Directional beam interference identification method and device, and storage medium

Abstract

The invention provides a directional beam interference identification method and device and a storage medium, and the method comprises the steps: constructing an initial directional beam interference identification model of a millimeter wave system; training an initial directional beam interference identification model; determining current resource allocation information of current equipment in the millimeter wave system according to the resource allocation scheme of the millimeter wave system; and inputting the current resource allocation information of the current equipment into the target directional beam interference identification model corresponding to the current equipment, and calculating to obtain a signal to interference plus noise ratio of the current equipment under the current resource allocation information. According to the method, the channel multipath effect and the cluster scattering characteristic of millimeter wave signal propagation are combined, the problem of real interference in millimeter wave directional transmission can be effectively solved, the short-term or instantaneous system performance can be predicted while directional beam interference can be accurately identified, and the system performance can be accurately identified. And the accuracy and the environmental adaptability of the millimeter wave interference identification process are improved.

Description

Technical Field

[0001] The present application relates to the field of millimeter-wave communication technologies, and in particular, to a method, device, and storage medium for identifying directional beam interference. Background Art

[0002] In recent years, with the rapid growth in the number of devices connected to wireless networks and the increasing demand for higher data rates, there is likely to be a serious shortage of network capacity in the near future. This has prompted researchers to effectively utilize the currently available frequency bands, namely the sub-6 GHz band, and turn their research attention to higher operating frequency bands, namely the millimeter-wave (mmWave) band. In addition, millimeter-wave communication is considered a key enabling technology to overcome the challenges of spectrum resource shortages in next-generation communication systems (Beyond Fifth Generation, B5G / 6th Generation Mobile Networks, 6G), which provides opportunities to achieve high-speed peak data rates, communication reliability, and ultra-low latency.

[0003] The essence of millimeter waves is directional transmission. Compared with traditional communication technologies, if the interference models of previous systems such as sub-6 GHz are adopted, accurate interference information cannot be obtained, which greatly limits the performance and capacity of the system. At the same time, the current statistical interference models in millimeter-wave scenarios generally reflect the long-term average system performance and are not applicable to dynamic factory environments. Therefore, the existing methods for obtaining millimeter-wave interference information have problems of low accuracy and poor environmental adaptability. Summary of the Invention

[0004] The purpose of the present application is to provide a method, device, and storage medium for identifying directional beam interference to solve the problems of low accuracy and poor environmental adaptability in the existing technology in view of the above deficiencies in the prior art.

[0005] To achieve the above object, the technical solutions adopted in the embodiments of the present application are as follows:

[0006] In a first aspect, an embodiment of the present application provides a method for identifying directional beam interference, the method including:

[0007] Construct an initial directional beam interference identification model for a millimeter-wave system, where the millimeter-wave system includes at least a plurality of terminal devices and a plurality of resource blocks, and the initial directional beam interference identification model is where is the current resource allocation information of device k in the millimeter-wave system, and γ k is the signal-to-interference-plus-noise ratio of device k under the current resource allocation information, is the signal-to-interference ratio of device k to interfering device h, is the signal-to-noise ratio of device k, and ε is the error parameter, is the beam used to serve interfering device h in the millimeter-wave system;

[0008] Train the initial directional beam interference recognition model to determine the corresponding value of and value of, and determine the target directional beam interference recognition model corresponding to each terminal device according to the value of and value of;

[0009] Determine the current resource allocation information of the current device in the millimeter-wave system according to the resource allocation scheme of the millimeter-wave system;

[0010] Input the current resource allocation information of the current device into the target directional beam interference recognition model corresponding to the current device, and calculate the signal-to-interference ratio of the current device under the current resource allocation information.

[0011] In a second aspect, another embodiment of the present application provides a directional beam interference recognition device, and the device includes:

[0012] A construction module for constructing an initial directional beam interference recognition model of a millimeter-wave system, where the millimeter-wave system includes at least multiple terminal devices and multiple resource blocks, and the initial directional beam interference recognition model is where is the current resource allocation information of device k in the millimeter-wave system, and γ k is the signal-to-interference ratio of device k under the current resource allocation information, is the signal-to-interference ratio of device k to interfering device g, is the signal-to-noise ratio of device k, and ε is the error parameter, is the beam used to serve interfering device h in the millimeter-wave system;

[0013] A training module for training the initial directional beam interference recognition model to determine the corresponding value of and value of, and determine the target directional beam interference recognition model corresponding to each terminal device according to the value of and value of;

[0014] A determination module for determining the current resource allocation information of the current device in the millimeter-wave system according to the resource allocation scheme of the millimeter-wave system;

[0015] A calculation module, configured to input the current resource allocation information of the current device into a target directional beam interference recognition model corresponding to the current device, and calculate a signal-to-interference-plus-noise ratio of the current device under the current resource allocation information.

[0016] In a third aspect, another embodiment of the present application provides an electronic device, including: a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the storage medium through the bus. The processor executes the machine-readable instructions to perform the steps of any method in the first aspect as described above.

[0017] In a fourth aspect, another embodiment of the present application provides a storage medium, on which a computer program is stored. When the computer program is run by a processor, it performs the steps of any method in the first aspect as described above.

[0018] The beneficial effects of the present application are as follows: By constructing an initial directional beam interference recognition model for a millimeter-wave system and training the initial directional beam interference recognition model, the corresponding value and value of each terminal device in the initial directional beam interference recognition model are obtained, and according to the value and value, the target directional beam interference recognition model corresponding to each terminal device is determined. According to the resource allocation scheme of the millimeter-wave system, the current resource allocation information of the current device in the millimeter-wave system is determined, and the current resource allocation information of the current device is input into the target directional beam interference recognition model corresponding to the current device, and the signal-to-interference-plus-noise ratio of the current device under the current resource allocation information is calculated. While realizing the framework of offline training and online prediction of directional beam interference recognition in the millimeter-wave system, the channel multipath effect is combined with the cluster scattering characteristics of millimeter-wave signal propagation, which can effectively solve the real interference problem in millimeter-wave directional transmission, can accurately identify the directional beam interference while predicting the short-term or instantaneous system performance, and improves the accuracy and environmental adaptability of the millimeter-wave interference recognition process. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1Schematic diagram of the application scenario of the directional beam interference recognition method provided by the embodiment of the present application;

[0021] Figure 2 Schematic diagram of the resource block allocation provided by the embodiment of the present application;

[0022] Figure 3 Schematic flow diagram of a directional beam interference recognition method provided by the embodiment of the present application;

[0023] Figure 4 Schematic flow diagram when determining the target directional beam interference recognition model corresponding to each terminal device in the directional beam interference recognition method provided by the embodiment of the present application;

[0024] Figure 5 Schematic flow diagram when constructing the initial directional beam interference recognition model of the millimeter wave system in the directional beam interference recognition method provided by the embodiment of the present application;

[0025] Figure 6 Schematic diagram of a scenario in the prior art without considering the multipath effect;

[0026] Figure 7 Schematic diagram of a scenario when constructing the received power expression of the terminal device in the directional beam interference recognition method provided by the embodiment of the present application;

[0027] Figure 8 Schematic flow diagram when determining the initial directional beam interference recognition model in the directional beam interference recognition method provided by the embodiment of the present application;

[0028] Figure 9 Schematic framework diagram of a directional beam interference recognition method provided by the embodiment of the present application;

[0029] Figure 10 Schematic diagram of the average RMSE of the SIR of 4 different algorithms when ρ = 6 and θ = 30° during the verification of the directional beam interference recognition method provided by the embodiment of the present application;

[0030] Figure 11 Schematic diagram of the average RMSE of the SIR under different density conditions when θ = 30° during the verification of the directional beam interference recognition method provided by the embodiment of the present application;

[0031] Figure 12 Schematic diagram of the average RMSE of the SINR of 4 different algorithms when ρ = 6 and θ = 30° during the simulation verification of the directional beam interference recognition method provided by the embodiment of the present application;

[0032] Figure 13Schematic diagram of the average RMSE of SINR under different density conditions when θ = 30° for simulating and verifying the directional beam interference recognition method provided by the embodiments of the present application;

[0033] Figure 14 Scatter plot of SIR when ρ = 6 for simulating and verifying the directional beam interference recognition method provided by the embodiments of the present application;

[0034] Figure 15 Scatter plot of SINR when ρ = 6 for simulating and verifying the directional beam interference recognition method provided by the embodiments of the present application;

[0035] Figure 16 Schematic diagram of the structure of the electronic device provided by the embodiments of the present application. Detailed implementation manners

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It should be understood that the accompanying drawings in the present application are only for the purposes of illustration and description, and are not used to limit the protection scope of the present application. In addition, it should be understood that the schematic drawings are not drawn according to the actual scale. The flowcharts used in the present application illustrate the operations implemented according to some embodiments of the present application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and the steps without logical context relationships may be reversed or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of the present application.

[0037] In addition, the described embodiments are only some embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application claimed, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present application.

[0038] It should be noted that the term "including" will be used in the embodiments of the present application to indicate the existence of the features stated hereinafter, but does not exclude adding other features.

[0039] In recent years, with the rapid growth in the number of devices connected to wireless networks and the increasing demand for higher data rates, there is likely to be a serious shortage of network capacity in the near future. This has prompted researchers to effectively utilize the currently available frequency bands, namely the sub-6GHz band, and turn their research attention to higher operating frequency bands, namely the millimeter wave (mmWave) band. In addition, millimeter wave communication is considered a key enabling technology to overcome the challenges of spectrum resource shortage in next-generation communication systems (Beyond Fifth Generation, B5G / 6th Generation Mobile Networks, 6G), which provides opportunities to achieve high-speed peak data rates, communication reliability, and ultra-low latency.

[0040] The millimeter wave band ranges from 30 to 300 GHz, corresponding to wavelengths of 1 to 10 mm, with a rich bandwidth of up to 252 GHz. Over time, assuming a reasonable availability of 40%, these millimeter wave bands may open up approximately 100 GHz of new spectrum for mobile broadband applications. In the millimeter wave band, not only can we have a huge available bandwidth, but also we can obtain smaller-sized antennas due to the small wavelength of millimeter waves, reducing the cost and power consumption of the antennas. This also means that more antennas can be accommodated within the same physical size, thus achieving a higher large-scale antenna array gain, improving the system's throughput, spectrum efficiency (SE), and energy efficiency (EE), and having the potential to increase the capacity of mobile networks.

[0041] Despite the many potential advantages of millimeter waves, they themselves suffer from severe attenuation due to rainfall, foliage, object blockage, and atmospheric absorption. Beamforming technology is to adjust the amplitude and phase of multi-antenna signals so that the finally radiated signal is concentrated in a certain direction for transmission or reception, which can provide a high beamforming gain. Therefore, equipping base stations and terminal devices with highly directional beamforming antennas can effectively mitigate the extremely high path loss and atmospheric attenuation of millimeter waves. The above reflects that the essence of millimeter wave communication is directional transmission.

[0042] From the perspective of radio resource management (RRM), due to the limitations of bandwidth and power in low-frequency systems (such as sub-6GHz), more and more researchers have begun to focus on resource scheduling in the millimeter-wave band. Traditional research on resource allocation for multiple cells generally assumes that interference information is known, but this is often difficult to hold in actual scenarios. Therefore, it is necessary to identify interference information from different sources before allocating resources for multi-cell multi-users. However, the accurate acquisition of interference information depends on accurate interference modeling and interference identification, that is, the basis for optimal scheduling of system resources is to establish an accurate interference model and accurately identify interference.

[0043] In addition, millimeter waves are considered to be the most promising solution for ultra-reliable low-latency communication (URLLC) intelligent factories. Future factories need to adapt to various extreme communication requirements, not only reliability and low latency, but also data rate and scalability. Although low-frequency sub-6GHz provides a good coverage for wireless connections, due to bandwidth limitations, these bands cannot serve robots or sensors simultaneously and still meet the capacity requirements. Therefore, it is crucial to utilize the large bandwidth of the millimeter-wave band to achieve connectionless in future factories. In millimeter-wave systems, the commonly used statistical interference model is applicable to analyzing long-term rather than short-term or instantaneous system performance. Here, long-term performance refers to the steady-state performance of the system under long-term operation or stable channel state information conditions, especially under various interferences and noises, and can be used for resource optimization of long-term systems, such as capacity and EE. While short-term or instantaneous performance refers to the dynamic performance under conditions where the channel state may change at any time, and is used to respond to sudden situations of the system in a timely manner to make reasonable resource allocation strategies. Due to the dynamic nature of the intelligent factory environment, the traditional method of establishing a long-term interference model is no longer applicable to the deterministic analysis of the factory environment. At this time, it is urgent to pay attention to the research of short-term (instantaneous) systems, consider the multipath effect, and analyze the current short-term (instantaneous) performance in order to better adapt to the dynamic environment and make better decisions and adjustments for the current resource scheduling.

[0044] For low-frequency systems, resources are generally orthogonal within a cell, that is, there is no interference within the cell, only interference between cells. Most existing work in the millimeter-wave band will only consider interference modeling within the cell or only consider the modeling of interference between cells, and the situation of analyzing interference by considering both will increase.

[0045] At present, most millimeter-wave systems establish statistical interference models, which can be used to analyze long-term system performance but are not suitable for short-term or instantaneous systems. For example: the two-lobe gain model or the three-gain-level beam model. These models are convenient for analysis and calculation, but only use a limited number of beam gain values to calculate the interference model, that is: use the gain in the strongest beam direction to represent the main lobe beam gain, and use smaller values to represent the weak side lobe gain. Although the model is simple, it overestimates the main lobe gain and underestimates the side lobe gain, and cannot reflect the true beam gain. The above models are all applicable to long-term system interference models and can reflect the average performance of the system. However, in a dynamically changing factory environment, the movement of objects or the change of the atmosphere may cause changes in the signal propagation conditions and affect the interference situation. Therefore, establishing a short-term or instantaneous model can better meet the interference situation in the current dynamic factory environment, and the influence of the multipath effect needs to be considered. At the same time, the above work assumes that the Channel State Information (CSI) is known, which is incorrect for short-term or instantaneous scenarios including some burst applications.

[0046] In summary, the method for obtaining millimeter-wave interference information in the prior art has problems of low accuracy and poor environmental adaptability.

[0047] Based on the above problems, the embodiments of the present application construct an initial directional beam interference recognition model for a millimeter-wave system and train the initial directional beam interference recognition model to determine the value and the value corresponding to each terminal device in the initial directional beam interference recognition model, and determine the target directional beam interference recognition model corresponding to each terminal device according to the value and the value. According to the resource allocation scheme of the millimeter-wave system, determine the current resource allocation information of the current device in the millimeter-wave system, input the current resource allocation information of the current device into the target directional beam interference recognition model corresponding to the current device, and calculate the signal-to-interference-plus-noise ratio of the current device under the current resource allocation information. While realizing the framework of offline training and online prediction of directional beam interference recognition in the millimeter-wave system, the multipath effect of the channel is combined with the cluster scattering characteristics of millimeter-wave signal propagation, which can effectively model the complex interference problems in millimeter-wave directional transmission and achieve accurate recognition of directional beam interference.

[0048] First, the related application scenarios involved in the directional beam interference recognition method provided by the embodiments of the present application are described.

[0049] Figure 1 It is a schematic diagram of the application scenario of the directional beam interference recognition method provided by the embodiments of the present application. Refer to Figure 1As shown, the directional beam interference recognition method provided by the embodiments of the present application can achieve the recognition of directional beam interference in a millimeter-wave system. Exemplarily, the millimeter-wave system in the embodiments of the present application can be applied to an indoor factory scenario, that is, the millimeter-wave system in the embodiments of the present application can be a millimeter-wave system under a millimeter-wave indoor factory (mmWave Indoor Factory, abbreviated as mInF) model.

[0050] Continuing to refer to Figure 1 As shown, inside the factory, there are multiple workshops separated by walls, and a corridor is reserved for transporting goods. In the millimeter-wave system, M millimeter-wave base stations (mmWave Base Station, abbreviated as MBS) are evenly and randomly placed on both sides of the corridor, and the minimum distance between any two MBSs is d min meters. There is one MBS in each workshop, that is, there are also M workshops or cells in the millimeter-wave system.

[0051] Continuing to refer to Figure 1 As shown, at this time, each MBS has Q M associated industrial devices (IndustrialDevice, abbreviated as InduDev). Therefore, there are K = MQ M devices in the millimeter-wave system. Each workshop is square with a side length of S meters. Each MBS and InduDev are equipped with directional antennas and can achieve beamforming. Each MBS can simultaneously transmit multiple directional beams to serve multiple users. The MBSs in the millimeter-wave system in the embodiments of the present application adopt a single-beam service single-user mode.

[0052] Continuing to refer to Figure 1 As shown, the minimum spectral unit in the millimeter-wave system in the embodiments of the present application is a resource block (Resource Block, abbreviated as RB), that is, the basic unit of physical resources is RB, and the same RB can be shared within and between cells. When the same RB is allocated to devices under different beams, beam-to-beam interference will occur.

[0053] Exemplarily, Figure 2 is a schematic diagram of resource block allocation provided by the embodiments of the present application. Referring to Figure 2 As shown, in the millimeter-wave system in the embodiments of the present application, each base station set MBSs can be represented as C = {C 1 , C 2 ,..., C m ,..., C M}, and at the same time, it also represents the workshop set. The beam set emitted by all MBSs is B = {B 1 , B 2 ,..., B n,...,B N}。The industrial equipment set in the millimeter-wave system in the embodiments of this application can be expressed as U = {U 1 , U 2 ,..., U k ,..., U K}}. Each MBS can utilize all the RB sets R = {RB 1 , RB 2 ,..., RB g ,..., RB G} in the wireless resource pool to serve all InduDevs.

[0054] Exemplarily, continuing to refer to Figure 2 shown, Figure 2 shows the transmit beam set of each MBS and the set of devices served by the current beam set. is the transmit beam set of MBS C m . For example, in an Entry#1, i.e. {1, 1, 0, 1,…, 1,…, 1, 0, 0, 0,…, 1, 14.7634}, "1" indicates that this device and device U 1 share the same RB 1 , and vice versa. And, under the current RB allocation relationship, the downlink SINR (DL-SINR) of device U 1 is 14.7634 dB.

[0055] The following details the directional beam interference identification method provided in the embodiments of this application in combination with multiple embodiments.

[0056] Figure 3 is a schematic flowchart of a directional beam interference identification method provided in the embodiments of this application. Referring to Figure 3 shown, the execution subject of this method can be any electronic device with processing capabilities, such as a millimeter-wave test device communicatively connected to the millimeter-wave system, etc. This method includes:

[0057] S301. Construct an initial directional beam interference identification model for the millimeter-wave system.

[0058] Among them, the millimeter-wave system includes at least multiple terminal devices and multiple resource blocks. The initial directional beam interference identification model is Among them, is the current resource allocation information of device k in the millimeter-wave system, γ k is the signal-to-interference-plus-noise ratio of device k under the current resource allocation information, is the signal-to-interference ratio between device k and interfering device h, is the signal-to-noise ratio of device k, ε is the error parameter, It is the beam for the service interference device h in the millimeter-wave system.

[0059] Among them, the resource block is the above-mentioned resource block, and the current resource allocation information is the set corresponding to the resource block allocation. Continue to refer to Figure 2 As shown, taking device U 1 as an example, the current resource allocation information can be the above-mentioned {1, 1, 0, 1, …, 1, …, 1, 0, 0, 0, …, 1}, and ε is the error parameter, which is used to capture the fast fading generated by the conversion from additive to multiplicative.

[0060] Optionally, the millimeter-wave system is a system that uses electromagnetic waves in the millimeter-wave band for applications such as communication, detection, and measurement. The millimeter-wave system in the embodiments of this application can be the above-mentioned millimeter-wave system. Device k is any terminal device in the millimeter-wave system. For example, it can be any of the above-mentioned industrial devices. The interference device h is any terminal device other than device k. For example, it can be other industrial devices corresponding to any of the above-mentioned industrial devices.

[0061] Optionally, the Gaussian antenna model and the millimeter-wave channel model can be determined, and combined with the multipath effect of the channel, so as to determine the initial directional beam interference recognition model of the millimeter-wave system.

[0062] Exemplarily, the Gaussian antenna beam model can be determined first, and the channel simulator (NYU Channel Model Simulator, abbreviated as NYUSIM) is used as the millimeter-wave multipath channel model, and combined with the multipath effect of the channel, the initial directional beam interference recognition model of the millimeter-wave system is constructed.

[0063] S302. Train the initial directional beam interference recognition model to determine the corresponding value and value in the initial directional beam interference recognition model of each terminal device, and determine the target directional beam interference recognition model corresponding to each terminal device according to the value and value.

[0064] Optionally, after obtaining the initial directional beam interference recognition model, the initial directional beam interference recognition model can be trained by a machine learning algorithm to determine the corresponding value and value in the initial directional beam interference recognition model of each terminal device, so as to determine according to the corresponding value and Based on the values, the target directional beam interference recognition models corresponding to each terminal device are determined. Among them, the machine learning algorithm can be the Nonlinear Least-squares Regression (NLR) or the Neural Network (NN).

[0065] Exemplarily, it can be trained by including the training data of each terminal device, so as to respectively obtain the corresponding value and value in the initial directional beam interference recognition model of each terminal device, and determine the target directional beam interference recognition models corresponding to each terminal device according to the value and value.

[0066] Among them, taking the terminal device as device k as an example, the target directional beam interference recognition model corresponding to the terminal device is Among them, is the current resource allocation information of device k in the millimeter-wave system, and γ k is the signal-to-interference-plus-noise ratio of device k under the current resource allocation information, is the signal-to-interference ratio between device k and interfering device h, is the signal-to-noise ratio of device k, is the beam used to serve interfering device h in the millimeter-wave system.

[0067] S303. According to the resource allocation scheme of the millimeter-wave system, determine the current resource allocation information of the current device in the millimeter-wave system.

[0068] It can be understood that in the millimeter-wave system, the millimeter-wave system includes multiple MBSs and multiple InduDevs. Each MBS can use all the RB sets in the wireless resource pool to serve all InduDevs, which means that each MBS has the ability to use all the RB resources to communicate with any InduDev.

[0069] Optionally, after obtaining the target directional beam interference recognition models of each terminal device, the current resource allocation information of the current device in the millimeter-wave system can be determined according to the resource allocation scheme of the millimeter-wave system. Among them, the current device can be any terminal device.

[0070] Among them, the resource allocation scheme of the millimeter-wave system is used to indicate which RBs each MBS in the millimeter-wave system should use to communicate with which InduDevs. The resource allocation scheme of the millimeter wave can be determined based on the quality of service (QoS) requirements of the terminal device, channel state information, system load, terminal device priority, and resource allocation algorithm.

[0071] Exemplarily, in a millimeter-wave system, the current resource allocation information of the current device can be determined from the resource allocation scheme of the millimeter-wave system through the identification of the current device.

[0072] S304. Input the current resource allocation information of the current device into the target directional beam interference recognition model corresponding to the current device, and calculate the signal-to-interference-plus-noise ratio of the current device under the current resource allocation information.

[0073] Optionally, after obtaining the current resource allocation information of the current device and the target directional beam interference recognition models corresponding to each terminal device, the current resource allocation information of the current device can be input into the target directional beam interference recognition model corresponding to the current device, so as to calculate the signal-to-interference-plus-noise ratio γ of the current device under the current resource allocation information. k 。

[0074] In this embodiment, by constructing an initial directional beam interference recognition model of the millimeter-wave system and training the initial directional beam interference recognition model, the corresponding value and value of each terminal device in the initial directional beam interference recognition model are obtained, and according to the value and value, the target directional beam interference recognition model corresponding to each terminal device is determined. According to the resource allocation scheme of the millimeter-wave system, the current resource allocation information of the current device in the millimeter-wave system is determined, and the current resource allocation information of the current device is input into the target directional beam interference recognition model corresponding to the current device, and the signal-to-interference-plus-noise ratio of the current device under the current resource allocation information is calculated. While realizing the framework of offline training and online prediction of directional beam interference recognition in the millimeter-wave system, the channel multipath effect is combined with the cluster scattering characteristics of millimeter-wave signal propagation, which can effectively solve the real interference problem in millimeter-wave directional transmission, can accurately identify the directional beam interference while realizing the prediction of short-term or instantaneous system performance, and improves the accuracy and environmental adaptability of the millimeter-wave interference recognition process.

[0075] In a possible implementation manner, Figure 4 This is a schematic flowchart of a process for determining the target directional beam interference recognition model corresponding to each terminal device in the directional beam interference recognition method provided by the embodiments of the present application. Referring to Figure 4 as shown, in the above S302, the initial directional beam interference recognition model is trained to determine the corresponding value and value of each terminal device in the initial directional beam interference recognition model, and according to the value and value, the target directional beam interference recognition model corresponding to each terminal device is determined, including:

[0076] S401. Construct a model to be solved based on the initial directional beam interference recognition model and a preset regression model.

[0077] Optionally, taking the training of the initial directional beam interference recognition model using the Nonlinear Least-squares Regression (NLR) algorithm as an example, the beam interference information matrix can be obtained through the NLR algorithm.

[0078] Optionally, the preset regression model can be an NLR algorithm model, which can be expressed as: where is the independent variable (regressor) of the i-th sample, y (i) is the dependent variable (regressand) of the i-th sample, and h(x; θ) is a nonlinear regression problem, that is, a known function with unknown parameters θ = (θ 1 , θ 2 , …, θ K ), where K is the number of unknown parameters.

[0079] Optionally, the solution can be expressed as: where the unknown parameters are subject to the constraint condition where ρ(·) is a metric loss function.

[0080] Optionally, based on the initial directional beam interference recognition model and the NLR algorithm model, a model to be solved can be constructed. where x is the independent variable of the model to be solved (i.e., the current resource allocation information of the beam interference user), is the position parameter in the function of the model to be solved. Among them, the model to be solved includes and

[0081] S402. Obtain the training data of the terminal device.

[0082] Among them, the training data includes the historical resource allocation information of the terminal device and the actual signal-to-interference-plus-noise ratio of the terminal device.

[0083] Optionally, taking one terminal device as an example, the training data of the terminal device can be obtained to solve the model to be solved.

[0084] S403. Input the historical resource allocation information in the training data into the model to be solved, obtain the predicted signal-to-interference-plus-noise ratio output by the model to be solved, and according to the actual signal-to-interference-plus-noise ratio, the predicted signal-to-interference-plus-noise ratio, the preset constraint condition, and the preset loss function, for the The value of is iteratively corrected, and after the iteration ends, the value of corresponding to the terminal device is obtained.

[0085] Optionally, input the historical resource allocation information in the training data into the model to be solved, so that by mining the training data and the NLR algorithm, the value of in the model to be solved is obtained, and an iterative method such as the trust region is used to solve the minimization problem, and the parameter value that minimizes the sum of prediction deviations is obtained.

[0086] Exemplarily, input the historical resource allocation information in the training data into the model to be solved, obtain the predicted signal-to-interference-plus-noise ratio output by the model to be solved, and based on the actual signal-to-interference-plus-noise ratio, the predicted signal-to-interference-plus-noise ratio, the preset constraint conditions, and the preset loss function, the value of in the model to be solved is iteratively corrected, and after the iteration ends, the value of corresponding to the terminal device is obtained.

[0087] Exemplarily, the specific solution can be expressed as: And add constraint conditions to each unknown parameter:

[0088]

[0089] It can be understood that in order to make each unknown parameter have its due physical meaning, that is, the power ratio must be positive, additional constraint conditions need to be added when solving the problem.

[0090] S404. Substitute the value of corresponding to the terminal device and the value of

[0091] into the initial directional beam interference recognition model, and delete the error parameters to obtain the target directional beam interference recognition model corresponding to the terminal device. Optionally, substitute the value of

[0092] corresponding to the terminal device and the value of

[0093] into the initial directional beam interference recognition model, and delete the error parameters in the initial directional beam interference recognition model to obtain the target directional beam interference recognition model corresponding to the terminal device. In a possible implementation manner, the loss function includes a first sub-loss function and a second sub-loss function, the first sub-loss function is a squared loss function, and the second sub-loss function is a linear loss function.Optionally, in an actual wireless network, the fluctuations of small-scale fading are often very severe. Therefore, in order to better adapt to the characteristics of the drastic changes in the wireless network channel, the Huber function can be used as the loss function, which can effectively weaken the influence of outliers and is applicable to the actual wireless network.

[0094] Optionally, the loss function includes a first sub-loss function and a second sub-loss function. The first sub-loss function is a squared loss function, and the second sub-loss function is a linear loss function, as follows:

[0095]

[0096] Optionally, the loss function uses the squared loss function when the error value is less than δ, and uses the linear function when it is greater than δ.

[0097] Through the initial directional beam interference recognition model and the preset regression model, a model to be solved is constructed, and the training data of the terminal device is obtained. The historical resource allocation information in the training data is input into the model to be solved, and the predicted signal-to-interference-plus-noise ratio (SINR) output by the model to be solved is obtained. Then, according to the actual SINR, the predicted SINR, the preset constraint conditions, and the preset loss function, the value and value in the model to be solved are iteratively corrected. After the iteration ends, the value and value corresponding to the terminal device are obtained. The value and value corresponding to the terminal device are substituted into the initial directional beam interference recognition model, and the error parameters are deleted to obtain the target directional beam interference recognition model corresponding to the terminal device, which can enable the obtained target directional beam interference recognition model to identify the interference from different sources and obtain accurate beam interference data.

[0098] In a possible implementation manner, Figure 5 is a schematic flowchart of a process for constructing the initial directional beam interference recognition model of the millimeter-wave system in the directional beam interference recognition method provided by the embodiments of the present application. Referring to Figure 5 shown, the above S301 for constructing the initial directional beam interference recognition model of the millimeter-wave system includes:

[0099] S501. Determine the SINR expression of the terminal device in the millimeter-wave system.

[0100] Optionally, for the downlink millimeter-wave system, for device k, that is, the terminal device U k the SINR is:

[0101]

[0102] Among them, Indicates in factory building C m of the MBSC m Transmission beam to serve device U k Received power, n k Indicates U k Index of the associated service beam. Indicates the in-cell interference power, that is, indicates in factory building C m of the MBSC m Transmission beam To serve device U u The beam of, to the terminal device U k Interference power; Indicates the inter-cell interference power, similarly to σ 2 Is the additive noise power. Indicates the same cell C m Internal interfering device U u And the terminal device U k Whether to share the same RB, Indicates different cells C i Interfering device U between s And the terminal device U k Whether to share the same RB.

[0103] Optionally, after obtaining the SINR of the terminal device U k Since in the SINR of the terminal device U k Among them, the first term of the denominator is the total in-cell interference, and the second term is the total inter-cell interference, then according to the SINR of the terminal device U k The signal-to-interference-plus-noise ratio expression is obtained as follows:

[0104]

[0105] Among them, Indicates the total beam interference to the terminal device U k That is, except for the target beam The beam other than generates beam interference to the terminal device U k Among them, the signal-to-interference-plus-noise ratio expression is used to indicate the main lobe interference (Inter-beam Interference from Mainlobe, abbreviated as IBIM) and sidelobe interference (Inter-beam Interference from Sidelobe, abbreviated as IBIS) of the terminal device U k .

[0106] S502. Construct the received power expression of the terminal device.

[0107] Among them, the received power expression includes parameters related to the beam directions and transmission gains of each beam path corresponding to the terminal device.

[0108] Optionally, Figure 6 Fig. is a schematic diagram of a scenario in the prior art without considering multipath effects. Refer to Figure 6 As shown, when the signal arrives at the receiving end InduDevs from the transmitting end MBSs without considering multipath effects, the received signal can be expressed as:

[0109]

[0110] Among them, represents the Close-in (CI) Free Space Reference Distance Path Loss (FSPL) model. P m represents the transmission power of MBSC m and represents the directional beam transmission gain from MBSC m to InduDevU k , while represents the directional beam reception gain.

[0111] Optionally, Figure 7 Fig. is a schematic diagram of a scenario for constructing the received power expression of the terminal device in the directional beam interference identification method provided in the embodiments of the present application. Refer to Figure 7 As shown, the signal from MBSC m experiences scattering of the cluster and finally reaches InduDevU k through multipath. Therefore, it can be assumed that there are a total of P multipaths in a relatively short time, and the received power is expressed as the sum of the directional antenna beam gains of each multipath power weighted, and the received power expression of the terminal device is constructed.

[0112] Optionally, the received power expression of InduDevU k is:

[0113]

[0114] Among them, is the received power of device k, is the close-in free space reference distance path loss model, is the set of clusters from the mth base station C m of the millimeter wave system to the transmitting beam reaching device k, and the set of multipaths within the cluster is The number of multipaths is Y. For the signal transmitted from the m-th base station C of the millimeter wave system m in the cluster the multipaths power For the m-th base station C of the millimeter wave system m the multipaths arriving at device k the directional beam transmission gain of the angle of departure is the directional beam reception gain of the angle of arrival.

[0115] Optionally, the received power expression of InduDevU can also be obtained u as:

[0116]

[0117] Optionally, the received power expression of InduDevU can also be obtained s as:

[0118]

[0119] S503. Determine the initial directional beam interference recognition model according to the signal-to-interference-plus-noise ratio expression and the received power expression.

[0120] Optionally, after obtaining the signal-to-interference-plus-noise ratio expression and the received power expression, operations can be performed according to the signal-to-interference-plus-noise ratio expression and the received power expression, so as to obtain the initial directional beam interference recognition model.

[0121] By determining the signal-to-interference-plus-noise ratio expression and constructing the received power expression, and determining the initial directional beam interference recognition model according to the signal-to-interference-plus-noise ratio expression and the received power expression, a joint intra-cell and inter-cell interference recognition model based on directional beams can be obtained in the millimeter wave scenario, so that the obtained initial directional beam interference recognition model comprehensively considers the characteristics of the antenna model, the channel model, and the multipath effect respectively, helps to obtain the short-term or instantaneous rather than long-term average system performance, improves the environmental adaptability of the obtained initial directional beam interference recognition model, and makes the millimeter wave interference recognition process more in line with the actual application.

[0122] In a possible implementation manner, Figure 8 is a schematic flowchart of a process for determining the initial directional beam interference recognition model in the directional beam interference recognition method provided by the embodiments of the present application. Referring to Figure 8 shown, the above S503 determines the initial directional beam interference recognition model according to the signal-to-interference-plus-noise ratio expression and the received power expression, including:

[0123] S801. Separate the signal-to-interference-plus-noise ratio (SINR) expression to obtain a first intermediate SINR expression and a second intermediate SINR expression.

[0124] Optionally, separate the interference and noise terms in the SINR expression to obtain a first intermediate SINR expression and a second intermediate SINR expression.

[0125] Optionally, the separation of the SINR expression in S801 to obtain a first intermediate SINR expression and a second intermediate SINR expression includes:

[0126] Take the reciprocal of the SINR expression and separate it to obtain a first intermediate SINR expression and a second intermediate SINR expression.

[0127] Exemplarily, take the reciprocal of the SINR expression to separate the interference and noise terms in the SINR expression as follows:

[0128]

[0129] Exemplarily, the first intermediate SINR expression is obtained as follows:

[0130]

[0131] Exemplarily, the second intermediate SINR expression is obtained as follows:

[0132]

[0133] S802. Input the received power expression into the first intermediate SINR expression to obtain a third intermediate SINR expression.

[0134] Exemplarily, input the received power expression of InduDevU k into the first intermediate SINR expression to obtain the third intermediate SINR expression as follows:

[0135]

[0136] where and represent the ratio of the target power to the intra-cell interference power (intra-Signal-to-Interference Ratio, abbreviated as intra-SIR). and represent the ratio of the target power to the inter-cell interference power (inter-Signal-to-Interference Ratio, abbreviated as inter-SIR). and is the ratio of the target power to the noise (the Signal-to-Noise Ratio, abbreviated as SNR).

[0137] S803. Input the second intermediate signal-to-interference-plus-noise ratio expression and the third intermediate signal-to-interference-plus-noise ratio expression into the signal-to-interference-plus-noise ratio expression to obtain an initial directional beam interference recognition model.

[0138] Optionally, the above S803 includes: inputting the second intermediate signal-to-interference-plus-noise ratio expression and the third intermediate signal-to-interference-plus-noise ratio expression into the signal-to-interference-plus-noise ratio expression to obtain a fourth intermediate signal-to-interference-plus-noise ratio expression, and simplifying the fourth intermediate signal-to-interference-plus-noise ratio expression to obtain an initial directional beam interference recognition model.

[0139] Exemplarily, let γ k = SINR k And Input the second intermediate signal-to-interference-plus-noise ratio expression and the third intermediate signal-to-interference-plus-noise ratio expression into the signal-to-interference-plus-noise ratio expression to obtain the following fourth intermediate signal-to-interference-plus-noise ratio expression:

[0140]

[0141]

[0142] where the potential factors may affect the measured SINR to some extent, so an error term ε' is introduced to capture this effect. denotes for U k except for all beams of, that is, the interference beam set of U k , h = u + s represents the interference device U s including the in-cell device Uu and the inter-cell device U h , and at the same time Cm + Ci represents the overall plant set for all MBSs.

[0143] Exemplarily, the values in each SINR dataset are in dB units, and the difference between the maximum and minimum values is at most 30 dB (10 3 ), so the above fourth intermediate signal-to-interference-plus-noise ratio expression can also be converted to dB and simplified to obtain an initial directional beam interference recognition model to avoid the disproportionate dataset affecting the training of the model. At the same time, the fast fading can be converted from additive to multiplicative and captured by the error parameter ε.

[0144] Based on the signal-to-interference-plus-noise ratio (SINR) expression and the received power expression, an initial directional beam interference recognition model is determined, such that the obtained initial directional beam interference recognition model comprehensively considers the characteristics of the antenna model, the channel model, and the multipath effect respectively, which helps to obtain the system performance in the short term or instantaneously rather than the long-term average, improves the environmental adaptability of the obtained initial directional beam interference recognition model, and makes the millimeter-wave interference recognition process more in line with practical applications.

[0145] It can be understood that Figure 9 is a schematic framework diagram of the directional beam interference recognition method provided by the embodiments of this application. Referring to Figure 9 as shown, the directional beam interference recognition method provided by the embodiments of this application includes an offline training stage and an online prediction stage.

[0146] In the training stage, based on the Gaussian antenna beam model and the system-level NYUSIM channel, an initial directional beam interference recognition model of the millimeter-wave system is constructed, and machine learning algorithms are used for offline training to obtain the target directional beam interference recognition model corresponding to each terminal device.

[0147] In the prediction stage, the current resource allocation information of each terminal device is input into the target directional beam interference recognition model corresponding to each terminal device to accurately predict the signal-to-interference-plus-noise ratio of each terminal device under the current resource allocation information.

[0148] Optionally, the following is an exemplary description of the simulation verification process of the directional beam interference recognition method provided by the embodiments of this application. Through the following verification process, it can be determined that the directional beam interference recognition method provided by the embodiments of this application has achieved good results in both the accuracy of interference recognition and the accuracy of SINR prediction.

[0149] Exemplarily, the NYUSIM multipath channel and the Gaussian antenna beam model are selected, and the NLR algorithm is used. TLM-NLR, TLM-NN, and OAM are used as comparison algorithms, and the same solver is used for simulation verification on the same machine.

[0150] Specifically, in the TLM-NLR algorithm, a two-lobe model is used for interference modeling, and NLR is used during interference training. In the TLM-NN algorithm, a two-lobe model is used for interference modeling, and the neural network algorithm (Neural Network, NN) is used during interference training. In the OAM algorithm, the same multipath channel model as DBII is used, but the beam model is an omnidirectional antenna model instead of the directional antenna in DBII.

[0151] Exemplarily, NYUSIM is used to generate all the data for training and validating the models and algorithms in this patent. Continuing to refer to Figure 1As shown, NYUSIM is applied to the millimeter-wave industrial frequency band. Under the condition of LoS, the frequency band is 140 GHz. The standard temperature is set to 17 °C, and the thermal noise is defined as σ 2 =-174 [dBm / Hz]+10log 10 (system bandwidth)+NF [dB]. The reference distance in any scenario is 1 meter. The PL model from the transmitter to the receiver is expressed as follows:

[0152]

[0153] where f is the frequency in GHz, d is the two-dimensional Transmitter-Receiver (T-R) separation distance with d ≥ 1 m, FSPL(f, 1m) is the free space path loss at the carrier frequency f and T-R distance of 1 m, η is the path loss exponent (PLE). AT (Atmosphere Attenuation term) is the attenuation term caused by the atmosphere. O2I (Out door to Indoor Penetration Loss) represents the loss caused by the signal propagating from the outdoor to the indoor environment, and is implemented using the high and low loss parabolic models of building penetration loss. FL (Foliage Loss) is the foliage attenuation term. χ is a zero-mean Gaussian random variable, and σ is the standard deviation in dB, representing the shadow fading of the distance-related average PL value.

[0154] The antenna model adopted in this application is a non-two-lobe gain antenna model that is more in line with the true directional beam. The formula for the Gaussian antenna gain is expressed as:

[0155]

[0156] where (θ, φ) is the offset angle of the direction angle and elevation angle from the center axis of the beam width, (θ 3dB , φ 3dB ) represent the main lobe beam widths of the direction angle and elevation angle respectively, both in degrees. G 0 is the maximum directional gain, that is, the gain at the beam width center axis angle. The values of (A, B) depend on the beam widths θ 3dB and φ 3dB , and η is the typical average antenna efficiency.

[0157] Exemplarily, other simulation parameters can be as shown in Table 1 below:

[0158] Table 1 Simulation Parameter Table

[0159]

[0160]

[0161] Exemplarily, in the foregoing steps, the beam interference information matrix can be trained through the NLR algorithm To verify the performance of interference recognition, the difference between the currently predicted beam interference matrix and the ideal value can be calculated to measure the prediction error. Therefore, the root mean square error (RMSE) of any industrial device InduDevU k can be expressed as:

[0162]

[0163] where U h is the interfering device, represents the ideal SIR of U k , and represents the predicted SIR. The specific calculation is as follows:

[0164]

[0165] where is the ideal received power of the target device U k , the serving beam is and the serving base station is C m . And represents the received power of the interfering device in cell C m or the received power of the interfering device between cells C or the ideal value of the received power of the interfering device between cells C i , and all can be calculated through the foregoing formulas.

[0166] Exemplarily, Figure 10 is a schematic diagram of the average RMSE of the SIR of four different algorithms when ρ = 6 and θ = 30° during the verification of the directional beam interference recognition method provided by the embodiment of the present application, Figure 11 is a schematic diagram of the average RMSE of the SIR under different density conditions when θ = 30° during the verification of the directional beam interference recognition method provided by the embodiment of the present application. Refer to Figure 10 Figure 11 and​As shown in the figure, the number of devices in each plant is the density, denoted by ρ, and the main lobe beam width of the transmitting beam is denoted by θ. When the density ρ = 6 and the main lobe beam width θ = 30°, the prediction accuracy of the directional beam interference recognition method provided by the embodiments of the present application is better than that of other algorithms, while TLM-NN is the worst algorithm. And as the training data increases, the performance of the directional beam interference recognition method provided by the embodiments of the present application becomes better and better.

[0167] For ρ = 2 and ρ = 4, both the directional beam interference recognition method provided by the embodiments of the present application and TLM-NLR perform poorly under extremely weak interference conditions. For millimeter waves, both the transmitted and received signals use directional beams, and the characteristic of directional beams is that the service range is precise, so the overall interference will be low, making it difficult to identify interference and reducing the prediction accuracy of interference. The directional beam interference recognition method provided by the embodiments of the present application can have a training data set of 150,000 and an RMSE of 4.850 dB, indicating that the directional beam interference recognition method provided by the embodiments of the present application constructs an accurate interference recognition model and has excellent interference recognition ability.

[0168] In summary, both TLM-NLR and TLM-NN do not consider the multipath effect and are more suitable for reflecting the long-term average performance of the system rather than performing deterministic analysis on short-term or instantaneous systems. And OAM uses an omnidirectional antenna beam model, and the training fitting effect is also not good. Therefore, the directional beam interference recognition method provided by the embodiments of the present application has the best accuracy.

[0169] The following is an exemplary description of the process of evaluating the adaptability of the directional beam interference recognition method provided by the embodiments of the present application to different conditions by inputting different current resource allocation information.

[0170] Exemplarily, based on an accurate and fine-grained beam interference recognition matrix By matching the current resource allocation variables The SINR of the device side can be obtained. Similarly, in order to verify the error between the predicted value and the ideal value, the average RMSE of SINR is calculated as follows:

[0171]

[0172] Among them, U k is the target device, represents the ideal SINR from base station C m to the target device U k and represents the predicted SINR.

[0173] Exemplarily, Figure 12Schematic diagram of the average RMSE of SINR of four different algorithms when ρ = 6 and θ = 30° for simulating and verifying the directional beam interference recognition method provided in the embodiments of the present application Figure 13 Schematic diagram of the average RMSE of SINR under different density conditions when θ = 30° for simulating and verifying the directional beam interference recognition method provided in the embodiments of the present application. Refer to Figure 12 and Figure 13 as shown Figure 12 The curve trend of Figure 10 is similar to Figure 13 The curve trend of Figure 11 is similar. Due to the increase in the training data set, the prediction ability of the algorithm gradually improves. In terms of predicting SINR, the directional beam interference recognition method provided in the embodiments of the present application shows the best prediction performance among the four algorithms, while the prediction ability of TLM-NN is the worst. Compared with Figure 10 , it is obvious that the performance of all algorithms in predicting SINR is better than that in recognizing SIR. This is because the service range of the directional beam is precise, so the beam interference is weak, making the noise and interference equivalent in affecting the target device. Therefore, noise plays an important role in predicting SINR, and adding noise improves the prediction performance. In Figure 12 , when ρ = 6 and the training data set is 150,000, the directional beam interference recognition method provided in the embodiments of the present application is 0.159 dB, which is 0.210 dB lower than TLM-NLR, improving the prediction performance by 4.72%; when the training data set is 1,000, the directional beam interference recognition method provided in the embodiments of the present application is 2.094 dB, which is 1.876 dB lower than TLM-NN, improving the prediction performance by 35.08%.

[0174] Exemplarily Figure 14 Scatter plot of SIR when ρ = 6 for simulating and verifying the directional beam interference recognition method provided in the embodiments of the present application Figure 15 Scatter plot of SINR when ρ = 6 for simulating and verifying the directional beam interference recognition method provided in the embodiments of the present application. Refer to Figure 14 and Figure 15 as shown Figure 14 and Figure 15 show the interference recognition and SINR prediction performance of the directional beam interference recognition method provided in the embodiments of the present application when ρ = 6. Eight colors represent eight different sizes of training data sets. In Figure 14 , there are 16×96 points, and each point represents the SIR of device U k ∈U. It is obvious that as the SIR increases, at a certain target device power, the interference will decrease, and the recognition difficulty will increase, resulting in a decline in the interference recognition performance. Therefore Figure 14The scatter plot of Figure 14 shows a "broom" shape. Compared with Figure 15 , Figure 15 exhibits better SINR prediction ability. Because in the case of less interference, the influence of noise increases, so interference plus noise can reduce the difficulty of prediction. In

[0175] In summary, the numerical results show that compared with the benchmark scheme, the directional beam interference recognition method provided by the embodiments of the present application can improve the prediction accuracy by up to 4.72% - 35.08%.

[0176] Based on the same inventive concept, the embodiments of the present application also provide a directional beam interference recognition device corresponding to the directional beam interference recognition method. Since the principle of solving problems by the device in the embodiments of the present application is similar to the above-mentioned directional beam interference recognition method in the embodiments of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0177] The directional beam interference recognition device provided by the embodiments of the present application includes: a construction module, a training module, a determination module, and a calculation module;

[0178] The construction module is used to construct an initial directional beam interference recognition model for the millimeter-wave system. Among them, the millimeter-wave system includes at least multiple terminal devices and multiple resource blocks, and the initial directional beam interference recognition model is Among them, is the current resource allocation information of device k in the millimeter-wave system, and γ k is the signal-to-interference-plus-noise ratio of device k under the current resource allocation information, is the signal-to-interference ratio between device k and interfering device h, is the signal-to-noise ratio of device k, ε is the error parameter, is the beam used to serve interfering device h in the millimeter-wave system;

[0179] The training module is used to train the initial directional beam interference recognition model to determine the value and value corresponding to each terminal device in the initial directional beam interference recognition model, and determine the target directional beam interference recognition model corresponding to each terminal device according to the value and value;

[0180] The determination module is used to determine the current resource allocation information of the current device in the millimeter-wave system according to the resource allocation scheme of the millimeter-wave system;

[0181] A computing module, configured to input the current resource allocation information of the current device into a target directional beam interference recognition model corresponding to the current device, and calculate the signal-to-interference-plus-noise ratio of the current device under the current resource allocation information.

[0182] Optionally, a training module, specifically configured to: construct a model to be solved according to an initial directional beam interference recognition model and a preset regression model, where the model to be solved includes and Obtain training data of the terminal device, where the training data includes historical resource allocation information of the terminal device and the actual signal-to-interference-plus-noise ratio of the terminal device; input the historical resource allocation information in the training data into the model to be solved, obtain the predicted signal-to-interference-plus-noise ratio output by the model to be solved, and according to the actual signal-to-interference-plus-noise ratio, the predicted signal-to-interference-plus-noise ratio, preset constraint conditions, and a preset loss function, perform iterative correction on the value of in the model to be solved and the value of , and after the iteration ends, obtain the value of corresponding to the terminal device and the value of ; substitute the value of corresponding to the terminal device and the value of into the initial directional beam interference recognition model, and delete the error parameters to obtain the target directional beam interference recognition model corresponding to the terminal device.

[0183] Optionally, the loss function includes a first sub-loss function and a second sub-loss function, the first sub-loss function is a squared loss function, and the second sub-loss function is a linear loss function.

[0184] Optionally, a construction module, specifically configured to: determine a signal-to-interference-plus-noise ratio expression of a terminal device in a millimeter wave system, where the signal-to-interference-plus-noise ratio expression is used to indicate the main lobe interference and side lobe interference of the terminal device; construct a received power expression of the terminal device, where the received power expression includes parameters related to the beam direction and transmission gain of each beam path corresponding to the terminal device; determine the initial directional beam interference recognition model according to the signal-to-interference-plus-noise ratio expression and the received power expression.

[0185] Optionally, the received power expression is: Where is the received power of device k, is the free space reference distance path loss model at close range, is from the mth base station C of the millimeter wave system m Transmitting beam The set of clusters reaching device k, and the cluster The multipath set inside is The number of multipaths is Y, The power of the multipath m in the cluster of the signal transmitted from the m-th base station C in the millimeter-wave system; The directional beam transmission gain of the angle of departure of the multipath m from the m-th base station C in the millimeter-wave system arriving at device k; The directional beam reception gain of the angle of arrival.

[0186] Optionally, the determining module is specifically configured to: separate the signal-to-interference-plus-noise ratio expression to obtain a first intermediate signal-to-interference-plus-noise ratio expression and a second intermediate signal-to-interference-plus-noise ratio expression; input the received power expression into the first intermediate signal-to-interference-plus-noise ratio expression to obtain a third intermediate signal-to-interference-plus-noise ratio expression; input the second intermediate signal-to-interference-plus-noise ratio expression and the third intermediate signal-to-interference-plus-noise ratio expression into the signal-to-interference-plus-noise ratio expression to obtain an initial directional beam interference recognition model.

[0187] Optionally, the determining module is specifically configured to: take the reciprocal of the signal-to-interference-plus-noise ratio expression and separate it to obtain a first intermediate signal-to-interference-plus-noise ratio expression and a second intermediate signal-to-interference-plus-noise ratio expression.

[0188] Optionally, the determining module is specifically configured to: input the second intermediate signal-to-interference-plus-noise ratio expression and the third intermediate signal-to-interference-plus-noise ratio expression into the signal-to-interference-plus-noise ratio expression to obtain a fourth intermediate signal-to-interference-plus-noise ratio expression; simplify the fourth intermediate signal-to-interference-plus-noise ratio expression to obtain an initial directional beam interference recognition model.

[0189] For the processing flow of each module in the device and the interaction flow between modules, reference may be made to the relevant descriptions in the above method embodiments, which will not be elaborated here.

[0190] An embodiment of the present application further provides an electronic device, as Figure 16 shown, Figure 16 which is a schematic structural diagram of the electronic device provided by the embodiment of the present application, including: a processor 1601, a memory 1602, and optionally, a bus 1603 may also be included. The memory 1602 stores machine-readable instructions executable by the processor 1601. When the electronic device runs, the processor 1601 communicates with the memory 1602 through the bus 1603, and when the machine-readable instructions are executed by the processor 1601, the steps of the above-mentioned directional beam interference recognition method are executed.

[0191] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the steps of the above-mentioned directional beam interference recognition method are executed.

[0192] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the method embodiments, and will not be elaborated herein. In several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of the devices or modules can be in electrical, mechanical or other forms.

[0193] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks or optical discs that can store program codes.

[0194] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all should be covered by the protection scope of the present application.

Claims

1. A directional beam interference identification method, characterized in that: include: Constructing an initial directional beam interference identification model for a millimeter wave system, wherein the millimeter wave system includes at least a plurality of terminal devices and a plurality of resource blocks, and the initial directional beam interference identification model is in, is the current resource allocation information of device k in the millimeter wave system, γ k is the signal-to-interference-noise ratio of device k under the current resource allocation information, is the signal-to-interference ratio of device k to interfering device h, is the signal-to-noise ratio of device k, ε is the error parameter, is a beam used to serve the interference device h in the millimeter wave system; Train the initial directional beam interference identification model to determine the corresponding directional beam interference of each terminal device in the initial directional beam interference identification model. The value and The value of The value and The value of determines the target directional beam interference identification model corresponding to each of the terminal devices; Determining current resource allocation information of a current device in the millimeter wave system according to the resource allocation scheme of the millimeter wave system; The current resource allocation information of the current device is input into a target directional beam interference recognition model corresponding to the current device, and a signal to interference plus noise ratio of the current device under the current resource allocation information is calculated.

2. The directional beam interference identification method according to claim 1, characterized in that: The initial directional beam interference identification model is trained to determine the corresponding directional beam interference of each terminal device in the initial directional beam interference identification model. The value and The value of The value and The target directional beam interference identification model corresponding to each of the terminal devices is determined by the value of, including: According to the initial directional beam interference identification model and the preset regression model, a model to be solved is constructed, and the model to be solved includes as well as Acquire training data of a terminal device, wherein the training data includes historical resource allocation information of the terminal device and an actual signal to interference and noise ratio of the terminal device; Inputting historical resource allocation information in the training data into the model to be solved, obtaining a predicted signal to interference plus noise ratio output by the model to be solved, and performing a signal to interference plus noise ratio calculation on the model to be solved according to the actual signal to interference plus noise ratio, the predicted signal to interference plus noise ratio, a preset restriction condition, and a preset loss function. The value of The value of is iteratively corrected, and the corresponding terminal device is obtained after the iteration. The value of The value of The terminal device corresponds to The value of The value of is substituted into the initial directional beam interference recognition model, and the error parameter is deleted to obtain the target directional beam interference recognition model corresponding to the terminal device.

3. The directional beam interference identification method according to claim 2, characterized in that: The loss function includes a first sub-loss function and a second sub-loss function, the first sub-loss function is a square loss function, and the second sub-loss function is a linear loss function.

4. The directional beam interference identification method according to claim 1, characterized in that: The construction of the initial directional beam interference identification model of the millimeter wave system includes: Determine a signal to interference plus noise ratio expression for a terminal device in the millimeter wave system, wherein the signal to interference plus noise ratio expression is used to indicate main lobe interference and side lobe interference of the terminal device; Constructing a receiving power expression of the terminal device, wherein the receiving power expression includes parameters related to the beam direction and transmission gain of each beam path corresponding to the terminal device; The initial directional beam interference identification model is determined according to the signal to interference plus noise ratio expression and the received power expression.

5. The directional beam interference identification method according to claim 4, characterized in that: The received power expression is: in, is the received power of device k, is the short-range free-space reference distance path loss model, is the mth base station C of the millimeter wave system m Transmit beam The set of clusters that reach device k, and cluster The multipath set in The number of multipaths is Y, is the mth base station C of the millimeter wave system m The transmitted signal is in the cluster multipath within power, is the mth base station C of the millimeter wave system m Multipath to device k The directional beam transmission gain at the departure angle is is the directional beam receiving gain at the arrival angle.

6. The directional beam interference identification method according to claim 4, characterized in that: The determining the initial directional beam interference identification model according to the signal to interference plus noise ratio expression and the received power expression includes: Separating the signal to interference plus noise ratio expression to obtain a first intermediate signal to interference plus noise ratio expression and a second intermediate signal to interference plus noise ratio expression; Inputting the received power expression into the first intermediate signal to interference plus noise ratio expression to obtain a third intermediate signal to interference plus noise ratio expression; The second intermediate signal to interference plus noise ratio expression and the third intermediate signal to interference plus noise ratio expression are input into the signal to interference plus noise ratio expression to obtain the initial directional beam interference identification model.

7. The directional beam interference identification method according to claim 6, characterized in that: The step of separating the signal to interference plus noise ratio expression to obtain a first intermediate signal to interference plus noise ratio expression and a second intermediate signal to interference plus noise ratio expression includes: The signal to interference plus noise ratio expression is reciprocated and separated to obtain a first intermediate signal to interference plus noise ratio expression and a second intermediate signal to interference plus noise ratio expression.

8. The directional beam interference identification method according to claim 6, characterized in that: The step of inputting the second intermediate signal to interference plus noise ratio expression and the third intermediate signal to interference plus noise ratio expression into the signal to interference plus noise ratio expression to obtain the initial directional beam interference identification model comprises: Inputting the second intermediate signal to interference plus noise ratio expression and the third intermediate signal to interference plus noise ratio expression into the signal to interference plus noise ratio expression to obtain a fourth intermediate signal to interference plus noise ratio expression; The fourth intermediate signal to interference plus noise ratio expression is simplified to obtain the initial directional beam interference identification model.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor executes the machine-readable instructions to perform the steps of the method for directional beam interference identification as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the directional beam interference identification method as claimed in any one of claims 1 to 8 are executed.

Images