Information processing apparatus, information processing method, and information processing program

Abstract

An information processing apparatus (10) acquires a reference value related to operation of a system requested by a user, generates source code using a generated model trained to output source code of software for controlling the system based on input of operation information related to operation of the system, compiles the generated source code, generates the software, evaluates whether a control state of the system using the generated software satisfies the reference value, and provides the generated software to the user if the control state satisfies the reference value.

Description

Technical Field

[0001] This invention relates to an information processing apparatus, an information processing method, and an information processing program. Background Technology

[0002] Currently, in communication systems, developers implement software development, while operators implement the application of that software. On the other hand, as communication systems become more complex, the importance of "DevOps"—which enables close collaboration between software development and application to improve productivity—is increasing.

[0003] Existing technical documents Patent documents Patent Document 1: Japanese Patent Application Publication No. 2023-183447 Summary of the Invention The problem that the invention aims to solve However, there is room for improvement in software development and application. For example, in communication systems, because the developers and implementers of the software are different, the lead time from software development to application can sometimes be lengthy. Furthermore, human error and other malfunctions are more likely to occur.

[0004] The present invention was made in view of the above circumstances, and its object is to provide an information processing apparatus, an information processing method, and an information processing program that can shorten the process from software development to application.

[0005] Technical solutions for solving the problem One aspect of the present invention provides an information processing apparatus comprising: an acquisition unit that acquires reference values ​​related to the use of a system requested by a user; a code unit that generates source code based on the acquired reference values ​​using a trained generative model, wherein the trained generative model outputs source code of software controlling the system based on input application information related to the use of the system; a compilation unit that compiles the generated source code to generate the software; and an evaluation unit that evaluates whether the control status of the system using the generated software meets the reference values, and if the control status meets the reference values, provides the generated software to the user.

[0006] One aspect of the present invention provides an information processing method executed by an information processing device, the method comprising: an acquisition step of acquiring a baseline value related to the use of a system requested by a user; a coding step of generating the source code based on the baseline value using a trained generative model, wherein the trained generative model outputs source code of software controlling the system based on input of use information related to the use of the system; a compilation step of compiling the generated source code to generate the software; and an evaluation step of evaluating whether the control status of the system using the generated software meets the baseline value, and if the control status meets the baseline value, providing the generated software to the user.

[0007] One aspect of the present invention provides an information processing program that causes a computer to perform: an acquisition process, acquiring a baseline value related to the use of a system requested by a user; a coding process, using a trained generative model to generate the source code based on the baseline value, the trained generative model outputting source code of software controlling the system based on input of use information related to the use of the system; a compilation process, compiling the generated source code to generate the software; and an evaluation process, evaluating whether the control status of the system using the generated software meets the baseline value, and if the control status meets the baseline value, providing the generated software to the user. Attached Figure Description

[0008] Figure 1 This is a diagram showing an overview of the information processing system implemented in this way.

[0009] Figure 2 This is a block diagram illustrating a structural example of an information processing apparatus for an information processing system implemented in an embodiment.

[0010] Figure 3 This is a diagram illustrating an example of the training data storage unit of an information processing apparatus according to an embodiment.

[0011] Figure 4 This is a diagram illustrating an example of the code information storage unit of an information processing apparatus according to an embodiment.

[0012] Figure 5 This is a diagram illustrating an example of the software storage unit of an information processing apparatus according to an embodiment.

[0013] Figure 6 This is a diagram illustrating an example of the generation model storage unit of the information processing apparatus in an embodiment.

[0014] Figure 7 This is a diagram illustrating a specific example 1 of an information processing system implemented in this way.

[0015] Figure 8 This is a diagram illustrating a specific example 2 of an information processing system implemented in this way.

[0016] Figure 9 This is a diagram illustrating a specific example 3 of an information processing system implemented in this way.

[0017] Figure 10 This is a flowchart illustrating an example of the processing flow of an information processing system implemented in this way.

[0018] Figure 11 This is a hardware structure diagram of an example of a computer that implements the functions of an information processing device in an embodiment. Detailed Implementation

[0019] Hereinafter, various embodiments of the present invention will be described using the accompanying drawings, but the present invention is not limited to the following embodiments. Furthermore, regarding the description of the drawings, the same reference numerals are used for the same parts and repeated descriptions are omitted; descriptions of components with the same function and the same processing are also omitted.

[0020] [1. Introduction] use Figure 1 An overview of the information processing system 100 of the implementation method will be described. Figure 1 This is a diagram showing an outline of the information processing system 100 according to an implementation method. Figure 1 In this system, the information processing system 100 consists of an information processing device 10 and an operator 0.

[0021] Information processing device 10 is a device that accepts input from operator O and performs various information processing tasks. It may be a RIC (Radio Access Network Intelligent Controller) or a server, etc. RIC is an information processing device that performs various information estimations and other tasks based on RAN and AI (Artificial Intelligence) and other computing resources.

[0022] First, the information processing device 10 trains the generative model M (step S1). For example, the information processing device 10 trains the generative model M by inputting training data based on the usage information of the communication system into the generative model M. Here, the generative model M is a machine learning model that outputs the source code of the software controlling the communication system based on the input of the usage information. In addition, the usage information is information related to the use of the communication system collected from the base station of the communication system, such as the communication system's evaluation indicators, i.e., KPIs (Key Performance Indicators) (e.g., abnormal disconnection rate, handover success rate), alarm information related to communication system anomalies (e.g., hardware alarms, software alarms, KPI alarms), and traffic volume information related to the communication data volume of the communication system (e.g., the communication data volume of each wireless terminal), etc.

[0023] Second, operator O instructs information processing device 10 to place a service order (step S2). For example, operator O inputs a benchmark value such as the KPI or standard of the communication system requested by operator O as a service order.

[0024] Third, the information processing device 10 generates source code using the generation model M (step S3). For example, the information processing device 10 inputs the service order obtained from the operator O into the generation model M and obtains the source code of the RAN software output by the generation model M. At this time, the information processing device 10 generates, for example, the source code of the RAN software used to implement various processes such as downlink and uplink in the communication system.

[0025] Fourth, the information processing device 10 uses a compiler to generate software (step S4). For example, the information processing device 10 compiles the source code of the RAN software output by the generation model M to generate the RAN software.

[0026] Fifth, the information processing device 10 uses a digital twin to evaluate the software (step S5). For example, the information processing device 10 executes the generated RAN software through the communication system in the virtual space of the digital twin, obtains the usage information of the communication system in the virtual space, and compares it with the service order to perform the evaluation.

[0027] Sixth, the information processing device 10 provides feedback on the evaluation results (step S6). For example, if the evaluation results meet the service order, the information processing device 10 publishes the generated RAN software. On the other hand, if the evaluation results do not meet the service order, the information processing device 10 retrains the generated model M using training data based on the application information of the communication system in virtual space. At this time, the information processing device 10 repeatedly executes the above-described steps S3 to S5 until it can generate RAN software whose evaluation results meet the service order.

[0028] By executing steps S1 to S6 above, the information processing system 100 can shorten the lead time from RAN software development to application, completing the process in a short period. Furthermore, the information processing system 100 can reduce human error and other malfunctions that occur during the RAN software development and application process. In other words, by fully utilizing the generative model M, the information processing system 100 can automatically implement "DevOps," enabling close collaboration between RAN software development and application.

[0029] [2. Structure and processing of information processing device 10] use Figure 2 The structure and processing of the information processing device 10 will be explained. Figure 2 This is a block diagram illustrating a structural example of the information processing device 10 of the information processing system 100 according to an embodiment. The information processing device 10 comprises an input unit 11, an output unit 12, a communication unit 13, a storage unit 14, and a control unit 15.

[0030] (Input Section 11) The input unit 11 is responsible for inputting various types of information into the information processing device 10. For example, the input unit 11 is implemented by a mouse and keyboard, etc., and accepts various types of information input into the information processing device 10.

[0031] (Output Section 12) The output unit 12 is responsible for outputting various types of information from the information processing device 10. For example, the output unit 12 is implemented by a display or the like, displaying various types of information stored in the information processing device 10.

[0032] (Ministry of Communications 13) The communication unit 13 is responsible for data communication with other devices. For example, the communication unit 13 performs data communication with various communication devices via routers or the like. Additionally, the communication unit 13 can perform data communication with a terminal (not shown).

[0033] (Storage Section 14) The storage unit 14 is implemented using semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or storage devices such as hard disks and optical disks. The storage unit 14 includes a training data storage unit 14a, a code information storage unit 14b, a software storage unit 14c, and a generated model storage unit 14d. Furthermore, the storage unit 14 stores various information referenced by the control unit 15 during operation, as well as various information acquired by the control unit 15 during operation.

[0034] (Training data storage unit 14a) Training data storage unit 14a stores training data. Here, it is used... Figure 3 This describes an example of the information stored in the training data storage unit 14a. Figure 3 This is a diagram illustrating an example of the training data storage unit 14a of the information processing apparatus 10 according to an embodiment. Figure 3 In the example, the training data storage unit 14a has a "training data" item.

[0035] "Training data" refers to the training data used by the training unit 15e, which will be described later. Examples include application information related to the use of the communication system collected from the base station of the communication system, and datasets of the source code of the RAN software controlling the communication system. Application information includes, for example, evaluation metrics (KPIs) of the communication system, alarm information related to abnormal alarms of the communication system, and traffic volume information related to the communication data volume of the communication system.

[0036] exist Figure 3 The example shown is of training data stored in training data storage unit 14a as “training data #1”, “training data #2”, “training data #3”, etc.

[0037] (Code Information Storage Section 14b) The code information storage unit 14b stores code information, including source code, etc. Here, we use... Figure 4 This describes an example of the information stored in the code information storage section 14b. Figure 4 This is a diagram illustrating an example of the code information storage unit 14b of the information processing apparatus 10 according to an embodiment. Figure 4 In the example, the code information storage unit 14b has a "code information" item.

[0038] "Code information" refers to the source code generated by the code section 15b described later, such as the source code of the RAN software. Additionally, "code information" refers to information referenced by the compilation section 15c described later; for example, it may include programs, programming languages ​​such as C and Python, and algorithms used when compiling the RAN software source code.

[0039] exist Figure 4 The example shown is code information stored in code information storage unit 14b as "code information #1", "code information #2", "code information #3", ...

[0040] (Software Storage Section 14c) The software storage unit 14c stores the software generated by the compilation unit 15c, which will be described later. Here, we use... Figure 5 This describes an example of information stored in the software storage unit 14c. Figure 5 This is a diagram illustrating an example of the software storage unit 14c of the information processing apparatus 10 according to an embodiment. Figure 5In the example, the software storage unit 14c has a "software" item.

[0041] "Software" refers to software generated by the compiler 15c described later, such as RAN software.

[0042] exist Figure 5 The example shown is of software stored in software storage unit 14c as "Software #1", "Software #2", "Software #3", ...

[0043] (Generated model storage section 14d) The generated model M is stored in the model storage unit 14d. Here, we use... Figure 6 This illustrates an example of the information stored in the model storage unit 14d. Figure 6 This is a diagram illustrating an example of the generation model storage unit 14d of the information processing apparatus 10 in an embodiment. Figure 6 In the example, the generative model storage unit 14d has a "generative model" item.

[0044] "Generative model" refers to the model data of a machine learning model trained by the training section 15e described later and used by the code section 15b described later. For example, it is the execution data of the algorithm used to execute the generative model M, including model parameters, hyperparameters, etc., as set values.

[0045] exist Figure 6 The example shown is of multiple trained generative models M, namely "Generative Model #1", "Generative Model #2", ..., stored in the generative model storage section 14d.

[0046] (Control Department 15) The control unit 15 is implemented, for example, by using a CPU (Central Processing Unit) or MPU (Micro Processing Unit) as the operating area to execute various programs (equivalent to an information processing program) stored in the storage device inside the information processing apparatus 10, with RAM as the operating area. Alternatively, the control unit 15 may be implemented using an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or a FPGA (Field Programmable Gate Array).

[0047] like Figure 2 As shown, the control unit 15 includes an acquisition unit 15a, a code unit 15b, a compilation unit 15c, an evaluation unit 15d, and a training unit 15e, and performs the information processing functions and roles described below. Furthermore, the internal structure of the control unit 15 is not limited to... Figure 2 The structure shown can be any structure that performs the information processing described later, or it can be any other structure. Furthermore, the connection relationships between the various processing units within the control unit 15 are not limited to... Figure 2 The connection relationship shown can also be other connection relationships.

[0048] (Acquisition Section 15a) The acquisition unit 15a acquires benchmark values ​​related to the application of the system requested by the user. Here, the system is, for example, a communication system. For instance, the acquisition unit 15a acquires benchmark values ​​related to the application of the communication system requested by the communication system administrator, i.e., operator O, as a service order. Furthermore, as benchmark values ​​related to the application of the communication system requested by operator O, the acquisition unit 15a acquires KPIs or standards as service orders. In addition, the acquisition unit 15a can also store the acquired service orders in the storage unit 14.

[0049] (Code Section 15b) Code unit 15b uses a trained generative model M to generate software source code based on acquired benchmark values. The trained generative model M outputs the source code of the control system software based on input application information related to the system's operation. Here, the software is, for example, RAN software. For instance, code unit 15b uses a trained generative model M that outputs the source code of the RAN software controlling the communication system based on input application information related to the communication system's operation, and generates the RAN software source code based on acquired benchmark values. Furthermore, the application information includes at least one of KPIs, alarm information, and traffic volume information collected from the communication system's base station. KPIs include, for example, indicators such as abnormal disconnection rate and handover success rate in the communication system. Alarm information includes, for example, abnormal values ​​in the communication system's hardware or software, and information related to abnormal KPI values. Traffic volume information includes, for example, the amount of communication data per terminal user in the communication system's wireless terminals. Furthermore, code unit 15b stores the generated software source code in code information storage unit 14b.

[0050] (Compilation Department 15c) The compilation unit 15c compiles the source code of the generated software to generate the software. For example, the compilation unit 15c compiles the source code of the generated RAN software using an automatic compiler to generate the RAN software. Furthermore, as the RAN software, the compilation unit 15c generates software that executes various processes related to the downlink and uplink of the communication system. In addition, the compilation unit 15c stores the generated software in the software storage unit 14c.

[0051] (Assessment Department 15d) The evaluation department 15d assesses whether the control status of the system using the generated software meets the baseline values. If the control status meets the baseline values, the generated software is provided to the user. For example, the evaluation department 15d assesses whether the communication control status of the communication system using the generated RAN software meets the baseline values ​​requested by operator O. If the control status meets the baseline values, the generated RAN software is provided to operator O.

[0052] Furthermore, the evaluation unit 15d executes the generated software through a system constructed on the virtual space of the digital twin and evaluates whether the control status meets the benchmark values. For example, the evaluation unit 15d executes the generated RAN software through a communication system constructed on the virtual space of the digital twin and evaluates whether the communication control status meets the benchmark values. In addition, the evaluation unit 15d refers to the software stored in the software storage unit 14c.

[0053] (Training Department 15e) Training unit 15e trains the generative model M. For example, training unit 15e trains the generative model M using training data. At this time, training unit 15e uses a dataset of communication system operation information collected from the base station of the communication system and the source code of the RAN software controlling the communication system as training data to train the generative model M. In addition, training unit 15e refers to the training data stored in training data storage unit 14a.

[0054] Furthermore, the training unit 15e retrains the generative model M when the control status does not meet the baseline value. For example, when the communication control status of the RAN software executed by the communication system constructed in the virtual space of the digital twin does not meet the baseline value, the training unit 15e retrains the generative model M using training data. In this case, the training unit 15e can also create training data based on the application information collected from the communication system constructed in the virtual space of the digital twin, and use the created training data to retrain the generative model M.

[0055] [3. Specific examples of information processing system 100] use Figures 7-9 Specific examples of the information processing system 100 according to the embodiments will be described below. Hereinafter, specific examples 1 to 3 of the information processing system 100 according to the embodiments will be described.

[0056] (Concrete example 1) use Figure 7 A specific example 1 of the information processing system 100 of the implementation method will be described. Figure 7 This is a diagram illustrating a specific example 1 of the information processing system 100 implemented in this way. Hereinafter, as specific example 1, a processing example of the information processing system 100 as a whole will be described.

[0057] like Figure 7 As shown in (1), the operator O trains the "GenAI engine" of the information processing device 10 by inputting the analysis results of "KPI / ALM (alarm information)" and "business volume information" as training data into the "GenAI engine" of the information processing device 10, i.e., the generation model M.

[0058] like Figure 7 As shown in (2), the operator O inputs the baseline value requested by the operator O, namely "SO (service order)," into the "GenAI engine" of the information processing device 10, namely the trained generative model M, to develop "firmware / setting information" as software.

[0059] like Figure 7 As shown in (3), the information processing device 10 generates source code based on the baseline value "SO" requested by the operator O via the "GenAI engine", i.e., the trained generative model M.

[0060] like Figure 7 As shown in (4), the information processing device 10 generates software from a source code pool and an algorithm library, which include the generated source code, via an "automatic compiler".

[0061] like Figure 7 As shown in (5), the information processing device 10 evaluates the generated software via an "automatic evaluation system" based on digital twins and then releases the software.

[0062] As described above, the information processing system 100 of the implementation method can automate "DevOps" by making full use of generated AI, enabling close collaboration between the development and application of RAN software.

[0063] (Concrete example 2) use Figure 8 A specific example 2 of the information processing system 100 of the implementation method will be described. Figure 8 This is a diagram illustrating a specific example 2 of the information processing system 100 of the implementation method. Hereinafter, as specific example 2, a processing example 1 of the information processing device 10 of the information processing system 100 will be described.

[0064] like Figure 8 As shown in (1), the information processing device 10 obtains a "service order" containing KPIs or standards from the operator O. In addition, the information processing device 10 inputs the obtained "service order" into the "GenAI engine", that is, the trained generative model M.

[0065] like Figure 8As shown in (2), when generating source code or software, the information processing device 10 collaborates with the "RAN software code" containing the source code of RAN software, the "algorithm library" containing the program that executes the algorithm, and the "language DB (database)" containing programming languages ​​such as C or Python.

[0066] like Figure 8 As shown in (3), the information processing device 10 generates source code based on the "service order" obtained from the operator O via the "GenAI engine". In addition, the information processing device 10 inputs the generated source code into the "automatic compiler".

[0067] like Figure 8 As shown in (4), the information processing device 10 generates software from the generated source code via an "automatic compiler". In addition, the information processing device 10 inputs the generated software into an "automatic evaluation system".

[0068] like Figure 8 As shown in (5), the information processing device 10 evaluates the generated software via an "automatic evaluation system" based on digital twins, and releases the software if the evaluation results meet the benchmark values.

[0069] like Figure 8 As shown in (6), the information processing device 10 evaluates the generated software via an "automatic evaluation system" based on digital twins, and provides feedback on the evaluation results if the evaluation results do not meet the benchmark values.

[0070] As described above, the information processing system 100 of the implementation method not only performs simple code generation by generating AI, but also combines code generation, automatic compiler and automatic evaluation system to automatically realize "DevOps" that enables close collaboration between the development and use of RAN software.

[0071] (Concrete example 3) use Figure 9 Example 3 of the information processing system 100 of the implementation method will be described. Figure 9 This is a diagram illustrating a specific example 3 of the information processing system 100 of the implementation method. Hereinafter, as specific example 3, a processing example 2 of the information processing device 10 of the information processing system 100 will be described.

[0072] like Figure 9 As shown in (1), the information processing device 10 obtains a "service order" containing KPIs or standards from the operator O. In addition, the information processing device 10 inputs the obtained "service order" into the "GenAI engine", that is, the trained generative model M.

[0073] like Figure 9As shown in (2), when generating source code or software, the information processing device 10 collaborates with the "RAN software code" containing the source code of RAN software, the "algorithm library" containing the program that executes the algorithm, and the "language DB (database)" containing programming languages ​​such as C or Python.

[0074] like Figure 9 As shown in (3), the information processing device 10 generates source code based on the "service order" obtained from the operator O via the "GenAI engine". In addition, the information processing device 10 inputs the generated source code into the "automatic compiler".

[0075] like Figure 9 As shown in (4), the information processing device 10 generates software from the generated source code via an "automatic compiler". In addition, the information processing device 10 inputs the generated software into an "automatic evaluation system".

[0076] like Figure 9 As shown in (5), the information processing device 10 evaluates the generated software via an "automatic evaluation system" based on digital twins, and releases the software if the evaluation results meet the benchmark values.

[0077] like Figure 9 As shown in (6), the information processing device 10 evaluates the generated software via an "automatic evaluation system" based on digital twins, and provides feedback on the evaluation results if the evaluation results do not meet the benchmark values.

[0078] like Figure 9 As shown in (7), when evaluating the generated software, the information processing device 10 compares the evaluation results with the SO standard, which serves as a benchmark value.

[0079] like Figure 9 As shown in (8), if the evaluation results meet the SO standard, the information processing device 10 outputs a notification.

[0080] like Figure 9 As shown in (9), if the evaluation result does not meet the SO standard, the information processing device 10 performs changes to the source code, changes to the parameters of the generated model M, etc.

[0081] As described above, the information processing system 100 of the implementation method can automate "DevOps" by cyclically releasing and processing until software that satisfies the service order is generated, thus enabling close collaboration between the development and use of RAN software.

[0082] [4. Processing flow of information processing system 100] use Figure 10 The processing flow of the information processing system 100 in the implementation method will be described. Figure 10This is a flowchart illustrating an example of the processing flow of the information processing system 100 in this embodiment. Furthermore, the processes described below, S101 to S106, can be executed in different orders. Additionally, some steps in S101 to S106 may be omitted.

[0083] (Generative model training process) First, the information processing device 10 performs generative model training processing (step S101). For example, the information processing device 10 trains the generative model M by inputting training data based on usage information (e.g., KPI, alarm information, traffic volume information) collected from the base station of the communication system into the generative model M.

[0084] (Service order instruction processing) Second, operator O executes the service order instruction processing (step S102). For example, operator O inputs the KPI or standard benchmark value of the communication system requested by operator O as a service order.

[0085] (Source code generation process) Third, the information processing device 10 performs source code generation processing (step S103). For example, the information processing device 10 inputs the service order obtained from the operator O into the generation model M, and obtains the source code of the RAN software output by the generation model M.

[0086] (Software generation and processing) Fourth, the information processing device 10 performs software generation processing (step S104). For example, the information processing device 10 compiles the source code of the RAN software output by the generation model M using an automatic compiler to generate the RAN software.

[0087] (Software evaluation and processing) Fifth, the information processing device 10 performs software evaluation processing (step S105). For example, by executing the generated RAN software through the communication system in the virtual space of the digital twin, the device obtains the usage information of the communication system in the virtual space and compares it with the service order, thereby performing an evaluation.

[0088] (Assessment Result Feedback Processing) Sixth, the information processing device 10 performs evaluation result feedback processing (step S106). For example, if the evaluation result meets the service order, the information processing device 10 publishes the generated RAN software and ends the processing. On the other hand, if the evaluation result does not meet the service order, the information processing device 10 retrains the generator model M using training data based on the application information of the communication system in virtual space, and repeatedly executes the above-described steps S103 to S105 until RAN software that meets the service order evaluation result can be generated.

[0089] [5. Effects of the Implementation Method] The effects of the implementation method will be explained. Hereinafter, the effects 1 to 5 corresponding to each process in the implementation method will be explained.

[0090] (Effect 1) First, in the process of the above-described embodiment, the information processing device 10 acquires reference values ​​related to the use of the system requested by the operator O, uses a generative model M trained on and outputting the source code of the software controlling the system based on the input of the application information related to the system's use, generates source code based on the acquired reference values, compiles the generated source code, generates software, evaluates whether the control status of the system using the generated software meets the reference values, and provides the generated software to the operator O if the control status meets the reference values. Therefore, in this process, by making full use of the generative model M, the process from software development to application can be shortened.

[0091] (Effect 2) Second, in the above-described implementation, if the control conditions do not meet the baseline value, the information processing device 10 retrains the generative model M. Therefore, in this process, by fully utilizing the generative model M which has received feedback on the evaluation results, the process from software development to application can be shortened.

[0092] (Effect 3) Third, in the above-described implementation, the information processing device 10 executes the generated software through a system constructed on the virtual space of the digital twin to evaluate whether the control status meets the benchmark value. Therefore, in this process, it is not necessary to evaluate the generated software in a real environment, thus shortening the process from software development to application.

[0093] (Effect 4) Fourth, in the above-described implementation method, the system is a communication system, and the application information includes at least one of KPIs, alarm information, and traffic volume information collected from the base station of the communication system. Therefore, in this process, by fully utilizing the generative model M, the process from software development to application in the communication system can be shortened.

[0094] (Effect 5) Fifth, in the above-described implementation method, the software is RAN software. Therefore, in this method, by fully utilizing the generation model M, the process from the development to the application of RAN software in the communication system can be shortened.

[0095] [6. Hardware Structure] The information processing device 10 of the above-described embodiments is, for example, composed of... Figure 11The computer 1000 with the structure shown is implemented. Hereinafter, the information processing device 10 will be described by example. Figure 11 This is a hardware structure diagram illustrating an example of a computer that implements the functions of the information processing apparatus 10 of the embodiment. The computer 1000 includes a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and storage medium interface (I / F) 1700.

[0096] CPU 1100 operates based on programs stored in ROM 1300 or HDD 1400, controlling various components. ROM 1300 stores the boot program executed by CPU 1100 when computer 1000 starts up, programs dependent on the hardware of computer 1000, etc.

[0097] HDD1400 stores programs executed by CPU1100 and data used by those programs. Communication interface 1500 receives data from other devices and sends it to CPU1100 via a specified communication network, and sends data generated by CPU1100 to other devices via the specified communication network.

[0098] CPU 1100 controls output devices such as monitors and printers, and input devices such as keyboards and mice, via input / output interface 1600. CPU 1100 acquires data from input devices via input / output interface 1600. Additionally, CPU 1100 outputs generated data to output devices via input / output interface 1600.

[0099] The storage medium interface 1700 reads the program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 into the RAM 1200 via the storage medium interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as DVD (Digital Versatile Disc), PD (Phase Change Rewritable Disk), an optical-magnetic recording medium such as MO (Magneto-Optical Disk), magnetic tape, magnetic recording media, or semiconductor memory.

[0100] For example, when the computer 1000 functions as the information processing device 10 in this embodiment, the CPU 1100 of the computer 1000 performs the functions of the control unit 15 by executing programs loaded into the RAM 1200. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may also be obtained from other devices via a defined communication network.

[0101] [7. Other] In the processes described in the above embodiments, all or part of the processes described as automatically performed can be performed manually, or all or part of the processes described as manually performed can be performed automatically using known methods. Furthermore, the processing procedures, specific names, and information containing various data or parameters shown in the above documents and figures can be arbitrarily changed, except where specifically stated. For example, the various information shown in the figures is not limited to the information illustrated.

[0102] Furthermore, the constituent elements of the devices illustrated are functional concepts and do not necessarily need to be physically configured as shown in the illustrations. That is, the specific way in which the devices are distributed / integrated is not limited to the way illustrated, and they can be distributed / integrated functionally or physically in any unit according to various loads or usage conditions.

[0103] Furthermore, the above-described embodiments can be appropriately combined without causing contradictions in the processing content.

[0104] The above description of some embodiments of this application is based on the accompanying drawings. However, these are examples and can be represented by the manner described in the disclosure of the invention. Various modifications and alterations can be implemented in other ways based on the knowledge of those skilled in the art.

[0105] Furthermore, the term "section, module, unit" mentioned above can be replaced with "unit" or "circuit," etc. For example, the receiving section can be replaced with a receiving unit or a receiving circuit.

[0106] Explanation of reference numerals in the attached figures 10 Information Processing Devices 11 Input Section 12 Output Section 13Ministry of Communications 14 Storage Department 14a Training Data Storage Department 14b Code Information Storage Department 14c Software Storage Unit 14D Generative Model Storage 15 Control Department 15a Acquisition Department 15b Code Section 15c Compilation Department 15d Assessment Department 15e Training Department 100 Information Processing System

Claims

1. An information processing device, comprising: The acquisition unit acquires baseline values ​​related to the user's request for system usage. The code department uses a trained generative model to generate source code based on the obtained benchmark values. The trained generative model outputs source code of the software controlling the system based on input application information related to the use of the system. The compilation department compiles the generated source code to produce the software; as well as The evaluation department evaluates whether the control status of the system using the generated software meets the benchmark value, and if the control status meets the benchmark value, provides the generated software to the user.

2. The information processing apparatus according to claim 1, wherein, The information processing device further includes a training unit, which retrains the generative model when the control condition does not meet the baseline value.

3. The information processing apparatus according to claim 1, wherein, The evaluation unit executes the generated software through the system constructed on the virtual space of the digital twin to evaluate whether the control status meets the benchmark value.

4. The information processing apparatus according to any one of claims 1 to 3, wherein, The system is a communication system. The application information includes at least one of the key performance indicators (KPIs), alarm information, and traffic volume information collected from the base stations of the communication system.

5. The information processing apparatus according to any one of claims 1 to 3, wherein, The software in question is Radio Access Network (RAN) software.

6. An information processing method, executed by an information processing device, the method comprising: The acquisition process involves obtaining baseline values ​​related to the user's request for system usage. The coding process involves using a trained generative model to generate the source code based on the baseline value. The trained generative model outputs the source code of the software controlling the system based on the input of application information related to the application of the system. The compilation process involves compiling the generated source code to produce the software. as well as The evaluation process assesses whether the control status of the system using the generated software meets the benchmark value. If the control status meets the benchmark value, the generated software is provided to the user.

7. An information processing program that causes a computer to execute: The acquisition process involves obtaining baseline values ​​related to the user's request for system usage. In the coding process, the trained generative model is used to generate the source code based on the benchmark value. The trained generative model outputs the source code of the software controlling the system based on the input of application information related to the application of the system. The compilation process involves compiling the generated source code to produce the software. as well as The evaluation process assesses whether the control status of the system using the generated software meets the benchmark value. If the control status meets the benchmark value, the generated software is provided to the user.

Images