Machine learning intelligent method and system for estimating gain control contour
Through machine learning intelligent methods and systems, the problem of high processing burden and cost of DSP components in multi-standard multi-band wireless communication systems is solved, and automated and intelligent gain control profile prediction is realized, improving signal quality and system performance.
Patent Information
- Application Number
- CN202410999078.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-12-06
AI Technical Summary
In multi-standard and multi-band wireless communication systems, digital signal processing DSP components are difficult to flexibly handle different wireless RF standard specifications and parameters, resulting in high design and manufacturing costs and lack of flexibility.
Using machine learning intelligent methods and systems, we can acquire and organize the total gain adjustment amount and amplifier-related data, establish basic algorithms and estimate model databases, machine learning instruction sets and data set storage devices, and use data processors to train the estimate model. Optimize the estimate model to determine the gain profile data of each amplifier, enabling automated and intelligent gain control.
It reduces the time and cost of manual debugging and testing, improves signal quality and stability, realizes accurate prediction and adjustment of the gain control profile, reduces the need for manual intervention, and improves the automation and intelligence level of the system.
Smart Images

Figure CN119089295A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of artificial intelligence machine learning technology, and in particular to a machine learning intelligent method and system for estimating a gain control profile. Background Art
[0002] Wireless communication systems are usually designed for a specific standard, such as GSM (Global System for Mobile Communications), W-CDMA (Wideband Code Division Multiple Access), LTE (Long Term Evolution), etc. In today's wireless service convergence market, users generally require that they can use the same wireless device, such as a smartphone, to obtain communication services from different operators with different communication system standards. This demand is also driving the development of multi-standard and multi-band transceivers, which can send / receive information from multiple types of radio signals in the entire wireless communication spectrum. The requirement of using the same device to adapt to transceivers of multiple standards means that the RF front end (the circuit between the transceiver antenna and the digital baseband stage) needs to provide a flexible automatic gain control (AGC) function so that the amplifier gain can be optimally adjusted according to different standards in different bands; the function of "determining what is the optimal amplifier gain value" is undertaken by the baseband digital signal processing (DSP) component to which it belongs; and in the case of multiple bands and multiple standards, this traditional method is not conducive to the design and manufacture of DSP components, and increases the processing burden of digital signal processing DSP components for different wireless RF standard specifications and required parameters, lacks flexibility, and is also costly. Summary of the invention
[0003] The present invention provides a machine learning intelligent method and system for estimating a gain control profile, which solves the technical problems that digital signal processing (DSP) components have a processing burden for different wireless radio frequency standards and required parameters, lack flexibility, and have high costs.
[0004] In order to solve the above technical problems, the technical solution of the present invention is as follows: In a first aspect, a machine learning intelligent method for estimating a gain control profile, the method comprising: Obtaining the total gain adjustment amount and data related to each amplifier to obtain input data; Establishing a basic algorithm and estimation model database and a machine learning instruction set and data set storage device, wherein the basic algorithm and estimation model database is used to store basic algorithms and various estimation models as well as updated estimation models continuously evolved through machine learning, and the machine learning instruction set and data set storage device is used to store instruction sets, amplifier reference data sets and training data sets; According to the basic algorithm and the estimation model database and the machine learning instruction set and the data set storage device, the estimation model training is performed using a data processor to obtain a training model; Use the training model to estimate the input data, and test and analyze the estimated results to obtain analysis results; According to the analysis results, model selection, estimation and testing are repeated to obtain the optimized estimation model; According to the optimized prediction model, the gain profile data group of each amplifier is determined, and the gain profile data group is output to the gain control preprocessing unit.
[0005] Furthermore, the total gain adjustment amount and data related to each amplifier are obtained to obtain input data, including: The gain control preprocessing unit obtains the total gain adjustment amount from the baseband digital signal processor; The gain control preprocessing unit obtains data related to each amplifier from the RFIC chip, including performance requirements required by the currently active wireless communication standards, received signal parameters related to the amplifier, and current environmental parameters; The gain control preprocessing unit organizes and weaves the total gain adjustment and the data related to each amplifier to obtain input data; The machine learning intelligence system receives input data from the gain control pre-processing unit.
[0006] Furthermore, a basic algorithm and prediction model database and a machine learning instruction set and data set storage device are established. The basic algorithm and prediction model database is used to store basic algorithms and various prediction models as well as updated prediction models obtained through continuous evolution of machine learning. The machine learning instruction set and data set storage device is used to store instruction sets, amplifier reference data sets and training data sets, including: Obtain basic algorithms and estimation models; Continuously evolve and optimize the estimation model through machine learning to obtain an updated estimation model; Storing the basic algorithm, the estimation model, and the updated estimation model in a basic algorithm and estimation model database; Obtain machine learning instruction sets and amplifier reference datasets; Get the training dataset; The instruction set, the amplifier reference data set, and the training data set are stored in a machine learning instruction set and data set storage device.
[0007] Further, according to the basic algorithm and the estimation model database and the machine learning instruction set and the data set storage device, the estimation model training is performed using a data processor to obtain a training model, including: Obtain training data from a machine learning instruction set and data set storage device; Select a basic algorithm and an estimation model from a basic algorithm and estimation model database; Using a data processor to train an estimation model based on the training data and the selected basic algorithm and estimation model; Acquire an amplifier reference data set from a machine learning instruction set and a data set storage device, and perform verification using an estimation model trained with the amplifier reference data set to obtain a verification result; The model parameters are adjusted according to the verification results to obtain the training model.
[0008] Furthermore, the training model is used to estimate the input data, and the estimated results are tested and analyzed to obtain analysis results, including: The machine learning intelligent system receives input data from the gain control preprocessing unit; Use the training model to estimate the input data to obtain the estimated results; The estimated results are tested and analyzed to obtain analytical results.
[0009] Furthermore, based on the analysis results, model selection, estimation and testing are repeated to obtain an optimized estimation model, including: According to the analysis results, different basic algorithms and estimation models are selected; Use the selected estimation model to perform estimation to obtain estimation results; Testing and evaluating the estimated results to obtain testing and evaluation results; According to the results of detection and evaluation, the model is adjusted and optimized to obtain an optimized estimation model.
[0010] Further, according to the optimization prediction model, the gain profile data group of each amplifier is determined, and the gain profile data group is output to the gain control preprocessing unit, including: According to the input data of the gain control preprocessing unit, an optimized prediction model is used to perform prediction to obtain a prediction result; According to the estimation results, the gain profile data set currently required for each amplifier is determined.
[0011] In a second aspect, a machine learning intelligent system for estimating a gain control profile comprises: Basic algorithm and estimation model database, used to store various estimation models and updated estimation models that are continuously evolved through machine learning; A machine learning instruction set and data set storage device for storing instruction sets and training data sets for storing characteristics of each individual amplifier stage and performance requirements of various wireless standards and signal characteristics and gain value data in the geographic and physical environment in which the amplifier was historically located; A data processor is used to infer the gain profile data set.
[0012] According to a third aspect, a computing device includes: one or more processors; The storage system is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the above method.
[0013] In a fourth aspect, a computer-readable storage medium stores a program, and the program implements the above method when executed by a processor.
[0014] The above solution of the present invention includes at least the following beneficial effects: In the above scheme of the present invention, the machine learning method can automatically learn and explore the relationship between input data and output results through model training and data analysis, and machine learning can more accurately predict the gain control profile; the machine learning model can be adaptively adjusted according to different input data, so the gain control profile can be estimated according to different amplifier characteristics and environmental conditions, so that the machine learning method can cope with various complex amplifier scenarios and changes; using the machine learning method for estimation can greatly reduce the time and cost of manual debugging and testing, and quickly find the optimal solution through automated model training and estimation process; the machine learning method can identify the factors and features that affect the gain control profile by analyzing a large amount of data and patterns, and can improve the gain control performance of the amplifier and improve the quality and stability of the signal by accurately predicting these factors and features; the machine learning method can realize automated and intelligent gain control profile estimation, and by establishing an estimation model, machine learning can automatically estimate the gain control profile in real-time or offline environments, reducing the need for manual intervention and monitoring, and improving the automation and intelligence level of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 It is a flow chart of a machine learning intelligent method for estimating a gain control profile provided by an embodiment of the present invention.
[0016] Figure 2It is a schematic diagram of a machine learning intelligent system for estimating a gain control profile provided by an embodiment of the present invention.
[0017] Reference numerals: 1304, gain control preprocessing unit; 1320, data processor; 1330, machine learning instruction set and data set storage device; 1332, instruction set; 1334, amplifier reference data set; 1336, training data set; 1340, basic algorithm and estimation model database. DETAILED DESCRIPTION
[0018] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0019] like Figure 1 As shown, an embodiment of the present invention proposes a machine learning intelligent method for estimating a gain control profile, the method comprising: Step 11, obtaining the total gain adjustment amount and data related to each amplifier to obtain input data; Step 12, establishing a basic algorithm and estimation model database 1340 and a machine learning instruction set and data set storage device 1330, wherein the basic algorithm and estimation model database 1340 is used to store basic algorithms and various estimation models as well as updated estimation models continuously evolved through machine learning, and the machine learning instruction set and data set storage device 1330 is used to store instruction sets, amplifier reference data sets and training data sets; Step 13, using the data processor 1320 to perform estimation model training according to the basic algorithm and estimation model database 1340 and the machine learning instruction set and data set storage device 1330 to obtain a training model; Step 14, using the training model to estimate the input data, and testing and analyzing the estimated results to obtain analysis results; Step 15, based on the analysis results, repeatedly perform model selection, estimation and testing to obtain an optimized estimation model; Step 16: determine the gain profile data group of each amplifier according to the optimization prediction model, and output the gain profile data group to the gain control preprocessing unit 1304.
[0020] In the embodiment of the present invention, the total gain adjustment amount and data related to each amplifier are collected as input data for subsequent model training and prediction; a basic algorithm and estimation model database 1340 and a machine learning instruction set and data set storage device 1330 are established, and the basic algorithm and estimation model database 1340 is used to store basic algorithms, various estimation models, and updated estimation models obtained through machine learning evolution. The prediction model can be developed using various machine learning algorithms, such as linear regression, logistic regression, decision tree, neural network, etc., and selected according to different data features and requirements to obtain the best prediction effect; the machine learning instruction set and data set storage device 1330 is used to store the instruction set, the amplifier reference data set and the training data set; the prediction model is trained using the data processor 1320 to obtain the training model, the trained model is used to predict the input data, and the prediction results are tested and analyzed to obtain the analysis results; through the training model and the analysis results, an optimized prediction model is obtained, thereby determining the gain profile data group of each amplifier, which can accurately predict the gain profile of each amplifier and achieve precise adjustment of the gain control; through the machine The model trained by the learning algorithm can estimate the input data, test and analyze the estimation results, and iterate the model selection, estimation and detection according to the analysis results to finally obtain the optimized estimation model; the optimized estimation model can better adapt to the actual situation and improve the accuracy and effect of gain control; by determining the gain profile data group of each amplifier and outputting it to the gain control preprocessing unit, the gain of the amplifier can be accurately controlled; the gain profile data group contains the gain value of each amplifier under different input conditions, which can be adjusted according to actual needs, so as to achieve accurate amplification and control of the signal; it can improve the performance and stability of the system, reduce signal distortion and interference, and improve the quality and reliability of signal transmission.
[0021] like Figure 1 As shown, step 11, obtaining the total gain adjustment amount and data related to each amplifier to obtain input data, including: Step 111, the gain control preprocessing unit 1304 obtains a total gain adjustment amount from the baseband digital signal processor; Step 112, the gain control preprocessing unit 1304 obtains data related to each amplifier from the RF integrated circuit chip, including performance requirements required by the currently active wireless communication standard, received signal parameters related to the amplifier, and current environmental parameters; Step 113, the gain control preprocessing unit 1304 arranges and weaves the total gain adjustment amount and the data related to each amplifier to obtain input data; In step 114 , the machine learning intelligent system receives input data from the gain control preprocessing unit 1304 .
[0022] In an embodiment of the present invention, the gain control preprocessing unit 1304 obtains the total gain adjustment amount from the baseband digital signal processor and obtains data related to each amplifier from the radio frequency integrated circuit chip; the data includes the performance requirements required by the current active wireless communication standard, the received signal parameters related to the amplifier, and the current environmental parameters; the gain control preprocessing unit 1304 organizes and weaves the data obtained from the baseband digital signal processor and the radio frequency integrated circuit chip to obtain input data; the input data will be passed to the machine learning intelligent system for processing and analysis; the gain control preprocessing unit is used to integrate data from different sources to provide comprehensive input data for the machine learning intelligent system; the total gain adjustment amount provides the overall gain control requirements, and the gain control requirements related to each amplifier are obtained. The data related to the amplifier provides key information such as amplifier performance requirements, received signal parameters and environmental parameters; by collating and weaving this data, it can provide accurate and comprehensive input for the machine learning intelligent system so that the system can perform effective prediction and control; it provides the input data required by the machine learning intelligent system, so that the system can be trained, estimated and optimized based on the data; by obtaining the total gain adjustment amount and the data related to each amplifier, and collating and weaving them into input data, the accuracy and reliability of the machine learning intelligent system can be improved; the system can better understand and adapt to different gain control requirements and environmental conditions, thereby achieving more accurate gain control profile estimation and adjustment; it plays an important role in performance optimization and signal quality improvement of wireless communication systems.
[0023] like Figure 1 As shown, step 12, establish a basic algorithm and estimation model database 1340 and a machine learning instruction set and data set storage device 1330, the basic algorithm and estimation model database 1340 is used to store basic algorithms and various estimation models and updated estimation models obtained by continuous evolution through machine learning, and the machine learning instruction set and data set storage device 1330 is used to store instruction sets, amplifier reference data sets and training data sets, including: Step 121, obtaining a basic algorithm and an estimation model; Step 122, continuously evolving and optimizing the estimation model through machine learning to obtain an updated estimation model; Step 123, storing the basic algorithm, the estimation model, and the updated estimation model in the basic algorithm and estimation model database 1340; Step 124, obtaining a machine learning instruction set and an amplifier reference data set; Step 125, obtaining a training data set; Step 126, store the instruction set, amplifier reference data set, and training data set in the machine learning instruction set and data set storage device 1330.
[0024] In the embodiment of the present invention, the basic algorithm and prediction model database 1340 is used to store various prediction models and updated prediction models obtained through continuous evolution through machine learning; the basic algorithms include probability theory, statistics, approximation theory, fuzzy logic, etc.; various basic algorithms and prediction models and updated prediction models obtained through continuous evolution through machine learning are used to provide to the data processor 1320 for use in the speculation process; the machine learning instruction set and data set storage device 1330 includes an instruction set 1332, an amplifier reference data set 1334 and a training data set 1336; the amplifier reference data set 1334 is used to store the characteristics of each individual amplifier at each level and the performance requirements of various wireless standards and the signal characteristics and gain value data of the geographic and physical environment where the amplifier was historically located; the characteristics of the amplifier include but are not limited to: gain, noise number, power efficiency and P1dB value (at 1 dB compression); various wireless standards may include but are not limited to: Global System for Mobile Communications (GSM), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), etc.; by analyzing the amplifier reference data set, it is possible to understand the gain changes and performance of the amplifier in different environments and different wireless standards, providing an important reference for estimating the gain control profile; the amplifier reference data set 1334 is used to help the machine learning intelligent system understand and predict the behavior of the amplifier; by analyzing and learning the amplifier reference data set 1334, the machine learning intelligent system can establish an understanding of the amplifier's behavioral patterns and apply them to the process of estimating the gain control profile; the accuracy and reliability of the estimation can be improved, so that the system can better adapt to the requirements of different environments and wireless standards; the existence of the amplifier reference data set plays an important role and effect in establishing an accurate estimation model and achieving accurate gain control profile estimation; it provides a rich data resource for the machine learning intelligent system, enabling the system to better predict and adjust the gain of the amplifier, thereby achieving precise control and optimization of the signal.
[0025] like Figure 1 As shown, step 13, according to the basic algorithm and estimation model database 1340 and the machine learning instruction set and data set storage device 1330, the data processor 1320 is used to perform estimation model training to obtain a training model, including: Step 131, obtaining training data from the machine learning instruction set and data set storage device 1330; Step 132, selecting a basic algorithm and a prediction model from a basic algorithm and prediction model database 1340; Step 133, using the data processor 1320 to perform estimation model training according to the training data and the selected basic algorithm and estimation model; Step 134, obtaining an amplifier reference data set from the machine learning instruction set and data set storage device 1330, and performing verification using the estimation model trained with the amplifier reference data set to obtain a verification result; Step 135, adjusting the model parameters according to the verification results to obtain a training model.
[0026] In an embodiment of the present invention, training data is obtained from the machine learning instruction set and data set storage device 1330 to obtain a data set for training a model; a basic algorithm and a prediction model are selected from the basic algorithm and prediction model database 1340, for example, fuzzy logic is selected as the basic algorithm, and a support vector machine is selected as the prediction model; the selected fuzzy logic algorithm and support vector machine model are input into the data processor 1320 together with the training data to perform model training; an amplifier reference data set is obtained from the machine learning instruction set and data set storage device 1330, and the prediction model trained with the amplifier reference data set is used for verification, which helps to evaluate the generalization ability and accuracy of the model; the parameters of the model are adjusted according to the verification results to optimize the performance and accuracy of the model to obtain a final training model, and the final training model can be used to estimate the gain control profile to achieve more accurate gain control.
[0027] like Figure 1 As shown, in step 14, the input data is estimated using the training model, and the estimated results are tested and analyzed to obtain analysis results, including: Step 141, the machine learning intelligent system receives input data from the gain control preprocessing unit 1304; Step 142, using the training model to estimate the input data to obtain an estimation result; Step 143, testing and analyzing the estimated results to obtain analysis results.
[0028] In an embodiment of the present invention, the machine learning intelligent system receives input data from the gain control preprocessing unit 1304; the gain control preprocessing unit organizes and weaves the total gain adjustment amount and the data related to each amplifier into input data, and transmits it to the machine learning intelligent system; the input data is estimated using the training model to obtain an estimated result; the training model is obtained through the previous training process, and can estimate the gain control profile according to the input data; the estimated result is tested and analyzed to obtain an analysis result; the estimated result is tested to compare the difference between the estimated value and the actual value to evaluate the accuracy of the estimate; at the same time, the estimated result can be further analyzed to explore The relationship and rules between data can be used to extract valuable information; the input data can be estimated by using the training model to obtain the estimated results; the accuracy and reliability of the estimation can be evaluated and the analysis results can be obtained by testing and analyzing the estimated results; the analysis results can be used to evaluate the performance of the gain control profile, understand the characteristics and behavior of the signal, and optimize the estimation model and adjust the gain control strategy; through this process, the machine learning intelligent system can estimate the input data according to the training model, and evaluate the accuracy and effect of the estimation through testing and analysis; it helps to improve the accuracy and stability of gain control, and provide more reliable gain control strategies and decision-making basis for practical applications.
[0029] like Figure 1 As shown, step 15, based on the analysis results, model selection, estimation and testing are repeatedly performed to obtain an optimized estimation model, including: Step 151, selecting different basic algorithms and estimation models according to the analysis results; Step 152, using the selected estimation model to perform estimation to obtain an estimation result; Step 153, testing and evaluating the estimated result to obtain a test and evaluation result; Step 154, adjusting and optimizing the model according to the detection and evaluation results to obtain an optimized prediction model.
[0030] In an embodiment of the present invention, different basic algorithms and estimation models are selected according to the analysis results; different basic algorithms and estimation models can be selected as needed according to the previous analysis results to adapt to different data characteristics and estimation requirements; the selected estimation model is used to perform estimation to obtain estimation results; the selected model is used to estimate the input data to obtain corresponding estimation results for estimating the gain control profile; the estimation results are tested and evaluated to obtain detection and evaluation results; the estimation results are tested and evaluated to compare the difference between the estimated value and the actual value to evaluate the accuracy and performance of the estimation; according to the detection and evaluation results, the model is adjusted and optimized to obtain an optimized estimation model; According to the detection and evaluation results, the parameters of the model can be adjusted and optimized to further improve the model's predictive ability and accuracy; the estimation model can be optimized by repeated model selection, estimation and detection; the performance and accuracy of different models can be evaluated by continuously selecting different estimation models for estimation and detection; according to the detection and evaluation results, the model parameters are adjusted and optimized to further improve the model's predictive ability and accuracy; through this process, an optimized estimation model can be obtained, which can better adapt to actual conditions and improve the accuracy and effect of gain control. The optimized estimation model can more accurately estimate the gain control profile and provide more reliable gain control strategies and decision-making basis for practical applications.
[0031] like Figure 1 As shown, step 16, determining the gain profile data group of each amplifier according to the optimization prediction model, and outputting the gain profile data group to the gain control preprocessing unit 1304, includes: Step 161, using the optimized estimation model to perform estimation according to the input data of the gain control preprocessing unit 1304 to obtain an estimation result; Step 162: Determine the gain profile data set currently required for each amplifier based on the estimation result.
[0032] In an embodiment of the present invention, an optimized prediction model is used to perform prediction based on input data of the gain control preprocessing unit 1304 to obtain a prediction result; the gain control preprocessing unit 1304 provides input data, including a total gain adjustment amount and data related to each amplifier; based on these data, an optimized prediction model is used to perform prediction to obtain a gain control prediction result; based on the prediction result, a gain profile data group currently required for each amplifier is determined; based on the prediction result, the gain profile data group currently required for each amplifier can be determined to meet the performance requirements of the wireless communication standard and the received signal parameters of the amplifier; by analyzing the prediction result and the amplifier reference data set, the gain required for each amplifier can be determined. adjustment amount and profile data; determine the gain profile data group of each amplifier according to the optimized prediction model; by using the optimized prediction model to predict the input data, the prediction result of the gain control profile can be obtained; according to the prediction result, the gain profile data group currently required for each amplifier can be determined to meet the requirements of the wireless communication standard and the performance requirements of the amplifier; the gain adjustment amount and profile data required for each amplifier can be determined according to the real-time input data and the optimized prediction model; provide accurate control guidance for the gain control preprocessing unit 1304, and can adjust the gain of each amplifier in real time according to the prediction result to meet the performance requirements of the wireless communication system and improve the signal quality and communication reliability.
[0033] like Figure 2 As shown, an embodiment of the present invention further provides a machine learning intelligent system 20 for estimating a gain control profile, comprising: Basic algorithm and estimation model database 1340, used to store various estimation models and updated estimation models continuously evolved through machine learning; Machine learning instruction set and data set storage device 1330, used to store instruction sets and training data sets, used to store the characteristics of each individual amplifier stage and the performance requirements of various wireless standards and the signal characteristics and gain value data of the geographic and physical environment where the amplifier was historically located; The data processor 1320 is used to estimate the gain profile data set.
[0034] Optionally, a machine learning intelligent method for estimating a gain control profile, the method comprising: Obtaining the total gain adjustment amount and data related to each amplifier to obtain input data; Establishing a basic algorithm and estimation model database and a machine learning instruction set and data set storage device, wherein the basic algorithm and estimation model database is used to store basic algorithms and various estimation models as well as updated estimation models continuously evolved through machine learning, and the machine learning instruction set and data set storage device is used to store instruction sets, amplifier reference data sets and training data sets; According to the basic algorithm and the estimation model database and the machine learning instruction set and the data set storage device, the estimation model training is performed using a data processor to obtain a training model; Use the training model to estimate the input data, and test and analyze the estimated results to obtain analysis results; According to the analysis results, model selection, estimation and testing are repeated to obtain the optimized estimation model; According to the optimized prediction model, the gain profile data group of each amplifier is determined, and the gain profile data group is output to the gain control preprocessing unit.
[0035] Furthermore, the total gain adjustment amount and data related to each amplifier are obtained to obtain input data, including: The gain control preprocessing unit obtains the total gain adjustment amount from the baseband digital signal processor; The gain control preprocessing unit obtains data related to each amplifier from the RFIC chip, including performance requirements required by the currently active wireless communication standards, received signal parameters related to the amplifier, and current environmental parameters; The gain control preprocessing unit organizes and weaves the total gain adjustment and the data related to each amplifier to obtain input data; The machine learning intelligence system receives input data from the gain control pre-processing unit.
[0036] Furthermore, a basic algorithm and prediction model database and a machine learning instruction set and data set storage device are established, wherein the basic algorithm and prediction model database is used to store basic algorithms and various prediction models as well as updated prediction models continuously evolved through machine learning, and the machine learning instruction set and data set storage device is used to store instruction sets, amplifier reference data sets and training data sets; The basic algorithm and estimation model database is used to store various estimation models and updated estimation models that are continuously evolved through machine learning; Various basic algorithms and estimation models as well as updated estimation models that are continuously evolved through machine learning are provided to the data processor for use in the inference process; The machine learning instruction set and data set storage device includes an instruction set, an amplifier reference data set, and a training data set; The amplifier reference data set is used to store the characteristics of each individual amplifier stage and the performance requirements of various wireless standards and the signal characteristics and gain value data for the geographic and physical environment in which the amplifiers were historically located.
[0037] Further, according to the basic algorithm and the estimation model database and the machine learning instruction set and the data set storage device, the estimation model training is performed using a data processor to obtain a training model, including: Select one or more basic algorithms and prediction models from a basic algorithm and prediction model database; Obtaining a machine learning instruction set and a data set in a data set storage device; Using the data set and the selected basic algorithm and estimation model, performing training of the estimation model on the data processor; Evaluate and validate the training model to obtain validation results; According to the verification results, the parameters of the model are adjusted to obtain a training model.
[0038] Furthermore, the training model is used to estimate the input data, and the estimated results are tested and analyzed to obtain analysis results, including: The machine learning intelligent system receives input data from the gain control preprocessing unit; Use the training model to estimate the input data to obtain the estimated results; The estimated results are tested and analyzed to obtain analytical results.
[0039] Furthermore, based on the analysis results, model selection, estimation and testing are repeated to obtain an optimized estimation model, including: According to the analysis results, different basic algorithms and estimation models are selected; Use the selected estimation model to perform estimation to obtain estimation results; Testing and evaluating the estimated results to obtain testing and evaluation results; According to the results of detection and evaluation, the model is adjusted and optimized to obtain an optimized estimation model.
[0040] Further, according to the optimization prediction model, the gain profile data group of each amplifier is determined, and the gain profile data group is output to the gain control preprocessing unit, including: According to the input data of the gain control preprocessing unit, an optimized prediction model is used to perform prediction to obtain a prediction result; According to the estimation results, the gain profile data set currently required for each amplifier is determined.
[0041] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.
[0042] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.
[0043] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.
[0044] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0045] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0046] In the embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or units, which can be electrical, mechanical or other forms.
[0047] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0048] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0049] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.
[0050] In addition, it should be noted that in the system and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. Moreover, the steps of performing the above series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order, and some steps can be performed in parallel or independently of each other. For those of ordinary skill in the art, it is understandable that all or any steps or components of the method and system of the present invention can be implemented in any computing system (including processors, storage media, etc.) or a network of computing systems in hardware, firmware, software or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.
[0051] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing system. The computing system can be a well-known general system. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code that implements the method or system. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be pointed out that in the system and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.
[0052] The above is a preferred embodiment of the present invention. 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 of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A machine learning intelligent method for estimating a gain control profile, characterized in that: The method comprises: Obtaining the total gain adjustment amount and data related to each amplifier to obtain input data; Establishing a basic algorithm and prediction model database (1340) and a machine learning instruction set and data set storage device (1330), wherein the basic algorithm and prediction model database (1340) is used to store basic algorithms and various prediction models as well as updated prediction models continuously evolved through machine learning, and the machine learning instruction set and data set storage device (1330) is used to store instruction sets, amplifier reference data sets and training data sets; According to the basic algorithm and the estimation model database (1340) and the machine learning instruction set and data set storage device (1330), the data processor 1320 is used to perform estimation model training to obtain a training model; Use the training model to estimate the input data, and test and analyze the estimated results to obtain analysis results; According to the analysis results, model selection, estimation and testing are repeated to obtain the optimized estimation model; According to the optimized prediction model, the gain profile data group of each amplifier is determined, and the gain profile data group is output to the gain control preprocessing unit (1304).
2. The machine learning intelligent method for estimating a gain control profile according to claim 1, characterized in that: Obtain the total gain adjustment and data related to each amplifier to obtain input data, including: The gain control preprocessing unit (1304) obtains the total gain adjustment amount from the baseband digital signal processor; The gain control preprocessing unit (1304) obtains data related to each amplifier from the radio frequency integrated circuit chip, including performance requirements required by the current active wireless communication standard, received signal parameters related to the amplifier, and current environmental parameters; The gain control preprocessing unit (1304) arranges and weaves the total gain adjustment amount and the data related to each amplifier to obtain input data; The machine learning intelligence system receives input data from the gain control pre-processing unit (1304).
3. The machine learning intelligent method for estimating a gain control profile according to claim 2, characterized in that: Establishing a basic algorithm and prediction model database (1340) and a machine learning instruction set and data set storage device (1330), wherein the basic algorithm and prediction model database (1340) is used to store basic algorithms and various prediction models as well as updated prediction models continuously evolved through machine learning, and the machine learning instruction set and data set storage device (1330) is used to store instruction sets, amplifier reference data sets and training data sets, including: Obtain basic algorithms and estimation models; Continuously evolve and optimize the estimation model through machine learning to obtain an updated estimation model; Storing the basic algorithm, the estimation model, and the updated estimation model in a basic algorithm and estimation model database (1340); Obtain machine learning instruction sets and amplifier reference datasets; Get the training dataset; The instruction set, the amplifier reference data set, and the training data set are stored in a machine learning instruction set and data set storage device (1330).
4. The machine learning intelligent method for estimating a gain control profile according to claim 3, characterized in that: According to the basic algorithm and the estimation model database (1340) and the machine learning instruction set and data set storage device (1330), the estimation model training is performed using the data processor (1320) to obtain the training model, including: Obtaining training data from a machine learning instruction set and data set storage device (1330); selecting a basic algorithm and an estimation model from a basic algorithm and estimation model database (1340); Using a data processor (1320) to train an estimation model based on the training data and the selected basic algorithm and estimation model; Obtaining an amplifier reference data set from a machine learning instruction set and data set storage device (1330), and performing verification using an estimation model trained with the amplifier reference data set to obtain a verification result; The model parameters are adjusted according to the verification results to obtain the training model.
5. The machine learning intelligent method for estimating a gain control profile according to claim 4, characterized in that: Use the trained model to estimate the input data, and test and analyze the estimated results to obtain analytical results, including: The machine learning intelligent system receives input data from the gain control pre-processing unit (1304); Use the training model to estimate the input data to obtain the estimated results; The estimated results are tested and analyzed to obtain analytical results.
6. The machine learning intelligent method for estimating a gain control profile according to claim 5, characterized in that: Based on the analysis results, model selection, estimation and testing are repeated to obtain an optimized estimation model, including: According to the analysis results, different basic algorithms and estimation models are selected; Use the selected estimation model to perform estimation to obtain estimation results; Testing and evaluating the estimated results to obtain testing and evaluation results; According to the results of detection and evaluation, the model is adjusted and optimized to obtain an optimized estimation model.
7. The machine learning intelligent method for estimating a gain control profile according to claim 6, characterized in that: Determine the gain profile data group of each amplifier according to the optimization prediction model, and output the gain profile data group to the gain control preprocessing unit (1304), including: According to the input data of the gain control preprocessing unit (1304), an optimized prediction model is used to perform prediction to obtain a prediction result; According to the estimation results, the gain profile data set currently required for each amplifier is determined.
8. A machine learning intelligent system for estimating a gain control profile, characterized in that: include: A basic algorithm and estimation model database (1340) for storing various estimation models and updated estimation models that are continuously evolved through machine learning; A machine learning instruction set and data set storage device (1330) for storing instruction sets and training data sets for storing characteristics of each individual amplifier stage and performance requirements of various wireless standards and signal characteristics and gain value data in the geographic and physical environment in which the amplifier was historically located; A data processor (1320) is used to infer a gain profile data set.
9. A computing device, characterized in that include: one or more processors; A storage system for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.
