Method and device for adjusting wireless network access frequency band, and program product

Abstract

The present invention relates to the technical field of communications, and provides a wireless network access frequency band adjustment method and apparatus, a program product, and a controller applied to a wireless network, the wireless network further comprising sampling nodes and service nodes, the method comprising: receiving frequency band information of a corresponding wireless network frequency band collected by each sampling node within a preset time period; and predicting an optimal frequency band according to the frequency band information to obtain a predicted optimal frequency band and a predicted accuracy rate, and sending the predicted optimal frequency band and the predicted accuracy rate to each service node, so that the service node determines whether to adjust the frequency band of the wireless network currently accessed by the node according to the predicted optimal frequency band and the predicted accuracy rate. According to the invention, the frequency band can be dynamically adjusted.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method, device and program product for adjusting a wireless network access frequency band. Background Art

[0002] With the rapid development of wireless communication technology, wireless network has become the current mainstream communication technology. Wireless network equipment such as 5G routers work by loading 5G modules onto router devices. Router devices include wired Ethernet interfaces and 5G wireless interfaces. The wireless interface mainly searches for the 5G frequency band of the base station and registers with the network and the operator to establish a connection before forwarding data. In specific application scenarios, 5G wireless communication can solve the wiring problems of wired communication, has good mobility and coverage, and can withstand high-density connections. In practical applications, 5G wireless communication can be used as a backup line for wired networks for device switching.

[0003] The frequency band range for wireless network equipment is provided by operators. Operators build base stations, deploy networks, and provide communication services to users in different frequency bands based on their own network planning and business needs. With the continuous development of mobile communication technology, operators' frequency band usage is also constantly adjusted and maintained. For example, from 2G to 5G, operators need to continuously upgrade network equipment and optimize network architecture to adapt to new frequency bands and higher communication rate requirements. Therefore, the network environment will change, and the device needs to search for frequency bands and find a suitable frequency band for registration.

[0004] As the network environment changes in real time, the signal of the device in the working frequency band may weaken or even cause link interruption. Therefore, the device will search the network again. The network search process consumes a lot of time and cannot guarantee the signal quality of the next network segment.

[0005] Currently, there are two methods for obtaining the relatively optimal frequency band:

[0006] The first method is to obtain it through the operator. Before the device is online, the operator is consulted to obtain the relatively optimal network frequency band in the area at that time, and the device is fixedly configured to lock in this frequency band. This method can solve the frequency selection problem in a short time, but since the network frequency band is fixed, once the signal quality of the current network frequency band changes, it may affect the signal of the device.

[0007] The second method is to force the device to traverse the frequency band. Before the 5G module dials up to the Internet, the module traverses the frequency band by locking the frequency, records the signal strength under each frequency band, and selects the best frequency band by comparing the signal strength. This method is also affected by changes in the network environment, has large errors, and is inaccurate. Summary of the invention

[0008] The present invention aims to provide a method, device, network equipment and program product for adjusting a wireless network access frequency band, which can dynamically adjust the frequency band information so that the frequency band information can adapt to changes in the network environment.

[0009] The embodiments of the present invention can be implemented as follows:

[0010] In a first aspect, the present invention provides a method for adjusting a wireless network access frequency band, which is applied to a controller in a wireless network, wherein the wireless network further includes a sampling node and a service node, and the method includes:

[0011] Receiving frequency band information of a corresponding wireless network frequency band collected by each sampling node within a preset time period;

[0012] The optimal frequency band is predicted based on the frequency band information to obtain the predicted optimal frequency band and the prediction accuracy, and the predicted optimal frequency band and the prediction accuracy are sent to each of the service nodes so that the service node determines whether to adjust the wireless network frequency band currently accessed by the node based on the predicted optimal frequency band and the prediction accuracy.

[0013] In an optional implementation, there are multiple sampling nodes, and each of the sampling nodes is used to collect frequency band information of a wireless network frequency band corresponding to the node according to a preset sampling period;

[0014] The step of predicting the optimal frequency band according to the frequency band information to obtain the predicted optimal frequency band and the prediction accuracy includes:

[0015] Dividing the frequency band information of multiple wireless network frequency bands into a test set and a validation set;

[0016] Building a prediction model based on the frequency band information in the test set;

[0017] According to the preset optimal signal parameters, the prediction model is used to predict the optimal frequency band to obtain the predicted optimal frequency band;

[0018] The prediction accuracy of the prediction model is calculated based on the frequency band information in the verification set.

[0019] In an optional implementation manner, the frequency band information includes frequency band and signal parameters, and the step of constructing a prediction model according to the frequency band information in the test set includes:

[0020] According to the frequency band of the frequency band information in the test set, normalize each signal parameter of each frequency band to obtain the normalized signal parameter of each frequency band;

[0021] Calculating the centroid of each frequency band according to the mean of each normalized signal parameter of each frequency band;

[0022] Constructing the prediction model according to the centroids of all the frequency bands; and / or.

[0023] The prediction model includes the centroid of each frequency band, and the centroid of each frequency band is obtained by calculating the mean after normalizing the signal parameters of each frequency band in the test set;

[0024] The step of predicting the optimal frequency band using the prediction model according to the preset optimal signal parameters to obtain the predicted optimal frequency band comprises:

[0025] Calculating the spatial distance between the preset optimal signal parameter and the centroid of each of the frequency bands;

[0026] The frequency band corresponding to the minimum value of all the spatial distances is used as the predicted optimal frequency band.

[0027] In an optional implementation manner, the frequency band information includes frequency band and signal parameters;

[0028] The step of calculating the prediction accuracy of the prediction model according to the frequency band information in the verification set includes:

[0029] sequentially taking the signal parameters of each frequency band in the validation set as target signal parameters, and using the prediction model to perform frequency band prediction on the target signal parameters to obtain a predicted frequency band of the target signal parameters;

[0030] If the frequency band of the target signal parameter is consistent with the predicted frequency band, it is determined that the prediction of the target signal parameter is correct, otherwise it is determined that the prediction is wrong;

[0031] Counting the number of signal parameters of the frequency bands predicted correctly in the validation set;

[0032] The prediction accuracy of the prediction model is calculated based on the number of signal parameters of the frequency bands predicted correctly in the verification set and the total number of signal parameters of all frequency bands in the verification set.

[0033] In an optional implementation, according to the number of signal parameters of the frequency bands predicted correctly in the validation set and the total number of signal parameters of all frequency bands in the validation set, after the step of calculating the prediction accuracy of the prediction model, the method further includes:

[0034] If the prediction accuracy is lower than a preset accuracy threshold, increasing the sampling frequency of the sampling node; and / or

[0035] Each of the frequency band information includes a plurality of signal parameters. According to the number of signal parameters of the frequency bands predicted correctly in the verification set and the total number of signal parameters of all frequency bands in the verification set, after the step of calculating the prediction accuracy of the prediction model, the method further includes:

[0036] If the prediction accuracy is lower than a preset accuracy threshold, adjusting the weight of each signal parameter;

[0037] According to the adjusted weight of each signal parameter, the prediction model is reconstructed and the prediction accuracy is recalculated until the prediction accuracy reaches the preset accuracy threshold.

[0038] In a second aspect, the present invention provides a method for adjusting a wireless network access frequency band, which is applied to a service node in a wireless network, wherein the wireless network further includes a sampling node and a controller, and the method includes:

[0039] Receiving the predicted optimal frequency band and the prediction accuracy sent by the controller, wherein the predicted optimal frequency band and the prediction accuracy are obtained by the controller predicting the optimal frequency band according to the received frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period;

[0040] Determine whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

[0041] In an optional implementation manner, the step of determining whether to adjust the current access wireless network frequency band of the node according to the predicted optimal frequency band and the prediction accuracy includes:

[0042] Determine whether a signal parameter of the currently accessed wireless network frequency band is greater than a preset alarm threshold;

[0043] If the signal parameter of the currently accessed wireless network frequency band is lower than the preset alarm threshold, adjusting the currently accessed wireless network frequency band to the predicted optimal frequency band;

[0044] If the signal parameter of the currently accessed wireless network frequency band is greater than the preset alarm threshold, determining whether the predicted optimal frequency band is better than the currently accessed wireless network frequency band;

[0045] If the predicted optimal frequency band is better than the currently accessed wireless network frequency band, and the prediction accuracy is greater than a preset accuracy threshold, adjusting the currently accessed wireless network frequency band to the predicted optimal frequency band;

[0046] If the predicted optimal frequency band is not better than the currently accessed wireless network frequency band, the currently accessed wireless network frequency band remains unchanged.

[0047] In an optional embodiment, the method further comprises:

[0048] If the signal parameter of the currently accessed wireless network frequency band is lower than a preset alarm threshold, obtaining the predicted optimal frequency band from the controller, and adjusting the currently accessed wireless network frequency band to the predicted optimal frequency band; and / or

[0049] The method further comprises:

[0050] The predicted optimal frequency band received most recently is stored, so that when it is detected that the connection with the controller is disconnected, the wireless network frequency band currently accessed by the node is adjusted according to the stored predicted optimal frequency band.

[0051] In a third aspect, the present invention provides a wireless network access frequency band adjustment device, which is applied to a controller in a wireless network, wherein the wireless network further includes a sampling node and a service node, and the device includes:

[0052] A receiving module, used to receive the frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period;

[0053] A prediction module is used to predict the optimal frequency band according to the frequency band information, obtain the predicted optimal frequency band and the prediction accuracy, and send the predicted optimal frequency band and the prediction accuracy to each of the service nodes, so that the service node determines whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

[0054] In a fourth aspect, the present invention provides a wireless network access frequency band adjustment device, which is applied to a service node in a wireless network, wherein the wireless network further includes a sampling node and a controller, and the device includes:

[0055] A receiving module, used for receiving the predicted optimal frequency band and the prediction accuracy sent by the controller, wherein the predicted optimal frequency band and the prediction accuracy are obtained by the controller predicting the optimal frequency band according to the received frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period;

[0056] The adjustment module is used to determine whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

[0057] In a fifth aspect, the present invention provides a program product, which, when executed by a processor, implements the wireless network access frequency band adjustment method described in the first aspect of the aforementioned implementation manner, or implements the wireless network access frequency band adjustment method described in the second aspect of the aforementioned implementation manner.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] The present invention predicts the optimal frequency band based on the frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period, and obtains the predicted optimal frequency band and the prediction accuracy. Each sampling node then determines whether to adjust the wireless network frequency band currently accessed by the node based on the predicted optimal frequency band and the prediction accuracy, thereby realizing dynamic adjustment of the frequency band, enabling the frequency band to adapt to changes in the network environment, and ensuring the real-time nature of the frequency band update and the stability of the signal quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0061] Figure 1 This is an example diagram of an application scenario provided by this embodiment.

[0062] Figure 2 This is a block diagram of an example network device provided in this embodiment.

[0063] Figure 3 This is a flowchart of an exemplary method for adjusting a wireless network access frequency band applied to a controller provided in this embodiment.

[0064] Figure 4 This is an example diagram of the centroid of the frequency band provided in this embodiment.

[0065] Figure 5 This is an example diagram of signal parameters of a frequency band before normalization provided by this embodiment.

[0066] Figure 6 This is an example diagram of the signal parameters of the normalized frequency band provided in this embodiment.

[0067] Figure 7 A flowchart of a method for adjusting a wireless network access frequency band applied to a service node is provided for this embodiment.

[0068] Figure 8 This is a block diagram of an exemplary embodiment of a wireless network access frequency band adjustment device applied to a controller provided in this embodiment.

[0069] Fig. 9 This is a block diagram of an exemplary device for adjusting a wireless network access frequency band applied to a service node provided in this embodiment.

[0070] Fig.10 This is an example diagram of the overall structural framework provided for this embodiment.

[0071] Icons: 10 - network device; 11 - processor; 12 - memory; 13 - bus; 100 - wireless network access frequency band adjustment device; 110 - receiving module; 120 - prediction module; 200 - wireless network access frequency band adjustment device; 210 - receiving module; 220 - adjustment module. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0073] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0075] In the description of the present invention, it should be noted that if the terms "upper", "lower", "inside", "outside", etc. appear to indicate an orientation or position relationship, they are based on the orientation or position relationship shown in the accompanying drawings, or are the orientation or position relationship in which the product of the invention is usually placed when used. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0076] In addition, the terms “first”, “second”, etc., if used, are merely used to distinguish between the descriptions and should not be understood as indicating or implying relative importance.

[0077] It should be noted that, in the absence of conflict, the features in the embodiments of the present invention may be combined with each other.

[0078] Please refer to Figure 1 , Figure 1 This is an example diagram of an application scenario provided in this embodiment. Figure 1In the process, the collection nodes are evenly deployed in the network area where the business nodes are located. The collection nodes and business nodes establish network connections with the controller respectively. The network connections can be through 5G wireless connections or wired connections to ensure data interoperability.

[0079] The sampling node and the service node are registered to the device management structure module of the controller respectively. The sampling node sets a software timer for periodic sampling. The controller can set the sampling parameter configuration and send it to the sampling node. The configuration of the sampling node at least includes: (1) sampling period. The sampling period can be set according to actual needs. For example, the sampling period is set to 5 minutes and sampling is performed every 5 minutes; (2) sampling signal parameters. The signal parameters to be sampled can be set according to actual needs, such as SINR (Signal to Interference plus Noise Ratio), RSRP (Reference Signal Received Power), RSSI (Received Signal Strength Indicator), time (sampling timestamp). In addition, other signal parameters of interest can be expanded, such as location information and other parameters. Among them, SINR is an important parameter for measuring the quality of wireless signals. It represents the ratio of the received useful signal power to the sum of the interference signal power and the noise power. The higher the SINR value, the better the signal quality and the more stable the performance of the communication system. RSRP is an important parameter in the LTE (Long Term Evolution) network, which indicates the average power of the reference signal received from the serving cell within a specific measurement bandwidth. RSRP is an important indicator for measuring the quality and strength of wireless signal coverage and is used to evaluate the coverage and performance of the network. RSSI indicates the signal strength received by the receiving end. RSSI value is usually used to measure the signal quality in wireless communications and is one of the important indicators for judging the stability and reliability of communication links.

[0080] After the service node is started normally, it runs the service according to the automatically selected frequency band. The frequency band is also called the frequency band, or BAND, which is the frequency range used in wireless communication. In wireless communication systems, different frequency bands are used to transmit different signals or provide different services. The selection of frequency band has an important impact on the performance and coverage of wireless communication.

[0081] The sampling nodes periodically measure the surrounding wireless signal strength (such as RSSI, RSRP, SINR, etc.). The number of sampling nodes is consistent with the number of frequency bands of the wireless network, and each sampling node is responsible for collecting information of the corresponding frequency band.

[0082] Controller: Receives frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period, analyzes and processes it, and obtains the predicted optimal frequency band.

[0083] Service node: It can be a device used by users, such as a 5G router for home use. It receives the best frequency band recommendation from the controller and adjusts its own working frequency band accordingly. If it finds that the current connection signal is not good, it will also actively request the latest prediction results to optimize the configuration.

[0084] Sampling nodes, controllers, and service nodes are usually different physical devices or logical units. The roles of each physical device can be flexibly set according to the actual deployment situation. Usually, the role of each device is determined during the planning stage. However, the role of a physical device is not fixed. For example, a physical device deployed as a sampling node can be redeployed as a controller according to actual needs.

[0085] From a hardware perspective, sampling nodes and controllers can be servers, electronic terminals, routers, switches and other devices. Different functional modules are deployed on these devices to enable them to play different roles such as sampling, control and business forwarding to achieve corresponding functions.

[0086] based on Figure 1 This embodiment also provides a block diagram of a network device, please refer to Figure 2 , Figure 2 The network device 10 is a block diagram of an example of a network device provided in this embodiment. Figure 1 The controller in the Figure 1 The network device 10 includes a processor 11, a memory 12 and a bus 13, and the processor 11 and the memory 12 are connected through the bus 13.

[0087] The processor 11 may be an integrated circuit chip having the ability to process signals. In the implementation process, each step of the wireless network access frequency band adjustment method of this embodiment may be completed by an integrated logic circuit of hardware in the processor 11 or by instructions in the form of software. The above-mentioned processor 11 may be a general-purpose processor, including a CPU (Central Processing Unit), a NP (Network Processor), etc.; it may also be a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Logic Gate Array) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0088] The memory 12 is used to store a program for implementing a method for adjusting a wireless network access frequency band. The program may be a software function module stored in the memory 12 in the form of software or firmware or solidified in an OS (Operating System) of the network device 10.

[0089] After receiving the execution instruction, the processor 11 executes the program to implement the wireless network access frequency band adjustment method of the above embodiment.

[0090] based on Figure 1 and Figure 2 This embodiment provides a method for adjusting the wireless network access frequency band of a controller, please refer to Figure 3 , Figure 3 The flowchart of the wireless network access frequency band adjustment method applied to the controller provided in this embodiment includes the following steps:

[0091] Step S101: receiving frequency band information of a corresponding wireless network frequency band collected by each sampling node within a preset time period.

[0092] In this embodiment, the frequency band information includes but is not limited to the frequency band, sampling timestamp, RSSI, RSRP, SINR, etc. In order to collect frequency band information of different frequency bands, sampling nodes in different wireless network frequency bands can be deployed to collect frequency band information. The preset time period can be set to 2 hours. Of course, the controller of the present invention can also start the prediction of the optimal frequency band after receiving a preset amount of frequency band information.

[0093] In this embodiment, the sampling node can periodically collect frequency band information, and send the frequency band information to the controller in real time each time it is collected, or it can collect a preset number of periods and send them once. The specific setting can be based on actual conditions. For example, when the current bandwidth load is relatively large, the impact on the business bandwidth can be reduced by reducing the number of transmissions.

[0094] In this embodiment, as an implementation method, the sampling node can encapsulate the frequency band information in the Ethernet message data field and transmit it to the controller on the link. After receiving this type of message, the controller decapsulates the frequency band information and stores it in the local database.

[0095] Step S102, predict the optimal frequency band according to the frequency band information, obtain the predicted optimal frequency band and the prediction accuracy, and send the predicted optimal frequency band and the prediction accuracy to each service node, so that the service node determines whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

[0096] In this embodiment, the predicted optimal frequency band is a frequency band that is more consistent with the current network environment predicted by the controller based on the frequency band information collected by each sampling node within a preset time period. The prediction accuracy is obtained by the controller evaluating the adopted prediction method based on part of the frequency band information. This allows the service node to determine whether the predicted optimal frequency band is more consistent with the needs of the node based on the actual situation and refer to the prediction accuracy, and then decide whether to adjust the currently connected wireless network frequency band based on the predicted optimal frequency band.

[0097] In this embodiment, as an implementation method, the controller may periodically broadcast and notify each service node of the optimal frequency band predicted each time and the prediction accuracy.

[0098] The above method provided in this embodiment predicts the optimal frequency band based on the frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period, obtains the predicted optimal frequency band and the prediction accuracy, and then adjusts the frequency band of the currently accessed wireless network based on the predicted optimal frequency band and the prediction accuracy, thereby realizing dynamic adjustment of the frequency band, enabling the frequency band to adapt to changes in the network environment, and ensuring the real-time nature of the frequency band update and the stability of the signal quality.

[0099] In an optional implementation, there may be multiple sampling nodes, each sampling node collects frequency band information of the wireless network frequency band corresponding to the node. This embodiment provides an implementation method for obtaining a predicted optimal frequency band and a prediction accuracy rate:

[0100] First, the frequency band information of multiple wireless network frequency bands is divided into a test set and a validation set;

[0101] In this embodiment, the controller can divide the sample data collected by each sampling node within a preset time period, that is, the frequency band information of multiple wireless network frequency bands, into a test set and a verification set according to the proportion, and the sample data in the test set is used to build a prediction model, and the sample data in the verification set is used to calculate the prediction accuracy. In order to ensure that the frequency band information in the test set and the verification set covers each frequency band, and to ensure the consistency of the prediction results and the verification results, when dividing the test set and the verification set, it is necessary to ensure that both the test set and the verification set can cover each frequency band, and to ensure that the frequency band information of the test set and the verification set conforms to the principle of random selection as much as possible.

[0102] Secondly, a prediction model is constructed based on the frequency band information in the test set;

[0103] In this embodiment, a clustering method can be used to obtain the centroid of each frequency band. The centroid of each frequency band is a concentrated reflection of the frequency band information of the frequency band. The prediction model includes the centroids of each frequency band. This embodiment provides a method for constructing a prediction model:

[0104] (1) according to the frequency band of the frequency band information in the test set, normalize the signal parameters of each frequency band to obtain the normalized signal parameters of each frequency band;

[0105] In this embodiment, the frequency band information may include signal parameters of multiple dimensions. Since the parameter values ​​of the signal parameters with larger changes have a greater impact on the prediction results, for example, the variation range of the SINR signal is approximately about 10, while the variation range of the RSRP can reach about 50, the calculated result will be greatly affected by the RSRP value, but less affected by the SINR, and the importance of each signal parameter cannot be reflected in the same standard. Therefore, in order to ensure that the weights of the statistical data of each signal parameter are consistent, it is necessary to normalize the signal parameters of each frequency band. For example, for each signal parameter of each frequency band, the characteristic value of the signal parameter is scaled to a small specific interval between -3 and 3, and the characteristic value is converted to a distribution with a mean of 0 and a standard deviation of 1. As an implementation method, for each signal parameter (as a target signal parameter) of a certain frequency band (as a target frequency band), the following formula can be used to calculate its normalized attribute:

[0106]

[0107] Z is the normalized attribute of the target signal parameter, μ is the mean of all values ​​of the target signal parameter in the target frequency band, and σ is the standard deviation of all values ​​of the target signal parameter in the target frequency band.

[0108] μ is the average value (i.e. mean) of the population, and the calculation formula is:

[0109]

[0110] σ is the population standard deviation. The calculation formula for standard deviation σ is:

[0111]

[0112] Where N is the number of data in the population, x i It is each data point in the population, each data point is a signal parameter in each frequency band, and the population is the total amount of target signal parameters in the target frequency band in the set.

[0113] (2) Calculate the centroid of each frequency band based on the mean of each normalized signal parameter of each frequency band;

[0114] In this embodiment, for each signal parameter of each frequency band, its corresponding normalized signal parameter is calculated. For any frequency band, all its normalized signal parameters are averaged and vectorized to obtain the centroid of any frequency band. One normalized signal parameter corresponds to one dimension of the centroid.

[0115] As a specific implementation method, for any target normalized signal parameter of the target frequency band, the mean of all target normalized signal parameters can be calculated, and the obtained mean is used as the dimension corresponding to the target normalized signal parameter. The dimension corresponding to the mean of all normalized signal parameters of the target frequency band is represented by a vector to obtain the centroid of the target frequency band.

[0116] (3) Construct a prediction model based on the centroid of all frequency bands.

[0117] In this embodiment, the prediction model includes the centroid of all frequency bands. For example, three frequency bands: BAND1, BAND2, and BAND3, and the signal parameters of each frequency band include two dimensions: SINR and RSSI. Please refer to Figure 4 , Figure 4 This is an example diagram of the centroid of the frequency band provided in this embodiment. Figure 4 In the figure, triangles, rectangles and circles represent different frequency bands respectively. The frequency band information of each frequency band includes three dimensions: SINR, RSRP and RSSI, which correspond to the three coordinate axes of SINR, RSRP and RSSI in the coordinate system respectively. A triangle represents the vector representation corresponding to a frequency band information of BAND1, and the red triangle represents the centroid of BAND1. A rectangle represents the vector representation corresponding to a frequency band information of BAND2, and the red rectangle represents the centroid of BAND2. A circle represents the vector representation corresponding to a frequency band information of BAND3, and the red circle represents the centroid of BAND3.

[0118] Third, according to the preset signal parameters, the prediction model is used to predict the optimal frequency band to obtain the predicted optimal frequency band;

[0119] In this embodiment, based on the three dimensions of RSSI, RSRP and SINR, the signal quality can be divided into four levels: excellent, good, fair, and poor. The values ​​of the signal parameters corresponding to each level are as follows:

[0120] Excellent: RSSI:>-70 RSRP:>-70 SINR:>15

[0121] Good: RSSI: -85~-70 RSRP: -85~-70 SINR: 10-15

[0122] Normal: RSSI: -95~-85 RSRP: -95~-85 SINR: 5~10

[0123] Bad: RSSI: <-95 RSRP: <-95 SINR: <5

[0124] The preset optimal signal parameters can be determined from the value range corresponding to the excellent or good level according to actual business needs or business expectations. For example, if the actual business demand is the "good" level, a value is selected from the RSSI range corresponding to the "good" level as the RSSI dimension of the preset optimal signal parameters, and the other two dimensions RSRP and SINR are similar.

[0125] A method for implementing frequency band information prediction is:

[0126] (1) Calculate the spatial distance between the preset optimal signal parameters and the centroid of each frequency band;

[0127] (2) The frequency band corresponding to the minimum value of all spatial distances is taken as the optimal frequency band for prediction.

[0128] In this embodiment, the spatial distance may be a Euclidean distance, that is, the Euclidean distance between the preset optimal signal parameter and the centroid of each frequency band is calculated. The frequency band to which the centroid with the smallest spatial distance belongs is the predicted optimal frequency band, that is, the frequency band with the highest probability of matching the preset optimal signal parameter, and the corresponding signal parameter is the predicted optimal signal parameter.

[0129] Continue to view Figure 4 , Figure 4 In the figure, the green cross box indicates the preset optimal signal parameters, and d1~d3 respectively indicate the spatial distance between it and the centroid of BAND1~BAND3. Figure 4 It can be seen that d2 is the smallest, then BAND2 is the preset frequency band, and the signal parameters of BAND2 are the predicted optimal signal parameters.

[0130] This embodiment also takes the frequency band information of 100 samples of frequency band 0 and frequency band 1 collected in a certain period as an example, and gives an example of the prediction model construction process, taking the signal parameters including RSRP and SINR as an example. Figure 5 , Figure 5 This is an example diagram of the frequency band information before normalization provided in this embodiment. Figure 5 In the figure, the blue circle indicates the BAND0 frequency band, the orange circle indicates the BAND1 frequency band, the purple circle indicates the centroid of BAND0, the green circle indicates the centroid of BAND1, and the red circle indicates the preset optimal signal parameters. Figure 6 , Figure 6 The pair provided in this embodiment Figure 5 An example diagram of the frequency band information after the signal parameters in are normalized and the average is calculated. Figure 6 In the example, the centroid of band 0 is [RSRP, SINR] = [-0.74, 0.61], and the centroid of band 1 is [RSRP, SINR] = [0.74-0.61]). According to the signal quality level classification, if the preset optimal signal parameters (such as [RSRP, SINR] = [-60, 17], the coordinates obtained after normalization are [0.31, 1.41], such as Figure 6 As shown in the red circle in , [0.31, 1.41] is brought into the prediction model, and its distance to the two centroids (Euclidean distance) is calculated respectively. The calculation results are:

[0131] Distance to centroid of BAND 0 (the distance between the normalized preset signal parameters and the centroid of band 0): 1.32

[0132] Distance to centroid of BAND 1 (the distance between the normalized preset signal parameters and the centroid of band 1): 2.06

[0133] From the calculation results, we can see that the closer the preset signal parameters are to the centroid of frequency band 0, the greater the probability of achieving the high-quality signal parameter [RSRP, SINR] = [-60, 17] when the service node works in frequency band 0. Therefore, the frequency band accessed by the service node in this period is 0, and the signal quality may be better. At this time, frequency band 0 is the predicted optimal frequency band.

[0134] It should be noted that the controller can also record the centroid of each prediction model according to the time period (the time period is obtained by calculating the difference between the sampling timestamp in the earliest received frequency band information and the timestamp in the last received frequency band information within the preset time period) to form historical statistics, which is conducive to the later troubleshooting of network problems. For example, if the signal parameters of a certain frequency band in a certain period of time seriously deviate from the centroid, resulting in a worse signal, it can be estimated that the frequency band may be under maintenance.

[0135] Finally, the prediction accuracy of the prediction model is calculated based on the frequency band information in the validation set.

[0136] In this embodiment, since the frequency band and signal parameters in the frequency band information in the verification set are known, the signal parameters can be used as target signal parameters, and the corresponding predicted frequency band can be calculated using the prediction model. If the predicted frequency band is consistent with the known frequency band, the prediction model is judged to be correct, otherwise the prediction model is judged to be wrong. According to the number of correct predictions and the total number of predictions, the prediction accuracy can be obtained. The corresponding implementation method is:

[0137] (1) The signal parameters of each frequency band (as the target frequency band) in the validation set are taken as the target signal parameters in turn, and the frequency band of the target signal parameters is predicted using the prediction model to obtain the predicted frequency band of the target signal parameters;

[0138] In this embodiment, since the centroids of each frequency band included in the prediction model are known, the spatial distance between the target signal parameters and the centroids of each frequency band of the prediction model can be calculated, and the frequency band to which the centroid corresponding to the minimum spatial distance belongs is taken as the predicted frequency band.

[0139] (2) If the target frequency band of the target signal parameter is consistent with the predicted frequency band, it is determined that the prediction of the target signal parameter is correct; otherwise, it is determined that the prediction is wrong;

[0140] (3) Count the number of signal parameters in the correct frequency bands predicted in the validation set;

[0141] (4) The prediction accuracy of the prediction model is calculated based on the number of signal parameters of the frequency bands predicted correctly in the validation set and the total number of signal parameters of all frequency bands in the validation set.

[0142] In this embodiment, since the frequency band information in the test set for constructing the prediction model and the validation set for evaluating the prediction accuracy are collected from each sampling node within a preset time period, and the test set and the validation set are randomly divided, the consistency of the prediction results and the prediction accuracy can be better guaranteed.

[0143] In this embodiment, when the prediction accuracy is lower than the preset accuracy threshold, in order to improve the prediction accuracy of the prediction model, this embodiment provides at least two implementation methods:

[0144] Method 1:

[0145] If the prediction accuracy is lower than the preset accuracy threshold, the sampling frequency of the sampling node is increased.

[0146] In this embodiment, increasing the sampling frequency means that a larger number of samples are used when the prediction model is subsequently reconstructed, so that the prediction result obtained will be more accurate, thereby improving the prediction accuracy.

[0147] In addition to method 1, since the impact of various signal parameters on signal quality is different in different application scenarios, in order to make the signal parameters more adaptable to the current application scenario and more accurately predict the frequency band, the prediction accuracy can also be improved by adjusting the weight of the signal parameters, that is, method 2:

[0148] (1) If the prediction accuracy is lower than the preset accuracy threshold, the weight of each signal parameter is adjusted;

[0149] (2) Reconstruct the prediction model and recalculate the prediction accuracy according to the adjusted weight of each signal parameter until the prediction accuracy reaches a preset accuracy threshold.

[0150] It should be noted that the above-mentioned method 1 and method 2 are only two of multiple implementation methods. In fact, technical personnel in this field can also expand the above-mentioned implementation methods according to the needs of actual scenarios, for example, adding more signal parameters (such as the location information of the sampling nodes) to improve the prediction accuracy.

[0151] Based on the above wireless network access frequency band adjustment method applied to the controller, this embodiment also provides a wireless network access frequency band adjustment method applied to the service node, so as to reasonably adjust the wireless network access frequency band according to the actual situation of the service node. Figure 7 , Figure 7 The flowchart of the wireless network access frequency band adjustment method applied to the service node provided in this embodiment includes the following steps:

[0152] Step S201, receiving a predicted optimal frequency band and a prediction accuracy rate sent by a controller, wherein the predicted optimal frequency band and the prediction accuracy rate are obtained by predicting the optimal frequency band according to the received frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period;

[0153] Step S202: determining whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

[0154] In this embodiment, if the service node determines that the predicted optimal frequency band is better than the wireless network frequency band currently accessed by the node, the wireless network frequency band currently accessed by the node can be adjusted to the predicted optimal frequency band to obtain better signal quality. Otherwise, the service node may not make adjustments, but keep the currently accessed wireless network frequency band unchanged.

[0155] This embodiment provides an adjustment method for performing different processing according to different situations:

[0156] (1) determining whether the signal parameter of the wireless network frequency band currently accessed is greater than a preset alarm threshold;

[0157] In this embodiment, if the signal parameter of the currently accessed wireless network frequency band is lower than the preset alarm threshold, it means that the signal quality may be extremely poor at this time and the currently accessed wireless network frequency band needs to be adjusted. Otherwise, it means that the signal quality is acceptable at this time and can meet the current business needs. Whether it needs to be adjusted requires further judgment.

[0158] In this embodiment, since there can be multiple signal parameters, a corresponding preset alarm threshold can be set for each signal parameter, and the judgment condition of whether the signal parameter of the currently connected wireless network frequency band is greater than the preset alarm threshold can be determined according to the actual scenario needs. For example, if all signal parameters of the currently connected wireless network frequency band are lower than the corresponding preset alarm threshold, then it is determined that the signal parameter of the currently connected wireless network frequency band is lower than the preset alarm threshold. It can also be that if more than a preset number of signal parameters of the currently connected wireless network frequency band are lower than the corresponding preset alarm threshold, then it is determined that the signal parameter of the currently connected wireless network frequency band is lower than the preset alarm threshold.

[0159] (2) if the signal parameter of the wireless network frequency band currently accessed is lower than the preset alarm threshold, adjusting the wireless network frequency band currently accessed to the predicted optimal frequency band;

[0160] (3) if the signal parameter of the wireless network frequency band currently accessed is greater than the preset alarm threshold, determining whether the predicted optimal frequency band is better than the wireless network frequency band currently accessed;

[0161] In this embodiment, as an implementation method for determining whether the predicted optimal frequency band is better than the currently accessed wireless network frequency band, the levels of the two can be determined respectively, and the superiority can be determined according to the different levels. For example, if the signal parameter of the currently accessed wireless network frequency band is at the "general" level, and the near-end attribute of the predicted optimal frequency band is at the "good" level, then the latter is better than the former.

[0162] (4) If the predicted optimal frequency band is better than the currently connected wireless network frequency band and the prediction accuracy is greater than a preset accuracy threshold, the currently connected wireless network frequency band is adjusted to the predicted optimal frequency band;

[0163] (5) If the predicted optimal frequency band is not better than the currently connected wireless network frequency band, the currently connected wireless network frequency band remains unchanged.

[0164] In this embodiment, if the controller has not issued the predicted optimal frequency band, and the service node detects that the signal quality is poor, the service node can also actively obtain the predicted optimal frequency band from the controller, which is implemented as follows:

[0165] If the frequency band of the currently connected wireless network is not greater than the preset alarm threshold, the predicted optimal frequency band is obtained from the controller, and the frequency band of the currently connected wireless network is adjusted to the predicted optimal frequency band.

[0166] In this embodiment, in order to prevent the service node from failing to receive the predicted optimal frequency band sent by the controller due to network disconnection, this embodiment further provides a processing method:

[0167] The predicted optimal frequency band received most recently is stored, so that when it is detected that the connection with the controller is disconnected, the frequency band of the currently connected wireless network is adjusted according to the stored predicted optimal frequency band.

[0168] In order to execute the corresponding steps in the above embodiment and each possible implementation method, the following provides a method for implementing a wireless network access frequency band adjustment device 100, which is applied to a controller. Figure 8 , Figure 8 This is a block diagram of the wireless network access frequency band adjustment device provided in this embodiment. It should be noted that the basic principle and technical effect of the wireless network access frequency band adjustment device 100 provided by the present invention are the same as those of the corresponding above-mentioned embodiments, and are not mentioned in this embodiment for the sake of brief description.

[0169] The wireless network access frequency band adjustment device 100 includes a receiving module 110 and a prediction module 120 .

[0170] The receiving module 110 is used to receive the frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period;

[0171] The prediction module 120 is used to predict the optimal frequency band according to the frequency band information, obtain the predicted optimal frequency band and the prediction accuracy, and send the predicted optimal frequency band and the prediction accuracy to each service node, so that the service node determines whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

[0172] In an optional implementation, there are multiple sampling nodes, and each sampling node is used to collect frequency band information of the wireless network frequency band corresponding to the node according to a preset sampling period;

[0173] The prediction module 120 is specifically used to: divide the frequency band information of multiple wireless network frequency bands into a test set and a verification set; build a prediction model based on the frequency band information in the test set; use the prediction model to predict the optimal frequency band according to preset optimal signal parameters to obtain a predicted optimal frequency band; calculate the prediction accuracy of the prediction model based on the frequency band information in the verification set.

[0174] In an optional implementation, the frequency band information includes the frequency band and the signal parameters. When the prediction module 120 is specifically used to construct the prediction model according to the frequency band information in the test set, it is specifically used to: normalize the signal parameters of each frequency band according to the frequency band of the frequency band information in the test set to obtain the normalized signal parameters of each frequency band; calculate the centroid of each frequency band according to the mean of the normalized signal parameters of each frequency band; and construct the prediction model according to the centroid of all frequency bands;

[0175] and / or

[0176] The prediction model includes the centroid of at least one frequency band, and the centroid of each frequency band is obtained by calculating the mean after normalizing the signal parameters of each frequency band in the test set;

[0177] The prediction module 120 is specifically used to predict frequency band information based on preset signal parameters using a prediction model, and when obtaining the predicted optimal frequency band, it is also specifically used to: calculate the spatial distance between the preset optimal signal parameters and the centroid of each frequency band; and take the frequency band corresponding to the minimum value of all spatial distances as the predicted optimal frequency band.

[0178] In an optional implementation, the frequency band information includes frequency band and signal parameters;

[0179] When the prediction module 120 is specifically used to calculate the prediction accuracy of the prediction model based on the frequency band information in the verification set, it is specifically used to: take the signal parameters of each frequency band in the verification set as the target signal parameters in turn, use the prediction model to perform frequency band prediction on the target signal parameters, and obtain the predicted frequency band of the target signal parameters; if the target frequency band of the target signal parameter is consistent with the predicted frequency band, then it is determined that the prediction for the target signal parameter is correct, otherwise it is determined that the prediction is wrong; count the number of signal parameters of the frequency bands predicted correctly in the verification set; calculate the prediction accuracy of the prediction model based on the number of signal parameters of the frequency bands predicted correctly in the verification set and the total number of signal parameters of all frequency bands in the verification set.

[0180] In an optional embodiment, the prediction module 120 is specifically used to predict the number of signal parameters of the correct frequency band in the verification set and the total number of signal parameters of all frequency bands in the verification set, after being used to calculate the prediction accuracy of the prediction model, and is also specifically used to:

[0181] If the prediction accuracy is lower than the preset accuracy threshold, the sampling frequency of the sampling node is increased;

[0182] and / or

[0183] Each frequency band information includes a plurality of signal parameters. The prediction module 120 is specifically used to predict the number of signal parameters of the correct frequency band in the verification set and the total number of signal parameters of all frequency bands in the verification set, and after calculating the prediction accuracy of the prediction model, is also specifically used to:

[0184] If the prediction accuracy is lower than the preset accuracy threshold, the weight of each signal parameter is adjusted;

[0185] According to the adjusted weight of each signal parameter, the prediction model is rebuilt and the prediction accuracy is recalculated until the prediction accuracy reaches the preset accuracy threshold.

[0186] In order to execute the corresponding steps in the above embodiment and each possible implementation method, the following provides a method for implementing a wireless network access frequency band adjustment device 200, which is applied to a service node. Fig. 9 , Fig. 9 The block diagram of the wireless network access frequency band adjustment device 200 provided in this embodiment. It should be noted that the basic principle and technical effect of the wireless network access frequency band adjustment device 200 provided by the present invention are the same as those of the corresponding above-mentioned embodiments, which are not mentioned in this embodiment for the sake of brief description.

[0187] The wireless network access frequency band adjustment device 200 includes a receiving module 210 and an adjustment module 220 .

[0188] The receiving module 210 is used to receive the predicted optimal frequency band and the prediction accuracy sent by the controller. The predicted optimal frequency band, i.e., the prediction accuracy, is obtained by the controller predicting the optimal frequency band according to the frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period.

[0189] The adjustment module 220 is used to determine whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

[0190] In an optional implementation, the adjustment module 220 is specifically used to: determine whether the signal parameter of the currently accessed wireless network frequency band is greater than a preset alarm threshold; if the signal parameter of the currently accessed wireless network frequency band is lower than the preset alarm threshold, adjust the currently accessed wireless network frequency band to the predicted optimal frequency band; if the signal parameter of the currently accessed wireless network frequency band is greater than the preset alarm threshold, determine whether the predicted optimal frequency band is better than the currently accessed wireless network frequency band; if the predicted optimal frequency band is better than the currently accessed wireless network frequency band, and the prediction accuracy is greater than the preset accuracy threshold, adjust the currently accessed wireless network frequency band to the predicted optimal frequency band; if the predicted optimal frequency band is not better than the currently accessed wireless network frequency band, keep the currently accessed wireless network frequency band unchanged.

[0191] In an optional implementation, the adjustment module 220 is further configured to: if the signal parameter of the currently accessed wireless network frequency band is lower than a preset alarm threshold, obtain a predicted optimal frequency band from the controller, and adjust the currently accessed wireless network frequency band to the predicted optimal frequency band;

[0192] and / or

[0193] The adjustment module 220 is also used for:

[0194] The predicted optimal frequency band received most recently is stored, so that when it is detected that the connection with the controller is disconnected, the frequency band of the wireless network currently accessed by the node is adjusted according to the stored predicted optimal frequency band.

[0195] This embodiment also provides an example diagram of module division based on the overall system structure, please refer to Fig.10 , Fig.10 This is an example diagram of the overall structural framework provided in this embodiment. Fig.10 In the system, the system includes a sampling node module, a controller module and a business node module, which run on the sampling node, the controller and the business node respectively.

[0196] Among them, the sampling node module includes a data acquisition unit. The signal judgment standard can evaluate the signal quality level through the parameter values ​​of signal-to-noise ratio (SINR), signal strength (RSSI), and signal receiving rate (RSRP). This unit works on hardware and consists of multiple sampling nodes distributed in the network. The number of sampling nodes is set to the number of frequency bands in the area, and is distributed according to the location of the service nodes in the same area.

[0197] The controller module includes a data preprocessing unit, an intelligent prediction unit and a frequency band sending unit.

[0198] Data preprocessing unit: Since the prediction model makes predictions by calculating the Euclidean distance of the parameter space, parameter values ​​with large changes have a greater impact on the prediction results. In order to ensure the consistency of the weight of the statistical data, this unit will normalize the raw data uploaded by the sampling node and store it locally for subsequent analysis. The data preprocessing unit ensures the accuracy and consistency of the data and provides high-quality data input for the training of the prediction model.

[0199] Intelligent prediction unit: This unit is a software module located in the controller. It uses the knn algorithm to build a prediction model and uses the historical data provided by the data acquisition unit to train a prediction model that can predict the optimal frequency band. Through the prediction model, the system can identify potential frequency band congestion and signal interference problems in advance and adjust the frequency band allocation in advance, which can not only improve network performance and stability, but also provide users with better network services. At the same time, the model verification and optimization unit ensures the accuracy and reliability of the prediction model and ensures the continuous optimization of the system.

[0200] Frequency band sending unit: responsible for sending the optimal frequency band value output by the intelligent prediction unit to the service nodes in the network. The predicted optimal frequency band is transmitted to the service node through the connection protocol with the controller (local area network or point-to-point connection), and the integrity of the data is ensured. Through the frequency band sending unit, the system can quickly and accurately convey the predicted optimal frequency band to the service node, realizing the dynamic allocation of frequency band resources.

[0201] The service node module includes a frequency band adjustment execution unit.

[0202] Frequency band adjustment execution unit: This unit works on the service node and is responsible for monitoring the network status of the current working frequency band of the service equipment. Its main function is to actively or passively adjust its own frequency band and optimize the configuration parameters according to the comparison between the current working status and the frequency band parameters sent by the controller. Considering the possibility of network disconnection, the local record is used to send the frequency band. When the network is disconnected, it automatically switches to the most recently received frequency band to ensure the signal quality after re-online. At the same time, the frequency band adjustment can ensure that the service equipment continues to work on the frequency band with stable signal quality, avoiding problems such as frequency band conflicts or signal quality degradation.

[0203] The present invention provides a program product. When the program product is executed by a processor, the wireless network access frequency band adjustment method applied to a controller described in the above implementation manner is implemented, or the wireless network access frequency band adjustment method applied to a service node described in the above implementation manner is implemented.

[0204] In summary, an embodiment of the present invention provides a method, device and program product for adjusting a wireless network access frequency band, the method comprising: receiving frequency band information of a corresponding wireless network frequency band collected by each sampling node within a preset time period; predicting an optimal frequency band based on the frequency band information, obtaining a predicted optimal frequency band and a prediction accuracy, and sending the predicted optimal frequency band and the prediction accuracy to each service node, so that the service node determines whether to adjust the wireless network frequency band currently accessed by the node based on the predicted optimal frequency band and the prediction accuracy. Compared with the prior art, this embodiment has at least the following advantages: (1) It realizes dynamic adjustment of the frequency band, so that the frequency band can adapt to changes in the network environment, ensuring the real-time nature of the frequency band update and the stability of the signal quality; (2) It captures the signal-to-noise ratio, signal strength, signal reception rate, timestamp and other frequency band information of different network frequency bands in real time and sends them to the controller in time, ensuring that the data used in the prediction can accurately and comprehensively reflect the network status information; (3) Through normalization processing, it avoids the influence of large changes in signal parameters on the prediction results, thereby affecting the accuracy of the prediction results; (4) By constructing a prediction model, it can identify potential frequency band congestion and signal interference problems in advance, and adjust the frequency band allocation in advance, which can not only improve network performance and stability, but also provide users with better network services; (5) The accuracy of the prediction model is verified, so that the service node can make more reasonable wireless network access frequency band adjustments according to the accuracy of the prediction model; (6) The service node pre-stores the most recently received predicted optimal frequency band. When a network disconnection is detected, it can automatically switch to the most recently predicted optimal frequency band to ensure the signal quality after re-accessing the network. At the same time, ensure that business equipment continues to operate in a frequency band with stable signal quality to avoid problems such as frequency band conflicts or signal quality degradation.

[0205] The above are only various embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for adjusting a wireless network access frequency band, characterized in that: A controller applied to a wireless network, wherein the wireless network further includes a sampling node and a service node, and the method includes: Receiving frequency band information of a corresponding wireless network frequency band collected by each sampling node within a preset time period; The optimal frequency band is predicted based on the frequency band information to obtain the predicted optimal frequency band and the prediction accuracy, and the predicted optimal frequency band and the prediction accuracy are sent to each of the service nodes so that the service node determines whether to adjust the wireless network frequency band currently accessed by the node based on the predicted optimal frequency band and the prediction accuracy.

2. The wireless network access frequency band adjustment method according to claim 1, characterized in that: There are multiple sampling nodes, each of which is used to collect frequency band information of the wireless network frequency band corresponding to the node according to a preset sampling period; the step of predicting the optimal frequency band according to the frequency band information to obtain the predicted optimal frequency band and the prediction accuracy includes: Dividing the frequency band information of multiple wireless network frequency bands into a test set and a validation set; Building a prediction model based on the frequency band information in the test set; According to the preset optimal signal parameters, the prediction model is used to predict the optimal frequency band to obtain the predicted optimal frequency band; The prediction accuracy of the prediction model is calculated based on the frequency band information in the verification set.

3. The wireless network access frequency band adjustment method according to claim 2, characterized in that: The frequency band information includes frequency band and signal parameters, and the step of constructing a prediction model according to the frequency band information in the test set includes: According to the frequency band of the frequency band information in the test set, normalize each signal parameter of each frequency band to obtain the normalized signal parameter of each frequency band; Calculating the centroid of each frequency band according to the mean of each normalized signal parameter of each frequency band; constructing the prediction model based on the centroids of all the frequency bands; and / or The prediction model includes the centroid of each frequency band, and the centroid of each frequency band is obtained by calculating the mean after normalizing the signal parameters of each frequency band in the test set; the step of predicting the optimal frequency band using the prediction model according to the preset optimal signal parameters to obtain the predicted optimal frequency band includes: Calculating the spatial distance between the preset optimal signal parameter and the centroid of each of the frequency bands; The frequency band corresponding to the minimum value of all the spatial distances is used as the predicted optimal frequency band.

4. The wireless network access frequency band adjustment method according to claim 2, characterized in that: The frequency band information includes frequency band and signal parameters; the step of calculating the prediction accuracy of the prediction model according to the frequency band information in the verification set includes: sequentially taking the signal parameters of each frequency band in the validation set as target signal parameters, and using the prediction model to perform frequency band prediction on the target signal parameters to obtain a predicted frequency band of the target signal parameters; If the frequency band of the target signal parameter is consistent with the predicted frequency band, it is determined that the prediction of the target signal parameter is correct, otherwise it is determined that the prediction is wrong; Counting the number of signal parameters of the frequency bands predicted correctly in the validation set; The prediction accuracy of the prediction model is calculated based on the number of signal parameters of the frequency bands predicted correctly in the verification set and the total number of signal parameters of all frequency bands in the verification set.

5. The method for adjusting the wireless network access frequency band according to claim 4, characterized in that: According to the number of signal parameters of the frequency bands predicted correctly in the verification set and the total number of signal parameters of all frequency bands in the verification set, after the step of calculating the prediction accuracy of the prediction model, the method further includes: If the prediction accuracy is lower than a preset accuracy threshold, increasing the sampling frequency of the sampling node; and / or Each of the frequency band information includes a plurality of signal parameters. According to the number of signal parameters of the frequency bands predicted correctly in the verification set and the total number of signal parameters of all frequency bands in the verification set, after the step of calculating the prediction accuracy of the prediction model, the method further includes: If the prediction accuracy is lower than a preset accuracy threshold, adjusting the weight of each signal parameter; According to the adjusted weight of each signal parameter, the prediction model is reconstructed and the prediction accuracy is recalculated until the prediction accuracy reaches the preset accuracy threshold.

6. A method for adjusting a wireless network access frequency band, characterized in that: Applied to a service node in a wireless network, the wireless network also includes a sampling node and a controller, the method includes: Receiving the predicted optimal frequency band and the prediction accuracy sent by the controller, wherein the predicted optimal frequency band and the prediction accuracy are obtained by the controller predicting the optimal frequency band according to the received frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period; Determine whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

7. The method for adjusting the wireless network access frequency band according to claim 6, characterized in that: The step of determining whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy includes: Determine whether a signal parameter of the currently accessed wireless network frequency band is greater than a preset alarm threshold; If the signal parameter of the currently accessed wireless network frequency band is lower than the preset alarm threshold, adjusting the currently accessed wireless network frequency band to the predicted optimal frequency band; If the signal parameter of the currently accessed wireless network frequency band is greater than the preset alarm threshold, determining whether the predicted optimal frequency band is better than the currently accessed wireless network frequency band; If the predicted optimal frequency band is better than the currently accessed wireless network frequency band, and the prediction accuracy is greater than a preset accuracy threshold, adjusting the currently accessed wireless network frequency band to the predicted optimal frequency band; If the predicted optimal frequency band is not better than the currently accessed wireless network frequency band, the currently accessed wireless network frequency band remains unchanged.

8. The method for adjusting the wireless network access frequency band according to claim 6, characterized in that: The method further comprises: If the signal parameter of the currently accessed wireless network frequency band is lower than a preset alarm threshold, obtaining the predicted optimal frequency band from the controller, and adjusting the currently accessed wireless network frequency band to the predicted optimal frequency band; and / or The method further comprises: The predicted optimal frequency band received most recently is stored, so that when it is detected that the connection with the controller is disconnected, the wireless network frequency band currently accessed by the node is adjusted according to the stored predicted optimal frequency band.

9. A wireless network access frequency band adjustment device, characterized in that: A controller applied to a wireless network, wherein the wireless network further includes a sampling node and a service node, and the device includes: A receiving module, used to receive the frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period; A prediction module is used to predict the optimal frequency band according to the frequency band information, obtain the predicted optimal frequency band and the prediction accuracy, and send the predicted optimal frequency band and the prediction accuracy to each of the service nodes, so that the service node determines whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

10. A wireless network access frequency band adjustment device, characterized in that: The device is applied to a service node in a wireless network, wherein the wireless network further includes a sampling node and a controller, and the device includes: A receiving module, used for receiving the predicted optimal frequency band and the prediction accuracy sent by the controller, wherein the predicted optimal frequency band and the prediction accuracy are obtained by the controller predicting the optimal frequency band according to the received frequency band information of the corresponding wireless network frequency band collected by each sampling node within a preset time period; The adjustment module is used to determine whether to adjust the wireless network frequency band currently accessed by the node according to the predicted optimal frequency band and the prediction accuracy.

11. A program product, characterized in that When the program product is executed by a processor, the wireless network access frequency band adjustment method described in any one of claims 1 to 5 is implemented, or the wireless network access frequency band adjustment method described in any one of claims 6 to 8 is implemented.

Images