attenuation factor value is determined by using a spatial attenuation function based on the relative position of the array element of the SnS state cluster and the boundary of the visible area, and the power attenuation factor value is smoothly attenuated to 0 with spatial coordinates; the attenuation rate of the spatial attenuation function is a preset adjustable parameter.

[0025] Optionally, a second power attenuation factor is determined for each element of the user equipment, including:

[0026] The user's grip on the device is determined based on a preset typical ratio; the grip includes three situations: head and one hand covering, both hands covering, and one hand covering.

[0027] Based on the grip condition, the attenuation value of each antenna element of the user equipment is obtained;

[0028] Based on the attenuation value, a second power attenuation factor is determined for each element of the user equipment.

[0029] Optionally, channel coefficients are generated based on the first power attenuation factor and the second power attenuation factor, including:

[0030] Based on the first power attenuation factor and the second power attenuation factor, the target formula for generating channel coefficients is input to generate channel coefficients.

[0031] Secondly, embodiments of this application provide an ultra-large-scale MIMO channel modeling apparatus, comprising:

[0032] The first acquisition module is used to acquire the cluster power of each cluster according to a preset channel model;

[0033] The first determining module is used to determine the spatial non-stationary SnS state of each cluster based on the power of each cluster or a preset statistical distribution model.

[0034] The second determining module is used to determine the first power attenuation factor of different clusters at the base station based on the SnS state of each cluster.

[0035] The first processing module is used to determine the second power attenuation factor for each array element of the user equipment.

[0036] The second processing module is used to generate channel coefficients based on the first power attenuation factor and the second power attenuation factor.

[0037] Thirdly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0038] Fourthly, embodiments of this application provide a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0039] Compared with existing technologies, the ultra-large-scale MIMO channel modeling method, apparatus, medium, and product provided in this application obtain the cluster power of each cluster according to a preset channel model; determine the spatial non-stationary SnS state of each cluster according to the cluster power or a preset statistical distribution model, and determine whether a cluster has experienced SnS, thus solving the problem of determining the spatial stationarity of clusters; based on the SnS state of each cluster, for array elements outside the visible area, determine the first power attenuation factor of different clusters at the base station and the second power attenuation factor of each array element of the user equipment, realizing the smooth evolution of cluster power at the visible area boundary of the ultra-large-scale MIMO array, solving the problem of abrupt changes in cluster power between the visible area and the array elements at the base station; this application dynamically adjusts the attenuation factor according to the SnS state of the cluster to avoid model mismatch. This application associates the power attenuation factor with the position of the user equipment (UE) array elements, solving the problem of abrupt changes in channel coefficients caused by ignoring the spatial differences between array elements in traditional methods, and achieves smooth transition of channel parameters through refined array element-level mapping, enhancing the physical interpretability of the model. Attached Figure Description

[0040] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0041] Figure 1 A schematic diagram of the ultra-large-scale MIMO channel modeling method provided in the embodiments of this application;

[0042] Figure 2 A flowchart illustrating the overall process of the ultra-large-scale MIMO channel modeling method provided in this application embodiment;

[0043] Figure 3 A schematic diagram of the spatial non-stationary probability distribution of a cluster provided in an embodiment of this application;

[0044] Figure 4 A schematic diagram of the spatial nonstationary probability function of a cluster provided in an embodiment of this application;

[0045] Figure 5 A schematic diagram of the probability density function of VP provided in the embodiments of this application;

[0046] Figure 6 This is a structural diagram of the ultra-large-scale MIMO channel modeling device provided in an embodiment of this application. Detailed Implementation

[0047] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, without limiting the number of objects; for example, the first object can be one or more. Furthermore, "or" in this application indicates at least one of the connected objects. For example, "A or B" covers three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0048] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as one in which the sender explicitly informs the receiver of specific information, the operation to be performed, or the requested result, etc.; an indirect instruction can be understood as one in which the receiver determines the corresponding information based on the instruction sent by the sender, or makes a judgment and determines the operation to be performed or the requested result, etc., based on the judgment result.

[0049] As described in the background section, existing modeling methods suffer from difficulties in accurately describing the spatial nonstationarity of XL-MIMO channels, leading to problems such as difficulty in obtaining model parameters, unclear physical meaning, and sudden changes in cluster power. To address at least one of these problems, this application provides a method, apparatus, medium, and product for modeling ultra-large-scale MIMO channels, which has the advantages of simple implementation and practical physical meaning, and is expected to overcome the problems existing in existing methods.

[0050] Please refer to Figure 1 This application provides a method for modeling a very large-scale MIMO channel, comprising:

[0051] Step 11: Obtain the cluster power of each cluster according to the preset channel model.

[0052] In this embodiment of the application, the multipath channel is decomposed into several clusters according to the cluster generation rules defined in the relevant standard protocol. Each cluster corresponds to a group of multipath components with similar delay and angle characteristics.

[0053] Specifically, using a preset channel model, the cluster power of each cluster is calculated using the formula for normalized relative power (Formula (1)).

[0054] , formula (1);

[0055] Among them, P n 'Indicates the current cluster power determined using the formula for normalized relative power; Let be the known delay of the nth cluster, DS be the delay spread value in 3GPP TR 38.901, and rτ be the delay distribution scaling factor. This represents the shadow fading term of the nth cluster, in [dB].

[0056] Furthermore, the current cluster power is normalized so that the sum of all cluster powers equals 1, thereby determining the cluster power P. n That is, the cluster power P is determined using formula (2). n .

[0057] , formula (2);

[0058] Step 12: Determine the spatial non-stationary SnS state of each cluster based on the power of each cluster or a preset statistical distribution model.

[0059] In this embodiment, based on ray tracing simulation, multiple users are simulated, each user corresponding to several clusters. Some of these clusters are spatially non-stationary, and some are spatially non-stationary. Therefore, a spatially non-stationary probability can be obtained for each user. The spatially non-stationary probabilities of all users are statistically fitted to a distribution, indicating that they may follow a normal distribution. A preset statistical distribution model is determined based on the normal statistical distribution for each user. The spatially non-stationary probability of users can be randomly generated using the preset statistical distribution model.

[0060] Step 13: Determine the first power attenuation factor of different clusters at the base station based on the SnS state of each cluster;

[0061] Step 14: Determine the second power attenuation factor for each element of the user equipment;

[0062] Step 15: Generate channel coefficients based on the first power attenuation factor and the second power attenuation factor.

[0063] In this embodiment, multipath clusters are extracted using a clustering algorithm based on a preset channel model (e.g., deterministic channel model 3), and the power distribution of each cluster is calculated. This power is formed by the superposition of multipath signals within the cluster, reflecting the characteristics of the local propagation environment. Based on the cluster power or preset statistical distribution from step 1, the spatial nonstationarity (SnS) of the clusters is analyzed. For example, the stability of the clusters is determined by statistical model parameters (e.g., delay spread, angle spread), distinguishing between static and dynamic propagation scenarios. Based on the SnS state of each cluster, combined with the spatial correlation of the base station antenna array (e.g., polarization modeling, antenna spacing influence), the path loss factor of different clusters is calculated. For example, a steeper attenuation model is used in high-correlation scenarios. For each element of the user equipment (UE) (e.g., millimeter-wave MIMO antenna), the path loss and shadow fading at the element level are modeled based on the propagation environment (e.g., scatterer distribution, multipath density) and the device's mobility characteristics. By combining the first attenuation factor at the base station and the second attenuation factor at the user end, along with multipath parameters (such as delay, angle, and polarization), the final channel coefficient matrix is ​​generated through the channel model for link-level simulation or system performance evaluation.

[0064] Optionally, step 13 above includes:

[0065] When the SnS state indicates that it is in a non-stationary state, the visibility probability and visibility area of ​​the corresponding SnS state cluster are calculated, and the first power attenuation factor of the SnS state cluster is determined based on the visibility probability and the visibility area.

[0066] When the SnS state cluster represents a stable state, the first power attenuation factor of all array elements of the corresponding non-SnS state cluster is determined to be 1.

[0067] In this embodiment, the SnS state includes both a non-stationary state and a stationary state. The processing when the SnS state is non-stationary: When a cluster is determined to be in a spatially non-stationary (SnS) state, its visibility probability and visible area need to be dynamically calculated. The first power attenuation factor within the visible area is set to 1, and the first power attenuation factor outside the visible area is set to a roll-off attenuation value. The further away an element is from the visible area, the closer its first power attenuation factor is to 0. Visibility probability represents the ratio of the number of visible elements of a cluster in a large-scale MIMO array to the total number of elements. Visible area refers to the specific observable region of the cluster in a very large-scale MIMO array.

[0068] Handling of stationary state (non-SnS cluster): When the cluster is in a stationary state, the cluster is visible at all elements of the ultra-large scale MIMO, and the first power attenuation factor of all elements is 1.

[0069] Optionally, step 12 of this application includes:

[0070] Based on the power of each cluster, the spatial nonstationary probability of each cluster is calculated using at least one of the following methods: multi-user joint analysis to obtain the probability distribution, single-user independent analysis to obtain the probability distribution, and cluster normalized power correlation modeling.

[0071] Based on the spatial non-stationary probability of each cluster, a binarization decision is made for each cluster to determine the spatial non-stationary state of each cluster.

[0072] In this embodiment, the specific steps for calculating the spatial non-stationary probability of each cluster are as follows: Define the SnS probability to model whether a cluster is non-stationary. Assume the total number of elements in the base station (BS) XL-MIMO array is N. all The number of visible array elements in a certain cluster is N n If a cluster is visible to some elements of the array (i.e., N elements are visible to others), then... n <N all If a cluster is visible to all elements in the array (N), then the cluster is spatially nonstationary; conversely, if a cluster is visible to all elements in the array (N), then the cluster is spatially nonstationary. n =N all If the cluster is not spatially stationary, then it is not spatially nonstationary.

[0073] Based on ray tracing simulation or channel measurement data, the SnS probability of the cluster is calculated using formula (3). :

[0074] , formula (3);

[0075] in, This represents the SNS probability of a cluster, where D is a certain probability distribution type (such as normal distribution, log-normal distribution, etc.). and This represents the mean and variance of the distribution. express It can also be a function related to cluster normalized power.

[0076] The specific implementation methods include the following three technical approaches:

[0077] Method 1 (Probability distribution obtained from multi-user joint analysis): Multi-user data preprocessing: The normalized power of clusters detected by all users in the target area is sorted in descending order; Power domain segmentation: The sorted cluster sequence is divided into K equal power intervals; Probability statistics calculation: In each power interval, the probability of a cluster satisfying the SnS condition is calculated, that is, the probability of a SnS cluster being the total number of clusters in that power interval; Distribution fitting: Maximum likelihood estimation is performed on the datasets in the K power intervals to verify that they follow a specific probability distribution type (such as normal distribution, log-normal distribution, etc.).

[0078] Method 2 (Probability distribution obtained by independent analysis of a single user): Calculation of SnS probability for a single user cluster: Calculate the SnS probability based on the cluster data of user i: that is, the ratio of the number of non-stationary clusters of user i to the total number of clusters detected by user i; Multi-user distribution modeling: Perform maximum likelihood estimation on the SnS probability dataset of I users to verify that it follows a specific probability distribution type (such as normal distribution, log-normal distribution, etc.).

[0079] Method 3 (Cluster Normalized Power Correlation Modeling): Global Data Integration: Aggregate detection clusters from all users, normalize the power of each cluster and sort them in descending order to construct a unified power sequence across users; Dynamic Power Segmentation: Divide the power value into K continuous intervals, and the median of each interval must satisfy the statistical distribution constraints of the power value in that interval; Power Interval Probability Calculation: Calculate the proportion of clusters with SnS characteristics in each power interval as the SnS probability observation value for that power segment; Functional Relationship Fitting: Use weighted regression analysis based on sample size to establish a continuous function model of SnS probability changing with normalized power, including but not limited to: exponential decay model, polynomial model, and piecewise linear model.

[0080] Finally, based on the cluster power Substitute into formula (3) to calculate the SnS probability of the cluster.

[0081] Based on the spatial nonstationary probability of each cluster, a binarized decision is made for each cluster, and the spatial nonstationary state of each cluster is determined using formula (4): that is:

[0082] , formula (4);

[0083] in, This indicates that the cluster has produced SnS; otherwise, SnS has not been produced. x is a randomly generated number that follows a certain pattern. .

[0084] Optionally, the visibility probability and visible region of the corresponding SnS state cluster are calculated, including:

[0085] Determine a first ratio between the number of visible array elements in the SnS state cluster and the total number of array elements in the VLS; establish a functional relationship between the visibility probability and the cluster power based on ray tracing simulation or channel measurement data; calculate the visibility probability of the SnS state cluster according to the functional relationship; or, calculate the visibility probability of the SnS state cluster according to the preset statistical distribution model.

[0086] Based on the visibility probability, the visible area of ​​the SnS state cluster is calculated using at least one of the following methods: center-radial method, corner-triggered method, and near-field preferred method.

[0087] In this embodiment of the application, the specific steps for calculating the visible probability (VP) of a SnS-state cluster are as follows: Define the number N of visible array elements in the SnS-state cluster. n The total number of array elements N in a very large-scale multiple-input multiple-output system all The first ratio. Based on ray tracing simulation or channel measurement data, a function of VP is established, expressed by formula (5): V=f(p), where V is a function of cluster power P; the preset statistical distribution model is expressed as V:Y(0,1), where V follows a statistical distribution of the form Y, which is greater than 0 and less than 1.

[0088] According to formula (5) or a preset statistical distribution model, the VP of cluster n, i.e., V, can be generated. n ,Right now: .

[0089] Where Pn is the power of cluster n, and x is a randomly generated random number that follows the law of... .

[0090] It should be noted that generating visible probabilities using the preset statistical distribution model requires "using a random number generator to produce random variables that follow a uniform distribution." Generating visible probabilities using functional relationships can be done directly based on the dependency between cluster power and the visible probability function.

[0091] For a cluster of SnS states, the specific steps for generating the visible range (VR) of the SnS state cluster are as follows: based on its visibility probability... The visible region (VR) of the SnS state cluster can be calculated using at least one of the following methods: center-radial method, corner-triggered method, and near-field preferred method.

[0092] In one implementation, the visible region of the SnS state cluster is calculated using a center-radial method, including:

[0093] In N all An initial element is randomly selected from the array elements as the geometric center; the length of the VR rectangle is randomly generated, with its minimum and maximum values ​​being respectively... and It follows a uniform distribution within this range. The number of covering array elements reaches N. n =V n ×N all Given the length of the rectangle, generate the width of rectangle VR. Based on the width and length of VR, expand outwards in both directions with the initial array element as the geometric center until the required number of visible array elements is met; if one end reaches the array boundary during bidirectional expansion, then perform unidirectional expansion.

[0094] In another implementation, a corner-triggered method is used to calculate the visible area of ​​the SnS state cluster, including: randomly selecting an expansion starting point from the four corner elements of the array, or selecting an expansion starting point based on a certain probability (for example, for a horizontal uniform planar array, the probability of selecting left and right vertices is 50%, and the probability of selecting top and bottom vertices is 80% and 20%, respectively); performing directional area expansion: randomly generating the length of the VR rectangle, with its minimum and maximum values ​​being respectively... and It follows a uniform distribution within this range; based on the number of covering array elements reaching N n =V n ×N all The width of rectangle VR is generated by combining the length of the rectangle with the width of the rectangle. The area enclosed by the length and width of the rectangle is VR.

[0095] In another implementation, a near-field preferred method is used to calculate the visible area of ​​the SnS state cluster, including: calculating the distance between all array elements and the user; selecting the array element with the smallest distance among all elements as the center and performing radial expansion; when the cumulative number of array elements N... n =V n ×N all At that point, the expansion stops, and the last complete circular area serves as the VR boundary.

[0096] Optionally, determining the first power attenuation factor of the SnS state cluster based on the visibility probability and the visible area includes:

[0097] If the SnS state cluster is within the visible area, the first power attenuation factor of the SnS state cluster is determined to be 1 according to a preset spatial attenuation function;

[0098] If the SnS state cluster is outside the visible area, a first power attenuation factor value is determined by using a spatial attenuation function based on the relative position of the array element of the SnS state cluster and the boundary of the visible area, and the power attenuation factor value is smoothly attenuated to 0 with spatial coordinates; the attenuation rate of the spatial attenuation function is a preset adjustable parameter.

[0099] In this embodiment, for a cluster in the SnS state, the calculation of the first power attenuation factor (PAF) for different elements of the XL-MIMO is as follows: within the VR (Vibration Zone), the cluster power remains constant, and PAF = 1; outside the VR, the cluster PAF is modeled as a function that slowly decreases to 0 between XL-MIMO elements to ensure a smooth transition between the visible and invisible zones. The function for establishing the PAF can be expressed as:

[0100] ; where f(x,y) represents a function of PAF, which is a function of VR in the horizontal (x) and vertical (y) directions in a uniform planar matrix coordinate system.

[0101] Optionally, step 14 above includes:

[0102] The user's grip on the device is determined based on a preset typical ratio; the grip includes three situations: head and one hand covering, both hands covering, and one hand covering.

[0103] Based on the grip condition, the attenuation value of each antenna element of the user equipment is obtained;

[0104] Based on the attenuation value, a second power attenuation factor is determined for each element of the user equipment.

[0105] In this embodiment, the physical occlusion scenarios of the user equipment (UE) are identified, including three typical cases: a first state of head and single-hand occlusion (such as the device being close to the head and held with one hand during a call); a second state of double-hand occlusion (such as both hands covering the sides of the device during landscape gaming); and a third state of single-hand occlusion (such as the single-hand holding causing part of the antenna to be blocked by the palm). For example, the probability of occurrence of the three states is set to 13%, 29%, and 58%. During the generation process, a value in the range of 0-1 is randomly generated, i.e., x~U(0,1). It is determined which range x falls into: 0-13%, 13%-42%, or 42%-100%, and that is the state it belongs to. Based on the state corresponding to the physical occlusion scenario, a preset table is consulted to determine the fixed attenuation factor of each antenna element of the UE, and it is converted to a linear value (i.e., falling between 0 and 1). Therefore, this application provides a processing method for obtaining the attenuation value of each antenna element of the user equipment based on the holding situation and using a lookup table.

[0106] Optionally, step 15 above includes:

[0107] Based on the first power attenuation factor and the second power attenuation factor, the target formula for generating channel coefficients is input to generate channel coefficients.

[0108] In this embodiment, the target formula is preferably the "target formula of the three-dimensional wireless channel model of 3GPP TR 38.901". The first power attenuation factor and the second power attenuation factor are input into the target formula for generating channel coefficients to obtain the generated channel coefficients. Here, the channel coefficients are preferably generated for line-of-sight (LOS) and non-line-of-sight (NLOS).

[0109] The formula for LOS is expressed as follows: ;

[0110] The formula for NLOS is expressed as: ;

[0111] Where, in the formula, and These represent the impulse responses of LOS and NLOS conditional channels, respectively. t represents the propagation delay and the time of propagation; Represents Rice's K-factor; N represents the total number of clusters. and This represents the impulse response of the LOS cluster and the nth NLOS cluster. All of the above parameters can be obtained from the "3GPP TR 38.901 Three-Dimensional Wireless Channel Model". The first power attenuation factor between the cluster and the base station antenna element is represented, where s represents the base station antenna element index and n represents the cluster index. This represents the second power attenuation factor on the user side (in some cases, the second power attenuation factor is the antenna element attenuation value), and u represents the UE antenna element index.

[0112] Reference Figure 2 As shown in the illustration, this application provides a specific flowchart, including the following steps:

[0113] Step 1, obtain cluster power;

[0114] Step 2, calculate the SnS probability of the cluster;

[0115] Step 3, determine the SnS state of the cluster;

[0116] Step 4: For clusters that are not in the SnS state, set the PAF of all XL-MIMO array elements to 1;

[0117] Step 5: For a SnS-state cluster, calculate the VP of the SnS-state cluster;

[0118] Step 6: For SnS-state clusters, generate the VR of the SnS-state clusters;

[0119] Step 7: For clusters in SnS state, calculate the PAF of different array elements;

[0120] Step 8: Generate PAFs for different array elements of the UE;

[0121] Step 9: Substitute the PAF values ​​to generate channel coefficients.

[0122] In a specific embodiment of the spatial non-stationary channel modeling scheme based on ray tracing simulation data, the specific simulation process and parameters are as follows:

[0123] Step 1, Obtain Cluster Power. In this step, the cluster power is generated according to Step 6 and Step 12 in "3GPP TR 38.901". .

[0124] Step 2, calculate the SnS probability (Pr) of the cluster. SNS Based on ray tracing simulation results, Pr SNS The proportion of SNS clusters in all samples within a specific power range is calculated. Since the clusters are unevenly distributed within the linear power domain (most clusters have low power, a few clusters have high power), the spatial non-stationary probability of clusters within the specific power range is statistically analyzed in the dB domain of max(Pn)−Pn (where Pn is the power of the nth cluster, in dB). (Refer to...) Figure 3 As shown, it can be observed that the SNS probability of a cluster follows a truncated normal distribution. Therefore, the generation of the SNS probability of a single cluster can be expressed as: .

[0125] Among them, Pr SNS This represents the probability that the nth cluster has the SNS property. This probability follows a truncated normal distribution with mean and variance , and its value range is limited to the interval [0,1].

[0126] Furthermore, a functional relationship can be established between the nonstationary probability of cluster space and cluster power, such as... Figure 4 As shown.

[0127] Step 3: Determine the SnS state of the cluster. This is based on the cluster's power. The probability of SnS generated for a cluster is used to determine whether the cluster generates SnS.

[0128] Step 4: For clusters not in the SnS state, set the PAF of all XL-MIMO elements to 1; if Then, the PAF of all array elements in XL-MIMO is set to 1, that is... .

[0129] Step 5: For clusters of SnS states, calculate the VP of the SnS state cluster. One option is to generate the VP based on the statistical distribution of the simulation data, referring to... Figure 5 As shown, it was found that as VP increases, its probability density gradually decreases, approximately following an exponential distribution. The mean of its distribution is .

[0130] Step 6: For SnS-state clusters, generate the VR of the SnS-state clusters;

[0131] Step 7: For clusters in SnS state, calculate the PAF of different elements in XL-MIMO. One option is to generate the PAF of different elements in XL-MIMO according to the following formula.

[0132] .

[0133] in, , These are the coordinates of the other corner point on the diagonal of the antenna array (VR reference corner point is...). ); These are the coordinates of the other corner point on the diagonal of the visible area (VR) of the rectangle. This represents the minimum distance between the s-th antenna element and the visible region (VR); C is the roll-off factor between the visible and invisible regions. Different forms of visible regions and slow changes in power between array elements can be achieved by adjusting the roll-off factor, ensuring the consistency of power changes between array elements.

[0134] Step 8: Generate PAFs for different array elements of the UE;

[0135] Step 9: Generate a channel with spatial non-stationary effects by combining the PAF on the base station side and the PAF on the UE side according to the formulas for LOS and NLOS.

[0136] In summary, this application proposes a stochastic-based approach for spatial non-stationarity modeling (SA-SnS), which solves the problem of determining the spatial stationarity of clusters by introducing the SnS probability of power-dependent clusters.

[0137] The SA-SnS method proposed in this application, targeting SnS clusters, solves the problem of obtaining parameters in VR and BD methods by defining VR through VP. The SA-SnS method introduces a distance-dependent power attenuation factor for array elements outside VR, achieving smooth evolution of cluster power at the visible region boundary and resolving the problem of abrupt cluster power changes in VR and BD methods. Furthermore, the SA-SnS method represents a forward extension of the existing 3GPP TR 38.901 channel model, requiring only the addition of a power attenuation factor to model spatial non-stationary effects in the original channel coefficient generation.

[0138] The various methods described above are based on embodiments of this application. Apparatus for implementing the above methods will now be provided.

[0139] Reference Figure 6 As shown, this application also provides a very large-scale MIMO channel modeling apparatus, comprising:

[0140] The first acquisition module 61 is used to acquire the cluster power of each cluster according to a preset channel model;

[0141] The first determining module 62 is used to determine the spatial non-stationary SnS state of each cluster based on the power of each cluster or a preset statistical distribution model.

[0142] The second determining module 63 is used to determine the first power attenuation factor of different clusters at the base station based on the SnS state of each cluster.

[0143] The first processing module 64 is used to determine the second power attenuation factor for each array element of the user equipment.

[0144] The second processing module 65 is used to generate channel coefficients based on the first power attenuation factor and the second power attenuation factor.

[0145] Optionally, the second determining module 63 includes:

[0146] The first determining unit is configured to calculate the visibility probability and visible area of ​​the corresponding SnS state cluster when the SnS state representation is in a non-stationary state, and determine the first power attenuation factor of the SnS state cluster based on the visibility probability and the visible area.

[0147] The second determining unit is used to determine the first power attenuation factor of all array elements of the corresponding non-SnS state cluster as 1 when the SnS state cluster is in a stable state.

[0148] Optionally, the second determining module 63 includes:

[0149] The third determining unit is used to calculate the spatial nonstationary probability of each cluster based on the power of each cluster using at least one of the following methods: obtaining the probability distribution through multi-user joint analysis, obtaining the probability distribution through single-user independent analysis, and cluster normalized power correlation modeling.

[0150] The fourth determining unit is used to perform a binarization decision on each cluster based on the spatial non-stationary probability of each cluster, and determine the spatial non-stationary SnS state of each cluster.

[0151] Optionally, the first determining unit is specifically used for:

[0152] Determine a first ratio between the number of visible array elements in the SnS state cluster and the total number of array elements in the VLS; establish a functional relationship between the visibility probability and the cluster power based on ray tracing simulation or channel measurement data; calculate the visibility probability of the SnS state cluster according to the functional relationship; or, calculate the visibility probability of the SnS state cluster according to the preset statistical distribution model.

[0153] Based on the visibility probability, the visible area of ​​the SnS state cluster is calculated using at least one of the following methods: center-radial method, corner-triggered method, and near-field preferred method.

[0154] Optionally, the first determining unit is further specifically used for:

[0155] If the SnS state cluster is within the visible area, the first power attenuation factor of the SnS state cluster is determined to be 1 according to a preset spatial attenuation function;

[0156] If the SnS state cluster is outside the visible area, a first power attenuation factor value is determined by using a spatial attenuation function based on the relative position of the array element of the SnS state cluster and the boundary of the visible area, and the power attenuation factor value is smoothly attenuated to 0 with spatial coordinates; the attenuation rate of the spatial attenuation function is a preset adjustable parameter.

[0157] Optionally, the first processing module 64 includes:

[0158] The first processing unit is used to determine the user's grip on the device according to a preset typical ratio; the grip includes three situations: head and one hand covering, both hands covering, and one hand covering.

[0159] The acquisition unit is used to acquire the attenuation value of each antenna element of the user equipment based on the holding situation;

[0160] The second processing unit is used to determine the second power attenuation factor for each array element of the user equipment based on the attenuation value.

[0161] Optionally, the second processing module 65 includes:

[0162] The third processing unit is used to generate channel coefficients by inputting a target formula for generating channel coefficients based on the first power attenuation factor and the second power attenuation factor.

[0163] It should be noted that the device in this embodiment corresponds to the device applied to the method described above. The implementation methods in each of the above embodiments are also applicable to the embodiments of this device and can achieve the same technical effect. The device provided in this application embodiment can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0164] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the above-described embodiments of the ultra-large-scale MIMO channel modeling method and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0165] This application also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, they implement the various processes of the above-described embodiment of the ultra-large-scale MIMO channel modeling method and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0166] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0167] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0168] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for modeling ultra-large-scale MIMO channels, characterized in that, include: Based on the preset channel model, obtain the cluster power of each cluster; Based on the power of each cluster or a preset statistical distribution model, the spatial non-stationary SnS state of each cluster is determined; wherein, determining the spatial non-stationary SnS state of each cluster based on the power of each cluster includes: calculating the spatial non-stationary probability of each cluster using at least one of the following methods: obtaining a probability distribution through multi-user joint analysis, obtaining a probability distribution through single-user independent analysis, and cluster normalized power correlation modeling; and performing a binarized decision on each cluster based on the spatial non-stationary probability of each cluster to determine the spatial non-stationary SnS state of each cluster. Based on the SnS state of each cluster, the first power attenuation factor of different clusters at the base station is determined, including: when the SnS state indicates a non-stationary state, calculating the visibility probability and visible area of ​​the corresponding SnS state cluster; if the SnS state cluster is within the visible area, determining the first power attenuation factor of the SnS state cluster as 1 according to a preset spatial attenuation function; if the SnS state cluster is outside the visible area, using the spatial attenuation function, based on the relative position of the array elements of the SnS state cluster and the boundary of the visible area, determining the first power attenuation factor value that smoothly attenuates to 0 with spatial coordinates; the attenuation rate of the spatial attenuation function is a preset adjustable parameter; when the SnS state cluster indicates a stationary state, determining the first power attenuation factor of 1 for all array elements of the corresponding non-SnS state cluster. Determining the second power attenuation factor for each array element of the user equipment includes: determining the user equipment's grip condition according to a preset typical ratio; the grip condition includes three situations: head and one hand blocking, two hands blocking, and one hand blocking; based on the grip condition, obtaining the attenuation value of each antenna element of the user equipment; and determining the second power attenuation factor for each array element of the user equipment based on the attenuation value. Based on the first power attenuation factor and the second power attenuation factor, channel coefficients are generated; The calculation of the visibility probability and visible region of the corresponding SnS state cluster includes: Determine a first ratio between the number of visible array elements in the SnS state cluster and the total number of array elements in the VLS; establish a functional relationship between the visibility probability and the cluster power based on ray tracing simulation or channel measurement data; calculate the visibility probability of the SnS state cluster according to the functional relationship; or, calculate the visibility probability of the SnS state cluster according to the preset statistical distribution model. Based on the visibility probability, the visible area of ​​the SnS state cluster is calculated using at least one of the following methods: center-radial method, corner-triggered method, and near-field preferred method.

2. The method according to claim 1, characterized in that, Based on the first power attenuation factor and the second power attenuation factor, channel coefficients are generated, including: Based on the first power attenuation factor and the second power attenuation factor, the target formula for generating channel coefficients is input to generate channel coefficients.

3. A device for modeling ultra-large-scale MIMO channels, characterized in that, include: The first acquisition module is used to acquire the cluster power of each cluster according to a preset channel model; The first determining module is used to determine the spatial non-stationary SnS state of each cluster based on the power of each cluster or a preset statistical distribution model. The second determining module is used to determine the first power attenuation factor of different clusters at the base station based on the SnS state of each cluster. The first processing module is used to determine the second power attenuation factor for each array element of the user equipment. The second processing module is used to generate channel coefficients based on the first power attenuation factor and the second power attenuation factor; The second determining module includes: a first determining unit, configured to calculate the visibility probability and visible area of ​​the corresponding SnS state cluster when the SnS state representation is in a non-stationary state, and determine the first power attenuation factor of the SnS state cluster based on the visibility probability and the visible area; and a second determining unit, configured to determine the first power attenuation factor of 1 for all array elements of the corresponding non-SnS state cluster when the SnS state cluster representation is in a stationary state. The first determining module includes: a third determining unit, used to calculate the spatial non-stationary probability of each cluster based on the power of each cluster using at least one of the following methods: obtaining a probability distribution through multi-user joint analysis, obtaining a probability distribution through single-user independent analysis, and cluster normalized power correlation modeling; and a fourth determining unit, used to perform binarization decision on each cluster based on the spatial non-stationary probability of each cluster to determine the spatial non-stationary SnS state of each cluster. The first determining unit is configured to: determine a first ratio between the number of visible array elements in the SnS state cluster and the total number of array elements in the VLS; establish a functional relationship between the visibility probability and the cluster power based on ray tracing simulation or channel measurement data; calculate the visibility probability of the SnS state cluster according to the functional relationship; or, calculate the visibility probability of the SnS state cluster according to the preset statistical distribution model. Based on the visibility probability, the visible area of ​​the SnS state cluster is calculated using at least one of the following methods: center-radial method, corner-triggered method, and near-field preferred method. The first determining unit is further configured to: if the SnS state cluster is within the visible area, determine a first power attenuation factor of 1 for the SnS state cluster according to a preset spatial attenuation function; if the SnS state cluster is outside the visible area, determine a first power attenuation factor value that smoothly attenuates to 0 with spatial coordinates based on the relative position of the array elements of the SnS state cluster and the boundary of the visible area using the spatial attenuation function; the attenuation rate of the spatial attenuation function is a preset adjustable parameter; The first processing module includes: a first processing unit, configured to determine the user equipment's grip condition according to a preset typical ratio; the grip condition includes three situations: head and one hand covering, two hands covering, and one hand covering; an acquisition unit, configured to acquire the attenuation value of each antenna element of the user equipment based on the grip condition; and a second processing unit, configured to determine a second power attenuation factor for each array element of the user equipment based on the attenuation value.

4. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 2.

5. A computer program product, characterized in that, It includes computer instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 2.

Images