Method, device, equipment and readable storage medium for processing hierarchical model information

Abstract

The present application provides a method, device, equipment and readable storage medium for processing hierarchical model information, which relates to the field of wireless technology and solves the problem of poor adaptability between the extracted features and transmission of the original source in the related technology. The method of an embodiment of the present application includes: obtaining the user's pre-scheduling information, the pre-scheduling information is used to indicate the relationship between the amount of data of the user's service data and the amount of transmission data supported by the air interface within the pre-scheduling time; according to the pre-scheduling information, selecting a target hierarchical model from multiple hierarchical models, wherein different hierarchical models correspond to different compression ratio ranges; and sending information of the target hierarchical model. The present application can select a hierarchical model with a suitable compression ratio range based on different pre-scheduling information, and compress the service data based on the selected hierarchical model with a suitable compression ratio range, which can effectively improve the adaptability of the compressed data and the transmission of the compressed service data.

Description

Technical Field

[0001] The present application relates to the field of wireless technology, and in particular to a method, device, equipment and readable storage medium for processing hierarchical model information. Background Art

[0002] Semantic communication is an emerging communication paradigm that focuses more on the semantic content of transmitted information. The core goal of semantic communication is to accurately convey the intention and context of information during the communication process. In the research of the sixth generation of mobile communication technology (6th generation wireless systems, 6G), semantic communication is regarded as one of the key technologies that will support the needs of intelligent ubiquity and sustainable development. Figure 1 As shown in the figure, the basic process of semantic communication includes the sending end extracting features from the original source, and then transmitting it to the receiving end through the channel after joint processing of the source and channel. After receiving the data transmitted through the channel, the receiving end performs joint processing of the source and channel on the received data, and then performs feature fusion on the processed data to restore the source.

[0003] In the prior art, when extracting features from the original information source, a unified method is used to extract features from the original information source without considering the transmission information corresponding to the original information source, resulting in poor compatibility between the extracted features and the transmission. Summary of the invention

[0004] The purpose of the present application is to provide a method, device, equipment and readable storage medium for processing hierarchical model information to solve the problem of poor compatibility between the extracted features and transmission of the original information source in the related art.

[0005] To achieve the above object, an embodiment of the present application provides a method for processing hierarchical model information, which is applied to a base station. The method includes:

[0006] Acquire pre-scheduling information of the user, where the pre-scheduling information is used to indicate the relationship between the amount of service data of the user within the pre-scheduling time and the amount of transmission data supported by the air interface;

[0007] According to the pre-scheduling information, a target hierarchical model is selected from a plurality of hierarchical models, wherein different hierarchical models correspond to different compression ratio ranges;

[0008] The information of the target hierarchical model is sent.

[0009] Optionally, the amount of compressed data obtained after compressing the service data using the target classification model is less than or equal to the amount of transmission data supported by the air interface.

[0010] Optionally, obtaining the pre-scheduling information of the user includes:

[0011] Acquire channel state information within a pre-scheduling time and service attribute information corresponding to the user within the pre-scheduling time;

[0012] Obtaining historical scheduling information of multiple users within the pre-scheduling time;

[0013] The pre-scheduling information of the user is generated according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time.

[0014] Optionally, the pre-scheduling information includes a pre-scheduling ratio interval;

[0015] The generating the pre-scheduling information of the user according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time includes:

[0016] Acquire a first data volume and a second data volume according to the service attribute information, wherein the first data volume is a data volume of a service whose service priority in the service attribute information is higher than a preset priority, and the second data volume is a total data volume of the service in the service attribute information;

[0017] The transmission data volume supported by the air interface is obtained according to the sum of the target data volumes corresponding to N moments within the pre-scheduling time, wherein the target data volume corresponding to each moment is the product of the probability that the base station schedules resources for the user at the moment and the air interface transmission data volume of the transmission channel of the user at the moment, the probability that the base station schedules resources for the user at the moment is predicted according to the historical scheduling information, the air interface transmission data volume of the transmission channel of the user at the moment is predicted according to the channel state information, and N is a positive integer;

[0018] The ratio of the transmission data amount supported by the air interface to the first data amount is determined as the pre-scheduling ratio upper limit of the pre-scheduling ratio interval, and the ratio of the transmission data amount supported by the air interface to the second data amount is determined as the pre-scheduling ratio lower limit of the pre-scheduling ratio interval.

[0019] Optionally, the pre-scheduling information includes a pre-scheduling ratio interval;

[0020] Selecting a target hierarchical model from a plurality of hierarchical models according to the pre-scheduling information includes:

[0021] Selecting a target compression ratio range matching the pre-scheduling ratio interval according to compression ratio ranges corresponding to a plurality of hierarchical models, wherein the pre-scheduling ratio interval is within the target compression ratio range, or a pre-scheduling ratio lower limit of the pre-scheduling ratio interval is within the target compression ratio range;

[0022] The classification model corresponding to the target compression ratio range is determined as the target classification model.

[0023] Optionally, the method of the embodiment of the present application further includes:

[0024] Acquire hierarchical model capability information reported by the terminal, where the hierarchical model capability information is used to indicate the hierarchical models supported by the terminal;

[0025] The sending of the target classification model information includes:

[0026] In a case where the hierarchical model capability information indicates that the terminal supports the target hierarchical model, sending index information of the target hierarchical model;

[0027] When the hierarchical model capability information indicates that the terminal does not support the target hierarchical model, sending model information of the target hierarchical model or sending model information of a target recovery model corresponding to the target hierarchical model;

[0028] The target recovery model is used to recover the data compressed by the target classification model.

[0029] Optionally, the method of the embodiment of the present application further includes:

[0030] receiving feedback information sent by a terminal, the feedback information being used to indicate that the terminal has configured a corresponding target recovery model according to information of the target hierarchical model, the target recovery model being used to recover data compressed by the target hierarchical model;

[0031] In response to the feedback information, compressed data is sent, where the compressed data is obtained by compressing the service data using the target classification model.

[0032] The present application also provides a method for processing hierarchical model information, which is applied to a terminal, and the method includes:

[0033] Acquire information of a target hierarchical model, where the target hierarchical model is a hierarchical model selected from a plurality of hierarchical models based on pre-scheduling information of the user, where different hierarchical models correspond to different compression ratio ranges, and the pre-scheduling information is used to indicate a relationship between a data volume of a user's service data and a transmission data volume supported by an air interface within a pre-scheduling time;

[0034] According to the information of the target hierarchical model, a target recovery model equivalent to the target hierarchical model is configured, and the target recovery model is used to recover the data compressed by the target hierarchical model.

[0035] Optionally, the method of the embodiment of the present application further includes:

[0036] Sending hierarchical model capability information, where the hierarchical model capability information is used to indicate the hierarchical models supported by the terminal;

[0037] The obtaining of information of the target classification model includes:

[0038] Acquire index information of a target hierarchical model, where the index information is sent when the hierarchical model capability information indicates that the terminal supports the target hierarchical model;

[0039] Alternatively, model information of the target hierarchical model or model information of the target recovery model corresponding to the target hierarchical model is obtained, wherein the information of the target hierarchical model or the model information of the target recovery model is sent when the hierarchical model capability information indicates that the terminal does not support the target hierarchical model.

[0040] Optionally, the method of the embodiment of the present application further includes:

[0041] Feedback information is sent, where the feedback information is used to indicate that the terminal has configured a corresponding target recovery model according to the information of the target hierarchical model.

[0042] The embodiment of the present application further provides a device for processing hierarchical model information, which is applied to a base station, and the device includes:

[0043] A first processing module, used to obtain pre-scheduling information of the user, where the pre-scheduling information is used to indicate the relationship between the amount of service data of the user within the pre-scheduling time and the amount of transmission data supported by the air interface;

[0044] A second processing module is used to select a target hierarchical model from a plurality of hierarchical models according to the pre-scheduling information, wherein different hierarchical models correspond to different compression ratio ranges;

[0045] The first transceiver module is used to send information of the target hierarchical model.

[0046] Optionally, the amount of compressed data obtained after compressing the service data using the target classification model is less than or equal to the amount of transmission data supported by the air interface.

[0047] Optionally, the first processing module is used to:

[0048] Acquire channel state information within a pre-scheduling time and service attribute information corresponding to the user within the pre-scheduling time;

[0049] Obtaining historical scheduling information of multiple users within the pre-scheduling time;

[0050] The pre-scheduling information of the user is generated according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time.

[0051] Optionally, the pre-scheduling information includes a pre-scheduling ratio interval;

[0052] The first processing module is used to: obtain a first data volume and a second data volume according to the service attribute information, the first data volume being the data volume of the service with a service priority higher than the preset priority in the service attribute information, and the second data volume being the total data volume of the service in the service attribute information;

[0053] The transmission data volume supported by the air interface is obtained according to the sum of the target data volumes corresponding to N moments within the pre-scheduling time, wherein the target data volume corresponding to each moment is the product of the probability that the base station schedules resources for the user at the moment and the air interface transmission data volume of the transmission channel of the user at the moment, the probability that the base station schedules resources for the user at the moment is predicted according to the historical scheduling information, the air interface transmission data volume of the transmission channel of the user at the moment is predicted according to the channel state information, and N is a positive integer;

[0054] The ratio of the transmission data amount supported by the air interface to the first data amount is determined as the pre-scheduling ratio upper limit of the pre-scheduling ratio interval, and the ratio of the transmission data amount supported by the air interface to the second data amount is determined as the pre-scheduling ratio lower limit of the pre-scheduling ratio interval.

[0055] Optionally, the pre-scheduling information includes a pre-scheduling ratio interval;

[0056] The second processing module is used for:

[0057] Selecting a target compression ratio range matching the pre-scheduling ratio interval according to compression ratio ranges corresponding to a plurality of hierarchical models, wherein the pre-scheduling ratio interval is within the target compression ratio range, or a pre-scheduling ratio lower limit of the pre-scheduling ratio interval is within the target compression ratio range;

[0058] The classification model corresponding to the target compression ratio range is determined as the target classification model.

[0059] Optionally, the device of the embodiment of the present application further includes:

[0060] A second transceiver module is used to obtain hierarchical model capability information reported by the terminal, where the hierarchical model capability information is used to indicate the hierarchical model supported by the terminal;

[0061] The first transceiver module is used for:

[0062] In a case where the hierarchical model capability information indicates that the terminal supports the target hierarchical model, sending index information of the target hierarchical model;

[0063] When the hierarchical model capability information indicates that the terminal does not support the target hierarchical model, sending model information of the target hierarchical model or sending model information of a target recovery model corresponding to the target hierarchical model;

[0064] The target recovery model is used to recover the data compressed by the target classification model.

[0065] Optionally, the device of the embodiment of the present application further includes:

[0066] A third transceiver module is used to receive feedback information sent by a terminal, wherein the feedback information is used to indicate that the terminal has configured a corresponding target recovery model according to the information of the target hierarchical model, and the target recovery model is used to recover the data compressed by the target hierarchical model;

[0067] The fourth transceiver module is used to send compressed data in response to the feedback information, where the compressed data is obtained by compressing the business data using the target classification model.

[0068] The embodiment of the present application further provides a device for processing hierarchical model information, which is applied to a terminal, and the device includes:

[0069] a fifth transceiver module, configured to obtain information of a target hierarchical model, wherein the target hierarchical model is a hierarchical model selected from a plurality of hierarchical models based on pre-scheduling information of the user, wherein different hierarchical models correspond to different compression ratio ranges, and the pre-scheduling information is used to indicate a relationship between a data volume of a user's service data within a pre-scheduling time and a transmission data volume supported by an air interface;

[0070] The third processing module is used to configure a target recovery model equivalent to the target hierarchical model according to information of the target hierarchical model, and the target recovery model is used to recover data compressed by the target hierarchical model.

[0071] Optionally, the device of the embodiment of the present application further includes:

[0072] a sixth transceiver module, configured to send hierarchical model capability information, wherein the hierarchical model capability information is used to indicate a hierarchical model supported by the terminal;

[0073] The fifth transceiver module is used for:

[0074] Acquire index information of a target hierarchical model, where the index information is sent when the hierarchical model capability information indicates that the terminal supports the target hierarchical model;

[0075] Alternatively, model information of the target hierarchical model or model information of the target recovery model corresponding to the target hierarchical model is obtained, wherein the information of the target hierarchical model or the model information of the target recovery model is sent when the hierarchical model capability information indicates that the terminal does not support the target hierarchical model.

[0076] An embodiment of the present application also provides a device for processing hierarchical model information, including: a transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; when the processor executes the program or instruction, the steps in the method for processing hierarchical model information as described above are implemented.

[0077] The embodiment of the present application further provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps in the method for processing hierarchical model information as described above are implemented.

[0078] An embodiment of the present application further provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps in the method for processing hierarchical model information as described above.

[0079] The beneficial effects of the above technical solution of the present application are as follows:

[0080] In an embodiment of the present application, the base station selects a target hierarchical model from multiple hierarchical models based on the user's pre-scheduling information. The pre-scheduling information is used to indicate the relationship between the data volume of the user's service data within the pre-scheduling time and the transmission data volume supported by the air interface. Since different hierarchical models correspond to different compression ratio ranges, it is possible to select a hierarchical model with a suitable compression ratio range based on different pre-scheduling information, and compress the service data based on the selected hierarchical model with a suitable compression ratio range, which can effectively improve the compatibility of the compressed data (i.e., the extracted service data) with the transmission after compression. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 It is the basic block diagram of semantic communication;

[0082] Figure 2 A schematic diagram of a flowchart of a method for processing hierarchical model information in some embodiments of the present application;

[0083] Figure 3 This is a schematic diagram of processing pre-scheduling information in some embodiments of the present application;

[0084] Figure 4 A schematic diagram of a flowchart of a method for processing hierarchical model information in some embodiments of the present application;

[0085] Figure 5 An interactive schematic diagram of a method for processing hierarchical model information in some embodiments of the present application;

[0086] Figure 6 A schematic diagram of a module of a device for processing hierarchical model information in some embodiments of the present application;

[0087] Figure 7 A schematic diagram of a module of a device for processing hierarchical model information in some embodiments of the present application;

[0088] Figure 8 A structural block diagram of a device for processing hierarchical model information in some embodiments of the present application;

[0089] Fig. 9 This is a structural block diagram of a device for processing hierarchical model information in some embodiments of the present application. DETAILED DESCRIPTION

[0090] In order to make the technical problems, technical solutions and advantages to be solved by the present application clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0091] It should be understood that the references to "one embodiment" or "an embodiment" throughout the specification mean that the specific features, structures, or characteristics associated with the embodiment are included in at least one embodiment of the present application. Therefore, the references to "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0092] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of one type, and the number of objects is not limited, for example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, for example, the protection scope of "A and / or B" covers at least three schemes, namely, Scheme 1: including A and excluding B; Scheme 2: including B and excluding A; Scheme 3: including both A and B. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0093] In the various embodiments of the present application, it should be understood that the size of the serial numbers of the following processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0094] Additionally, the terms "system" and "network" are often used interchangeably herein.

[0095] In the embodiments provided in this application, it should be understood that "B corresponding to A" means that B is associated with A and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, but B can also be determined according to A and / or other information.

[0096] like Figure 2 As shown, an embodiment of the present application provides a method for processing hierarchical model information, which is executed by a base station, and the method includes:

[0097] Step 201: obtaining pre-scheduling information of a user, where the pre-scheduling information is used to indicate the relationship between the amount of service data of the user within a pre-scheduling time and the amount of transmission data supported by the air interface.

[0098] For example, the pre-scheduling information is used to indicate the ratio of the number of transmissions supported by the air interface within the pre-scheduling time to the amount of service data of the user within the pre-scheduling time.

[0099] The data volume of the user's service data within the pre-scheduling time includes at least one of the total data volume of the user's service data within the pre-scheduling time and the data volume of service data with a service priority higher than a preset priority within the pre-scheduling time.

[0100] Optionally, the above pre-scheduling time can be described as a pre-scheduling time window T.

[0101] Optionally, the user in the embodiments of the present application can be understood as an end user.

[0102] Step 202: According to the pre-scheduling information, a target hierarchical model is selected from a plurality of hierarchical models, wherein different hierarchical models correspond to different compression ratio ranges.

[0103] The hierarchical model in the embodiment of the present application is used to perform feature extraction processing, that is, compression processing, on the source data based on the compression ratio range.

[0104] Optionally, the hierarchical model in the embodiment of the present application can be understood as a compression model or a feature extraction model.

[0105] In the embodiment of the present application, the above-mentioned multiple multi-level models can be a plurality of pre-set hierarchical models. For example, the above-mentioned multiple hierarchical models include four hierarchical models, and the corresponding compression ratio ranges are (0.9, 1], (0.7, 0.9], (0.5, 0.7], and (0, 0.5] respectively.

[0106] In addition, in the embodiment of the present application, different hierarchical models correspond to different compression strategies. During training, each hierarchical model is trained according to a different compression strategy, which is highly targeted and has high performance when used.

[0107] Based on the above compression ratio range, the target classification model can use a target compression strategy to compress the business data, and the target compression strategy includes different compression ratios corresponding to business data of different priorities, wherein the compression ratio of business data of the first priority is higher than the compression ratio of business data of the second priority, and wherein the first priority is higher than the second priority. For example, the compression ratio range corresponding to the target classification model is (0.9, 1], when compressing the business data of the first priority, a compression ratio of 1 is used for compression, that is, lossless compression is used for the business data of the first priority, and when compressing the business data of the second priority, a compression ratio of 0.9 is used for compression, that is, the amount of business data after compression is 0.9 times the amount of business data before compression.

[0108] The target compression strategy can also be described as a compression method using an importance gradient. That is, the service data is arranged in order from high priority to low priority. When compressing, lossless compression is used for high-priority service data, and lossy compression is used for low-priority service data. When the user's service data volume is greater than the transmission data volume supported by the air interface, the data with low priority is compressed first.

[0109] Step 203: Send the information of the target classification model.

[0110] Here, the information of the target hierarchical model is sent to the terminal so that the terminal can configure a corresponding target recovery model based on the target hierarchical model information, thereby facilitating the terminal to recover data compressed by the target hierarchical model based on the target recovery model.

[0111] Optionally, in an embodiment of the present application, the base station also selects a source-channel joint processing model, and sends information of the selected source-channel joint processing model to the terminal.

[0112] Optionally, the selection process of the above-mentioned source-channel joint processing model may refer to the selection process of the existing source-channel joint processing model, which will not be repeated here.

[0113] In an embodiment of the present application, the base station selects a target hierarchical model from multiple hierarchical models based on the user's pre-scheduling information. The pre-scheduling information is used to indicate the relationship between the amount of data of the user's service data within the pre-scheduling time and the amount of transmission data supported by the air interface. Since different hierarchical models correspond to different compression ratio ranges, it is possible to select a target hierarchical model with a suitable compression ratio range based on the user's pre-scheduling information to compress the user's service data, thereby effectively improving the compatibility of the compressed data (i.e., the extracted service data) after compression by the target hierarchical model with the transmission.

[0114] Optionally, the amount of compressed data obtained after compressing the service data using the target classification model is less than or equal to the amount of transmission data supported by the air interface.

[0115] In an embodiment of the present application, a suitable compression ratio is selected within the compression ratio range corresponding to the target hierarchical model to compress the user's service data, so that the amount of compressed data after compression processing is less than or equal to the amount of transmission data supported by the air interface, thereby enabling the compressed data to meet the amount of transmission data supported by the air interface.

[0116] Optionally, obtaining the user's pre-scheduling information includes:

[0117] Acquire channel state information within a pre-scheduling time and service attribute information corresponding to the user within the pre-scheduling time;

[0118] Obtaining historical scheduling information of multiple users within the pre-scheduling time;

[0119] The pre-scheduling information of the user is obtained according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time.

[0120] like Figure 3 As shown, in an embodiment of the present application, the base station obtains channel state information (Channel State Information, CSI) reported by the terminal, and obtains service attribute information sent by the service source, and generates the above-mentioned pre-scheduling information based on the channel state information, service attribute information and historical scheduling information of multiple users within the pre-scheduling time stored by the base station itself.

[0121] Optionally, the above historical scheduling information includes but is not limited to the following: service attribute information of the user within the corresponding scheduling time, the channel state channel corresponding to the user, and the probability of the base station scheduling resources for the user.

[0122] Optionally, the service attribute information includes but is not limited to service type, total data volume of the service, ratio information of high priority service volume to low priority service volume, etc. Among them, the high priority service volume may include the service volume of services whose service priority is higher than the preset priority, and the low priority service volume includes the service volume of services whose service priority is lower than or equal to the preset priority.

[0123] Here, based on the channel state information, service attribute information and historical scheduling information within the pre-scheduling time, the pre-scheduling information of the user is generated, so as to select a hierarchical model with a suitable compression ratio range based on the pre-scheduling information of the user.

[0124] Optionally, the pre-scheduling information includes a pre-scheduling ratio interval, and the pre-scheduling ratio interval includes an upper limit of the pre-scheduling ratio and a lower limit of the pre-scheduling ratio;

[0125] The generating the pre-scheduling information of the user according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time includes:

[0126] Acquire a first data volume and a second data volume according to the service attribute information, wherein the first data volume is a data volume of a service whose service priority in the service attribute information is higher than a preset priority, and the second data volume is a total data volume of the service in the service attribute information;

[0127] The transmission data volume supported by the air interface is obtained according to the sum of the target data volumes corresponding to N moments within the pre-scheduling time, wherein the target data volume corresponding to each moment is the product of the probability that the base station schedules resources for the user at the moment and the air interface transmission data volume of the transmission channel of the user at the moment, the probability that the base station schedules resources for the user at the moment is predicted according to the historical scheduling information, the air interface transmission data volume of the transmission channel of the user at the moment is predicted according to the channel state information, and N is a positive integer;

[0128] The ratio of the transmission data amount supported by the air interface to the first data amount is determined as the pre-scheduling ratio upper limit of the pre-scheduling ratio interval, and the ratio of the transmission data amount supported by the air interface to the second data amount is determined as the pre-scheduling ratio lower limit of the pre-scheduling ratio interval.

[0129] In some embodiments of the present application, the above-mentioned service attribute information includes service type, total data volume of the service, and ratio information of high priority service volume to low priority service volume. Based on the ratio information and the total data volume of the service, the above-mentioned first quantity can be obtained, and the above-mentioned second data volume can be obtained based on the total quantity of the service.

[0130] From the historical scheduling information of multiple users, obtain the target user that is closest to the service attribute information and the channel state information corresponding to the above user, obtain the probability that the base station schedules resources for the target user, and obtain the probability that the base station schedules resources for the above user at a certain time.

[0131] In the embodiment of the present application, different channel state information corresponds to different air interface transmission data volumes, and based on the corresponding relationship, the air interface transmission data volume corresponding to the channel state information corresponding to a certain moment is obtained.

[0132] In some embodiments of the present application, the above pre-scheduling ratio interval is obtained by the following formula:

[0133] ;

[0134] ;

[0135] in, represents the upper limit of the pre-scheduling ratio of the i-th user within the pre-scheduling time window T, Indicates the amount of transmission data supported by the air interface. represents the first data volume, represents the lower limit of the pre-scheduling ratio of the i-th user within the pre-scheduling time window T, represents the second data volume, represents the probability that the base station schedules resources for the i-th user at time t, represents the air interface data transmission volume of the transmission channel of the i-th user at time t.

[0136] In summary, the pre-scheduling information is .

[0137] The base station compares the above-mentioned upper limit of the pre-scheduling ratio and the lower limit of the pre-scheduling ratio with the compression ratio ranges of each hierarchical model to select the closest and most appropriate hierarchical model.

[0138] Optionally, the pre-scheduling information includes a pre-scheduling ratio interval;

[0139] Selecting a target hierarchical model from a plurality of hierarchical models according to the pre-scheduling information includes:

[0140] Selecting a target compression ratio range matching the pre-scheduling ratio interval according to compression ratio ranges corresponding to a plurality of hierarchical models, wherein the pre-scheduling ratio interval is within the target compression ratio range, or a pre-scheduling ratio lower limit of the pre-scheduling ratio interval is within the target compression ratio range;

[0141] The classification model corresponding to the target compression ratio range is determined as the target classification model.

[0142] In the embodiment of the present application, the upper limit of the pre-scheduling ratio is used to indicate the ratio of the amount of transmission data supported by the air interface to the amount of high-priority service data transmitted by the user within the pre-scheduling time, and the lower limit of the pre-scheduling ratio is used to indicate the ratio of the amount of transmission data supported by the air interface within the pre-scheduling time to the total amount of data transmitted by the user. The lower limit of the pre-scheduling ratio can be understood as the maximum compression degree of the service data, and the upper limit of the pre-scheduling ratio can be understood as the minimum compression degree of the service data. In the embodiment of the present application, it is first determined whether the amount of data of the service data compressed according to the minimum compression degree is less than or equal to the amount of transmission data supported by the air interface. If it is less than or equal to the amount of transmission data supported by the air interface, the compression ratio range including the above-mentioned pre-scheduling ratio interval is selected as the target compression ratio range. If it is greater than the amount of transmission data supported by the air interface, it is further determined whether the amount of data of the service data compressed according to the minimum compression degree is less than or equal to the amount of transmission data supported by the air interface. If it is less than or equal to the number of transmissions supported by the air interface, the compression ratio range including the above-mentioned lower limit of the pre-scheduling ratio is selected as the target compression ratio range.

[0143] In the embodiment of the present application, assuming that the total data volume is 330MB, the transmission data volume supported by the air interface within the pre-scheduling time is 150MB, and the ratio of high priority traffic to low priority traffic is 6:4, then , , then the compression range of (0,0.5] can be selected from the four compression ratio ranges of (0.9,1], (0.7,0.9], (0.5,0.7], and (0,0.5], and the model corresponding to the selected compression range is determined as the target classification model.

[0144] Optionally, the method of the embodiment of the present application further includes:

[0145] Acquire hierarchical model capability information reported by the terminal, where the hierarchical model capability information is used to indicate the hierarchical models supported by the terminal;

[0146] The sending of the target classification model information includes:

[0147] In a case where the hierarchical model capability information indicates that the terminal supports the target hierarchical model, sending index information of the target hierarchical model;

[0148] When the hierarchical model capability information indicates that the terminal does not support the target hierarchical model, the model information of the target hierarchical model is sent or the model information of the target recovery model corresponding to the target hierarchical model is sent; wherein the target recovery model is used to recover the data compressed by the target hierarchical model.

[0149] In an embodiment of the present application, when the hierarchical model capability information indicates that the terminal supports the target hierarchical model, it indicates that the terminal can obtain the model information of the target hierarchical model. Therefore, in this case, the base station does not need to transmit the model information of the target hierarchical model to the terminal. The terminal can determine the model information of the corresponding target hierarchical model based on the index information of the target hierarchical model transmitted by the base station, and then can determine the model information of the corresponding target recovery model. By only transmitting the index information, the signaling transmission overhead between the base station or the network is reduced. When the hierarchical model capability information indicates that the terminal does not support the target hierarchical model, it indicates that the terminal cannot obtain the model information of the target hierarchical model. At this time, the base station needs to transmit the model information of the target hierarchical model so that the terminal can configure the corresponding target recovery model based on the model information of the target hierarchical model, or the base station directly transmits the model information of the target recovery model to the terminal.

[0150] For example, the terminal's hierarchical model capability information indicates that the terminal supports hierarchical model 1, hierarchical model 2, and hierarchical model 3, and the target hierarchical model is hierarchical model 2. The base station can send index information of hierarchical model 2 to the terminal to save signaling transmission overhead.

[0151] Optionally, the method of the embodiment of the present application further includes:

[0152] Acquire feedback information sent by the terminal, the feedback information is used to indicate that the terminal has configured a corresponding target recovery model according to information of the target hierarchical model, and the target recovery model is used to recover data compressed by the target hierarchical model;

[0153] The compressed data is sent according to the feedback information.

[0154] Here, after configuring the corresponding target recovery model based on the target hierarchical model information sent by the base station, the terminal sends feedback information to the base station so that the base station can determine that the terminal can recover the data compressed by the target hierarchical model, and then start sending compressed data based on the feedback information.

[0155] In an embodiment of the present application, a base station selects a target hierarchical model from a plurality of hierarchical models based on the user's pre-scheduling information, and the pre-scheduling information is used to indicate the relationship between the amount of service data of the user within the pre-scheduling time and the amount of transmission data supported by the air interface. The amount of compressed data obtained based on the target hierarchical model is less than or equal to the amount of transmission data supported by the air interface, so that the amount of data compressed based on the selected target hierarchical model can meet the amount of transmission data supported by the air interface, effectively improving the adaptability of the compressed data (i.e., the extracted service data) to the transmission. In an embodiment of the present application, hierarchical models with different compression ratio ranges are used for feature extraction for different pre-scheduling information, instead of using a unified compression ratio for feature extraction, effectively improving the extraction performance and coding efficiency.

[0156] like Figure 4 As shown, the embodiment of the present application also provides a method for processing hierarchical model information, which is applied to a terminal, and the method includes:

[0157] Step 401: Obtain information of a target classification model, where the target classification model is a classification model selected from a plurality of classification models based on the user's pre-scheduling information, wherein different classification models correspond to different compression ratio ranges, and the pre-scheduling information is used to indicate the relationship between the amount of user service data within the pre-scheduling time and the amount of transmission data supported by the air interface.

[0158] The hierarchical model in the embodiment of the present application is used to perform feature extraction processing, ie, compression processing, on the source data based on the compression ratio range. Optionally, the hierarchical model in the embodiment of the present application can be understood as a compression model or a feature extraction model.

[0159] For example, the pre-scheduling information is used to indicate the ratio of the number of transmissions supported by the air interface within the pre-scheduling time to the amount of service data of the user within the pre-scheduling time.

[0160] The data volume of the user's service data within the pre-scheduling time includes at least one of the total data volume of the user's service data within the pre-scheduling time and the data volume of service data with a service priority higher than a preset priority within the pre-scheduling time.

[0161] Optionally, the above pre-scheduling time can be described as a pre-scheduling time window T.

[0162] Optionally, the user in the embodiments of the present application can be understood as an end user.

[0163] In addition, in the embodiment of the present application, different hierarchical models correspond to different compression strategies. During training, each hierarchical model is trained according to a different compression strategy, which is highly targeted and has high performance when used.

[0164] Step 402: According to the information of the target hierarchical model, a target recovery model equivalent to the target hierarchical model is configured, and the target recovery model is used to recover the data compressed by the target hierarchical model.

[0165] Optionally, in an embodiment of the present application, the terminal also obtains information about the source channel joint processing model sent by the base station, and configures a corresponding source channel joint processing model based on the information.

[0166] In an embodiment of the present application, the target hierarchical model is a hierarchical model selected from multiple hierarchical models based on the user's pre-scheduling information, wherein different hierarchical models correspond to different compression ratio ranges, and thus the target compression ratio range of the target hierarchical model is a compression ratio range applicable to the user's pre-scheduling information, thereby effectively improving the adaptability of the data compressed by the selected hierarchical model (i.e., the extracted business data) to transmission.

[0167] Optionally, the amount of compressed data obtained after compressing the service data using the target classification model is less than or equal to the amount of transmission data supported by the air interface.

[0168] In an embodiment of the present application, a suitable compression ratio is selected within the compression ratio range corresponding to the target hierarchical model to compress the user's service data, so that the amount of compressed data after compression processing is less than or equal to the amount of transmission data supported by the air interface, thereby enabling the compressed data to meet the amount of transmission data supported by the air interface.

[0169] In an embodiment of the present application, a target classification model is selected from multiple classification models based on the user's pre-scheduling information, and the pre-scheduling information is used to indicate the relationship between the data volume of the user's service data within the pre-scheduling time and the transmission data volume supported by the air interface. The data volume of compressed data obtained based on the target classification model is less than or equal to the transmission data volume supported by the air interface, so that the data volume compressed based on the selected target classification model can meet the transmission data volume supported by the air interface, effectively improving the adaptability of the compressed data (i.e., the extracted service data) and the transmission.

[0170] Optionally, the method of the embodiment of the present application further includes:

[0171] Sending hierarchical model capability information, where the hierarchical model capability information is used to indicate the hierarchical models supported by the terminal;

[0172] The obtaining of information of the target classification model includes:

[0173] Acquire index information of a target hierarchical model, where the index information is sent when the hierarchical model capability information indicates that the terminal supports the target hierarchical model;

[0174] Alternatively, model information of the target hierarchical model or model information of the target recovery model corresponding to the target hierarchical model is obtained, wherein the information of the target hierarchical model or the model information of the target recovery model is sent when the hierarchical model capability information indicates that the terminal does not support the target hierarchical model.

[0175] Here, the terminal sends the hierarchical model capability information to the base station, so that the base station can subsequently send the index information of the target hierarchical model or the model information of the target hierarchical model to the terminal based on the hierarchical model capability information.

[0176] In an embodiment of the present application, when the hierarchical model capability information indicates that the terminal supports the target hierarchical model, it indicates that the terminal can obtain the model information of the target hierarchical model. Therefore, in this case, the base station does not need to transmit the model information of the target hierarchical model to the terminal. The terminal can determine the model information of the corresponding target hierarchical model based on the index information of the target hierarchical model transmitted by the base station, and then can determine the model information of the corresponding target recovery model. By only transmitting the index information, the signaling transmission overhead between the base station or the network is reduced. When the hierarchical model capability information indicates that the terminal does not support the target hierarchical model, it indicates that the terminal cannot obtain the model information of the target hierarchical model. At this time, the base station needs to transmit the model information of the target hierarchical model so that the terminal can configure the corresponding target recovery model based on the model information of the target hierarchical model, or the base station directly transmits the model information of the target recovery model to the terminal.

[0177] For example, the terminal's hierarchical model capability information indicates that the terminal supports hierarchical model 1, hierarchical model 2, and hierarchical model 3, and the target hierarchical model is hierarchical model 2. The base station can send index information of hierarchical model 2 to the terminal to save signaling transmission overhead.

[0178] Optionally, the method of the embodiment of the present application further includes:

[0179] Feedback information is sent, where the feedback information is used to indicate that the terminal has configured a corresponding target recovery model according to the information of the target hierarchical model.

[0180] Here, after configuring the corresponding target recovery model based on the target hierarchical model information sent by the base station, the terminal sends feedback information to the base station so that the base station can determine that the terminal can recover the data compressed by the target hierarchical model, and then start sending compressed data based on the feedback information.

[0181] The following describes the interaction process between the terminal and the base station in the present application in conjunction with the embodiments.

[0182] like Figure 5 The interaction process includes:

[0183] Step 1: The terminal sends hierarchical model capability information to the base station in response to the capability query instruction of the base station.

[0184] Step 2: The terminal sends channel state information to the base station.

[0185] Step 3: The service information source sends the above service attribute information to the base station.

[0186] Step 3 can be performed simultaneously with step 2, or before step 2. There is no restriction on the order of step 2.

[0187] Step 4: The base station obtains the above pre-scheduling information based on the channel state information, service attribute information and historical scheduling information.

[0188] Step 5: Select a target hierarchical model based on the pre-scheduling information and the compression ratio range of the hierarchical model, and then perform feature extraction based on the target hierarchical model and complete the source-channel joint processing model selection.

[0189] The above feature extraction can be understood as compressing the source data based on the target classification model.

[0190] The above-mentioned source-channel joint processing model may be the same as the existing model and will not be described in detail here.

[0191] Step 6: According to the hierarchical model capability information, send the index information or model information of the target hierarchical model.

[0192] Step 7: The terminal configures the equivalent target recovery model and source channel joint processing model, and sends feedback information after the configuration is completed.

[0193] Step 8: The base station transmits the compressed data.

[0194] Specifically, assuming that the model capability information reported by the terminal includes model 1, model 2 and model 3, the service type in the service attribute information is HD video, the total data volume is 330MB, the transmission data volume supported by the air interface within the pre-scheduling time is 150MB, the ratio of high priority service volume to low priority service volume is 6:4, the compression ratio range corresponding to model 1 is (0.9, 1], the compression ratio range corresponding to model 2 is (0.7, 0.9], the compression ratio range corresponding to model 3 is (0.5, 0.7], and the compression ratio range corresponding to model 4 is (0, 0.5], and the pre-scheduling range obtained based on the above information is (0.45, 0.77], then the above model 4 is selected as the target hierarchical model, and based on this, the source data is feature extracted, and the model information of model 4 is sent to the terminal. The terminal configures the corresponding recovery model based on the model information of model 4 and sends feedback information, and then starts transmitting data based on the feedback information.

[0195] In an embodiment of the present application, for semantic communication, model selection is required for hierarchical feature extraction, and the basis for selection comes from pre-scheduling information, which is generated by the base station. The base station collects the current CSI of the UE, the historical scheduling information of multiple users within the base station pre-scheduling time window T, and the current service attributes of the UE (service type, total quantity value, proportion of high and low priority services). The hierarchical feature extraction model is selected in combination with the source channel coding model, and synchronized to the terminal, which sends feedback information. The base station sends service data that has completed feature extraction based on the selected hierarchical model. The hierarchical feature extraction in this scheme will greatly improve the extraction performance and transmission adaptation level, and the present application provides a method and basis for the classification and selection of feature extraction models, which can make more flexible and effective use of wireless resources.

[0196] like Figure 6 As shown, the embodiment of the present application also provides a processing device for hierarchical model information, which is applied to a base station, and the device includes:

[0197] The first processing module 601 is used to obtain pre-scheduling information of the user, where the pre-scheduling information is used to indicate the relationship between the amount of service data of the user within the pre-scheduling time and the amount of transmission data supported by the air interface;

[0198] A second processing module 602 is used to select a target hierarchical model from a plurality of hierarchical models according to the pre-scheduling information, wherein different hierarchical models correspond to different compression ratio ranges;

[0199] The first transceiver module 603 is used to send the information of the target classification model.

[0200] Optionally, the amount of compressed data obtained after compressing the service data using the target classification model is less than or equal to the amount of transmission data supported by the air interface.

[0201] Optionally, the first processing module is used to:

[0202] Acquire channel state information within a pre-scheduling time and service attribute information corresponding to the user within the pre-scheduling time;

[0203] Obtaining historical scheduling information of multiple users within the pre-scheduling time;

[0204] The pre-scheduling information of the user is generated according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time.

[0205] Optionally, the pre-scheduling information includes a pre-scheduling ratio interval;

[0206] The first processing module is used to: obtain a first data volume and a second data volume according to the service attribute information, the first data volume being the data volume of the service with a service priority higher than the preset priority in the service attribute information, and the second data volume being the total data volume of the service in the service attribute information;

[0207] The transmission data volume supported by the air interface is obtained according to the sum of the target data volumes corresponding to N moments within the pre-scheduling time, wherein the target data volume corresponding to each moment is the product of the probability that the base station schedules resources for the user at the moment and the air interface transmission data volume of the transmission channel of the user at the moment, the probability that the base station schedules resources for the user at the moment is predicted according to the historical scheduling information, the air interface transmission data volume of the transmission channel of the user at the moment is predicted according to the channel state information, and N is a positive integer;

[0208] The ratio of the transmission data amount supported by the air interface to the first data amount is determined as the pre-scheduling ratio upper limit of the pre-scheduling ratio interval, and the ratio of the transmission data amount supported by the air interface to the second data amount is determined as the pre-scheduling ratio lower limit of the pre-scheduling ratio interval.

[0209] Optionally, the pre-scheduling information includes a pre-scheduling ratio interval;

[0210] The second processing module is used for:

[0211] Selecting a target compression ratio range matching the pre-scheduling ratio interval according to compression ratio ranges corresponding to a plurality of hierarchical models, wherein the pre-scheduling ratio interval is within the target compression ratio range, or a pre-scheduling ratio lower limit of the pre-scheduling ratio interval is within the target compression ratio range;

[0212] The classification model corresponding to the target compression ratio range is determined as the target classification model.

[0213] Optionally, the device of the embodiment of the present application further includes:

[0214] A second transceiver module is used to obtain hierarchical model capability information reported by the terminal, where the hierarchical model capability information is used to indicate the hierarchical model supported by the terminal;

[0215] The first transceiver module is used for:

[0216] In a case where the hierarchical model capability information indicates that the terminal supports the target hierarchical model, sending index information of the target hierarchical model;

[0217] When the hierarchical model capability information indicates that the terminal does not support the target hierarchical model, sending model information of the target hierarchical model or sending model information of a target recovery model corresponding to the target hierarchical model;

[0218] The target recovery model is used to recover the data compressed by the target classification model.

[0219] Optionally, the device of the embodiment of the present application further includes:

[0220] A third transceiver module is used to receive feedback information sent by a terminal, wherein the feedback information is used to indicate that the terminal has configured a corresponding target recovery model according to the information of the target hierarchical model, and the target recovery model is used to recover the data compressed by the target hierarchical model;

[0221] The fourth transceiver module is used to send compressed data in response to the feedback information, where the compressed data is obtained by compressing the business data using the target classification model.

[0222] It should be noted that the device is a device corresponding to the method embodiment executed by the above-mentioned base station. All implementation methods of the above-mentioned method embodiment can be applied to the device embodiment and can achieve the same technical effect, which will not be repeated here.

[0223] like Figure 7 As shown, the embodiment of the present application also provides a device for processing hierarchical model information, which is applied to a terminal, and the device includes:

[0224] The fifth transceiver module 701 is used to obtain information of a target hierarchical model, where the target hierarchical model is a hierarchical model selected from multiple hierarchical models based on the pre-scheduling information of the user, wherein different hierarchical models correspond to different compression ratio ranges, and the pre-scheduling information is used to indicate the relationship between the data volume of the user's service data within the pre-scheduling time and the transmission data volume supported by the air interface;

[0225] The third processing module 702 is used to configure a target recovery model equivalent to the target hierarchical model according to the information of the target hierarchical model, and the target recovery model is used to recover the data compressed by the target hierarchical model.

[0226] Optionally, the device of the embodiment of the present application further includes:

[0227] a sixth transceiver module, configured to send hierarchical model capability information, wherein the hierarchical model capability information is used to indicate a hierarchical model supported by the terminal;

[0228] The fifth transceiver module is used for:

[0229] Acquire index information of a target hierarchical model, where the index information is sent when the hierarchical model capability information indicates that the terminal supports the target hierarchical model;

[0230] Alternatively, model information of the target hierarchical model or model information of the target recovery model corresponding to the target hierarchical model is obtained, wherein the information of the target hierarchical model or the model information of the target recovery model is sent when the hierarchical model capability information indicates that the terminal does not support the target hierarchical model.

[0231] Optionally, the device of the embodiment of the present application further includes:

[0232] The seventh transceiver module is used to send feedback information, where the feedback information is used to indicate that the terminal has configured a corresponding target recovery model according to the information of the target hierarchical model.

[0233] It should be noted that the device is a device corresponding to the method embodiment executed by the above-mentioned terminal. All implementation methods of the above-mentioned method embodiment can be applied to the device embodiment and can achieve the same technical effect, which will not be repeated here.

[0234] The embodiment of the present application provides a device for processing hierarchical model information, which may be specifically a base station, such as Figure 8 As shown, it includes a transceiver 810, a processor 800, a memory 820, and a program or instruction stored in the memory 820 and executable on the processor 800; when the processor 800 executes the program or instruction, the steps of the processing method of the hierarchical model information executed by the above-mentioned base station are implemented.

[0235] The transceiver 810 is used to receive and send data under the control of the processor 800 .

[0236] Among them, Figure 8 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking together various circuits of one or more processors represented by processor 800 and memory represented by memory 820. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 810 may be a plurality of components, i.e., including a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. The processor 800 is responsible for managing the bus architecture and general processing, and the memory 820 may store data used by the processor 800 when performing operations.

[0237] The present application embodiment provides a device for processing hierarchical model information, such as Fig. 9As shown, it includes a transceiver 910, a processor 900, a memory 920, and a program or instruction stored in the memory 920 and executable on the processor 900; when the processor 900 executes the program or instruction, the steps of the method for processing the hierarchical model information executed by the above terminal are implemented. The hierarchical model information processing device may be an electronic device. The electronic device may be a terminal, a computer device, a user device, etc.

[0238] The transceiver 910 is used to receive and send data under the control of the processor 900.

[0239] Among them, Fig. 9 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 900 and various circuits of memory represented by memory 920 are linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 910 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. For different user devices, the user interface 930 may also be an interface capable of externally and internally connecting required devices, and the connected devices include but are not limited to a keypad, a display, a speaker, a microphone, a joystick, and the like.

[0240] The processor 900 is responsible for managing the bus architecture and general processing, and the memory 920 can store data used by the processor 900 when performing operations.

[0241] An embodiment of the present application provides a system for processing hierarchical model information, including a terminal and a base station, wherein the base station can implement the steps of the method for processing hierarchical model information executed by the base station as described above, and the terminal can implement the steps of the method for processing hierarchical model information executed by the terminal as described above.

[0242] An embodiment of the present application also provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps in the method for processing hierarchical model information as described above are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0243] The embodiment of the present application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the various processes of the embodiment of the above-mentioned hierarchical model information processing method and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0244] The processor is a processor in the hierarchical model information processing device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0245] It should be further explained that the terminals described in this specification include but are not limited to smart phones, tablet computers, etc., and many of the functional components described are called modules in order to more particularly emphasize the independence of their implementation methods.

[0246] In the embodiment of the present application, module can be implemented with software so that it can be executed by various types of processors. For example, an executable code module of an identification can include one or more physical or logical blocks of computer instructions, for example, it can be constructed as an object, process or function. Nevertheless, the executable code of the identified module does not need to be physically located together, but can include different instructions stored in different positions, and when these instructions are logically combined together, it constitutes a module and realizes the specified purpose of the module.

[0247] In fact, executable code module can be a single instruction or many instructions, and can even be distributed on a plurality of different code segments, distributed among different programs, and distributed across a plurality of memory devices.Similarly, operation data can be identified in the module, and can be implemented and organized in the data structure of any appropriate type according to any appropriate form.Described operation data can be collected as a single data set, or can be distributed in different locations (including on different storage devices), and can only be present on a system or network as an electronic signal at least in part.

[0248] When a module can be implemented by software, considering the level of existing hardware technology, a person skilled in the art can build a corresponding hardware circuit to implement the corresponding function of the module that can be implemented by software without considering the cost. The hardware circuit includes a conventional very large scale integration (VLSI) circuit or gate array and existing semiconductors such as logic chips, transistors, or other discrete components. The module can also be implemented by a programmable hardware device, such as a field programmable gate array, a programmable array logic, a programmable logic device, etc.

[0249] The above exemplary embodiments are described with reference to the accompanying drawings, and many different forms and embodiments are feasible without departing from the spirit and teachings of the present application. Therefore, the present application should not be constructed as a limitation of the exemplary embodiments proposed herein. More specifically, these exemplary embodiments are provided so that the present application will be perfect and complete, and the scope of the present application will be conveyed to those who are familiar with the technology. In these figures, the component sizes and relative sizes may be exaggerated for clarity. The terms used here are only based on the purpose of describing specific exemplary embodiments and are not intended to be limiting. As used herein, unless the text clearly indicates otherwise, the singular forms "one", "an" and "the" are intended to include these multiple forms. It will be further understood that the terms "comprising" and / or "including" when used in this specification indicate the presence of the features, integers, steps, operations, components and / or components, but do not exclude the presence or increase of one or more other features, integers, steps, operations, components, components and / or their groups. Unless otherwise indicated, when stated, a range of values ​​includes the upper and lower limits of that range and any subranges therebetween.

[0250] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for processing hierarchical model information, characterized in that: Applied to a base station, the method comprises: Acquire pre-scheduling information of the user, where the pre-scheduling information is used to indicate the relationship between the amount of service data of the user within the pre-scheduling time and the amount of transmission data supported by the air interface; According to the pre-scheduling information, a target hierarchical model is selected from a plurality of hierarchical models, wherein different hierarchical models correspond to different compression ratio ranges; Sending information of the target classification model; The obtaining of the pre-scheduling information of the user includes: Acquire channel state information within a pre-scheduling time and service attribute information corresponding to the user within the pre-scheduling time; Obtaining historical scheduling information of multiple users within the pre-scheduling time; Generate the pre-scheduling information of the user according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time; The pre-scheduling information includes a pre-scheduling ratio interval; The generating the pre-scheduling information of the user according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time includes: Acquire a first data volume and a second data volume according to the service attribute information, wherein the first data volume is a data volume of a service whose service priority in the service attribute information is higher than a preset priority, and the second data volume is a total data volume of the service in the service attribute information; The transmission data volume supported by the air interface is obtained according to the sum of the target data volumes corresponding to N moments within the pre-scheduling time, wherein the target data volume corresponding to each moment is the product of the probability that the base station schedules resources for the user at the moment and the air interface transmission data volume of the transmission channel of the user at the moment, the probability that the base station schedules resources for the user at the moment is predicted according to the historical scheduling information, the air interface transmission data volume of the transmission channel of the user at the moment is predicted according to the channel state information, and N is a positive integer; The ratio of the transmission data amount supported by the air interface to the first data amount is determined as the pre-scheduling ratio upper limit of the pre-scheduling ratio interval, and the ratio of the transmission data amount supported by the air interface to the second data amount is determined as the pre-scheduling ratio lower limit of the pre-scheduling ratio interval.

2. The method according to claim 1, characterized in that The amount of compressed data obtained after compressing the service data using the target classification model is less than or equal to the amount of transmission data supported by the air interface.

3. The method according to claim 1 or 2, characterized in that: Selecting a target hierarchical model from a plurality of hierarchical models according to the pre-scheduling information includes: Selecting a target compression ratio range matching the pre-scheduling ratio interval according to compression ratio ranges corresponding to a plurality of hierarchical models, wherein the pre-scheduling ratio interval is within the target compression ratio range, or a pre-scheduling ratio lower limit of the pre-scheduling ratio interval is within the target compression ratio range; The classification model corresponding to the target compression ratio range is determined as the target classification model.

4. The method according to claim 1 or 2, characterized in that: Also includes: Acquire hierarchical model capability information reported by the terminal, where the hierarchical model capability information is used to indicate the hierarchical models supported by the terminal; The sending of the target classification model information includes: In a case where the hierarchical model capability information indicates that the terminal supports the target hierarchical model, sending index information of the target hierarchical model; When the hierarchical model capability information indicates that the terminal does not support the target hierarchical model, sending model information of the target hierarchical model or sending model information of a target recovery model corresponding to the target hierarchical model; The target recovery model is used to recover the data compressed by the target classification model.

5. The method according to claim 1 or 2, characterized in that: Also includes: receiving feedback information sent by a terminal, the feedback information being used to indicate that the terminal has configured a corresponding target recovery model according to information of the target hierarchical model, the target recovery model being used to recover data compressed by the target hierarchical model; In response to the feedback information, compressed data is sent, where the compressed data is obtained by compressing the service data using the target classification model.

6. A device for processing hierarchical model information, characterized in that: Applied to a base station, the device comprises: A first processing module, used to obtain pre-scheduling information of the user, where the pre-scheduling information is used to indicate the relationship between the amount of service data of the user within the pre-scheduling time and the amount of transmission data supported by the air interface; A second processing module is used to select a target hierarchical model from a plurality of hierarchical models according to the pre-scheduling information, wherein different hierarchical models correspond to different compression ratio ranges; A first transceiver module, used for sending information of the target classification model; The first processing module is used for: Acquire channel state information within a pre-scheduling time and service attribute information corresponding to the user within the pre-scheduling time; Obtaining historical scheduling information of multiple users within the pre-scheduling time; Generate the pre-scheduling information of the user according to the channel state information, the service attribute information and the historical scheduling information within the pre-scheduling time; The pre-scheduling information includes a pre-scheduling ratio interval; The first processing module is used to: obtain a first data volume and a second data volume according to the service attribute information, the first data volume being the data volume of the service with a service priority higher than the preset priority in the service attribute information, and the second data volume being the total data volume of the service in the service attribute information; The transmission data volume supported by the air interface is obtained according to the sum of the target data volumes corresponding to N moments within the pre-scheduling time, wherein the target data volume corresponding to each moment is the product of the probability that the base station schedules resources for the user at the moment and the air interface transmission data volume of the transmission channel of the user at the moment, the probability that the base station schedules resources for the user at the moment is predicted according to the historical scheduling information, the air interface transmission data volume of the transmission channel of the user at the moment is predicted according to the channel state information, and N is a positive integer; The ratio of the transmission data amount supported by the air interface to the first data amount is determined as the pre-scheduling ratio upper limit of the pre-scheduling ratio interval, and the ratio of the transmission data amount supported by the air interface to the second data amount is determined as the pre-scheduling ratio lower limit of the pre-scheduling ratio interval.

7. The device according to claim 6, characterized in that The amount of compressed data obtained after compressing the service data using the target classification model is less than or equal to the amount of transmission data supported by the air interface.

8. The device according to claim 6 or 7, characterized in that The second processing module is used for: Selecting a target compression ratio range matching the pre-scheduling ratio interval according to compression ratio ranges corresponding to a plurality of hierarchical models, wherein the pre-scheduling ratio interval is within the target compression ratio range, or a pre-scheduling ratio lower limit of the pre-scheduling ratio interval is within the target compression ratio range; The classification model corresponding to the target compression ratio range is determined as the target classification model.

9. The device according to claim 6 or 7, characterized in that Also includes: A second transceiver module is used to obtain hierarchical model capability information reported by the terminal, where the hierarchical model capability information is used to indicate the hierarchical model supported by the terminal; The first transceiver module is used for: In a case where the hierarchical model capability information indicates that the terminal supports the target hierarchical model, sending index information of the target hierarchical model; When the hierarchical model capability information indicates that the terminal does not support the target hierarchical model, sending model information of the target hierarchical model or sending model information of a target recovery model corresponding to the target hierarchical model; The target recovery model is used to recover the data compressed by the target classification model.

10. The device according to claim 6 or 7, characterized in that Also includes: A third transceiver module is used to receive feedback information sent by a terminal, wherein the feedback information is used to indicate that the terminal has configured a corresponding target recovery model according to the information of the target hierarchical model, and the target recovery model is used to recover the data compressed by the target hierarchical model; The fourth transceiver module is used to send compressed data in response to the feedback information, where the compressed data is obtained by compressing the service data using the target classification model.

11. A device for processing hierarchical model information, comprising: A transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; characterized in that when the processor executes the program or instruction, the steps in the method for processing hierarchical model information as described in any one of claims 1 to 5 are implemented.

12. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by a processor, the steps in the method for processing hierarchical model information according to any one of claims 1 to 5 are implemented.

13. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps in the method for processing hierarchical model information as claimed in any one of claims 1 to 5.

Images