Optimization method, communication equipment and storage medium for filter window selection model

Abstract

The embodiment of the present invention provides an optimization method for a filter window selection model, a communication device and a storage medium, the method comprising: processing the communication data according to the filter window selection model corresponding to the scheduling level value of the communication data to obtain a target filter window corresponding to the communication data; determining the classification accuracy of the filter window selection model according to the target filter window and a preset filter window label corresponding to the communication data; when the classification accuracy is less than or equal to the preset accuracy, adjusting the model parameters of the filter window selection model according to the preset filter window label corresponding to the communication data to obtain an optimized filter window selection model, so as to improve the convenience of optimizing the filter window selection model, improve the optimization effect of the filter window selection model, and thus improve the classification accuracy of filter window classification of the communication data.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to an optimization method for a filter window selection model, communication equipment and a storage medium. Background Art

[0002] In the process of channel estimation filtering, the communication system needs to select a suitable filter window to improve the accuracy of channel estimation filtering and weaken the influence of noise and interference signals in the channel. However, in the prior art, it is generally necessary to design or adjust the selection scheme of the filter window based on expert experience, which takes a lot of time and energy, and it is easy to lead to poor convenience in optimizing the selection scheme of the filter window, poor optimization effect of the selection scheme of the filter window, and thus poor classification accuracy of the filter window. Summary of the invention

[0003] The main purpose of the present invention is to provide an optimization method, communication equipment and storage medium for a filter window selection model, aiming to improve the convenience of optimizing the filter window selection model, improve the optimization effect of the filter window selection model, and improve the classification accuracy of the filter window classification of communication data.

[0004] In a first aspect, an embodiment of the present invention provides a method for optimizing a filter window selection model, comprising:

[0005] According to the filter window selection model corresponding to the scheduling level value of the communication data, the communication data is processed to obtain the target filter window corresponding to the communication data;

[0006] Determining the classification accuracy of the filter window selection model according to the target filter window and the preset filter window label corresponding to the communication data;

[0007] When the classification accuracy is less than or equal to the preset accuracy, the model parameters of the filter window selection model are adjusted according to the preset filter window label corresponding to the communication data to obtain an optimized filter window selection model.

[0008] In a second aspect, an embodiment of the present invention further provides a communication device, the communication device comprising a memory and a processor;

[0009] The memory is used to store computer programs;

[0010] The processor is used to execute the computer program and implement any one of the methods for optimizing the filter window selection model provided in the specification of the present invention when executing the computer program.

[0011] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program can be executed by a processor, and the processor implements any one of the optimization methods for the filter window selection model provided in the specification of the present invention.

[0012] The embodiment of the present invention provides an optimization method for a filter window selection model, a communication device and a storage medium. The embodiment of the present invention processes communication data according to a filter window selection model corresponding to a scheduling level value of the communication data to obtain a target filter window corresponding to the communication data; determines the classification accuracy of the filter window selection model according to the target filter window and a preset filter window label corresponding to the communication data; when the classification accuracy is less than or equal to the preset accuracy, adjusts the model parameters of the filter window selection model according to the preset filter window label corresponding to the communication data to obtain an optimized filter window selection model, which is conducive to improving the convenience of optimizing the filter window selection model and improving the optimization effect of the filter window selection model, so as to improve the classification accuracy of filter window classification of communication data. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0014] Figure 1 A schematic flow chart of a method for optimizing a filter window selection model provided by an embodiment of the present invention;

[0015] Figure 2 A schematic diagram of channel characteristic information of communication data involved in an embodiment of the present invention;

[0016] Figure 3 A schematic diagram of an application scenario of an optimization method for a filter window selection model according to another embodiment of the present invention;

[0017] Figure 4 A schematic block diagram of a communication device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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. 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.

[0019] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0020] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0021] The embodiment of the present invention provides a method for optimizing a filter window selection model, a communication device, and a storage medium. The method for optimizing a filter window selection model can be applied to a communication device. The communication device can be a base station, a server, an electronic device, etc., which is not limited here.

[0022] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0023] Exemplarily, the existing filter window selection scheme mostly adopts the method of selecting the filter window based on expert experience. When the channel environment changes, the filter window selection scheme needs to be adjusted manually, which easily leads to poor convenience in optimizing the filter window selection scheme. Moreover, in the process of manually adjusting the parameters related to the filter window selection, it is easy to affect the channel estimation process of the communication system, which easily leads to poor optimization effect of the filter window selection scheme and adversely affects the classification accuracy of the filter window.

[0024] Based on this, the filter window selection scheme can be optimized according to the optimization method of the filter window selection model provided in an embodiment of the present invention, for example, the filter window selection model can be optimized to improve the convenience of optimizing the filter window selection model, improve the optimization effect of the filter window selection model, and improve the classification accuracy of the filter window classification of communication data.

[0025] See also Figure 1 , Figure 1A flow chart of an optimization method for a filter window selection model provided in an embodiment of the present invention. Exemplarily, the optimization method for the filter window selection model is applied to a communication device. The communication device may be a base station. The communication device may also be an electronic device, and the electronic device may be a computer, a mobile phone or other device. The communication device may also be a server, and the server may be a separate server, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. The server may be a separate server or a server cluster, which is not limited here.

[0026] like Figure 1 As shown, the optimization method of the filter window selection model includes steps S101 to S103.

[0027] Step S101 : selecting a filter window model according to a scheduling level value of communication data, processing the communication data, and obtaining a target filter window corresponding to the communication data.

[0028] Exemplarily, when the communication data is acquired, the target filter window corresponding to the communication data can be determined according to the filter window selection model. For example, the communication data is input into the filter window selection model, and the filter window selection model can process the communication data to obtain the target filter window corresponding to the communication data.

[0029] In some embodiments, multiple filter window selection models may be provided, wherein different filter window selection models are used to select target filter windows for communication data with different scheduling level values. For example, for multiple communication data, when the scheduling level values ​​of the multiple communication data are obtained, the scheduling level values ​​of the multiple communication data may be clustered to determine the cluster center points corresponding to the scheduling level values ​​of the multiple communication data. According to the cluster center points, the multiple scheduling level values ​​may be divided to determine multiple scheduling level value ranges. Based on this, an association relationship between multiple scheduling level value ranges and multiple filter window selection models may be established.

[0030] For example, the filter window selection model corresponding to the scheduling level value of the communication data can be determined according to the association between the scheduling level value of the communication data and multiple scheduling level value ranges and multiple filter window selection models. For example, if the scheduling level value of the communication data is within the scheduling level value range corresponding to a certain filter window selection model, the filter window selection model can be determined as the filter window selection model corresponding to the scheduling level value of the communication data. Accordingly, when the filter window selection model corresponding to the scheduling level value of the communication data is determined, the filter window selection model can process the communication data to obtain the target filter window corresponding to the communication data.

[0031] In some implementations, the filter window selection model can perform feature extraction processing on the communication data to obtain channel feature information of the communication data. The channel feature information can be used to characterize the channel state. Figure 2 As shown, the channel characteristic information may include the ratio of the main path and secondary path of the channel signal to the noise estimate, the channel path number estimate, the channel time domain waveform characteristics, etc., of course, it is not limited to this, and no limitation is made here. For example, an extraction priority may be set between different channel characteristic information, and the extraction priority may be used to indicate that the corresponding channel characteristic information is the channel characteristic information that must be extracted, or to indicate that the corresponding channel characteristic information is the channel characteristic information that is selected to be extracted. The filter window selection model can perform feature extraction processing on the communication data according to the extraction priority between different channel characteristic information to obtain the channel characteristic information of the communication data. For example, the filter window selection model can preferentially extract the channel characteristic information that must be extracted related to the communication data, and then selectively extract the channel characteristic information that is selected to be extracted related to the communication data. Based on this, the filter window selection model can obtain part of the channel characteristic information related to the communication data, and can also obtain all the channel characteristic information related to the communication data.

[0032] For example, when part or all of the channel characteristic information related to the communication data is obtained, the obtained channel characteristic information can be calculated and stored according to a preset data processing algorithm. Figure 2 The data format shown is calculated and saved accordingly to establish an association relationship between the communication data and the channel characteristic information of the communication data. Of course, it is not limited to this and is not limited here.

[0033] In some embodiments, the filter window selection model can determine the target filter window corresponding to the communication data based on the channel characteristic information of the communication data. For example, the filter window selection model can determine the channel characteristic information of the communication data, compare it with the preset channel characteristic information of the communication data determined by each of the multiple preset filter windows corresponding to the filter window selection model, determine the degree of similarity between the channel characteristic information of the communication data and the preset channel characteristic information of the communication data, and select the preset filter window corresponding to the preset channel characteristic information with the highest degree of similarity, and determine it as the target filter window corresponding to the communication data.

[0034] In other embodiments, a neural network module may be provided in the filter window selection model. Based on the neural network module, the channel characteristic information of the communication data and the target filter window corresponding to the communication data may be determined according to the communication data. For example, the communication data is input into the neural network model, and the neural network module may use a convolutional neural network to perform feature extraction processing on the communication data to obtain the channel characteristic information of the communication data. Based on the channel characteristic information of the communication data, the neural network module may use a fully connected neural network to train and classify the communication data to obtain the target filter window corresponding to the communication data. Of course, it is not limited to this. For example, the neural network module service may be other classifiers, such as a support vector machine, a dendrogram, etc., which are not limited here.

[0035] Based on the determined target filter window corresponding to the communication data, the filter window selection model corresponding to the scheduling level value of the communication data can be adaptively optimized. For example, the classification accuracy of the filter window selection model can be determined based on the target filter window corresponding to the communication data determined by the filter window selection model to determine whether it is greater than or equal to the preset classification accuracy. If not, the communication data can be used as the optimization basis of the filter window selection model. Of course, it is not limited to this and is not limited here.

[0036] In this way, the communication data is processed according to the filter window selection model corresponding to the scheduling level value of the communication data to obtain the target filter window corresponding to the communication data, which can be used for subsequent adaptive optimization of the filter window selection model to improve the convenience of optimizing the filter window selection model, improve the optimization effect of the filter window selection model, and improve the classification accuracy of the filter window classification of the communication data.

[0037] Step S102: Determine the classification accuracy of the filter window selection model according to the target filter window and the preset filter window label corresponding to the communication data.

[0038] For example, the preset filter window label corresponding to the communication data may be determined by a plurality of preset filter windows corresponding to the filter window selection model. For example, one of the preset filter windows may be selected from the plurality of preset filter windows corresponding to the filter window selection model as the preset filter window label corresponding to the communication data. Among them, the plurality of preset filter windows corresponding to the filter window selection model may be designed according to different delay spread characteristics of the channel, and the preset filter window label may be used to indicate the ideal filter window corresponding to the communication data. For example, when the filter window selection model performs channel estimation filtering on the communication data through the preset filter window corresponding to the preset filter window label, the channel noise and interference signal involved in the communication data are best weakened.

[0039] In some implementations, it can be determined whether the target filter window corresponding to the communication data matches the preset filter window label corresponding to the communication data to determine the classification accuracy of the filter window selection model. In the case where the target filter window does not match the preset filter window label, channel estimation filtering of the communication data through the target filter window is likely to result in poor weakening of the channel noise and interference signals involved in the communication data. Based on this, it can be determined that the classification accuracy of the filter window selection model is poor, which can be used as a basis for subsequent adaptive optimization of the filter window selection model.

[0040] For example, the filter window selection model can be adaptively optimized according to a preset optimization frequency. The preset optimization frequency may include optimization once a week, once every two weeks, once a month, etc. The preset optimization frequency may be pre-set or set by the user, which is not limited here.

[0041] Taking the preset optimization frequency of once a week as an example, when the current time is the optimization time of the filter window selection model, the communication data set corresponding to the filter window selection model can be obtained. The communication data set may include multiple communication data, and the multiple communication data are all communication data processed by the filter window selection model in the past week. For example, it can be determined whether the target filter window corresponding to each communication data in the communication data set is consistent with the preset filter window corresponding to each communication data. According to the ratio between the number of communication data whose target filter window in the communication data set is consistent with the preset filter window and the total number of communication data in the communication data set, the classification accuracy of the filter window selection model can be determined. Based on the determination of the classification accuracy, it can be used as a basis for adaptive optimization of the filter window selection model in the future. Of course, the preset optimization frequency is not limited to optimization once a week, and is not limited here.

[0042] In this way, the classification accuracy of the filter window selection model is determined according to the preset filter window label corresponding to the target filter window and the communication data. The classification accuracy can be used as the basis for adaptive optimization of the filter window selection model in the future, so as to improve the convenience of optimizing the filter window selection model, improve the optimization effect of the filter window selection model, and improve the classification accuracy of filter window classification of communication data.

[0043] In some implementations, the communication data is decoded based on multiple preset filter windows corresponding to the filter window selection model to obtain decoding feature information corresponding to the communication data; based on the decoding feature information, a preset filter window label corresponding to the communication data is determined.

[0044] For example, the filter window selection model may be provided with a plurality of preset filter windows. The plurality of preset filter windows may respectively perform decoding processing on the communication data to obtain decoding characteristic information corresponding to the communication data. The decoding characteristic information may be used to indicate the decoding performance of the plurality of preset filter windows on the communication data. The better the decoding performance of the preset filter window on the communication data, the better the weakening effect of the preset filter window on the channel noise and interference signal involved in the communication data.

[0045] Based on this, the decoding performance of multiple preset filter windows on the communication data can be compared according to the decoding characteristic information to obtain a comparison result. According to the comparison result, the preset filter window with the best decoding performance on the communication data is selected from the multiple preset filter windows as the preset filter window label corresponding to the communication data.

[0046] In this way, according to the multiple preset filter windows corresponding to the filter window selection model, the communication data is decoded and processed to obtain the decoding feature information corresponding to the communication data. The decoding feature information can be used as a basis for selecting the preset filter window with the best weakening effect on the channel noise and interference signal involved in the communication data from multiple preset filter windows, thereby determining the preset filter window label corresponding to the communication data, which is conducive to improving the convenience of determining the preset filter window label corresponding to the communication data. Based on the determination of the preset filter window label, the subsequent filter window selection model can determine whether adaptive optimization is required based on whether the target filter window matches the preset filter window label, which is conducive to improving the convenience of optimizing the filter window selection model, improving the optimization effect of the filter window selection model, and improving the classification accuracy of the filter window classification of the communication data.

[0047] Exemplarily, the decoding characteristic information may include the result check values ​​of the communication data in multiple preset filter windows, the scheduling level values ​​of the communication data in multiple preset filter windows, and the number of decoding iterations of the communication data in multiple preset filter windows. For example, the result check value, the scheduling level value, and the number of decoding iterations can be used to characterize the channel state corresponding to the preset filter window. The better the channel state corresponding to the preset filter window, the better the decoding performance of the preset filter window, and accordingly, the better the effect of the preset filter window on weakening the channel noise and interference signals involved in the communication data.

[0048] In some embodiments, based on at least one of the result check values ​​of the communication data in multiple preset filter windows, the scheduling level values ​​of the communication data in multiple preset filter windows, and the number of decoding iterations of the communication data in multiple preset filter windows, one of the preset filter windows is selected from the multiple preset filter windows to determine the preset filter window label corresponding to the communication data.

[0049] For example, the comparison order between the result check values ​​of the communication data in multiple preset filter windows, the scheduling level values ​​of the communication data in multiple preset filter windows, and the number of decoding iterations of the communication data in multiple preset filter windows can be determined when determining the preset filter window label corresponding to the communication data based on the influence of the result check value, the scheduling level value, and the number of decoding iterations on the channel state corresponding to the preset filter window. In some embodiments, the influence of the result check value on the channel state corresponding to the preset filter window is higher than the influence of the scheduling level value on the channel state corresponding to the preset filter window, and the influence of the scheduling level value on the channel state corresponding to the preset filter window is higher than the influence of the number of decoding iterations on the channel state corresponding to the preset filter window. Then, at least one of the result check values ​​of the communication data in multiple preset filter windows, the scheduling level values ​​of the communication data in multiple preset filter windows, and the number of decoding iterations of the communication data in multiple preset filter windows can be compared in sequence.

[0050] For example, a plurality of preset filter windows corresponding to the filter window selection model include preset filter window 1 to preset filter window 7. When the result check value of the communication data in preset filter window 7 is normal, and the result check values ​​of the communication data in preset filter window 1 to preset filter window 6 are all wrong, the preset filter window label corresponding to the communication data can be determined according to preset filter window 7.

[0051] When the result check values ​​of the communication data in the preset filter window 1 and the preset filter window 7 are normal, and the result check values ​​of the communication data in the preset filter window 2 to the preset filter window 6 are all wrong, the scheduling level values ​​of the communication data in the preset filter window 1 and the preset filter window 7 can be further compared. In the case where the scheduling level values ​​of the communication data in the preset filter window 1 and the preset filter window 7 are different, the preset filter window label corresponding to the communication data can be determined according to one of the preset filter window 1 and the preset filter window 7. For example, according to the scheduling level value of the communication data in the preset filter window 1 being higher than the scheduling level value of the communication data in the preset filter window 7, the preset filter window 1 is selected to determine the preset filter window label corresponding to the communication data. Of course, it is not limited to this. For example, the scheduling level value of the communication data in the preset filter window 1 and the scheduling level value of the communication data in the preset filter window 7 can be compared with the difference between the scheduling level value of the communication data. If the difference between the dispatching level value of the communication data in the preset filter window 1 and the dispatching level value of the communication data is less than the difference between the dispatching level values ​​of the communication data in the preset filter window 7, the preset filter window 1 can be selected to determine the preset filter window label corresponding to the communication data, which is not limited here. The method of determining the preset filter window label corresponding to the communication data according to the dispatching level values ​​of the communication data in multiple preset filter windows can be pre-set or set by the user, which is not limited here.

[0052] When the result check values ​​of the communication data in the preset filter window 1 and the preset filter window 7 are normal, and the dispatch level values ​​of the communication data in the preset filter window 1 and the preset filter window 7 are the same, the number of decoding iterations of the communication data in the preset filter window 1 and the preset filter window 7 can be further compared. When the number of decoding iterations of the communication data in the preset filter window 1 and the preset filter window 7 are different, the preset filter window label corresponding to the communication data can be determined according to one of the preset filter window 1 and the preset filter window 7. For example, according to the number of decoding iterations of the communication data in the preset filter window 1 being lower than the number of decoding iterations of the communication data in the preset filter window 7, the preset filter window 7 is selected to determine the preset filter window label corresponding to the communication data. Of course, it is not limited to this. According to the number of decoding iterations of the communication data in multiple preset filter windows, the method of determining the preset filter window label corresponding to the communication data can be pre-set or set by the user, which is not limited here.

[0053] In this way, according to at least one of the result check values ​​of the communication data in multiple preset filter windows, the scheduling level values ​​of the communication data in multiple preset filter windows, and the number of decoding iterations of the communication data in multiple preset filter windows, one of the preset filter windows is selected from the multiple preset filter windows to determine the preset filter window label corresponding to the communication data. The preset filter window label corresponding to the communication data can be determined based on the decoding performance of the communication data by the multiple preset filter windows, which is conducive to improving the convenience of determining the preset filter window label corresponding to the communication data. Based on the determination of the preset filter window label, the subsequent filter window selection model can determine whether adaptive optimization is required based on whether the target filter window matches the preset filter window label, which is conducive to improving the convenience of optimizing the filter window selection model, improving the optimization effect of the filter window selection model, and improving the classification accuracy of the filter window classification of the communication data.

[0054] In some embodiments, when the result check values ​​of the communication data in some preset filter windows among multiple preset filter windows are normal, the scheduling level values ​​of the communication data in some preset filter windows are the same, and the number of decoding iterations of the communication data in some preset filter windows are the same, the preset filter window label corresponding to the communication data is determined based on the comparison result of the scheduling level value of the communication data and the preset scheduling level value.

[0055] For example, the preset scheduling level value is set based on the experience of selecting the signal estimation window. Since the scheduling level value of the communication data can be used to characterize the size of the channel capacity, the preset scheduling level value can be determined based on the channel capacity and channel delay extension properties corresponding to each of the multiple preset filter windows corresponding to the filter window selection model. Based on the setting of the preset scheduling level value, when some of the multiple preset filter windows have consistent decoding performance for the communication data, the filter window selection model can modulate the preset filter window label corresponding to the communication data based on the preset scheduling level value.

[0056] Take the multiple preset filter windows corresponding to the filter window selection model, including preset filter window 1 to preset filter window 7, as an example. When the result check values ​​of the communication data in preset filter window 2 and preset filter window 3 are normal, the scheduling level values ​​of the communication data in preset filter window 2 and preset filter window 3 are the same, and the number of decoding iterations of the communication data in preset filter window 2 and preset filter window 3 are the same, it can be determined that the preset filter window 2 and the preset filter window 3 have the same decoding performance for the communication data. However, if the filter window selection model determines the preset filter window label corresponding to the communication data according to the preset filter window and the preset filter window 3, then the subsequent filter window selection model determines the classification accuracy of the filter window selection model according to the preset filter window label corresponding to the communication data and the target filter window corresponding to the communication data. For example, the filter window selection model will repeatedly compare different preset filter window labels of the same communication data with the target filter window corresponding to the communication data, thereby causing an adverse effect on the classification accuracy of the filter window selection model.

[0057] Based on this, when there are different preset filter windows with consistent decoding performance for communication data, it is necessary to select one of the preset filter windows from the different preset filter windows with consistent decoding performance to determine the preset filter window label corresponding to the communication data. For example, based on the comparison result between the scheduling level value of the communication data and the preset scheduling level value, one of the preset filter windows can be selected from the different preset filter windows with consistent decoding performance to determine the preset filter window label corresponding to the communication data. For example, in the case of different comparison results, one of the preset filter windows can be selected from the different preset filter windows with consistent decoding performance through different selection methods to determine the preset filter window label corresponding to the communication data.

[0058] In this way, based on the comparison result between the scheduling level value of the communication data and the preset scheduling level value, the preset filter window label corresponding to the communication data is determined, and the filter window selection model can determine, based on the comparison result, to select one of the preset filter windows from some preset filter windows with consistent decoding performance for the communication data, and determine the preset filter window label corresponding to the communication data, which is conducive to improving the convenience and flexibility of determining the preset filter window label corresponding to the communication data.

[0059] For example, when the scheduling level value is less than the preset scheduling level value, the preset filter window label corresponding to the communication data is determined according to the preset filter window with the smallest preset window width information in some preset filter windows.

[0060] Take the case where some preset filter windows include preset filter windows 1 to preset filter windows 3, the preset window width information of preset filter window 1 is 56Ts, the preset window width information of preset filter window 2 is 80Ts, and the preset window width information of preset filter window 3 is 120Ts. Ts is used to indicate the sampling period of orthogonal frequency division multiplexing (OFDM) symbols. In the case where the scheduling level value is less than the preset scheduling level value, since the preset window width information of preset filter window 1 in some preset filter windows is less than the preset window width information of preset filter window 2, and the preset window width information of preset filter window 1 is less than the preset window width information of preset filter window 3, the preset filter window label corresponding to the communication data can be determined based on preset filter window 1.

[0061] For example, when the scheduling level value is greater than or equal to the preset scheduling level value, the preset filter window label corresponding to the communication data is determined according to the preset filter window with the largest preset window width information in some preset filter windows.

[0062] Take as an example that some preset filter windows include preset filter windows 4 to preset filter windows 7, the preset window width information of preset filter window 4 is 180Ts, the preset window width information of preset filter window 5 is 250Ts, the preset window width information of preset filter window 6 is 360Ts, and the preset window width information of preset filter window 7 is 480Ts. In the case where the scheduling level value is greater than or equal to the preset scheduling level value, since the preset window width information of preset filter window 7 in some preset filter windows is greater than the preset window width information of preset filter windows 4 to preset filter windows 6 respectively, the preset filter window label corresponding to the communication data can be determined according to preset filter window 7.

[0063] In this way, when the comparison result between the scheduling level value of the communication data and the preset scheduling level value is different, different preset filter window labels can be used to determine the strategy, and one of the preset filter windows can be selected from some preset filter windows to determine the preset filter window label corresponding to the communication data, which is conducive to improving the convenience and flexibility of determining the preset filter window label corresponding to the communication data.

[0064] In some embodiments, communication data whose decoding characteristic information meets preset data elimination conditions are eliminated from the communication data; the preset data elimination conditions include at least one of the following: the result check values ​​of the communication data in multiple preset filter windows are all wrong, the scheduling level values ​​of the communication data in multiple preset filter windows are all the same, and the number of decoding iterations of the communication data in multiple preset filter windows are all the same.

[0065] For example, when the decoding characteristic information corresponding to the communication data meets at least one of the following conditions: the result check values ​​of the communication data in multiple preset filter windows are all wrong, the scheduling level values ​​of the communication data in multiple preset filter windows are all the same, and the number of decoding iterations of the communication data in multiple preset filter windows are all the same, if the preset filter window corresponding to the communication data is determined based on one of the multiple preset filter windows, it will easily lead to an adverse impact on the classification accuracy of the filter window selection model.

[0066] Taking multiple preset filter windows including preset filter windows 1 to preset filter windows 7 as an example, if the decoding characteristic information of the communication data in preset filter windows 1 to preset filter windows 7 is consistent, if one of the preset filter windows is selected to determine the preset filter window label corresponding to the communication data, the probability that the target filter window corresponding to the communication data matches the preset filter window label corresponding to the communication data is less than the probability that the target filter window corresponding to the communication data does not match the preset filter window label corresponding to the communication data. In addition, as the number of communication data with consistent decoding performance in multiple preset filter windows increases, it is easy to lower the classification accuracy of the filter window selection model, resulting in an adverse effect on the classification accuracy of the filter window selection model.

[0067] Based on this, a preset data rejection condition can be set to reject communication data with consistent decoding performance in multiple preset filter windows, so as to avoid adversely affecting the classification accuracy of the filter window selection model.

[0068] Take the multiple preset filter windows corresponding to the filter window selection model including preset filter window 1 to preset filter window 7 as an example. If the result check values ​​of the communication data in the preset filter window 1 to the preset filter window 7 are all wrong, which is equivalent to the preset filter window 1 to the preset filter window 7 The decoding performance of the communication data is not good, and the communication data cannot be used to optimize the filter window selection model, then there is no need to determine the preset filter window label corresponding to the communication data. Accordingly, when the result check values ​​of the communication data in the preset filter window 1 to the preset filter window 7 are correct, if the scheduling level values ​​of the communication data in the preset filter window 1 to the preset filter window 7 are the same, and / or the number of decoding iterations of the communication data in the preset filter window 1 to the preset filter window 7 are the same, which is equivalent to the preset filter window 1 to the preset filter window 7. The decoding performance of the communication data is consistent, then there is no need to determine the preset filter window label corresponding to the communication data. Based on this, when there is communication data that has not been determined to have a preset filter window label, it is subsequently impossible to determine the classification accuracy of the filter window selection model based on the preset filter window label corresponding to the communication data and the target filter window corresponding to the communication data. In this case, it can be determined that the decoding feature information of the communication data meets the preset data rejection conditions, and the communication data can be rejected to avoid the adverse effects of the communication data that meets the preset rejection conditions on the classification accuracy of the filter window selection model.

[0069] Of course, it is not limited to this. If the decoding feature information corresponding to the communication data meets the preset data elimination conditions, the preset filter window label corresponding to the communication data can be determined as a prompt filter window label. The prompt filter window label can be different from the preset filter window label determined according to any one of the multiple preset filter windows, so as to distinguish the communication data that meets the preset data elimination conditions and the communication data that does not meet the preset data elimination conditions. Based on the communication data that meets the preset data elimination conditions, the prompt filter window label is set. Before the filter window selection model determines the classification accuracy of the filter window selection model according to the target filter window corresponding to the communication data and the preset filter window label corresponding to the communication data, the communication data with the prompt filter window label can be eliminated to avoid the communication data that meets the preset elimination conditions. The adverse effect caused when determining the classification accuracy of the filter window selection model.

[0070] In this way, based on the setting of the preset data elimination conditions, the communication data whose decoding characteristic information meets the preset data elimination conditions can be eliminated to avoid the adverse effect of the communication data whose decoding characteristic information meets the preset data elimination conditions on the classification accuracy of the filter window selection model, which is conducive to improving the accuracy of optimizing the filter window selection model.

[0071] Step S103: When the classification accuracy is less than or equal to the preset accuracy, the model parameters of the filter window selection model are adjusted according to the preset filter window label corresponding to the communication data to obtain an optimized filter window selection model.

[0072] For example, when the classification accuracy of the filter window selection model is less than or equal to the preset accuracy, it can be determined that the classification accuracy of the target filter window corresponding to the communication data determined by the filter window selection model from the multiple preset filter windows corresponding to the filter window selection model is insufficient, and the filter window selection model cannot accurately judge the channel state of all communication data after being processed by their respective corresponding target filter windows. Based on this, it is necessary to adjust the model parameters of the filter window selection model to optimize the filter window selection model.

[0073] In some embodiments, the model parameters of the filter window selection model can be adjusted according to the preset filter window label corresponding to the communication data. Since the preset filter window label can be used to indicate the ideal filter window corresponding to the communication data, the multiple preset filter windows corresponding to the filter window selection model can be screened according to the preset filter window label to reduce the total number of preset filter windows corresponding to the filter window selection model. By reducing the total number of preset filter windows corresponding to the filter window selection model, the model parameters of the filter window selection model are adjusted to obtain an optimized filter window selection model. The optimized filter window selection model can determine the target filter window corresponding to the communication data from fewer preset filter windows. Based on this, the possibility that the target filter window corresponding to the communication data determined according to the optimized filter window selection model is consistent with the preset filter window label corresponding to the communication data will be higher than the possibility that the target filter window corresponding to the communication data determined according to the filter window selection model before optimization is consistent with the preset filter window label corresponding to the communication data. Correspondingly, the possibility that the target filter window corresponding to the communication data determined according to the optimized filter window selection model does not match the preset filter window label corresponding to the communication data will be lower than the possibility that the target filter window corresponding to the communication data determined according to the filter window selection model before optimization is consistent with the preset filter window label corresponding to the communication data.

[0074] In this way, when the classification accuracy is less than or equal to the preset accuracy, the model parameters of the filter window selection model are adjusted according to the preset filter window label corresponding to the communication data to obtain the optimized filter window selection model. Then, the preset filter window label corresponding to the communication data can be used to adjust the multiple preset filter windows corresponding to the filter window selection model, and the model parameters of the filter window selection model are adjusted based on the adjustment of the preset filter windows to obtain the optimized filter window selection model, which is conducive to improving the convenience of optimizing the filter window selection model.

[0075] In some embodiments, based on the preset filter window label corresponding to the communication data, a preset number of preset filter windows are selected from multiple preset filter windows corresponding to the filter window selection model; based on the preset number of preset filter windows, the model parameters of the filter window selection model are adjusted to obtain an optimized filter window selection model.

[0076] For example, a preset number of preset filter windows may be selected from a plurality of preset filter windows corresponding to the filter window selection model according to the proportion of the preset filter window labels corresponding to the communication data in the preset filter window labels corresponding to all communication data.

[0077] Taking the multiple preset filter windows corresponding to the filter window selection model including preset filter window 1 to preset filter window 7, and the preset number is 4 as an example, the preset filter window label corresponding to the communication data may include a preset filter window label determined according to one of the preset filter window 1 to preset filter window 7. For all communication data processed by the filter window selection model, the preset filter window label corresponding to 25% of the communication data is determined according to the preset filter window 1, the preset filter window label corresponding to 20% of the communication data is determined according to the preset filter window 4, the preset filter window label corresponding to 15% of the communication data is determined according to the preset filter window 6, the preset filter window label corresponding to 15% of the communication data is determined according to the preset filter window 7, and the preset filter window labels of the remaining 25% of the communication data are respectively determined according to the preset filter window 2, the preset filter window 3 and the preset filter window 5, and the proportion information is less than 15%. Based on this, 4 preset filter windows corresponding to the preset filter window labels with higher proportion information in the preset filter window labels corresponding to all communication data can be selected from the preset filter window 1 to the preset filter window 7 as the basis for adjusting the model parameters of the filter window selection model. For example, preset filter window 1, preset filter window 4, preset filter window 6, and preset filter window 7 may be selected as a basis for adjusting the model parameters of the filter window selection model.

[0078] Of course, it is not limited to this. The preset number can be pre-set or set by the user, which is not limited here. In an exemplary embodiment, when the model parameters of the filter window selection model are adjusted according to the preset number of preset filter windows, and the optimized filter window selection model is obtained, if the classification accuracy of the communication data by the optimized filter window selection model is still less than or equal to the preset accuracy, the number of preset filter windows corresponding to the optimized filter window selection model can be further reduced to reduce the classification difficulty of the target filter window corresponding to the communication data determined by the optimized filter window selection model. For example, when the number of preset filter windows corresponding to the optimized filter window selection model is 4, the number of preset filter windows can be reduced to 3 according to the preset filter window labels corresponding to the communication data processed by the optimized filter window selection model, and so on. Of course, it is not limited to this, and it is not limited here. In an exemplary embodiment, when the classification accuracy of the filter window selection model is less than or equal to the preset accuracy, and the number of multiple preset filter windows corresponding to the filter window selection model is less than or equal to the preset filter window number, the filter window selection model can be replaced with the original experience model. The original experience model can be a filter window selection model designed according to expert experience. Of course, it is not limited to this and no limitation is made here.

[0079] In this way, by adjusting the model parameters of the filter window selection model based on a preset number of preset filter windows, an optimized preset filter window is obtained. The optimized preset filter window can determine the target filter window corresponding to the communication data from fewer preset filter windows, which is beneficial to improving the classification accuracy of the filter window classification of the communication data, thereby helping to improve the optimization effect of the filter window selection model.

[0080] In some implementations, when there are multiple filter window selection models each corresponding to the same preset number of preset filter windows, the multiple filter window selection models are merged to obtain a merged filter window selection model.

[0081] For example, in the case where the scheduling level values ​​of different communication data correspond to different filter window selection models, taking different filter window selection models including filter window selection model 1 to filter window selection model 4 as an example, if the classification accuracy of filter window selection model 1 to filter window selection model 4 is less than or equal to the preset accuracy, it is necessary to optimize filter window selection model 1 to filter window selection model 4 respectively. In some embodiments, it is necessary to select a preset number of preset filter windows from a plurality of preset filter windows corresponding to filter window selection model 1 to filter window selection model 4 according to the preset filter window labels corresponding to the communication data processed by each of filter window selection model 1 to filter window selection model 4. For example, the preset number of preset filter windows corresponding to filter window selection model 1 is preset filter window 2 to preset filter window 4, the preset number of preset filter windows corresponding to filter window selection model 2 is preset filter window 3 to preset filter window 5, the preset number of preset filter windows corresponding to filter window selection model 3 is preset filter window 1 to preset filter window 3, and the preset number of preset filter windows corresponding to filter window selection model 4 is preset filter window 2 to preset filter window 4, then it can be determined that the preset number of preset filter windows corresponding to filter window selection model 1 and filter window selection model 4 are the same. Based on this, the filter window selection model 1 and the filter window selection model 4 can be merged to obtain a merged filter window selection model. In an exemplary embodiment, the model parameters of the merged filter window selection model can be adjusted according to the preset filter windows 2 to 4 to optimize the merged filter window selection model. Of course, it is not limited to this and is not limited here.

[0082] In this way, by merging multiple filter window selection models, when the accuracy of multiple filter window selection models in classifying the target filter window corresponding to the communication data is less than or equal to the preset classification accuracy, the merged filter window selection model can be optimized to optimize multiple filter window selection models at the same time, which is beneficial to improving the optimization efficiency of the filter window selection model.

[0083] Exemplarily, taking the application of the optimization method of the filter window selection model to communication equipment as an example, the optimization method of the filter window selection model can be used to upgrade the signal estimation filter module of the communication equipment.

[0084] For example, Figure 3 As shown, when the communication device receives the communication data, the process of processing the communication data may include a channel filtering process and a model optimization process. For example, according to the channel filtering process, the pilot channel of the communication data can be calculated, and the channel filtering window corresponding to the communication data, that is, the target filtering window corresponding to the communication data, can also be determined based on the filter window selection model and the pilot channel of the communication data. According to the model optimization process, the communication data can be decoded to determine the preset filter window label corresponding to the communication data, and the classification accuracy of the filter window selection model can be determined according to the target filter window corresponding to the communication data processed by the filter window selection model and the preset filter window label corresponding to the communication data, so as to adaptively update the filter window selection model according to the classification accuracy of the filter window selection model. Based on this, the channel filtering process and the model optimization process can be independent of each other, and the model optimization process will not increase the timing and computational complexity of the channel filtering process. For example, an optimization cycle can be set for the model optimization process. The model optimization process can then determine whether it is necessary to optimize the filter window selection model based on the communication data processed by the channel filtering process within the optimization cycle, and optimize the filter window selection model when it is necessary to optimize the filter window selection model, so that the filter window selection model can track environmental changes and adaptively update the target filter window corresponding to the communication data without the need for manual parameter debugging.

[0085] The optimization method of the filter window selection model provided in the above embodiment processes the communication data according to the filter window selection model corresponding to the scheduling level value of the communication data to obtain the target filter window corresponding to the communication data; determines the classification accuracy of the filter window selection model according to the target filter window and the preset filter window label corresponding to the communication data; when the classification accuracy is less than or equal to the preset accuracy, adjusts the model parameters of the filter window selection model according to the preset filter window label corresponding to the communication data to obtain the optimized filter window selection model. Based on this, the classification accuracy of the filter window selection model for classifying the target filter window of the communication data can be evaluated according to the target filter window corresponding to the communication data determined by the filter window selection model and the preset filter window label corresponding to the communication data determined by the multiple preset filter windows corresponding to the filter window selection model, thereby determining the optimization strategy for optimizing the current filter window selection model according to the classification accuracy, thereby improving the convenience of optimizing the filter window selection model, improving the optimization effect of the filter window selection model, and improving the classification accuracy of the filter window classification of the communication data.

[0086] The method of the present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0087] Exemplarily, the above method may be implemented in the form of a computer program, and the computer program may be run on a communication device.

[0088] See also Figure 4 , Figure 4 A schematic block diagram of a communication device provided in an embodiment of the present invention.

[0089] like Figure 4 As shown, the communication device 300 includes a processor 301 and a memory 302 , and the processor 301 and the memory 302 are connected via a bus 303 , such as an I2C (Inter-integrated Circuit) bus.

[0090] Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire communication device. The processor 301 can be a central processing unit (CPU), and the processor 301 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0091] Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0092] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the embodiment of the present invention, and does not constitute a limitation on the communication device to which the embodiment of the present invention is applied. The specific server may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0093] The processor is used to execute a computer program stored in the memory, and implement any one of the methods for optimizing the filter window selection model provided by the embodiments of the present invention when executing the computer program.

[0094] In one embodiment, the processor is used to execute a computer program stored in the memory, and implements the following steps when executing the computer program:

[0095] According to the filter window selection model corresponding to the scheduling level value of the communication data, the communication data is processed to obtain the target filter window corresponding to the communication data;

[0096] Determining the classification accuracy of the filter window selection model according to the target filter window and the preset filter window label corresponding to the communication data;

[0097] When the classification accuracy is less than or equal to the preset accuracy, the model parameters of the filter window selection model are adjusted according to the preset filter window label corresponding to the communication data to obtain an optimized filter window selection model.

[0098] It should be noted that those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the communication device described above can refer to the corresponding process in the aforementioned optimization method embodiment of the filter window selection model, and will not be repeated here.

[0099] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. The computer program can be executed by a processor to implement an optimization method for any filter window selection model provided in the description of the embodiment of the present invention.

[0100] The storage medium may be an internal storage unit of the communication device described in the above embodiment, such as a hard disk or memory of the communication device. The storage medium may also be an external storage device of the communication device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the communication device.

[0101] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0102] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0103] The serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for optimizing a filter window selection model, characterized in that: include: According to the filter window selection model corresponding to the scheduling level value of the communication data, the communication data is processed to obtain the target filter window corresponding to the communication data; Determining the classification accuracy of the filter window selection model according to the target filter window and the preset filter window label corresponding to the communication data; When the classification accuracy is less than or equal to the preset accuracy, the model parameters of the filter window selection model are adjusted according to the preset filter window label corresponding to the communication data to obtain an optimized filter window selection model.

2. The optimization method according to claim 1, characterized in that: The step of adjusting the model parameters of the filter window selection model according to the preset filter window label corresponding to the communication data to obtain the optimized filter window selection model includes: According to the preset filter window label corresponding to the communication data, a preset number of preset filter windows are selected from a plurality of preset filter windows corresponding to the filter window selection model; According to the preset number of preset filter windows, the model parameters of the filter window selection model are adjusted to obtain an optimized filter window selection model.

3. The optimization method according to claim 2, characterized in that: After selecting a preset number of preset filter windows from a plurality of preset filter windows corresponding to the filter window selection model according to the preset filter window label of the communication data, the method further includes: In the case that there are multiple filter window selection models, each of which corresponds to the same preset number of preset filter windows, the multiple filter window selection models are merged to obtain a merged filter window selection model.

4. The optimization method according to any one of claims 1 to 3, characterized in that: Before determining the classification accuracy of the filter window selection model according to the target filter window and the preset filter window label corresponding to the communication data, the method further includes: Decoding the communication data according to a plurality of preset filter windows corresponding to the filter window selection model to obtain decoding feature information corresponding to the communication data; A preset filter window label corresponding to the communication data is determined according to the decoding characteristic information.

5. The optimization method according to claim 4, characterized in that: The step of determining a preset filter window label corresponding to the communication data according to the decoding characteristic information includes: According to at least one of the result check values ​​of the communication data in multiple preset filter windows, the scheduling level values ​​of the communication data in multiple preset filter windows, and the number of decoding iterations of the communication data in multiple preset filter windows, one preset filter window is selected from the multiple preset filter windows to determine the preset filter window label corresponding to the communication data.

6. The optimization method according to claim 4, characterized in that: The step of determining a preset filter window label corresponding to the communication data according to the decoding characteristic information includes: When the result check values ​​of some preset filter windows of the communication data in the multiple preset filter windows are normal, the scheduling level values ​​of the communication data in the some preset filter windows are the same, and the number of decoding iterations of the communication data in the some preset filter windows are the same, the preset filter window label corresponding to the communication data is determined according to the comparison result of the scheduling level value of the communication data and the preset scheduling level value.

7. The optimization method according to claim 6, characterized in that: The determining, according to a comparison result between the scheduling level value of the communication data and the preset scheduling level value, a preset filtering window label corresponding to the communication data comprises: In the case where the scheduling level value is less than the preset scheduling level value, determining the preset filter window label corresponding to the communication data according to the preset filter window with the smallest preset window width information in the part of the preset filter windows; When the scheduling level value is greater than or equal to the preset scheduling level value, the preset filter window label corresponding to the communication data is determined according to the preset filter window with the largest preset window width information in the part of the preset filter windows.

8. The optimization method according to claim 4, characterized in that: Before determining the classification accuracy of the filter window selection model according to the target filter window corresponding to the communication data and the preset filter window label corresponding to the communication data, the method further includes: Eliminating, from the communication data, communication data whose decoding characteristic information meets a preset data elimination condition; The preset data elimination conditions include at least one of the following: the result check values ​​of the communication data in the multiple preset filter windows are all wrong, the scheduling level values ​​of the communication data in the multiple preset filter windows are all the same, and the decoding iteration numbers of the communication data in the multiple preset filter windows are all the same.

9. A communication device, characterized in that: The communication device includes a memory and a processor; The memory is used to store computer programs; The processor is used to execute the computer program and implement the optimization method of the filter window selection model according to any one of claims 1 to 8 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program can be executed by a processor, and the processor implements the optimization method of the filter window selection model described in any one of claims 1 to 8.

Images