Deployment method, system and equipment of heterogeneous network in indoor distribution system and medium

By collecting and optimizing real-time environmental data in the room system, building a neural network model, and dynamically adjusting the deployment of heterogeneous network nodes, the problem of signal coverage in heterogeneous network environment is solved, and the optimal signal and network performance is achieved. Improves user experience and system efficiency.

CN118984461AActive Publication Date: 2024-11-19BEIJING XINRUNTONG TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411225872.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-03
Publication Date
2024-11-19
Estimated Expiration
2044-09-03

AI Technical Summary

Technical Problem

The existing room separation systems have unevenness in signal coverage and user experience, especially in heterogeneous network environments, which are difficult to ensure the uniformity of signal coverage and network performance.

Method used

By collecting real-time environmental data, building and optimizing neural network models, using Energy Valley optimization algorithm to predict signal coverage, and dynamically adjust network node deployment plans, and monitoring data feedback in real time to optimize deployment plans.

Benefits of technology

Ensure the best signal and network performance under different environmental conditions, adapt to environmental changes and user needs, and improve user experience and system efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118984461A_ABST
    Figure CN118984461A_ABST
Patent Text Reader

Abstract

The invention relates to a method, a system and equipment for deploying a heterogeneous network in an indoor distribution system and a medium. The deployment method of the heterogeneous network in the indoor distribution system comprises the following steps: collecting real-time environment data of an indoor area; preprocessing the real-time environment data; constructing a neural network model; optimizing the neural network model by using an energy valley optimization algorithm; inputting the preprocessed real-time environment data into the optimized neural network model, and generating a deployment scheme of the heterogeneous network in the indoor area; performing network node deployment on the indoor area based on the deployment scheme; acquiring monitoring data of each deployed network node in real time; the monitoring data are fed back to the optimized neural network model, and the optimized neural network model dynamically adjusts the deployment scheme of the heterogeneous network of the indoor area based on the monitoring data; and adjusting the deployment of each network node based on the dynamically adjusted deployment scheme. According to the invention, the optimal signal and network performance can be provided under different environmental conditions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of communications, and in particular to a method, system, device and medium for deploying heterogeneous networks in a room-distributed system. Background Art

[0002] Indoor distributed systems (indoor distribution systems) are widely used in environments such as office buildings, large shopping malls or stadiums to ensure high-quality coverage of wireless signals in the environment. However, the existing indoor distributed systems still have many problems in terms of signal coverage and user experience. For example, uneven signal coverage causes weak or no signal in some areas of the indoor area where the indoor distributed system is located, affecting the normal use of users and affecting user experience. In particular, with the introduction of heterogeneous networks (different types of networks, such as WiFi, 4G or 5G, etc.), the complexity of heterogeneous networks makes it more difficult to ensure the uniformity of signal coverage of indoor distributed systems, network performance in indoor distributed systems, and good user experience in indoor areas. Summary of the invention

[0003] Based on this, it is necessary to provide a method, system, device and medium for deploying heterogeneous networks in indoor distributed systems that provide optimal signal coverage and network performance in response to the above technical problems.

[0004] In a first aspect, the present invention provides a method for deploying a heterogeneous network in a distributed indoor system, the method for deploying a heterogeneous network in the distributed indoor system comprising:

[0005] Collecting real-time environmental data of the indoor area, the real-time environmental data including: real-time temperature of the indoor area, real-time humidity of the indoor area, real-time signal strength of the indoor area and real-time location information of users in the indoor area;

[0006] Preprocessing the real-time environmental data;

[0007] Build a neural network model;

[0008] Optimizing the neural network model using an energy valley optimization algorithm;

[0009] The preprocessed real-time environmental data is input into the optimized neural network model, and the optimized neural network model predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, and generates a deployment plan of the heterogeneous network in the indoor area based on the signal coverage and the network type of the heterogeneous network in the indoor area;

[0010] Deploy network nodes in the indoor area based on the deployment scheme;

[0011] Acquire monitoring data of each of the deployed network nodes in real time, the monitoring data including: signal strength of each of the network nodes in the indoor area, user connection status of each of the network nodes, and network load of each of the network nodes;

[0012] Feeding back the monitoring data to the optimized neural network model, the optimized neural network model dynamically adjusting the deployment scheme of the heterogeneous network in the indoor area based on the monitoring data;

[0013] The deployment of each of the network nodes is adjusted based on the dynamically adjusted deployment plan.

[0014] Optionally, constructing a neural network model includes:

[0015] Build an initial neural network model;

[0016] Acquire historical environmental data of the indoor area and the network type of the heterogeneous network in the indoor area, the historical environmental data including: historical temperature of the indoor area, historical humidity of the indoor area, historical signal strength of the indoor area, and historical location information of users in the indoor area;

[0017] Training the initial neural network model based on the historical environment data and the network type;

[0018] The trained initial neural network model is evaluated based on the loss function, and the trained initial neural network module is optimized based on the evaluation result to obtain the neural network model.

[0019] Optionally, the optimizing the neural network model by using an energy valley optimization algorithm includes:

[0020] Randomly generate a parameter set of the neural network model, the parameter set including a plurality of parameters, each of the parameters being a weight or a bias of the neural network model;

[0021] Obtaining energy of the neural network model based on the parameter set;

[0022] Performing iterative optimization based on the parameter set and the energy to obtain an optimal parameter set;

[0023] The optimal parameter set is updated to the neural network model to optimize the neural network model.

[0024] Optionally, the iterative optimization based on the parameter set and the energy to obtain an optimal parameter set includes:

[0025] Randomly perturbing the parameter set so that at least one parameter in the parameter set increases by a random value to form a new parameter set;

[0026] Obtaining new energy of the neural network model based on the new parameter set;

[0027] If the new energy is less than the current energy, the new parameter set is updated to the current parameter set; if the new energy is greater than or equal to the current energy, the new parameter set is updated to the current parameter set based on the acceptance probability obtained by the following probability formula

[0028] New is the current parameter set; the probability formula is: Wherein, P is the acceptance probability, T is the current temperature of the indoor area, and △E is the difference between the current energy and the new energy;

[0029] gradually lowering the current temperature of the indoor area;

[0030] Repeat the above steps until the maximum number of iterations is reached or the energy is less than the set threshold, and the current parameter set at this time is taken as the optimal parameter set.

[0031] Optionally, the optimized neural network model predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, including: the optimized neural network model predicts the signal strength of the indoor area based on the following formula:

[0032]

[0033] in, is the signal strength of the indoor area, X is the input real-time environmental data, W is the weight of the optimized neural network, and b is the bias of the optimized neural network.

[0034] Optionally, the deployment plan includes: a deployment location of each of the network nodes, a power of each of the network nodes, and a network type of the network nodes.

[0035] Optionally, the optimized neural network model dynamically adjusts the deployment scheme of the heterogeneous network in the indoor area based on the monitoring data, including:

[0036] The optimized neural network model evaluates the monitoring data;

[0037] The optimized neural network model dynamically adjusts the deployment scheme of the heterogeneous network in the indoor area based on the evaluation results.

[0038] In a second aspect, the present invention further provides a deployment system for heterogeneous networks in a room-based system, wherein the deployment system for heterogeneous networks in the room-based system comprises:

[0039] A first acquisition module is used to collect real-time environmental data of an indoor area, wherein the real-time environmental data includes: real-time temperature of the indoor area, real-time humidity of the indoor area, real-time signal strength of the indoor area, and real-time location information of users in the indoor area;

[0040] A preprocessing module, used for preprocessing the real-time environmental data;

[0041] Model building module, used to build neural network models;

[0042] A model optimization module, used to optimize the neural network model using an energy valley optimization algorithm to obtain an optimized neural network model, wherein the optimized neural network module predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, and generates a deployment plan for the heterogeneous network in the indoor area based on the signal coverage and the network type of the heterogeneous network in the indoor area;

[0043] A second acquisition module is used to obtain monitoring data of each of the deployed network nodes in real time, wherein the monitoring data includes: signal strength of each of the network nodes in the indoor area, user connection status of each of the network nodes, and network load of each of the network nodes;

[0044] A feedback module is used to feed back the monitoring data to the optimized neural network model; the optimized neural network model dynamically adjusts the deployment plan of the heterogeneous network in the indoor area based on the monitoring data.

[0045] In a third aspect, the present invention further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the processor, when executing the computer program, implements the steps of the method for deploying a heterogeneous network in a distributed indoor system as described in any one of the first aspects.

[0046] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the method for deploying a heterogeneous network in a distributed indoor system as described in any one of the first aspects are implemented.

[0047] The above-mentioned method, system, device and medium for deploying heterogeneous networks in the indoor distributed system optimize the neural network model by adopting the energy valley optimization algorithm, and then use the optimized neural network model to deploy the heterogeneous network in the indoor distributed system, which can ensure that the best signal and network performance can be provided under different environmental conditions, so that the user experience is the best; in addition, by real-time feedback of monitoring data to the optimized neural network model, the neural network model dynamically adjusts the deployment plan, and can continuously and dynamically optimize the deployment plan of the heterogeneous network in the indoor distributed system, which can adapt to the ever-changing environmental conditions and user needs, thereby improving the overall user experience and system efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic flow chart of a method for deploying a heterogeneous network in a room distribution system provided in another embodiment of the present invention;

[0049] Figure 2 A structural block diagram of a deployment system for heterogeneous networks in a room-distributed system provided in yet another embodiment of the present invention;

[0050] Figure 3 This is a diagram of the internal structure of a computer device provided in yet another embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0052] In one embodiment, Figure 1 As shown, a method for deploying a heterogeneous network in a room-distributed system is provided. The method for deploying a heterogeneous network in a room-distributed system may include the following steps: S10 to S18.

[0053] S10: Collecting real-time environmental data of the indoor area, the real-time environmental data including: real-time temperature of the indoor area, real-time humidity of the indoor area, real-time signal strength of the indoor area and real-time location information of users in the indoor area.

[0054] S11: Preprocessing the real-time environmental data.

[0055] S12: Build a neural network model.

[0056] S13: Optimize the neural network model using an energy valley optimization algorithm.

[0057] S14: Input the preprocessed real-time environmental data into the optimized neural network model. The optimized neural network model predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, and generates a deployment plan for the heterogeneous network in the indoor area based on the signal coverage and the network type of the heterogeneous network in the indoor area.

[0058] S15: Deploy network nodes in the indoor area based on the deployment plan.

[0059] S16: Acquire monitoring data of each of the deployed network nodes in real time, wherein the monitoring data includes: signal strength of each of the network nodes in the indoor area, user connection status of each of the network nodes, and network load of each of the network nodes.

[0060] S17: Feedback the monitoring data to the optimized neural network model, and the optimized neural network model dynamically adjusts the deployment plan of the heterogeneous network in the indoor area based on the monitoring data.

[0061] S18: Adjust the deployment of each of the network nodes based on the dynamically adjusted deployment plan.

[0062] The method for deploying heterogeneous networks in a distributed indoor system of the present invention optimizes the neural network model by adopting an energy valley optimization algorithm, and then uses the optimized neural network model to deploy the heterogeneous networks in the distributed indoor system, thereby ensuring that the best signal and network performance can be provided under different environmental conditions, so that the user experience is optimal; in addition, by feeding back monitoring data to the optimized neural network model in real time, the neural network model dynamically adjusts the deployment plan, and can continuously and dynamically optimize the deployment plan of the heterogeneous network in the distributed indoor system, and can adapt to changing environmental conditions and user needs, thereby improving the overall user experience and system efficiency.

[0063] In step S10, refer to Figure 1 In step S10, real-time environmental data of the indoor area is collected, and the real-time environmental data includes: real-time temperature of the indoor area, real-time humidity of the indoor area, real-time signal strength of the indoor area and real-time location information of users in the indoor area.

[0064] As an example, the indoor area is the indoor area where the indoor distribution system is located, such as the indoor area of ​​an office building where the indoor area is arranged, the indoor area of ​​a large shopping mall, or the indoor area of ​​a sports stadium, etc.

[0065] As an example, a variety of sensors may be arranged in the indoor area to collect real-time environmental data of the indoor area. Specifically, the various sensors may include temperature sensors, humidity sensors, signal strength meters, and user positioning devices, etc. The temperature sensor may obtain the real-time temperature of each area in the indoor area; the humidity sensor may obtain the real-time humidity of each area in the indoor area; the signal strength meter may include but is not limited to an RSSI measuring device, which may be used to obtain the real-time signal strength of each area in the indoor area, and the real-time signal strength may be used to evaluate the coverage of the current heterogeneous network; the user positioning device may include a Bluetooth positioning device or a WiFi positioning device, etc., and the user positioning device may allow users to obtain real-time location information of users in the indoor area, so as to analyze user distribution and user needs.

[0066] As an example, in step S10, the network types currently available in the indoor area (such as WiFi, 4G or 5G, etc.) can also be recorded to facilitate consideration of characteristics of different network types in subsequent analysis.

[0067] In step S11, refer to Figure 1 In step S11, the real-time environment data is preprocessed.

[0068] As an example, step 11 may include the following steps: S111 to S113.

[0069] S111: Clean the real-time environment data. By cleaning, missing values ​​and outliers in the real-time environment data can be removed. Specifically, for missing values, interpolation or mean filling can be used for processing; for outliers, thresholds can be set or traditional methods (such as Z-score) can be used for identification and processing.

[0070] S112: Normalize the cleaned data. Normalization can convert data of different dimensions to the same size to eliminate the impact of the dimension on the subsequent neural network model training. Common normalization methods can include minimum-maximum normalization and Z-score normalization. Taking minimum-maximum normalization as an example, its corresponding formula can be: Among them, X is the original data, X' is the normalized data, and X max is the maximum value of the current data, X min The minimum value of the current data.

[0071] S113: Extract features from the normalized data. Specifically, extract useful features from the normalized data to facilitate the subsequent training of the neural network model. Feature extraction can be performed using methods such as principal component analysis (CPA); principal component analysis can reduce the dimensionality of high-dimensional data while retaining the main information of the data.

[0072] In one example, in a large shopping mall, sensors are first arranged in different areas of the indoor area to ensure that the acquisition areas of all sensors can cover the important areas of the indoor area after superposition. Sensors can collect real-time environmental data regularly (for example, every minute) and upload it. During the data cleaning process, if the temperature sensor in a certain area fails, resulting in missing data, the system will automatically identify and fill it with the average temperature of the area; then, all data are normalized to ensure that different features are on the same scale; finally, the first two principal components are extracted through PCA as input features of the subsequent neural network model. This series of steps can ensure the high quality of the pre-processed real-time environmental data and lay the foundation for the subsequent training and optimization of the neural network model.

[0073] In step S12, refer to Figure 1 In step S12, a neural network model is constructed.

[0074] As an example, in step S12, constructing a neural network model may include the following steps:

[0075] S121: Construct an initial neural network model.

[0076] S122: Acquire historical environmental data of the indoor area and the network type of the heterogeneous network in the indoor area, wherein the historical environmental data includes: historical temperature of the indoor area, historical humidity of the indoor area, historical signal strength of the indoor area, and historical location information of users in the indoor area.

[0077] S123: Training the initial neural network model based on the historical environment data and the network type.

[0078] S124: Evaluate the trained initial neural network model based on the loss function, and optimize the trained initial neural network module based on the evaluation result to obtain the neural network model.

[0079] As an example, before step S121, step S12 may further include step S120, and step S120 may include: dividing the preprocessed real-time environment data into a training set, a validation set, and a test set. Typically, the data in the training set accounts for 70% of the total data, the data in the validation set accounts for 15% of the total data, and the data in the test set accounts for 15% of the total data. The above divisions can ensure that the initial neural network model can be trained and evaluated on different data.

[0080] As an example, the initial neural network model in step S121 may include a convolutional layer, a pooling layer, and a fully connected layer.

[0081] Specifically, there may be multiple convolutional layers, and the convolutional layers are used to extract features of input data; each convolutional layer may be followed by an activation function (such as ReLU) to introduce nonlinearity. The formula for the convolution operation may be: Z = W × X + b, where Z is the convolution result, W is the convolution kernel, X is the input data, and b is the bias term. In the convolutional layer, the size and stride of the convolution kernel may be adjusted according to the features of the input data.

[0082] Specifically, adding the pooling layer (e.g., the maximum pooling layer) after the convolution layer can reduce the dimension of the feature map and reduce the computational complexity while retaining important features. The formula of the maximum pooling layer can be: Among them, Y i,j is the feature value of position (i, j) in the feature map after pooling, X i+m,j+n is the feature value of position (i+m, j+n) in the input feature map, It means to traverse all elements in the pooling window and take the maximum value, m represents the offset of the pooling window in the height direction, and n represents the offset of the pooling window in the width direction.

[0083] Specifically, after the convolution layer and the pooling layer, the fully connected layer is added, and the fully connected layer can integrate features and make final predictions. The output of the fully connected layer can be processed by an activation function (such as Softmax or Sigmoid) to generate a final prediction result.

[0084] As an example, in step S122, the sensor for collecting the real-time environmental data can be used to collect and save data at different historical nodes to obtain the historical environmental data. The specific method of obtaining the network type of the heterogeneous network in the indoor area is well known to those skilled in the art and will not be repeated here.

[0085] As an example, in step S124, a suitable loss function (such as mean square error MES function or cross entropy loss function) can be selected to evaluate the prediction performance of the trained initial neural network model. The formula of the loss function can be:

[0086]

[0087] in, is the loss value, y i is the true value of the i-th record, is the predicted value output by the initial neural network model, and N is the number of samples in the initial neural network model.

[0088] As an example, in step S124, an optimization algorithm (such as Adam or RMSprop) can be used to optimize the trained initial neural network module based on the evaluation result, and the parameters of the trained initial neural network module can be updated to obtain the neural network model.

[0089] In step S13, refer to Figure 1 In step S13, the neural network model is optimized using an energy valley optimization algorithm.

[0090] As an example, step S13 may include the following steps: S131 to S134.

[0091] S131: Randomly generate a parameter set of the neural network model, wherein the parameter set includes a plurality of parameters, each of which is a weight or a bias of the neural network model.

[0092] S132: Obtaining the energy of the neural network model based on the parameter set.

[0093] S133: Perform iterative optimization based on the parameter set and the energy to obtain an optimal parameter set.

[0094] S134: updating the optimal parameter set to the neural network model to optimize the neural network model.

[0095] As an example, in step S131, the parameter set may include weights and biases of the convolutional layer and the fully connected layer. The parameters of the parameter set may be generated by normal distribution or uniform distribution to ensure that the neural network model has a certain degree of randomness in the early stage of training to avoid falling into a local optimal solution.

[0096] As an example, in step S132, obtaining the energy of the neural network model based on the parameter set may include: obtaining the energy of the neural network model based on the following loss function:

[0097]

[0098] Among them, y i is the true value of the i-th record, is the predicted value output by the neural network model, N is the number of samples in the neural network model, and i is an integer greater than or equal to 1 and less than or equal to N.

[0099] It should be noted that energy E is the energy of the neural network model in the current state, y i is the true value of the i-th record, is the predicted value output by the neural network model of the current state, and N is the number of samples in the neural network model of the current state.

[0100] As an example, step S133 may include the following steps: S1331 to S1335.

[0101] S1331: Randomly perturb the parameter set so that at least one parameter in the parameter set increases by a random value to form a new parameter set.

[0102] S1332: Obtain new energy of the neural network model based on the new parameter set.

[0103] S1333: If the new energy is less than the current energy, the new parameter set is updated to the current parameter set; if the new energy is greater than or equal to the current energy, the new parameter set is updated to the current parameter set based on the acceptance probability obtained by the following probability formula; the probability formula is: Wherein, P is the acceptance probability, T is the current temperature of the indoor area, and ΔE is the difference between the current energy and the new energy.

[0104] S1334: gradually lowering the current temperature of the indoor area;

[0105] S1335: Repeat the above steps until the maximum number of iterations is reached or the energy is less than the set threshold, and the current parameter set at this time is used as the optimal parameter set.

[0106] As an example, in step S1331, by randomly perturbing the parameter set, at least one parameter in the parameter set can be increased by a random value, for example, a parameter in the parameter set can be increased by a random value, or some of the multiple parameters in the parameter set can be increased by a random value, or all the parameters in the parameter set can be increased by a random value. Since the parameters of the new parameter set are changed compared to the parameters of the original parameter set, the state of the neural network model corresponding to the new parameter set will also change, and will also be updated from the original current state to another different new state.

[0107] As an example, in step S1332, the specific method for obtaining the new energy of the neural network model based on the new parameter set can refer to the formula for obtaining the energy of the neural network model based on the parameter set in step S132, except that the value in the formula is the predicted value output by the neural network model of the new state corresponding to the new parameter set, and N is the number of samples in the neural network model of the new state corresponding to the new parameter set.

[0108] As an example, step S1333 is the receiving criterion of new energy.

[0109] As an example, in step S1334, the probability of accepting a poor solution is reduced by gradually reducing the current temperature of the indoor area. The reduction of the current temperature may be performed in an exponential decay or linear decay manner.

[0110] As an example, the optimized parameter set obtained in step S1335 is used as the optimal parameters of the neural network model. After step S1335, the optimized parameters in the obtained optimized parameter set can be input into the trained neural network model to obtain the optimized neural network model.

[0111] In step S14, refer to Figure 1 In step S14, the preprocessed real-time environmental data is input into the optimized neural network model, and the optimized neural network model predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, and generates a deployment plan for the heterogeneous network in the indoor area based on the signal coverage and the network type of the heterogeneous network in the indoor area.

[0112] As an example, in step S14, the optimized neural network model predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, which may include: the optimized neural network model predicts the signal strength of the indoor area based on the following formula:

[0113]

[0114] in, is the signal strength of the indoor area, X is the input real-time environmental data, W is the weight of the optimized neural network, and b is the bias of the optimized neural network. The signal coverage of the indoor area can be predicted according to the signal strength of each area in the indoor area.

[0115] As an example, in step S14, the generated deployment plan may include: the deployment location of each of the network nodes, the power of each of the network nodes, and the network type of each of the network nodes.

[0116] Specifically, based on the signal coverage and the network type of the heterogeneous network in the indoor area, a deployment plan for the heterogeneous network in the indoor area is generated, which may include at least one of the following plans: According to the signal coverage of the indoor area, the optimal position of each of the network nodes (such as WiFi access points, 4G base stations, 5G base stations, etc.) is determined to ensure the uniformity of signal coverage. According to the signal coverage of the indoor area, the power of each of the network nodes is dynamically adjusted. For example, for areas with weak signal strength, the power of the corresponding network node can be increased to improve signal coverage. Different types of network nodes are reasonably allocated according to user needs and network types; for example, 5G base stations are deployed preferentially in areas with dense users, and WiFi access points can be used in areas with lower coverage requirements.

[0117] In step S15, refer to Figure 1 In step S15, network nodes are deployed in the indoor area based on the deployment plan.

[0118] Specifically, the network nodes may be automatically deployed in the indoor area based on the deployment plan, or the network nodes may be manually deployed in the indoor area based on the deployment plan.

[0119] In step S16, refer to Figure 1 In step S16, monitoring data of each deployed network node is obtained in real time, and the monitoring data includes: signal strength of each network node in the indoor area, user connection status of each network node, and network load of each network node.

[0120] As an example, the signal strength of each network node in the indoor area, the user connection status of each network node, the network load of each network node and other data can be monitored in real time through sensors and a network management system.

[0121] Specifically, the signal strength of each network node at different locations in the indoor area can be expressed in dBm. The user connection status of each network node can include the number of connected users and their connection quality. The network load of each network node can include the usage status and delay time of each network node, etc.

[0122] As an example, the monitoring data of each of the network nodes can be expressed by the following formula:

[0123]

[0124] Wherein, D is the data set of monitoring data of each network node to be fed back, S i is the signal strength of the ith network node, U iis the number of users connected to the ith network node, L i is the network load of the ith network node, and N is the total number of network nodes in the indoor area.

[0125] In step S17, refer to Figure 1 In step S17, the monitoring data is fed back to the optimized neural network model, and the optimized neural network model dynamically adjusts the deployment plan of the heterogeneous network in the indoor area based on the monitoring data.

[0126] As an example, in step S17, the optimized neural network model dynamically adjusts the deployment plan of the heterogeneous network in the indoor area based on the monitoring data, which may include the following steps: the optimized neural network model evaluates the monitoring data; the optimized neural network model dynamically adjusts the deployment plan of the heterogeneous network in the indoor area based on the evaluation results.

[0127] Specifically, the optimized neural network model will evaluate whether the performance of the current network node meets the preset standards; for example, if the signal strength of a certain area in the indoor area is lower than -75dBm, the optimized neural network model will identify the area as a weak signal area. The optimized neural network model will adjust the parameters of the current network node according to the evaluation results, including the power of the current network node, the location of the current network node, and the network type of the current network node; for example, if the signal strength of a WiFi access point is insufficient, the optimized neural network model may recommend increasing the power of the WiFi access point or adding a new access point in the area.

[0128] In step S18, refer to Figure 1 In step S18, the deployment of each of the network nodes is adjusted based on the dynamically adjusted deployment plan.

[0129] Specifically, the network nodes may be automatically deployed in the indoor area based on the dynamically adjusted deployment plan, or the network nodes may be manually deployed in the indoor area based on the dynamically adjusted deployment plan.

[0130] More specifically, if the power needs to be adjusted, the power output of the network node can be adjusted in real time according to the dynamically adjusted deployment plan output by the optimized neural network model; for example, if the signal strength of a certain network node is still insufficient, the indoor distributed system can automatically increase its power from 20dBm to 25dBm. If the position of the network node needs to be adjusted, the position of the corresponding network node can be moved according to the dynamically adjusted deployment plan output by the optimized neural network model to improve signal coverage, and the indoor distributed system can physically move the network node according to actual conditions. If the network type needs to be reallocated, the network type of the corresponding network node can be reallocated according to the dynamically adjusted deployment plan output by the optimized neural network model; for example, if the network node in the user-dense area is a WiFi access point, the network node can be reconfigured as a 5G base station with stronger signal coverage.

[0131] In another embodiment, the present invention further provides a deployment system of heterogeneous networks in a room-based system, wherein the deployment system of heterogeneous networks in the room-based system comprises: a first acquisition module 10, a preprocessing module 20, a model building module 30, a model optimization module 40, a second acquisition module 50 and a feedback module 60; wherein the first acquisition module 10 is used to collect real-time environmental data of the indoor area, and the real-time environmental data comprises: the real-time temperature of the indoor area, the real-time humidity of the indoor area, the real-time signal strength of the indoor area and the real-time location information of the user in the indoor area; the preprocessing module 20 is used to preprocess the real-time environmental data; the model building module 30 is used to construct a neural network model; the model optimization module 40 is used to optimize the neural network using an energy valley optimization algorithm The model is optimized to obtain an optimized neural network model, the optimized neural network module predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, and generates a deployment plan for the heterogeneous network in the indoor area based on the signal coverage and the network type of the heterogeneous network in the indoor area; the second acquisition module 50 is used to obtain the monitoring data of each deployed network node in real time, and the monitoring data includes: the signal strength of each network node in the indoor area, the user connection status of each network node and the network load of each network node; the feedback module 60 is used to feed back the monitoring data to the optimized neural network model; the optimized neural network model dynamically adjusts the deployment plan of the heterogeneous network in the indoor area based on the monitoring data.

[0132] For the specific definition of the deployment system of heterogeneous networks in the indoor distributed system, please refer to the definition of the deployment method of heterogeneous networks in the indoor distributed system above, which will not be repeated here. Each module in the above-mentioned deployment system of heterogeneous networks in the indoor distributed system can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0133] In another embodiment, the present invention further provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as real-time network signal data, real-time data after preprocessing, real-time data after conversion, and network signal coverage prediction results. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it is implemented as follows Figure 1 The invention also provides steps of a method for deploying heterogeneous networks in a room-distributed system in accordance with an embodiment of the invention.

[0134] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0135] In another embodiment, the present invention further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following is achieved: Figure 1 The invention also provides steps of a method for deploying heterogeneous networks in a room-distributed system in accordance with an embodiment of the invention.

[0136] In yet another embodiment, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following Figure 1 The invention also provides steps of a method for deploying heterogeneous networks in a room-distributed system in accordance with an embodiment of the invention.

[0137] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0138] The technical features of the above embodiments may be combined arbitrarily. 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.

[0139] The above-mentioned embodiments only express several implementation methods of the present invention, 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 ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for deploying heterogeneous networks in a room-distributed system, characterized in that: include: Collecting real-time environmental data of the indoor area, the real-time environmental data including: real-time temperature of the indoor area, real-time humidity of the indoor area, real-time signal strength of the indoor area and real-time location information of users in the indoor area; Preprocessing the real-time environmental data; Build a neural network model; Optimizing the neural network model using an energy valley optimization algorithm; The preprocessed real-time environmental data is input into the optimized neural network model, and the optimized neural network model predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, and generates a deployment plan of the heterogeneous network in the indoor area based on the signal coverage and the network type of the heterogeneous network in the indoor area; Deploy network nodes in the indoor area based on the deployment scheme; Acquire monitoring data of each of the deployed network nodes in real time, the monitoring data including: signal strength of each of the network nodes in the indoor area, user connection status of each of the network nodes, and network load of each of the network nodes; Feeding back the monitoring data to the optimized neural network model, the optimized neural network model dynamically adjusting the deployment scheme of the heterogeneous network in the indoor area based on the monitoring data; The deployment of each of the network nodes is adjusted based on the dynamically adjusted deployment plan.

2. The method for deploying heterogeneous networks in a distributed indoor system according to claim 1, characterized in that: The construction of the neural network model includes: Build an initial neural network model; Acquire historical environmental data of the indoor area and the network type of the heterogeneous network in the indoor area, the historical environmental data including: historical temperature of the indoor area, historical humidity of the indoor area, historical signal strength of the indoor area, and historical location information of users in the indoor area; Training the initial neural network model based on the historical environment data and the network type; The trained initial neural network model is evaluated based on the loss function, and the trained initial neural network module is optimized based on the evaluation result to obtain the neural network model.

3. The method for deploying heterogeneous networks in a distributed indoor system according to claim 1, characterized in that: The method of optimizing the neural network model using an energy valley optimization algorithm includes: Randomly generate a parameter set of the neural network model, the parameter set including a plurality of parameters, each of the parameters being a weight or a bias of the neural network model; Obtaining energy of the neural network model based on the parameter set; Performing iterative optimization based on the parameter set and the energy to obtain an optimal parameter set; The optimal parameter set is updated to the neural network model to optimize the neural network model.

4. The method for deploying heterogeneous networks in a distributed indoor system according to claim 3, characterized in that: The iterative optimization based on the parameter set and the energy to obtain an optimal parameter set includes: Randomly perturbing the parameter set so that at least one parameter in the parameter set increases by a random value to form a new parameter set; Obtaining new energy of the neural network model based on the new parameter set; If the new energy is less than the current energy, the new parameter set is updated to the current parameter set; if the new energy is greater than or equal to the current energy, the new parameter set is updated to the current parameter set based on the acceptance probability obtained by the following probability formula; the probability formula is: Wherein, P is the acceptance probability, T is the current temperature of the indoor area, and △E is the difference between the current energy and the new energy; gradually lowering the current temperature of the indoor area; Repeat the above steps until the maximum number of iterations is reached or the energy is less than the set threshold, and the current parameter set at this time is taken as the optimal parameter set.

5. The method for deploying heterogeneous networks in a distributed indoor system according to claim 1, characterized in that: The optimized neural network model predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, including: the optimized neural network model predicts the signal strength of the indoor area based on the following formula: in, is the signal strength of the indoor area, X is the input real-time environmental data, W is the weight of the optimized neural network, and b is the bias of the optimized neural network.

6. The method for deploying heterogeneous networks in a distributed indoor system according to claim 1, characterized in that: The deployment scheme includes: the deployment location of each of the network nodes, the power of each of the network nodes and the network type of the network nodes.

7. The method for deploying heterogeneous networks in a distributed indoor system according to claim 1, characterized in that: The optimized neural network model dynamically adjusts the deployment scheme of the heterogeneous network in the indoor area based on the monitoring data, including: The optimized neural network model evaluates the monitoring data; The optimized neural network model dynamically adjusts the deployment scheme of the heterogeneous network in the indoor area based on the evaluation results.

8. A deployment system for heterogeneous networks in a distributed indoor system, characterized in that: The deployment system of the heterogeneous network in the indoor distributed system includes: A first acquisition module is used to collect real-time environmental data of an indoor area, wherein the real-time environmental data includes: real-time temperature of the indoor area, real-time humidity of the indoor area, real-time signal strength of the indoor area, and real-time location information of users in the indoor area; A preprocessing module, used for preprocessing the real-time environmental data; Model building module, used to build neural network models; A model optimization module, used to optimize the neural network model using an energy valley optimization algorithm to obtain an optimized neural network model, wherein the optimized neural network module predicts the signal coverage of the indoor area based on the preprocessed real-time environmental data, and generates a deployment plan for the heterogeneous network in the indoor area based on the signal coverage and the network type of the heterogeneous network in the indoor area; A second acquisition module is used to obtain monitoring data of each of the deployed network nodes in real time, wherein the monitoring data includes: signal strength of each of the network nodes in the indoor area, user connection status of each of the network nodes, and network load of each of the network nodes; A feedback module is used to feed back the monitoring data to the optimized neural network model; the optimized neural network model dynamically adjusts the deployment plan of the heterogeneous network in the indoor area based on the monitoring data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for deploying a heterogeneous network in a room distributed system according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for deploying a heterogeneous network in a room distributed system according to any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Lithium battery charge state prediction method and device based on model fusion

    CN117783875A

  • Energy storage power station safety monitoring method and device based on three-dimensional point cloud and voiceprint, and medium

    CN118038237A

  • Method and system for deploying and evaluating networks in indoor environments

    US20110032826A1