Indoor antenna system transmitting power prediction method based on deep learning

Abstract

The invention relates to the technical field of deep learning, in particular to an indoor antenna system transmitting power prediction method based on deep learning, and the method comprises the steps: building the communication between a digital beam controller and an antenna in a target place, and controlling the overall transmitting power of an antenna unit; a training data set containing multiple sets of training data is obtained, the training data comprises the passenger flow density and the corresponding real-time overall transmitting power, and the final optimal overall transmitting power corresponding to the passenger flow density in each set of training data is calculated based on the training data and a preset simulation model; training a preset neural network model to obtain a power prediction model; the passenger flow density in the target place is obtained and input to the power prediction model, the optimal overall transmitting power is output, and the digital beam controller generates corresponding digital control signals according to the optimal overall transmitting power and sends the digital control signals to all the antenna units. According to the invention, efficient management of the indoor antenna system can be realized.

Description

Technical Field

[0001] The present application relates to the field of deep learning technology, and in particular to a method for predicting the transmission power of an indoor antenna system based on deep learning. Background Art

[0002] In modern wireless communication systems, indoor antenna systems are widely used in large places such as shopping malls, underground parking lots, and hospitals to improve signal coverage quality and meet users' network needs. The passenger flow density in the above places usually has obvious time-varying characteristics. For example, during the peak hours of the day, there are dense crowds and high network loads, while at night when there are few people, the network demand drops significantly. This tidal change in network demand has brought severe challenges to the management of existing indoor antenna systems.

[0003] Traditional indoor antenna systems usually adopt a fixed power output design. The power output of indoor antenna systems is based on pre-set parameters and cannot be dynamically adjusted according to real-time network needs. As a result, during peak passenger flow periods, indoor antenna systems may not be able to effectively cope with high-load network demands, while during low-load periods, the high power output of indoor antenna systems may cause resource waste and increased energy consumption. Therefore, indoor antenna systems lack sufficient flexibility in dealing with ever-changing network demands and complex environments, making it difficult to achieve optimal resource utilization and signal coverage under changing conditions, and ultimately unable to achieve efficient management of indoor antenna systems. Summary of the invention

[0004] This application provides a method for predicting the transmission power of an indoor antenna system based on deep learning, which can solve the problem that the fixed power output of the existing antenna system cannot adapt to the changes in tidal network demand. This application provides the following technical solutions:

[0005] In a first aspect, the present application provides a method for predicting transmission power of an indoor antenna system based on deep learning, the method comprising:

[0006] Establishing communication between a digital beam controller and all antenna units in a target location, wherein the digital beam controller controls the overall transmit power of all antenna units through a digital signal;

[0007] Acquire a training data set including multiple sets of training data, wherein the training data include passenger flow density and its corresponding real-time overall transmission power, and calculate the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and a preset simulation model;

[0008] The passenger flow density of each group of training data in the training data set is used as input, and the final optimal overall transmission power corresponding to the passenger flow density is used as output, and the preset neural network model is trained to obtain a power prediction model;

[0009] The passenger flow density in the target place is obtained in real time and input into the power prediction model. The power prediction model outputs the optimal overall transmission power. The digital beam controller generates a corresponding digital control signal according to the optimal overall transmission power output by the power prediction model and sends it to all antenna units.

[0010] In a specific implementation scheme, establishing communication between the digital beam controller and all antenna units in the target location includes:

[0011] Arrange a digital beam controller and a plurality of antenna units in a target location, wherein the plurality of antenna units are distributed in different areas of the target location;

[0012] The digital beam controller and the antenna unit use the same communication protocol to establish wireless communication;

[0013] The digital beam controller generates corresponding digital control signals according to the overall transmission power input by the user and the distribution characteristics of the antenna units, and sends the signals to each antenna unit through wireless communication.

[0014] In a specific implementation scheme, the digital beam controller allocates the transmit power of each antenna unit according to the overall transmit power input by the user as follows:

[0015]

[0016] Among them, P i is the transmission power of the ith antenna unit, P total is the overall transmit power input by the user, G i is the gain of the ith antenna element, d ci is the distance between the digital beam controller and the i-th antenna element, α is the path loss factor, N is the number of antenna elements, and j represents the summation index.

[0017] In a specific implementation scheme, a calibration coefficient is introduced to adjust the calculated transmit power of each antenna unit as follows:

[0018]

[0019] in, is the adjusted transmit power of the ith antenna unit, C cal is the calibration factor; calibration factor C cal The calculation method is as follows:

[0020]

[0021] in, is the sum of the actual transmit powers of all antenna units.

[0022] In a specific possible implementation scheme, the calculating the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and the preset simulation model includes:

[0023] Each set of training data includes the passenger flow density of a certain place, the real-time overall transmission power of the antenna corresponding to the passenger flow density, the basic power of the place, and the maximum passenger flow density set for the place;

[0024] A simulation model is introduced to calculate the optimal overall transmission power corresponding to the passenger flow density in each set of training data. The simulation model is as follows:

[0025]

[0026] Among them, P best is the optimal overall transmission power, P base is the basic power in the training data, D is the passenger flow density in the training data, and D max is the maximum passenger flow density in the training data set, and γ is the nonlinear adjustment factor.

[0027] In a specific possible implementation scheme, the step of calculating the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and the preset simulation model further includes:

[0028] The relative error between the optimal overall transmission power and the real-time overall transmission power corresponding to the passenger flow density in each set of training data is calculated as follows:

[0029]

[0030] Where ΔP is the relative error, P real is the real-time overall transmit power in the set of training data;

[0031] The relative error is compared with a preset threshold. If the relative error is less than or equal to the preset threshold, the real-time overall transmission power is determined as the final optimal overall transmission power corresponding to the passenger flow density in the set of training data;

[0032] If the relative error is greater than a preset threshold, the optimal overall transmission power calculated using the simulation model is determined as the final optimal overall transmission power corresponding to the passenger flow density in the set of training data.

[0033] In a specific possible implementation scheme, the passenger flow density of each group of training data in the training data set is used as input, and the final optimal overall transmission power corresponding to the passenger flow density is used as output, and before the preset neural network model is trained to obtain the power prediction model, the method further includes:

[0034] The training data set is expanded. Assume that the training data set contains the passenger flow density D of each training data. n And the corresponding final optimal overall transmission power P best,n , we hope to generate the augmented training data D n,new and P best,n,new , the amplification formula is as follows:

[0035] D n,new =D n ×(1+∈);

[0036]

[0037] Among them, ∈ is the disturbance term, ∈∈[-0.1,0.1], and γ is the nonlinear adjustment factor;

[0038] The passenger flow density of each group of training data in the expanded training data set is taken as input, and the final optimal overall transmission power corresponding to the passenger flow density is taken as output. The preset neural network model is trained to obtain a power prediction model.

[0039] In the second aspect, the present application provides a deep learning-based indoor antenna system transmission power prediction system, which adopts the following technical solutions:

[0040] A deep learning-based indoor antenna system transmission power prediction system, comprising:

[0041] A communication establishment module, used to establish communication between a digital beam controller and all antenna units in a target location, wherein the digital beam controller controls the overall transmission power of all antenna units through digital signals;

[0042] A data acquisition module, used to acquire a training data set including a plurality of sets of training data, wherein the training data include passenger flow density and its corresponding real-time overall transmission power, and to infer the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and a preset simulation model;

[0043] A model training module, used to take the passenger flow density of each group of training data in the training data set as input, and the final optimal overall transmission power corresponding to the passenger flow density as output, and train a preset neural network model to obtain a power prediction model;

[0044] The power control module is used to obtain the passenger flow density in the target place in real time and input it into the power prediction model. The power prediction model outputs the optimal overall transmission power. The digital beam controller generates a corresponding digital control signal according to the optimal overall transmission power output by the power prediction model and sends it to all antenna units.

[0045] In a third aspect, the present application provides an electronic device, comprising a processor and a memory; the memory stores a program, and the program is loaded and executed by the processor to implement a method for predicting the transmission power of an indoor antenna system based on deep learning as described in the first aspect.

[0046] In a fourth aspect, the present application provides a computer-readable storage medium, wherein a program is stored in the storage medium, and when the program is executed by a processor, the program is used to implement a method for predicting the transmission power of an indoor antenna system based on deep learning as described in the first aspect.

[0047] In summary, the beneficial effects of this application include at least:

[0048] 1) By designing the power allocation formula between the digital beam controller and the antenna unit, the transmission power allocation in different scenarios can meet the signal coverage requirements while avoiding power waste. The formula not only takes into account the local characteristics of each antenna unit such as gain and distance, but also combines the limitations of the overall system transmission power to ensure the reasonable allocation of power. This method can dynamically adapt to environmental and scenario changes, optimize power allocation, and by introducing calibration coefficients, further eliminate minor errors caused by environmental factors or unbalanced configurations, ensuring that the overall transmission power is close to the target value entered by the user, thereby providing stable and efficient communication coverage.

[0049] 2) The simulation model and neural network model used enable the system to calculate the optimal overall transmission power based on real-time passenger flow density data. By introducing nonlinear adjustment factors and path loss factors into the training data, the complex nonlinear relationship between passenger flow density and power demand can be accurately reflected. At the same time, the amplification algorithm is introduced to expand the training data through perturbations, further improving the robustness and accuracy of the prediction model. This model can be flexibly adjusted in different scenarios, accurately predict the optimal transmission power, and ensure the system's adaptability and accuracy.

[0050] 3) In order to improve the generalization and robustness of the power prediction model, this application adopts a data augmentation method to generate more training data that conforms to actual application scenarios based on disturbance terms and nonlinear adjustment factors. This augmentation method not only takes into account the nonlinear relationship between passenger flow density and transmission power, but also simulates random fluctuations in actual scenarios, such as seasonal changes, emergencies, etc. In this way, the augmented training data set can be closer to the real scene, improve the accuracy of the model, avoid overfitting, and improve the adaptability of the system in different environments.

[0051] By establishing communication between the digital beam controller and all antenna units in the target location, the problem that the traditional indoor antenna system cannot adjust the power in real time according to the passenger flow density is first solved. Then, by collecting multiple sets of passenger flow density and corresponding transmission power training data, and combining the simulation model to deduce the optimal overall transmission power corresponding to each data point, it is ensured that the power output can be dynamically adjusted according to the change of passenger flow. Then, these training data are input into the preset neural network model for training to obtain a power prediction model. The model can predict the corresponding optimal overall transmission power based on the passenger flow density data in the target location obtained in real time. Finally, the digital beam controller generates the corresponding digital control signal based on the prediction result and sends it to all antenna units to achieve precise adjustment of the overall transmission power of the antenna unit. Through this process, the present application effectively solves the problem that the fixed power output of the existing antenna system cannot adapt to the changes in tidal network demand, and can adjust the power according to actual needs in different time periods, thereby improving the utilization efficiency of network resources, avoiding waste, meeting dynamically changing network needs, and achieving efficient management of indoor antenna systems.

[0052] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present application in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of a method for predicting transmission power of an indoor antenna system based on deep learning in an embodiment of the present application.

[0054] Figure 2 It is a schematic diagram of the overall process of the indoor antenna system transmission power prediction method based on deep learning in an embodiment of the present application.

[0055] Figure 3 It is a structural block diagram of the indoor antenna system transmission power prediction system based on deep learning in an embodiment of the present application.

[0056] Figure 4It is a block diagram of an electronic device for predicting the transmission power of an indoor antenna system based on deep learning in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0058] Optionally, the present application takes the deep learning-based indoor antenna system transmission power prediction method provided in each embodiment as an example of being used in an electronic device, where the electronic device is a terminal or a server. The terminal may be a mobile phone, a computer, a tablet computer, etc. This embodiment does not limit the type of electronic device.

[0059] Reference Figure 1 , is a flow chart of a method for predicting transmission power of an indoor antenna system based on deep learning provided by an embodiment of the present application, the method comprising at least the following steps:

[0060] Step S101: establish communication between a digital beam controller and all antenna units in a target location, and the digital beam controller controls the overall transmission power of all antenna units through digital signals.

[0061] In step S101, a digital beam controller and several antenna units are first deployed in the target location, wherein the several antenna units are distributed in different areas of the target location to ensure the comprehensiveness of signal coverage. As a core control device, the digital beam controller can generate digital control signals and send instructions to all antenna units. Specifically, a unified communication protocol is used to establish wireless communication between the digital beam controller and the antenna units, such as Wi-Fi, ZigBee or other low-latency wireless communication protocols suitable for indoor antenna systems.

[0062] In practice, after communication is established, the digital beam controller generates a corresponding digital control signal based on the overall transmit power input by the user and the distribution characteristics of the antenna units, and sends it to each antenna unit through wireless communication. After receiving the control signal, each antenna unit adjusts its transmit power according to the instruction, so that the overall transmit power of all antenna units is basically equal to the overall transmit power input by the user.

[0063] The digital beam controller allocates the transmit power of each antenna unit according to the overall transmit power input by the user as follows:

[0064]

[0065] Among them, P i is the transmission power of the ith antenna unit, P totalis the overall transmit power input by the user, G i is the gain of the ith antenna element, d ci is the distance between the digital beam controller and the i-th antenna unit, α is the path loss factor, which is used to characterize the attenuation of the signal in space and is a constant, N is the number of antenna units, and j represents the summation index.

[0066] In the above formula, an important advantage of the formula design is that when allocating power to each antenna unit, it not only takes into account the local characteristics of each antenna unit (such as gain and distance), but also takes into account the limitations of the transmit power of the entire system. By balancing the overall transmit power with local demand, it ensures that the system can maintain reasonable coverage and energy distribution in different areas and environments without causing the power in some areas to be too low, affecting the signal quality, or too high, wasting energy. Not only does it achieve precise power control, but it can also dynamically adapt to changes in demand in different scenarios. By introducing the distribution relationship between the overall power and each antenna unit, the formula can ensure that the system's resources are reasonably allocated in space, maximizing coverage requirements while avoiding resource waste.

[0067] In the prior art, although there are some methods that use weight factors to allocate antenna power, most of them are only based on simple weighted averages, ignoring the interactive effects of multiple complex factors. For example, many existing power allocation algorithms simply allocate power through linear weighting of distance and gain, failing to fully consider the nonlinear attenuation characteristics of signals in spatial propagation, as well as the complex changes that may occur in the environment. Therefore, the effect of this method may have certain errors in practical applications and cannot flexibly adapt to the needs of different scenarios. However, the formula designed above combines the path loss factor and gain, so that power allocation does not only depend on the distance between antennas, but also comprehensively considers the performance characteristics of each antenna unit and the actual propagation environment. This method can better adapt to environmental changes through precise mathematical modeling and avoids the problem of over-simplification in the prior art.

[0068] In addition, preferably, after calculating the transmit power of each antenna unit according to the above formula, the overall transmit power should be close to the overall transmit power input by the user, but not necessarily completely equal. Because the present application adopts a distribution method, there may be slight differences in the overall power due to site limitations, environmental factors or uneven configuration of antenna units. Therefore, in order to ensure that the overall transmit power is as close as possible to the power input by the user, and to make up for the slight differences caused by site limitations, environmental factors or uneven configuration of antenna units, a calibration coefficient is introduced to adjust the calculated transmit power of each antenna unit to achieve accurate matching of the overall transmit power.

[0069] Specifically, the calibration coefficient is set as a constant to adjust the ratio of the transmit power of all antenna units. The purpose is to make a proportional correction based on the difference between the sum of the actual transmit power and the target transmit power input by the user. The calibration coefficient is introduced to adjust the calculated transmit power of each antenna unit as follows:

[0070]

[0071] in, is the adjusted transmit power of the ith antenna unit, C cal is the calibration factor. Calibration factor C cal The calculation method is as follows:

[0072]

[0073] in, is the sum of the actual transmit powers of all antenna units.

[0074] The calibration coefficient ensures that even if there are slight differences in power due to environmental factors or uneven configuration, the overall transmit power can ultimately match the target transmit power input by the user. In this way, the overall transmit power of the system can be accurately adjusted to avoid errors caused by uneven configuration or external environmental factors as much as possible, ultimately making the system output stable and in line with expectations. In addition, the calibration coefficient calculation of the present application is based on the overall transmit power allocation, and a unified digital control signal is generated by an optimization formula for adjustment, rather than responding to signal changes in real time for each antenna, thus avoiding the high computational burden caused by frequent adjustments. In addition, the number of antennas is limited in the typical application scenarios of the present application, which further reduces the complexity and real-time requirements of the system.

[0075] Step S102, obtaining a training data set including multiple sets of training data, the training data including passenger flow density and its corresponding real-time overall transmission power, and calculating the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and a preset simulation model.

[0076] In step S102, a training data set is first collected from a public database, and the training data set includes multiple sets of training data, wherein each set of training data includes the passenger flow density of a certain place, the real-time overall transmission power of the antenna corresponding to the passenger flow density, the basic power of the place, and the maximum passenger flow density set for the place. The basic power of the place refers to the lowest power level of the system when the place is not affected by the passenger flow density.

[0077] It should be noted that the training data set of this application is obtained through a public database. Specifically, some websites may disclose passenger flow statistics of certain areas in the construction of smart cities, which is combined with the relevant power data of wireless communication base stations that have been disclosed to form a public database.

[0078] Then, based on the training data set and the preset simulation model, the optimal overall transmission power corresponding to each group of passenger flow density is calculated. Specifically, since the real-time overall transmission power of the antenna corresponding to the passenger flow density of a certain place in each set of training data only represents the overall transmission power of the antenna used in the actual implementation process, it does not represent the most preferred overall transmission power of the antenna. Therefore, it is necessary to introduce a simulation model to calculate the optimal overall transmission power corresponding to the passenger flow density in each set of training data. The simulation model is as follows:

[0079]

[0080] Among them, P best is the optimal overall transmission power, P base is the basic power in the training data, D is the passenger flow density in the training data, and D max is the maximum passenger flow density in this group of training data, γ is a nonlinear adjustment factor used to control the nonlinear growth degree of the impact of passenger flow density on transmission power. In this application, it is a fixed value, set to 1.2. It should be noted that the setting of its value is obtained through sufficient analysis and experimental verification of typical scenarios and has wide applicability. Although the nonlinear characteristics of different places may vary, the goal of this application is to provide a general, stable and easy-to-implement technical solution, rather than tailoring it for each scenario. The setting of fixed values ​​can achieve balanced performance in most scenarios and meet actual needs.

[0081] In the design process of the above simulation model, a nonlinear adjustment factor is introduced to describe the effect of passenger flow density on transmission power in power form, which has higher flexibility and adaptability. The core advantage of this design is that it avoids the limitation that the conventional linear weighted model may be too simple and cannot reflect the complex nonlinear relationship between passenger flow density changes and power demand in actual scenarios. The combination of passenger flow density and maximum passenger flow density in the model can not only dynamically adapt to the differences of different scales and venues, but also reflect the proportion of passenger flow density in venue capacity, and has upper and lower limits on the calculation of power demand. baseThe introduction of basic power ensures that the system can still provide stable basic communication coverage under low or even no passenger flow density, avoiding the signal blind spot problem caused by excessive power reduction. Compared with the conventional linear model, this design simulates the characteristics of the significant increase in demand for communication power under high passenger flow density through nonlinear growth. The setting of the nonlinear adjustment factor provides additional flexibility, allowing the model to adjust the slope of the power growth curve according to the actual characteristics of the venue. In addition, this design innovation is reflected in its gradual enhancement of the weight of the impact of passenger flow density, especially in the case of high density, the weighting of power allocation is more reasonable, which solves the defect that the conventional linear model cannot accurately reflect the surge in power demand. In summary, this model not only ensures the practical feasibility of power allocation, but also fully considers the complexity of passenger flow changes in the wireless communication environment.

[0082] Finally, the relative error between the optimal overall transmission power and the real-time overall transmission power corresponding to the passenger flow density in each set of training data is calculated as follows:

[0083]

[0084] Where ΔP is the relative error, P real is the real-time overall transmission power in the set of training data. The relative error is compared with the preset threshold. If the relative error is less than or equal to the preset threshold, the real-time overall transmission power is determined as the final optimal overall transmission power corresponding to the passenger flow density in the set of training data. If the relative error is greater than the preset threshold, the optimal overall transmission power calculated using the simulation model is determined as the final optimal overall transmission power corresponding to the passenger flow density in the set of training data.

[0085] In implementation, the relative error analysis between the optimal overall transmit power calculated by the simulation model and the real-time overall transmit power in the training data can be carried out through the above method, which can effectively combine the advantages of theoretical calculation and actual data, and ensure that the final optimal overall transmit power has both theoretical optimization effect and meets the needs of actual scenarios. If the relative error is small, it means that the real-time overall transmit power is close enough to the theoretical optimal value, and direct adoption can avoid the deviation that may be caused by over-reliance on the simulation model; if the relative error is large, it means that the real-time overall transmit power may not reach the optimal value. At this time, the use of simulation results can make up for the potential unreasonable configuration in the actual scenario, thereby ensuring accuracy while enhancing the system's adaptability and robustness to different scenarios.

[0086] Step S103: taking the passenger flow density of each group of training data in the training data set as input and the final optimal overall transmission power corresponding to the passenger flow density as output, training the preset neural network model to obtain a power prediction model.

[0087] In implementation, before training the preset neural network model, since the training data set usually comes from the collected data in the actual environment, and these data may not be sufficient in some cases, especially for complex nonlinear problems, it is necessary to expand the training data set, and take the passenger flow density of each group of training data in the expanded training data set as input, and the final optimal overall transmission power corresponding to the passenger flow density as output, and train the preset neural network model to obtain the power prediction model.

[0088] In practice, the augmented data set allows the power prediction model to be exposed to more different data points, avoiding the model being trained only in a local area of ​​the training data set, thereby improving its prediction accuracy on new data. Through data augmentation, the small range fluctuations in passenger flow density and transmission power in the real world can be taken into account, which can make the training data closer to actual application scenarios.

[0089] It should be noted that without data augmentation, the model may only be trained within the scope of the original data, which may lead to a decrease in its predictive ability when facing new and unseen combinations of passenger flow density and transmission power. However, the technical solution of the present application is also feasible without augmenting the training data set.

[0090] Specifically, when the training data set is expanded, it is assumed that the training data set contains the passenger flow density D of each training data. n And the corresponding final optimal overall transmission power P best,n , we hope to generate the augmented training data D n,new and P best,n,new , the amplification formula is as follows:

[0091] D n,new =D n ×(1+∈);

[0092]

[0093] Wherein, ∈ is a disturbance term, indicating the random change of passenger flow density, ∈∈[-0.1,0.1]. γ is a nonlinear adjustment factor, which has the same meaning as in step S102 and is set to 1.2.

[0094] In the design of the above-mentioned amplification formula, the amplification method utilizes the nonlinear relationship between passenger flow density and transmission power, and adjusts the existing data by introducing a disturbance term. The disturbance term is used to simulate the fluctuation of passenger flow density in actual scenarios (such as seasonal changes, emergencies, etc.), so that the data set is closer to the real scenario. The range of the disturbance term is set between -0.1 and 0.1, that is, a small random fluctuation is added on the basis of passenger flow density and optimal transmission power. The small-range disturbance can generate new data points near the original data distribution, ensuring that the amplified data will not deviate from the actual scenario and retain the authenticity and reliability of the original data. Small-range disturbances increase the diversity of data, but will not introduce extreme values, which can improve the generalization ability of the prediction model and avoid overfitting of the model due to data deviation. In addition, small disturbances can help the model learn the data distribution characteristics in a smoother way during the training process, avoiding increased training difficulty or unstable convergence due to excessive disturbances. In this way, the model can learn more changes, improve the robustness of the model, and avoid overfitting.

[0095] In summary, by introducing disturbance terms and nonlinear adjustment factors, the amplification formula allows the generated new data points to not only take into account the nonlinear relationship between passenger flow density and optimal transmission power, but also simulate the random fluctuations of data in the real environment. This method is more in line with the uncertainty and complexity of data in practical applications than traditional amplification methods, and can effectively improve the generalization ability and robustness of the model. Compared with conventional methods, this amplification method can better reflect complex nonlinear relationships, thereby improving the quality and diversity of training data and ensuring that the power prediction model performs more accurately and stably in actual scenarios.

[0096] Optionally, the neural network model preset in this application is a recurrent neural network model, which is a neural network suitable for processing regression problems, especially when there is a certain nonlinear relationship between input and output, the recurrent neural network model can capture complex mapping relationships through its multi-layer nonlinear structure. Other types of neural network models can also be selected, and this application does not limit the specific type of the preset neural network model.

[0097] Step S104: Obtain the passenger flow density in the target place in real time and input it into the power prediction model. The power prediction model outputs the optimal overall transmission power. The digital beam controller generates a corresponding digital control signal according to the optimal overall transmission power output by the power prediction model and sends it to all antenna units.

[0098] In step S104, first, the passenger flow density in the target place is obtained in real time through the sensors pre-deployed in the target place, for example, infrared sensors are arranged in the main channels or areas in the target place to detect the number of people passing through; or the existing Wi-Fi probe technology is used to estimate the passenger flow density by counting the number of mobile devices. In this application, the passenger flow density is usually expressed as the number of people per unit area or per square meter (for example, people / square meter). Then, the obtained passenger flow density in the target place is input into the power prediction model. The power prediction model outputs the optimal overall transmission power under the passenger flow density based on the input passenger flow density through the learned rules and experience, that is, the most suitable antenna transmission power level under the passenger flow density. Finally, the digital beam controller generates a corresponding digital control signal based on the optimal overall transmission power output by the power prediction model and the distribution characteristics of the antenna unit, and sends it to each antenna unit through wireless communication.

[0099] It should be noted that in practice, the passenger flow density in the target place is not obtained in real time and continuously and the transmission power is not adjusted. Instead, the passenger flow density in the target place is obtained at regular intervals and the overall transmission power of the antenna is controlled and adjusted based on the latest passenger flow density.

[0100] In addition, as a preferred embodiment, a loss function is designed, and the power prediction model parameters are updated using a gradient descent method based on the loss function until the loss function is minimized. As shown below:

[0101]

[0102] Where M is the number of data in the expanded training data set, that is, the number of groups of training data. best,n is the actual optimal overall transmission power of the nth training data calculated by the simulation model, is the predicted optimal overall transmit power of the nth training data output by the power prediction model, D max Represents the maximum passenger flow density in the training data. n represents the passenger flow density of the nth training data. ∈ is the disturbance term, which represents the random change of passenger flow density, ∈∈[-0.1,0.1].

[0103] In the design process of the above loss function, the introduction of relative error in the loss function helps to reduce the excessive punishment of some samples due to excessive absolute value differences, and avoids the model from unfairly favoring some data points due to excessive errors in the case of high transmit power. It ensures that the model focuses on relative prediction errors rather than just numerical differences. Data points with higher passenger flow density contribute more to the total loss, which makes the model pay more attention to the performance in high-density situations, which meets the needs of actual situations. For example, in high-density situations, the system requires more precise transmit power control, so high-density data points should have a greater impact on model training. The disturbance term can simulate the random fluctuation of the data, so that the model can better adapt to different environmental changes that may occur in practice and increase its generalization ability. By adding these disturbances to the training data, the model can be trained in environments that are not completely consistent or not completely ideal, which enhances the model's adaptability to future changes. Through the above loss function, the power prediction model not only relies on the impact of passenger flow density on transmit power, but also considers the relative error of actual power. This approach can guide the power prediction model to make more accurate and practical transmit power predictions, especially when the system faces a dynamically changing environment.

[0104] In summary, combined with Figure 2 , by establishing communication between the digital beam controller and all antenna units in the target place, the problem that the traditional indoor antenna system cannot adjust the power in real time according to the passenger flow density is first solved. Then, by collecting multiple sets of passenger flow density and corresponding transmission power training data, and combining the simulation model to deduce the optimal overall transmission power corresponding to each data point, it is ensured that the power output can be dynamically adjusted according to the change of passenger flow. Then, these training data are input into the preset neural network model for training to obtain a power prediction model. The model can predict the corresponding optimal overall transmission power based on the passenger flow density data in the target place obtained in real time. Finally, the digital beam controller generates the corresponding digital control signal based on the prediction result and sends it to all antenna units to achieve precise adjustment of the overall transmission power of the antenna unit. Through this process, the present application effectively solves the problem that the fixed power output of the existing antenna system cannot adapt to the changes in tidal network demand, and can adjust the power according to actual needs in different time periods, thereby improving the utilization efficiency of network resources, avoiding waste, meeting dynamically changing network needs, and achieving efficient management of indoor antenna systems.

[0105] Figure 3 This is a structural block diagram of a deep learning-based indoor antenna system transmission power prediction system provided by an embodiment of the present application. The system includes at least the following modules:

[0106] A communication establishment module is used to establish communication between the digital beam controller and all antenna units in the target location. The digital beam controller controls the overall transmission power of all antenna units through digital signals.

[0107] A data acquisition module is used to acquire a training data set including multiple sets of training data, the training data including passenger flow density and its corresponding real-time overall transmission power, and to deduce the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and a preset simulation model;

[0108] A model training module is used to take the passenger flow density of each group of training data in the training data set as input, and the final optimal overall transmission power corresponding to the passenger flow density as output, and train the preset neural network model to obtain a power prediction model;

[0109] The power control module is used to obtain the passenger flow density in the target place in real time and input it into the power prediction model. The power prediction model outputs the optimal overall transmission power. The digital beam controller generates a corresponding digital control signal according to the optimal overall transmission power output by the power prediction model and sends it to all antenna units.

[0110] For relevant details, refer to the above method embodiment.

[0111] Figure 4 4 is a block diagram of an electronic device provided by an embodiment of the present application. The device at least includes a processor 401 and a memory 402.

[0112] The processor 401 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 401 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 401 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 401 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 401 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0113] The memory 402 may include one or more computer-readable storage media, which may be non-transitory. The memory 402 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 402 is used to store at least one instruction, which is used to be executed by the processor 401 to implement the indoor antenna system transmission power prediction method based on deep learning provided in the method embodiment of the present application.

[0114] In some embodiments, the electronic device may further optionally include: a peripheral device interface and at least one peripheral device. The processor 401, the memory 402 and the peripheral device interface may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface via a bus, a signal line or a circuit board. Schematically, the peripheral devices include but are not limited to: a radio frequency circuit, a touch display screen, an audio circuit, and a power supply.

[0115] Of course, the electronic device may also include fewer or more components, which is not limited in this embodiment.

[0116] Optionally, the present application also provides a computer-readable storage medium, in which a program is stored, and the program is loaded and executed by a processor to implement the indoor antenna system transmission power prediction method based on deep learning of the above method embodiment.

[0117] Optionally, the present application also provides a computer product, which includes a computer-readable storage medium, in which a program is stored, and the program is loaded and executed by a processor to implement the indoor antenna system transmission power prediction method based on deep learning of the above method embodiment.

[0118] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for predicting the transmission power of an indoor antenna system based on deep learning, characterized in that: The method comprises: Establishing communication between a digital beam controller and all antenna units in a target location, wherein the digital beam controller controls the overall transmit power of all antenna units through a digital signal; Acquire a training data set including multiple sets of training data, wherein the training data include passenger flow density and its corresponding real-time overall transmission power, and calculate the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and a preset simulation model; The passenger flow density of each group of training data in the training data set is used as input, and the final optimal overall transmission power corresponding to the passenger flow density is used as output, and the preset neural network model is trained to obtain a power prediction model; The passenger flow density in the target place is obtained in real time and input into the power prediction model. The power prediction model outputs the optimal overall transmission power. The digital beam controller generates a corresponding digital control signal according to the optimal overall transmission power output by the power prediction model and sends it to all antenna units.

2. The method for predicting indoor antenna system transmission power based on deep learning according to claim 1, characterized in that: The establishment of communication between the digital beam controller and all antenna units in the target location includes: Arrange a digital beam controller and a plurality of antenna units in a target location, wherein the plurality of antenna units are distributed in different areas of the target location; The digital beam controller and the antenna unit use the same communication protocol to establish wireless communication; The digital beam controller generates corresponding digital control signals according to the overall transmission power input by the user and the distribution characteristics of the antenna units, and sends the signals to each antenna unit through wireless communication.

3. The method for predicting indoor antenna system transmission power based on deep learning according to claim 2, characterized in that: The digital beam controller allocates the transmit power of each antenna element according to the overall transmit power input by the user as follows: Among them, P i is the transmission power of the ith antenna unit, P total is the overall transmission power input by the user, G i is the gain of the ith antenna element, d ci is the distance between the digital beam controller and the i-th antenna element, α is the path loss factor, N is the number of antenna elements, and j represents the summation index.

4. The method for predicting the transmission power of an indoor antenna system based on deep learning according to claim 3 is characterized in that: The calibration coefficient adjusts the calculated transmit power of each antenna unit as follows: in, is the adjusted transmit power of the ith antenna unit, C cal is the calibration factor; calibration factor C cal The calculation method is as follows: in, is the sum of the actual transmit powers of all antenna units.

5. The method for predicting indoor antenna system transmission power based on deep learning according to claim 1, characterized in that: The method of calculating the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and the preset simulation model includes: Each set of training data includes the passenger flow density of a certain place, the real-time overall transmission power of the antenna corresponding to the passenger flow density, the basic power of the place, and the maximum passenger flow density set for the place; A simulation model is introduced to calculate the optimal overall transmission power corresponding to the passenger flow density in each set of training data. The simulation model is as follows: Among them, P best is the optimal overall transmission power, P base is the basic power in the training data, D is the passenger flow density in the training data, and D max is the maximum passenger flow density in the training data set, and γ is the nonlinear adjustment factor.

6. The method for predicting transmission power of indoor antenna system based on deep learning according to claim 5, characterized in that: The step of calculating the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and the preset simulation model also includes: The relative error between the optimal overall transmission power and the real-time overall transmission power corresponding to the passenger flow density in each set of training data is calculated as follows: Where ΔP is the relative error, P real is the real-time overall transmit power in the set of training data; The relative error is compared with a preset threshold. If the relative error is less than or equal to the preset threshold, the real-time overall transmission power is determined as the final optimal overall transmission power corresponding to the passenger flow density in the set of training data; If the relative error is greater than a preset threshold, the optimal overall transmission power calculated using the simulation model is determined as the final optimal overall transmission power corresponding to the passenger flow density in the set of training data.

7. The method for predicting indoor antenna system transmission power based on deep learning according to claim 1, characterized in that: The method further includes taking the passenger flow density of each group of training data in the training data set as input, taking the final optimal overall transmission power corresponding to the passenger flow density as output, and training the preset neural network model to obtain the power prediction model: The training data set is expanded. Assume that the training data set contains the passenger flow density D of each training data. n And the corresponding final optimal overall transmission power P best,n , we hope to generate the augmented training data D n,new and P best,n,new , the amplification formula is as follows: D n,new =D n ×(1+∈); Among them, ∈ is the disturbance term, ∈∈[-0.1,0.1], and γ is the nonlinear adjustment factor; The passenger flow density of each group of training data in the expanded training data set is taken as input, and the final optimal overall transmission power corresponding to the passenger flow density is taken as output. The preset neural network model is trained to obtain a power prediction model.

8. A deep learning-based indoor antenna system transmission power prediction system, characterized in that: The system comprises: A communication establishment module, used to establish communication between a digital beam controller and all antenna units in a target location, wherein the digital beam controller controls the overall transmission power of all antenna units through digital signals; A data acquisition module, used to acquire a training data set including a plurality of sets of training data, wherein the training data include passenger flow density and its corresponding real-time overall transmission power, and to infer the final optimal overall transmission power corresponding to the passenger flow density in each set of training data based on the training data and a preset simulation model; A model training module, used to take the passenger flow density of each group of training data in the training data set as input, and the final optimal overall transmission power corresponding to the passenger flow density as output, and train a preset neural network model to obtain a power prediction model; The power control module is used to obtain the passenger flow density in the target place in real time and input it into the power prediction model. The power prediction model outputs the optimal overall transmission power. The digital beam controller generates a corresponding digital control signal according to the optimal overall transmission power output by the power prediction model and sends it to all antenna units.

9. An electronic device, characterized in that: The device includes a processor and a memory; a program is stored in the memory, and the program is loaded and executed by the processor to implement a method for predicting the transmission power of an indoor antenna system based on deep learning as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The storage medium stores a program, which, when executed by a processor, is used to implement a method for predicting the transmission power of an indoor antenna system based on deep learning as described in any one of claims 1 to 7.

Images