System, method, and program

Abstract

A first system (10) comprises: an acquisition unit (11) that acquires wireless quality information from a wireless network such as a radio access network (RAN); and a generation unit (12) that generates on the basis of the wireless quality information acquired by the acquisition unit (11), a radio wave fluctuation model for estimating fluctuations in radio wave quality in accordance with the wireless quality information, and generates an error rate model for estimating an error rate such as a block error rate for each modulation and coding scheme (MCS) in accordance with the radio wave quality.

Description

SYSTEM, APPARATUS, METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM

[0001] The present disclosure relates to systems, devices, methods, programs, and non-transitory computer-readable media.

[0002] In wireless network systems such as 5G (5th Generation) and LTE (Long Term Evolution), a base station or a terminal device measures the wireless quality of a received reference signal and selects an MCS (Modulation and Coding Schemes) index to be used in wireless communication based on the measurement result. The MCS index is an index value indicating a combination of a coding scheme and a modulation scheme. Note that selecting (determining) an MCS index is sometimes referred to as selecting (determining) an MCS. For example, when a terminal device measures wireless quality, as described in Non-Patent Document 1, the terminal device reports a CQI (Channel Quality Indicator) index indicating the wireless quality of the measured reference signal to the base station via a CSI (Channel State Information) report. Patent Document 1 also describes using an error rate table including a channel state value dimension, a transmission parameter dimension, and an error rate element to select a TBS (Transport Block Size) that results in an appropriate error rate according to the CQI.

[0003] OLLA (Outer Loop Link Adaptation) is a known technique related to optimizing the MCS index. In OLLA, the MCS index is first roughly adjusted based on measurement results using a reference signal, and then gradually increased. If a block error occurs, the MCS index is decreased, and loop control is repeated to achieve a target block error rate (BLER) on average (e.g., 10%). A block error refers to a situation in which a transmitted transport block cannot be received correctly at the receiving end.

[0004] Special Publication No. 2012-509647

[0005] ShareTechnote, "5G / NR - CSI Report", [online], [Retrieved December 7, 2022], Internet<URL:https: / / www.sharetechnote.com / html / 5G / 5G_CSI_Report.html>

[0006] In OLLA, in order to determine an appropriate MCS (MCS Index) that satisfies a target BLER, trial-and-error loop control is repeated while detecting block errors, as described above. Since related techniques are premised on detecting block errors, it may be difficult to determine an appropriate MCS depending on the target BLER.

[0007] In view of the above, one object of the present disclosure is to provide a system, an apparatus, a method, a program, and a non-transitory computer-readable medium that enable determination of an appropriate MCS.

[0008] The system according to the present disclosure includes an acquisition means for acquiring wireless quality information from a wireless network, and a generation means for generating a radio wave fluctuation model based on the acquired wireless quality information, which estimates fluctuations in radio wave quality according to the wireless quality information, and for generating an error rate model for estimating an error rate for each modulation and coding method according to the radio wave quality.

[0009] Another system according to the present disclosure includes an acquisition means for acquiring radio quality information from a wireless network, and an estimation means for estimating an error rate for each modulation and coding scheme based on the acquired radio quality information, using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the radio quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality.

[0010] The device according to the present disclosure includes an acquisition means for acquiring wireless quality information from a wireless network, and a generation means for generating a radio wave fluctuation model based on the acquired wireless quality information, which estimates fluctuations in radio wave quality according to the wireless quality information, and for generating an error rate model for estimating an error rate for each modulation and coding method according to the radio wave quality.

[0011] Another device according to the present disclosure includes an acquisition means for acquiring radio quality information from a wireless network, and an estimation means for estimating an error rate for each modulation and coding scheme based on the acquired radio quality information, using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the radio quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality.

[0012] The method disclosed herein acquires wireless quality information from a wireless network, and based on the acquired wireless quality information, generates a radio wave fluctuation model that estimates fluctuations in radio wave quality according to the wireless quality information, and also generates an error rate model that estimates an error rate for each modulation and coding scheme according to the radio wave quality.

[0013] Another method according to the present disclosure involves acquiring radio quality information from a wireless network, and using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the radio quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality, estimating an error rate for each modulation and coding scheme based on the acquired radio quality information.

[0014] The non-transitory computer-readable medium according to the present disclosure is a non-transitory computer-readable medium storing a program for causing a computer to execute a process of acquiring radio quality information from a wireless network, generating a radio wave fluctuation model that estimates fluctuations in radio wave quality according to the acquired radio quality information, and generating an error rate model that estimates an error rate for each modulation and coding scheme according to the radio wave quality.

[0015] Another non-transitory computer-readable medium according to the present disclosure is a non-transitory computer-readable medium storing a program for causing a computer to execute a process of acquiring radio quality information from a wireless network, and estimating an error rate for each modulation and coding scheme based on the acquired radio quality information using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the radio quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality.

[0016] According to the present disclosure, it is possible to provide a system, an apparatus, a method, a program, and a non-transitory computer-readable medium that enable determination of an appropriate MCS.

[0017] 1 is a configuration diagram showing an overview of a first system according to an embodiment. 2 is a configuration diagram showing an overview of a second system according to an embodiment. 3 is a configuration diagram showing an overview of a system according to an embodiment. 4 is a configuration diagram showing an overview of a first device according to an embodiment. 5 is a configuration diagram showing an overview of a second device according to an embodiment. 6 is a configuration diagram showing an overview of a device according to an embodiment. 7 is a flowchart showing an overview of a first method according to an embodiment. 8 is a flowchart showing an overview of a second method according to an embodiment. 9 is a configuration diagram showing an example of a configuration of a RAN system according to embodiment 1. 10 is a configuration diagram showing an example of a configuration of each device of the RAN system according to embodiment 1. 11 is a diagram explaining an example of a radio wave fluctuation model according to embodiment 1. 12 is a diagram explaining an example of a transmission error model according to embodiment 1. 13 is a configuration diagram showing another example of a configuration of each device of the RAN system according to embodiment 1. 14 is a diagram explaining an overview of a method of generating a radio wave fluctuation model and a transmission error model according to embodiment 1. 15 is a diagram explaining another example of a radio wave fluctuation model according to embodiment 1. 16 is a diagram explaining another example of a transmission error model according to embodiment 1. 17 is a flowchart showing an operation example of a learning phase process in the RAN system according to embodiment 1. 18 is a diagram explaining a learning process of a radio wave fluctuation model according to embodiment 1. 19 is a diagram explaining a learning process of a transmission error model according to embodiment 1. FIG. 1 is a diagram illustrating a specific example of the learning process of the transmission error model according to embodiment 1. FIG. 2 is a diagram illustrating a specific example of the learning process of the transmission error model according to embodiment 1. FIG. 3 is a flowchart illustrating an example of operation of the inference phase process in the RAN system according to embodiment 1. FIG. 4 is a diagram illustrating the inference process of the radio wave fluctuation model and the transmission error model according to embodiment 1. FIG. 5 is a diagram illustrating another example of the inference process of the radio wave fluctuation model and the transmission error model according to embodiment 1. FIG. 6 is a configuration diagram illustrating an example of the configuration of each device in the RAN system according to embodiment 2. FIG. 7 is a flowchart illustrating an example of operation of the learning phase process in the RAN system according to embodiment 2. FIG. 8 is a diagram illustrating the learning process of the radio wave fluctuation learning model according to embodiment 2. FIG. 9 is a diagram illustrating the learning process of the transmission error learning model according to embodiment 2. FIG. 10 is a flowchart illustrating an example of operation of the inference phase process in the RAN system according to embodiment 2.Fig. 10 is a diagram for explaining the inference processing of a radio wave fluctuation learning model and a transmission error learning model according to embodiment 2. Fig. 11 is a configuration diagram showing another configuration example of Near-RT RIC according to embodiment 2. Fig. 12 is a configuration diagram showing a configuration example of each device of a RAN system according to embodiment 3. Fig. 13 is a configuration diagram showing an example of the configuration of each device of a RAN system according to embodiment 4. Fig. 14 is a configuration diagram showing an overview of the hardware of a computer according to the embodiments.

[0018] Hereinafter, embodiments will be described with reference to the drawings. In each drawing, the same elements are given the same reference numerals, and duplicate explanations will be omitted as necessary. Note that the arrows shown in each drawing are examples for explanatory purposes and do not limit the type or direction of data.

[0019] For example, high reliability and low latency are required in industrial network systems used in factories and manufacturing sites. Because transmission latency is significantly affected by delays due to retransmissions, it is desirable to suppress the occurrence of block errors and prevent retransmissions.

[0020] The inventors have studied methods for selecting an appropriate MCS when a low BLER is the goal for high reliability and low latency, and have found that with OLLA, a related technology, it is difficult to select an appropriate MCS that satisfies a low BLER.

[0021] Specifically, when targeting low-latency and highly reliable communication such as in an industrial network system, the target BLER is, for example, 10 －５ In this case, the trial and error loop in OLLA takes a long time, and the true value of BLER fluctuates during that time. Since BLER is determined by multiple factors such as fluctuations in radio wave quality and data size, OLLA adjusts the BLER to the target value by running a trial and error loop. For example, if the target BLER is set to 10 －５In this case, even if a terminal device occupies all of the radio resources, a block error will occur only once every minute to several minutes. Normally, a terminal device occupies only a portion of the radio resources, so a block error will occur only once every several tens of minutes, and when a block error occurs, the radio wave quality may have fluctuated significantly. For this reason, with OLLA, it is difficult to select an appropriate MCS when, for example, the target BLER is low and the radio wave quality fluctuates.

[0022] Therefore, in the embodiment, even when the target BLER is low and the radio wave quality fluctuates, it is possible to select an appropriate MCS.

[0023] (Outline of the embodiment) First, an outline of the embodiment will be described. FIG. 1 shows a schematic configuration of a first system 10 according to the embodiment, and FIG. 2 shows a schematic configuration of a second system 20 according to the embodiment. For example, the first system 10 and the second system 20 constitute a system that controls a wireless network such as a Radio Access Network (RAN). The first system 10 and the second system 20 may be RAN Intelligent Controllers (RICs) that perform intelligent control in an Open RAN (O-RAN), which opens up the RAN. For example, the first system 10 includes a Non-Real Time (Non-RT) RIC, and the second system 20 includes a Near-RT RIC, but are not limited thereto.

[0024] As shown in Fig. 1, the first system 10 includes an acquisition unit 11 and a generation unit 12. The acquisition unit 11 acquires radio quality information from a wireless network. For example, the acquisition unit 11 acquires radio wave quality information such as wideband CQI from a RAN including an O-RAN distributed unit (O-DU) and an O-RAN central unit (O-CU). The acquisition unit 11 may also acquire error rate information such as BLER for each MCS from the RAN.

[0025] The generation unit 12 generates an estimation model for estimating an error rate for each MCS based on the radio quality information acquired by the acquisition unit 11. The error rate for each MCS is an error rate associated with each MCS index, and is the error rate when data encoded and modulated using the encoding method and modulation method specified by the MCS index is transmitted. The error rate is, for example, a BLER. The estimation model may be located in the first system 10 or in an external system. The estimation model is a model for estimating an error rate for each MCS for the next time based on acquired past radio quality information. The next time may be any time after the acquired radio quality information. For example, the estimation model includes a radio wave fluctuation model and an error rate model. That is, the generation unit 12 generates a radio wave fluctuation model and an error rate model based on the radio quality information.

[0026] The radio wave fluctuation model is a model that estimates fluctuations in radio wave quality according to radio quality information. For example, the radio wave fluctuation model may be a calculation model that calculates fluctuations in radio wave quality using a predetermined calculation method. In this case, the generation unit 12 may set parameters that the radio wave fluctuation model uses to calculate fluctuations in radio wave quality based on the acquired radio quality information. The radio wave fluctuation model may also be a learning model that predicts fluctuations in radio quality according to the radio quality information. In this case, the generation unit 12 may use the acquired radio quality information to train the radio wave fluctuation model to learn fluctuations in radio wave quality according to the radio quality information.

[0027] The error rate model is a model that estimates an error rate for each MCS according to the radio wave quality whose fluctuations are estimated by the radio wave fluctuation model. For example, the error rate model may be a calculation model that calculates the error rate for each MCS using a predetermined calculation method. In this case, the generation unit 12 may set parameters that the error rate model uses to calculate the error rate for each MCS based on the estimation results of the generated radio wave fluctuation model and the acquired error rate information for each MCS. The error rate model may also be a learning model that predicts the error rate for each MCS according to the radio wave quality. In this case, the generation unit 12 may train the error rate model to learn the error rate for each MCS according to the radio wave quality using the estimation results of the generated radio wave fluctuation model and the acquired error rate information for each MCS.

[0028] 2, the second system 20 includes an acquisition unit 21 and an estimation unit 22. The acquisition unit 21 acquires wireless quality information from a wireless network. Similar to the acquisition unit 11 of the first system 10, the acquisition unit 21 acquires time-series data such as wideband CQI from the RAN.

[0029] The estimation unit 22 uses an estimation model to estimate an error rate for each MCS according to the radio quality information acquired by the acquisition unit 21. The estimation model may be located in the second system 20 or in an external system. The estimation unit 22 uses the estimation model to estimate an error rate for each MCS for the next time period from the acquired past radio quality information. For example, the estimation model is a model generated by the generation unit 12 of the first system and includes a radio wave fluctuation model and an error rate model, as described above. That is, the estimation unit 22 estimates an error rate for each MCS based on the radio quality information acquired by the acquisition unit 21, using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the radio quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality. The second system 20 may further include a determination unit that determines an MCS to be set in the wireless network based on the error rate for each MCS estimated by the estimation unit 22.

[0030] The units of the first system 10 and the second system 20 may constitute one system, or may constitute multiple systems. Fig. 3 shows an example of a system configuration according to an embodiment. For example, as shown in Fig. 3, a system 30 may include the units of the first system 10 and the second system 20. In this case, the acquisition unit 11 and the acquisition unit 21 may be combined into one acquisition unit. Furthermore, the units of the system 30 may be arranged in separate systems.

[0031] Furthermore, the first system 10 and the second system 20 may each be configured by one device or multiple devices. FIG. 4 shows an example configuration of a first device 40 according to an embodiment, and FIG. 5 shows an example configuration of a second device 50 according to an embodiment. As shown in FIG. 4, the first device 40 may include the acquisition unit 11 and the generation unit 12 shown in FIG. 1. As shown in FIG. 5, the second device 50 may include the acquisition unit 21 and the estimation unit 22 shown in FIG. 2. As with the first system 10 and the second system 20, for example, the first device 40 may be a non-RT RIC, and the second device 50 may be a near-RT RIC.

[0032] Similar to the configuration of each unit in the first system 10 and the second system 20, each unit in the first device 40 and the second device 50 may constitute a single device, or may constitute multiple devices. FIG. 6 shows an example of the configuration of a device according to an embodiment. For example, as shown in FIG. 6, a device 60 may include each unit of the first device 40 and the second device 50. In this case, the acquisition unit 11 and the acquisition unit 21 may be combined into a single acquisition unit. Alternatively, each unit of the device 60 may be arranged in a separate device.

[0033] Furthermore, part or all of the first system 10 and the second system 20 may be placed on an edge or in a cloud using virtualization technology or the like. Each unit may be placed in a specific location, or may be distributed across multiple locations. An edge is a location or platform on the base station side, including an O-DU and an O-CU. A cloud is a location or platform on the core network side, away from the base station. For example, the acquisition unit 11 and the generation unit 12 may be placed in the cloud, and the acquisition unit 21 and the estimation unit 22 may be placed on an edge. Furthermore, the acquisition unit 11, the generation unit 12, the acquisition unit 21, and the estimation unit 22 may each be distributed.

[0034] Fig. 7 shows a first method according to an embodiment, and Fig. 8 shows a second method according to an embodiment. For example, the first method is performed by the first system 10 of Fig. 1 or the first device 40 of Fig. 4. The second method is performed by the second system 20 of Fig. 2 or the second device 50 of Fig. 5.

[0035] As shown in FIG. 7 , the acquisition unit 11 acquires wireless quality information from a wireless network (S11). Next, the generation unit 12 generates an estimation model that estimates an error rate for each MCS based on the acquired wireless quality information (S12). For example, the generation unit 12 generates a radio wave fluctuation model and an error rate model using the acquired wireless quality information. The generation unit 12 may generate a radio wave fluctuation model using the acquired wireless quality information, and then generate an error rate model using an estimation result of the generated radio wave fluctuation model. The generation unit 12 may apply the generated estimation model including the radio wave fluctuation model and the error rate model to the second system 20.

[0036] As shown in FIG. 8 , the acquisition unit 21 acquires radio quality information from a wireless network (S21). Next, the estimation unit 22 uses an estimation model to estimate an error rate for each MCS according to the acquired radio quality information (S22). The estimation unit 22 estimates the error rate for each MCS using the radio wave fluctuation model and the error rate model generated by the generation unit 12. For example, the estimation unit 22 may use the generated radio wave fluctuation model to estimate fluctuations in radio wave quality according to the acquired radio quality information, and may use the generated error rate model to estimate the error rate for each MCS according to the radio wave quality whose fluctuations are estimated by the radio wave fluctuation model. Furthermore, the second system 20 may determine the MCS to be set in the RAN based on the estimated error rate for each MCS.

[0037] As described above, in the embodiment, an estimation model for estimating an error rate for each MCS is generated based on radio quality information acquired from a wireless network, and the error rate for each MCS is estimated using the generated estimation model. The generated estimation model includes a radio wave fluctuation model for estimating fluctuations in radio wave quality and an error rate model for estimating an error rate for each MCS. This allows for accurate estimation of the BLER for each MCS based on the radio quality information. For example, even if the radio wave quality fluctuates and the target BLER is low, the BLER for each MCS can be estimated. This makes it possible to select an appropriate MCS based on the estimated BLER for each MCS.

[0038] (First Embodiment) Next, a first embodiment will be described. In this embodiment, an example will be described in which an estimation model is generated by determining parameters of a predetermined calculation model based on wireless quality information collected from a RAN.

[0039] 9 shows an example of the configuration of a RAN system 1 according to this embodiment. As shown in FIG. 9, the RAN system 1 includes a Near-RT RIC 100, a Non-RT RIC 200, and an E2 node 300.

[0040] The Non-RT RIC 200 and the Near-RT RIC 100, and the Non-RT RIC 200 and the E2 node 300 are connected via the O1 interface so that they can communicate with each other. The O1 interface is an interface that mainly transmits and receives data and messages required for operation and management. Note that the interface is a connection interface defined by a communication protocol used to transmit and receive data and messages, and includes logical transmission paths and networks, as well as physical transmission paths and networks.

[0041] The Non-RT RIC 200 and the Near-RT RIC 100 are communicatively connected via an A1 interface. The Near-RT RIC 100 and the E2 node 300 are connected via an E2 interface. The A1 interface and the E2 interface are interfaces that mainly send and receive data and messages required for control purposes.

[0042] The E2 node 300 is a node that constitutes the RAN and includes an O-DU and an O-CU. Note that either the O-DU or the O-CU, or both, may be referred to as the E2 node 300. The RAN is a wireless network accessed by a UE (User Equipment) and is connected to a core network such as a 5G Core network (5GC) or an Evolved Packet Core (EPC). The RAN may include an O-RU (O-RAN Remote Unit) that constitutes an antenna. The UE is a terminal device that connects to the RAN and performs wireless communication, and may be a mobile phone, smartphone, tablet terminal, IoT (Internet of Things) terminal, or the like. The UE may also be an application device such as a robot, drone, or self-driving car that implements terminal functions.

[0043] The E2 node 300 including the O-DU and O-CU provides base station functionality. The base station is, for example, a gNB (next generation Node B) or an eNB (evolved Node B), but is not limited to these. The O-DU and O-CU are examples of nodes that provide base station functionality, and may be other network nodes.

[0044] The O-DU is a logical node that provides the radio signal control function and Layer 2 control function of the base station. The O-DU accommodates the O-RU, controls the antenna radio signal (beam) in the O-RU, and performs protocol processing such as MAC (Media Access Control) and RLC (Radio Link Control) required between the O-RU and O-CU.

[0045] The O-CU is a logical node that provides the base station's radio resource control function and data processing function higher than Layer 2. The O-CU accommodates the O-DU and performs data transmission and reception via the accommodated O-DU, QoS (Quality of Service) control, cell / UE management, handover control, and protocol processing such as PDCP (Packet Data Convergence Protocol), SDAP (Service Data Adaptation Protocol), and RRC (Radio Resource Control) required between the O-DU and the core network.

[0046] The E2 node 300 may include any number of O-DUs and O-CUs, one or more. In other words, it may include multiple base stations. The number of O-DUs and O-CUs does not necessarily have to be the same. The O-DUs and O-CUs may be located in different locations or in the same location. Furthermore, the O-DUs and O-CUs may be implemented by different virtual machines running on an edge virtualization platform, or by the same virtual machine. The O-DUs and O-CUs may be virtualized distributed units (vDUs) and virtualized central units (vCUs), and may constitute a virtual base station. The O-DUs and O-CUs may also be physical DUs and CUs. Furthermore, the E2 node 300 may be a base station device including the functions of the O-DUs and O-CUs.

[0047] The Near-RT RIC 100 is a logic function that controls and optimizes the RAN in near real time. The Near-RT RIC 100 controls the RAN in a short control period, for example, of 10 ms (milliseconds; the same applies hereinafter) or more and less than 1 s (seconds; the same applies hereinafter). The Near-RT RIC 100 collects and analyzes radio information from the E2 node 300, which includes either or both of the O-DU and O-CU, via the E2 interface, and controls the E2 node 300 according to the radio information. The Near-RT RIC 100 includes a machine learning model, which is a trained model, and may analyze the radio information and determine RAN control using the machine learning model. It is not limited to a machine learning model, and other models having similar functions may also be used.

[0048] For example, the Near-RT RIC 100 performs control according to radio information in accordance with a control policy acquired from the Non-RT RIC 200 via the A1 interface. The control policy is a policy related to RAN control, such as the A1 policy. The A1 policy is guidance used for RAN optimization defined in the A1 interface. The Near-RT RIC 100 is located in the same location as either or both of the O-DU and O-CU, or in a location close to either or both of the O-DU and O-CU. For example, the Near-RT RIC 100 may be implemented in the same edge virtual machine as either or both of the O-DU and O-CU.

[0049] The Non-RT RIC 200 is a logic function that controls and optimizes the RAN in non-real time. The Non-RT RIC 200 controls the RAN in a long control period of, for example, 1 second or more. The Non-RT RIC 200 manages control policies, manages the operations of the E2 node 300 and the Near-RT RIC 100, and performs learning (generation) and updating of machine learning models and other models.

[0050] For example, the Non-RT RIC 200 generates a control policy and notifies the Near-RT RIC 100 of the generated control policy via the A1 interface. The Non-RT RIC 200 also manages and sets the configuration information (Configuration) of the E2 node 300 based on data acquired from the E2 node 300 and the Near-RT RIC 100 via the O1 interface. The Non-RT RIC 200 is located in an SMO (Service Management and Orchestration) that manages and orchestrates the RAN. The SMO is located in a location away from the E2 node 300 and the Near-RT RIC 100, for example, on the cloud. The Non-RT RIC 200 may include the functionality of the SMO.

[0051] 10 shows an example of the configuration of the Near-RT RIC 100 and the E2 node 300 in the RAN system 1 according to this embodiment. As shown in FIG. 10, the E2 node 300 includes a radio information acquisition unit 310, a radio information transmission unit 320, a control information reception unit 330, and a RAN control unit 340.

[0052] The radio information acquisition unit 310 acquires radio information of the RAN. In response to an instruction from the Near-RT RIC 100, the radio information acquisition unit 310 acquires information stored in the O-DU or O-CU, and radio information from the UE or O-RU. The radio information acquisition unit 310 acquires, as the radio information, for example, radio quality information collected from the UE and radio quality information measured by the E2 node 300. The radio information is not limited to radio quality information, and may include other information necessary for controlling the RAN.

[0053] For example, the radio quality information includes radio wave quality such as wideband CQI. The wideband CQI is a CQI value measured by a UE and notified to the E2 node 300 via a CSI report, or a CQI value measured by the E2 node 300. For example, a base station including the E2 node 300 transmits a reference signal (CSI-RS: Channel State Information - Reference Signal) for measuring downlink channel quality to the UE, and the UE measures the quality of the received reference signal and notifies the base station of the measured CQI value via a CSI report. In addition, the UE transmits a reference signal (SRS: Sounding Reference Signal) for measuring uplink channel quality to the base station including the E2 node 300, and the base station measures the quality of the received reference signal and obtains a CQI value from the measurement result.

[0054] The radio wave quality may be subband CQI, SINR (Signal to Interference plus Noise power Ratio), RSRP (Reference Signal Received Power), RSRQ (Reference Signal Received Quality), RSSI (Received Signal Strength Indicator), etc.

[0055] The radio quality information also includes a BLER (transmission success rate) for each MCS. The BLER for each MCS is the number of block errors for each MCS measured by the UE and reported to the E2 node 300 over a predetermined period, or the number of block errors for each MCS or MIMO (Multiple-Input and Multiple-Output) multiplicity measured by the E2 node 300 over a predetermined period. For example, a base station including the E2 node 300 generates a transport block for downlink BLER measurement using a coding scheme and modulation scheme for each MCS index and transmits the generated transport block for each MCS index to the UE. The UE measures block errors in the received transport block for each MCS index and reports the measurement results to the base station. The UE also generates a transport block for uplink BLER measurement using a coding scheme and modulation scheme for each MCS index and transmits the generated transport block for each MCS index to the base station including the E2 node 300. The base station measures the block error of the transport block for each received MCS index.

[0056] The wireless information transmission unit 320 transmits wireless information including wireless quality information acquired by the wireless information acquisition unit 310 to the Near-RT RIC 100 via the E2 interface. For example, the wireless information transmission unit 320 transmits wireless information in response to an instruction from the Near-RT RIC 100.

[0057] The control information receiving unit 330 receives control information from the Near-RT RIC 100 via the E2 interface. The control information is radio control information for controlling the RAN according to radio information, such as an MCS index for each UE. The control information may also include radio resource allocation priority, parameters for handover control and beam control, etc.

[0058] The RAN control unit 340 controls the RAN based on the received control information. For example, the RAN control unit 340 sets an MCS index for each UE included in the received control information in the MCS control unit in the O-DU or O-CU. In addition, the RAN control unit 340 may set a radio resource allocation priority for each UE included in the received control information in the radio resource control unit in the O-DU or O-CU.

[0059] As shown in FIG. 10 , the Near-RT RIC 100 includes a radio information acquisition unit 110, a radio information recording unit 120, a model storage unit 130, a model calculation unit 140, a BLER estimation unit 150, an MCS determination unit 160, and a control information transmission unit 170.

[0060] The radio information acquisition unit 110 receives radio information including radio quality information from the E2 node 300 including either or both of an O-DU and an O-CU via the E2 interface. The radio information acquisition unit 110 also serves as a radio quality information acquisition unit that acquires radio quality information. During a learning phase process (during a process for generating an estimation model), the radio information acquisition unit 110 collects radio quality information from the E2 node 300 as learning data that the model calculation unit 140 uses to calculate parameters of the estimation model. Furthermore, during an inference phase process (during a process for estimating an estimation model), the radio information acquisition unit 110 collects radio quality information from the E2 node 300 as inference data that the BLER estimation unit 150 uses to estimate a BLER using the estimation model. For example, the radio information acquisition unit 110 may instruct the E2 node 300 on the data to be collected and the period of collection.

[0061] The wireless information recording unit 120 is a database that records (stores) wireless information received from the E2 node 300. The wireless information recording unit 120 accumulates wireless information including wireless quality information as time-series data. Note that the wireless information acquisition unit 110 may output the received wireless information to the model calculation unit 140 and the BLER estimation unit 150.

[0062] The model storage unit 130 stores an estimation model that the model calculation unit 140 calculates parameters for and that the BLER estimation unit 150 uses to estimate BLER. The estimation model is a model that estimates the BLER for each MCS at the next time based on time-series data of past wireless quality information. The function of the estimation model may be realized by one model or by multiple models. In this example, the model storage unit 130 stores a radio wave fluctuation model M1 and a transmission error model M2 as estimation models.

[0063] The radio wave fluctuation model M1 is a model that estimates fluctuations in radio quality (radio wave quality) from time-series data of radio quality information. Because radio wave quality fluctuates with time and frequency, the radio wave fluctuation model M1 estimates radio wave quality that fluctuates with time and frequency. The radio wave fluctuation model M1 estimates radio wave quality for the next time from time-series data of past radio wave quality. The radio wave fluctuation model M1 is a calculation model that calculates fluctuations (probability distribution) in radio wave quality by performing predetermined calculations using set parameters. For example, the radio wave fluctuation model M1 is a propagation model that calculates the propagation characteristics of a wireless transmission path (such as the amount of multipath).

[0064] For example, when a UE performs wireless communication while moving, the radio wave quality fluctuates due to fading. Therefore, the radio wave fluctuation model M1 may be a fading model that calculates propagation characteristics due to fading. The radio wave fluctuation model M1 is not limited to a fading model, and may be a model that calculates other propagation characteristics. The probability distribution p(r) of the fading model can be calculated using the following probability density function (Equation 1). Equation 1 is an example of a Nakagawa-Rice fading model. FIG. 11 shows the cumulative probability distribution according to the fading model of Equation 1, and indicates the cumulative probability with respect to radio wave strength r (SINR). For example, the cumulative probability distribution increases as the radio wave strength increases up to a certain radio wave strength, and then becomes a constant value.

[0065] In Equation 1, σ corresponds to the number of reflected waves, and the more reflected waves there are, the larger the value of σ becomes. σ also indicates the slope of the rising edge of the cumulative probability distribution in FIG. 11. In Equation 1, A defines the average value of radio wave strength. When radio wave strength decreases stochastically due to fading, A defines the position of the shoulder (peak) of the cumulative probability distribution in FIG. 11. For example, the slope σ of the cumulative probability distribution and the shoulder position A are parameters (radio wave fluctuation parameters) used to calculate the radio wave quality of the radio wave fluctuation model M1. The parameters may be either the slope σ of the cumulative probability distribution or the shoulder position A, or may include both. In Equation 1, I ０ is the zeroth-order modified Bessel function of the first kind.

[0066] The transmission error model M2 (error rate model) is a model that estimates the BLER for each MCS from the radio quality (radio wave quality) estimated by the radio wave fluctuation model M1. If the radio quality is given, the error rate (transmission error) does not vary (or varies slowly) with time or frequency, but depends on the base station (O-CU or O-DU), UE, etc. Therefore, the transmission error model M2 may estimate the BLER in accordance with these. The transmission error model M2 is a calculation model that calculates the BLER for each MCS by performing a predetermined calculation using set parameters.

[0067] Specifically, the transmission error model M2 is a model that calculates an error rate distribution (error rate characteristics) for radio wave quality (wireless quality) for each MCS. FIG. 12 shows the error rate distribution for radio wave quality (SINR) for a certain MCS. The SINR value can be estimated from the CQI value. For example, the error rate distribution remains constant up to a certain radio wave quality, and then decreases as the radio wave quality increases. The transmission error model M2 can be realized by a distribution function that calculates the error rate distribution for the radio wave quality shown in FIG. 12. For example, the distribution function includes β, which defines the slope of the fall of the error rate distribution in FIG. 12, and α, which defines the position of the shoulder (peak). For example, the slope β and the position α of the shoulder of the error rate distribution are parameters (error rate parameters) used to calculate the error rate of the transmission error model M2. The parameters may be either the slope β of the error rate distribution or the position α of the shoulder, or may include both.

[0068] The model calculation unit 140 calculates parameters of an estimation model based on time-series data of radio quality information (radio information) received from the E2 node 300 using the radio information acquisition unit 110 and recorded in the radio information recording unit 120. The model calculation unit 140 is a generation unit that generates an estimation model by calculating parameters. The model calculation unit 140 stores (sets) the calculated parameters in the model storage unit 130. The model calculation unit 140 includes a radio wave fluctuation model calculation unit 141 and a transmission error model calculation unit 142. The radio wave fluctuation model calculation unit 141 calculates parameters of a radio wave fluctuation model M1 (slope σ of the cumulative probability distribution and shoulder position A) using time-series data of radio quality information (wideband CQI, etc.). The transmission error model calculation unit 142 calculates parameters of a transmission error model M2 (slope β of the error rate distribution and shoulder position α) using the estimation result of the radio wave fluctuation model M1 for which parameters have been calculated and radio quality information (error rate for each MCS).

[0069] The BLER estimation unit 150 estimates the BLER for each MCS based on time-series radio quality information (radio information) received from the E2 node 300 using the radio information acquisition unit 110 and recorded in the radio information recording unit 120. The BLER estimation unit 150 estimates the BLER for each MCS using an estimation model whose parameters have been calculated by the model calculation unit 140 and stored in the model storage unit 130. The BLER estimation unit 150 estimates fluctuations (probability distribution) of radio wave quality using time-series data of radio quality information (such as wideband CQI) and a radio wave fluctuation model M1 whose parameters have been calculated. The BLER estimation unit 150 estimates the BLER for each MCS using a transmission error model M2 whose parameters have been calculated, using the fluctuations (probability distribution) of radio wave quality estimated by the radio wave fluctuation model M1.

[0070] The MCS determiner 160 determines the MCS to be set in the E2 node 300 based on the BLER for each MCS estimated by the BLER estimator 150. The MCS determiner 160 selects an MCS index that satisfies the target BLER from the estimated BLER for each MCS.

[0071] The control information transmitting unit 170 transmits the control information (MCS) determined by the MCS determining unit 160 to the E2 node 300. The control information transmitting unit 170 transmits the determined MCS index to the E2 node 300 including either or both of the O-DU and O-CU via the E2 interface.

[0072] 10 is an example, and other configurations may be used as long as the operation according to the present embodiment described below is possible. A portion of the configuration of the Near-RT RIC 100 may be arranged in the Non-RT RIC 200. For example, as shown in FIG. 13, the model calculation unit 140 may be arranged in the Non-RT RIC 200. The model storage unit 130 may be arranged in the Non-RT RIC 200.

[0073] 13, the Non-RT RIC 200 includes a radio information acquisition unit 210, a radio information recording unit 220, and a model calculation unit 140. The radio information acquisition unit 210 receives radio information including radio quality information from an E2 node 300 including either or both of an O-DU and an O-CU via the O1 interface. The radio information acquisition unit 210 may also receive radio information of the E2 node 300 from the Near-RT RIC 100 via the O1 interface. That is, the Near-RT RIC 100 may collect radio information from the E2 node 300 and transfer the collected radio information from the Near-RT RIC 100 to the Non-RT RIC 200.

[0074] Similar to the wireless information recording unit 120, the wireless information recording unit 220 records the wireless information received by the wireless information acquisition unit 210 from the E2 node 300. The model calculation unit 140 calculates parameters of an estimation model based on time-series data of wireless quality information (wireless information) received from the E2 node 300 using the wireless information acquisition unit 210 and recorded in the wireless information recording unit 220. The model calculation unit 140 transmits the calculated parameters (generated estimation model) to the Near-RT RIC 100 and stores them in the model storage unit 130.

[0075] 14 shows an outline of a method for generating the radio wave fluctuation model M1 and the transmission error model M2. During the learning phase process, the radio wave fluctuation model M1 first calculates the radio wave quality distribution (cumulative probability distribution) using the above formula 1 (S1). At this time, the parameters of the radio wave fluctuation model M1 (the slope σ of the cumulative probability distribution and the shoulder position A) may be predetermined initial values.

[0076] Next, the radio wave fluctuation model calculation unit 141 calculates a wideband CQI from the radio wave quality distribution calculated by the radio wave fluctuation model M1 (S2). Specifically, the radio wave quality for each RBG (Resource Block Group) is calculated from the radio wave quality distribution calculated by the radio wave fluctuation model M1. The radio wave quality may be calculated not only for each RBG but also for each RB (Resource Block). For example, the SINR of each RBG is determined from the frequency and time of each RBG, and the radio wave quality of each RBG is calculated by obtaining the cumulative probability at the determined SINR using Equation 1. Furthermore, the calculated radio wave qualities of each RBG are averaged to calculate a wideband CQI (CQI for the entire band). The number of RBGs may be estimated from the MCS or the number of active UEs.

[0077] Next, the radio wave fluctuation model calculation unit 141 calculates parameters of the radio wave fluctuation model M1 (the slope σ of the cumulative probability distribution and the shoulder position A) from the wideband CQI (averaged radio wave quality) calculated from the radio wave quality distribution of the radio wave fluctuation model M1 and time-series data of the wideband CQI (radio wave quality) acquired from the E2 node 300 (S3). The wideband CQI calculated by the radio wave fluctuation model M1 is compared with the time-series data of the wideband CQI acquired from the E2 node 300, and the parameters of the radio wave fluctuation model M1 are estimated using Bayesian estimation, a Kalman filter, or the like. For example, the parameters of the radio wave fluctuation model M1 are calculated so as to reduce the error between the wideband CQI calculated by the radio wave fluctuation model and the wideband CQI acquired from the E2 node 300. For example, the parameters may be calculated by accumulating the acquired time-series wideband CQIs to create a distribution histogram and fitting the created histogram to Equation 1. Alternatively, the estimated values ​​of the parameters may be calculated (corrected) sequentially from the observed wideband CQIs using a method such as a Kalman filter.

[0078] Next, the radio wave fluctuation model M1 calculates the radio wave quality distribution using the calculated parameters, and the transmission error model M2 calculates the error rate distribution for each MCS (S4). At this time, the parameters of the transmission error model M2 (the slope β of the error rate distribution and the shoulder position α) may be predetermined initial values.

[0079] Next, the transmission error model calculation unit 142 calculates the BLER for each MCS from the radio wave quality distribution (estimated result) calculated by the radio wave fluctuation model M1 and the error rate distribution calculated by the transmission error model M2 (S5). Specifically, the error rate for each RGB is calculated from the cumulative probability of each SINR calculated by the radio wave fluctuation model M1 and the error rate of the SNIR calculated by the transmission error model M2. For example, the SINR is determined from the frequency and time of each RBG, the cumulative probability at the determined SINR is determined using the radio wave fluctuation model M1, and the error rate at the determined SINR is determined using the transmission error model M2. The error rate for each RBG is then calculated by multiplying the determined cumulative probability by the error rate. The error rate may be determined not only for each RBG but also for each RB. Furthermore, the BLER is calculated by integrating the determined error rates for each RBG. The BLER for each MCS is calculated using the error rate distribution for each MCS calculated by the transmission error model M2. As described above, the block error probability at a certain point in time is calculated by multiplying the radio wave quality distribution (probability distribution of radio wave quality) according to the radio wave fluctuation model M1 by the error rate distribution according to the transmission error model M2. For example, if the probability that the radio wave strength reaches r1 is p1 and the error rate at r1 is P1, then p1 x P1 is the probability of an error. In reality, radio wave strength can take on other values, so the same process is repeated for other radio wave strengths; for example, the probability p2 that the radio wave strength reaches r2 is multiplied by the error rate P2 at that radio wave strength. The error probability when a certain MCS is used is calculated by adding these values ​​together (p1 x P1 + p2 x P2 + ...).

[0080] Next, the transmission error model calculation unit 142 calculates parameters of the transmission error model M2 (the slope β of the error rate distribution and the shoulder position α) from the BLER for each MCS calculated from the radio wave quality distribution of the radio wave fluctuation model M1 and the error rate distribution of the transmission error model M2, and from the time-series data of the number of block errors (BLER) for each MCS acquired from the E2 node 300 (S6). The BLER for each MCS calculated using the radio wave fluctuation model M1 and the transmission error model M2 is compared with the time-series data of the number of block errors for each MCS acquired from the E2 node 300, and the parameters of the transmission error model M2 are estimated using Bayesian estimation, a Kalman filter, or the like. The parameters of the transmission error model M2 are calculated so as to minimize the error between the BLER for each MCS calculated using the radio wave fluctuation model M1 and the transmission error model M2 and the number of block errors for each MCS acquired from the E2 node 300. The error rate can be calculated by accumulating the actual values ​​of the acquired number of block errors for a certain period of time. For example, if the parameters of the transmission error model M2 are correct, the error probability calculated from the transmission error model M2 and the error rate calculated from the actual number of block errors will be approximately the same. If there is an error in the parameters of the transmission error model M2, a difference will occur between the error probability calculated from the transmission error model M2 and the error rate calculated from the actual number of block errors. Therefore, by adjusting the parameters in a direction that reduces this difference, the parameters of the transmission error model M2 can be adjusted to more plausible values. For example, this adjustment is performed sequentially using a Kalman filter.

[0081] The radio wave fluctuation model calculation unit 141 may calculate the parameters of the radio wave fluctuation model M1 according to radio wave shielding or interference in the RAN. For example, the parameters of the radio wave fluctuation model M1 may be calculated by inputting the shielding strength and the change probability into a Kalman filter, or the parameters of the radio wave fluctuation model M1 may be calculated by inputting the interference strength and the change probability into a Kalman filter. The interference strength or shielding strength may be estimated from time-series data on the congestion level of adjacent cells or a fluctuation pattern of radio wave quality, and the parameters of the radio wave fluctuation model M1 may be calculated using the estimated interference strength or shielding strength. Since the value of A in Equation 1 (the position of the shoulder of the cumulative probability distribution) is affected by shielding or interference, a model that takes shielding or interference into account can be generated by incorporating changes in A into the model.

[0082] Fig. 15 shows an example of a radio wave fluctuation model M1 that takes shielding into account. As shown in Fig. 15, for example, the probability distribution may have a shape with a peak at the estimated value A calculated from the current radio wave strength and a tail (the shaded area in Fig. 15) that represents the possibility of deterioration due to shielding or interference. This makes it possible to take into account fluctuations due to shielding, interference, etc., thereby improving the accuracy of BLER prediction.

[0083] Furthermore, the radio wave fluctuation model calculation unit 141 may calculate the parameters of the radio wave fluctuation model M1 according to spatial correlation. The spatial correlation is the spatial correlation occurring between the antennas of the base stations (O-DU and O-CU). In MIMO, radio wave quality varies depending on the strength of spatial correlation, so the probability distribution is changed according to the spatial correlation. For example, the spatial correlation may be estimated from the fluctuation pattern of the radio wave quality, and the parameters of the radio wave fluctuation model M1 may be calculated according to the magnitude of the estimated spatial correlation. For example, when the spatial correlation is low (when the Rank is low), it is considered that the number of radio wave paths is small, and therefore it can be assumed that the spread of radio wave fluctuation is small. For this reason, when a radio wave fluctuation model that sequentially updates parameters using a Kalman filter or the like is used, the strength of spatial correlation may be detected, and the spread of the fluctuation of the model (the slope component of the cumulative distribution) may be quickly tracked according to the strength of the detected spatial correlation. Furthermore, the parameters of the radio wave fluctuation model M1 may be calculated using the observed values ​​of the wideband CQI for each Rank and the number of transmissions and errors for each MCS. When collecting the observed values ​​of the number of transmissions and the number of errors for each wideband CQI or MCS, the values ​​may be tallied for each spatial multiplicity, and input and output for parameter calculation may be performed for each multiplicity.

[0084] Furthermore, the transmission error model M2 may include an error rate distribution (error rate characteristics) for each number of retransmissions (number of transmissions). For example, as shown in FIG. 16, the transmission error model M2 may include an error rate distribution for the second to fourth retransmissions. As shown in FIG. 16, the error rate for a low SINR decreases as the number of retransmissions increases. For example, the transmission error model calculation unit 142 calculates parameters of the error rate distribution for each number of retransmissions (number of transmissions). During inference, the transmission error model M2 may estimate a BLER for each number of retransmissions from each error rate distribution. This makes it possible to estimate the BLER for each number of retransmissions (number of transmissions) for each MCS, and to determine the MCS from the estimated BLER, taking the number of retransmissions into consideration.

[0085] Next, the learning phase process and the inference phase process in the RAN system 1 will be described. The learning phase process is a process of calculating (learning) parameters of an estimation model and generating an estimation model with the parameters calculated (trained). The inference phase process is a process of estimating (inferring) the BLER for each MCS using the estimation model with the parameters calculated. For example, the inference phase process may be performed after the learning phase process, or the learning phase process and the inference phase process may be performed in parallel after the first learning phase process (after the parameters have been calculated once). For example, in the learning phase process, the estimation model may be updated in real time whenever new data (wireless quality information) is generated, or the learning process of the estimation model may be performed in advance (offline).

[0086] FIG. 17 shows an example of the operation of the learning phase process in the RAN system 1 according to this embodiment. As shown in FIG. 17 , the Near-RT RIC 100 acquires radio quality information and stores the acquired radio quality information (S101). For example, the radio information acquisition unit 110 receives radio quality information from the E2 node 300, and the radio information recording unit 120 stores the received radio quality information. The stored radio quality information is data for calculating (learning) parameters of the estimation model, and includes, for example, radio wave quality such as wideband CQI (or CQI for each RBG) and the number of block errors (transmission success rate) for each MCS. The radio wave quality such as wideband CQI is data used to calculate parameters of the radio wave fluctuation model M1, and the number of block errors for each MCS is data used to calculate parameters of the transmission error model M2.

[0087] Next, the Near-RT RIC 100 converts the stored wireless quality information into time series data (S102). For example, the wireless information recording unit 120 acquires the stored wireless quality information for a predetermined period of time and converts it into time series data. The model calculation unit 140 may convert the wireless quality information into time series data when calculating parameters.

[0088] Next, the Near-RT RIC 100 calculates parameters of an estimation model that estimates the BLER for each MCS from the time-series data of wireless quality information stored in the wireless information recording unit 120, and generates the estimation model (S103-S104). In this example, the learning process for the estimation model is performed in two steps: the parameters of the radio wave fluctuation model M1 are determined first, and the results are used to calculate the parameters of the transmission error model M2. That is, the radio wave fluctuation model calculation unit 141 calculates the parameters of the radio wave fluctuation model M1 (S103), and then the transmission error model calculation unit 142 calculates the parameters of the transmission error model M2 (S104). The radio wave fluctuation model calculation unit 141 and the transmission error model calculation unit 142 store the calculated parameters (learned models) in the model storage unit 130.

[0089] FIG. 18 shows the input and output of the radio wave fluctuation model calculation unit 141 in S103 (the first step of the learning process). As shown in FIG. 18, the radio wave fluctuation model calculation unit 141 acquires time-series data of radio wave quality, such as wideband CQI, from the radio information recording unit 120 and calculates parameters of the radio wave fluctuation model M1 based on the acquired time-series data of radio wave quality. As described in FIG. 14, the radio wave fluctuation model calculation unit 141 estimates the parameters of the radio wave fluctuation model M1 from the time-series data of radio wave quality using Bayesian estimation, a Kalman filter, or the like. For example, the radio wave fluctuation model calculation unit 141 calculates the slope σ of the cumulative probability distribution and the shoulder position A. Parameters related to radio wave fluctuation may include the average radio wave strength (A in Equation 1), the K factor, the probability and strength of occurrence of obstruction, and the probability and strength of occurrence of interference. The K factor is the ratio of line-of-sight waves to non-line-of-sight waves (reflected waves). For example, the average radio wave strength and the K factor may be calculated by fitting Equation 1, which is a radio wave fluctuation model M1 (Nakagawa-Rice fading model), to a histogram or the like generated from time-series radio wave quality. For the probability and strength of obstruction occurrence and the probability and strength of interference occurrence, parameters may be calculated to obtain the radio wave fluctuation model M1 shown in FIG. 15, for example.

[0090] FIG. 19 shows the input / output of the radio wave fluctuation model M1 and the transmission error model calculation unit 142 in S104 (the second step of the learning process). As shown in FIG. 19, time-series data on radio wave quality, such as wideband CQI, from the radio information recording unit 120 is input to the radio wave fluctuation model M1, whose parameters have been calculated in S103, to calculate a probability distribution (cumulative probability distribution) of the radio wave quality. Next, the transmission error model calculation unit 142 acquires time-series data on the number of block errors for each MCS from the radio information recording unit 120, and calculates parameters of the transmission error model M2 based on the probability distribution of radio wave quality calculated by the radio wave fluctuation model M1 and the acquired time-series data on the number of block errors for each MCS. As described with reference to FIG. 14, the transmission error model calculation unit 142 estimates the parameters of the transmission error model M2 using Bayesian estimation, a Kalman filter, or the like, based on the calculation results of the radio wave fluctuation model M1 and the number of block errors for each MCS. For example, the transmission error model calculation unit 142 calculates the slope β and the shoulder position α of the error rate distribution. The calculated parameters provide the block error rate for the combination of radio quality and MCS when there is no radio fluctuation in the frequency direction or time direction.

[0091] By calculating using the probability distribution of radio wave quality, it is possible to evaluate whether the block error that occurred is an error caused by fluctuations in radio wave quality (in the frequency and time directions) or an error caused by a low MCS relative to the radio wave quality value, and the parameters of the transmission error model M2 can be calculated according to the evaluation results. Specifically, as shown in Figure 20, by giving the probability of radio wave fluctuations (radio wave quality), it is possible to estimate, when an error occurs, whether the error occurred because the SINR fluctuated downward due to fluctuations (as shown by the arrow E1), or whether the error occurred because the transmission error model itself underestimated the error rate (as shown by the arrow E2). For example, if the probability of radio wave quality fluctuating downward due to fluctuations is high, the transmission error model itself is not corrected as in the case of E1. Conversely, if the probability of radio wave quality fluctuating downward is low, the transmission error model is corrected in a direction that does not underestimate the error rate as in the case of E2. Note that a high probability of radio wave quality fluctuating downward is a case where the probability increases toward lower radio wave quality. For example, as shown in FIG. 21, distribution R2 of radio wave quality has a wider base than distribution R1 of radio wave quality, and the probability of radio wave quality decreasing is higher, so the probability of radio wave quality decreasing is high.

[0092] Fig. 22 shows an example of the operation of the inference phase process in the RAN system 1 according to this embodiment. As shown in Fig. 22, the Near-RT RCI 100 acquires radio quality information and stores the acquired radio quality information (S201). As in S101 of Fig. 17, the radio information acquisition unit 110 receives radio quality information from the E2 node 300, and the radio information recording unit 120 stores the received radio quality information. The stored radio quality information is data for estimation (inference) of the estimation model, and includes radio wave quality such as wideband CQI. The radio wave quality such as wideband CQI is data used by the radio wave fluctuation model M1 to estimate radio wave fluctuations.

[0093] Next, the Near-RT RIC 100 converts the stored wireless quality information into time series data (S202). As in S102 of Fig. 17, the wireless information recording unit 120 acquires the stored wireless quality information for a predetermined period of time and converts it into time series data. When the BLER estimation unit 150 estimates the BLER, the wireless quality information may be converted into time series data.

[0094] Next, the Near-RT RIC 100 uses the estimation model with parameters calculated in the model storage unit 130 to estimate the BLER for each MCS from the time-series data of the wireless quality time information stored in the wireless information recording unit 120 (S203-S204). Specifically, the BLER estimation unit 150 estimates fluctuations in radio wave quality using a radio wave fluctuation model M1 with parameters calculated (S203), and estimates the BLER for each MCS using a transmission error model M2 with parameters calculated (S204).

[0095] 23 shows the input and output of the radio wave fluctuation model M1 and the transmission error model M2 in S203 and S204. As shown in FIG. 23, time-series data of radio wave quality, such as wideband CQI, from the radio information recording unit 120 is input to the radio wave fluctuation model M1, whose parameters have been calculated, to calculate the probability distribution (cumulative probability distribution) of radio wave quality. Next, the probability distribution of radio wave quality calculated by the radio wave fluctuation model M1 is input to the transmission error model M2, whose parameters have been calculated, to calculate the BLER for each MCS. For example, the BLER estimation unit 150 may calculate the BLER by multiplying the cumulative probability of radio wave quality in the SINR (SINR estimated from the CQI) calculated by the radio wave fluctuation model M1 by the error rate in the SINR calculated by the transmission error model M2, as in FIG. 14. Even if the most recent CQI has the same value, if the fluctuation is large (the probability of quality downward fluctuation is high), the estimated error rate will be high.

[0096] Furthermore, as shown in FIG. 24 , the BLER for each MCS may be calculated by inputting the TBS (Transport Block Size: number of allocated RBs) into the transmission error model M2. The TBS is obtained from the base station (O-DU or O-CU) via the E2 interface. Because the transmission error model M2 (transmission error probability) is affected by the TBS, the TBS is used to correct for this effect. Because the error rate varies depending on the number of RBs allocated per slot (TTI: Transmission Time Interval), the BLER is calculated based on the TBS. A larger number of RBs increases the frequency used. A wider frequency range averages out fluctuations due to fading, tending to lower the error rate. Therefore, the BLER may be lowered when the TBS is large. Furthermore, when the TBS (or CBS: Code Block Size) is small, the amount of redundant coded data decreases (probabilistic bias is more likely to occur), resulting in a higher error probability. For example, two transmission error models, one for a small TBS and one for a large TBS, may be used to determine the model parameters for each TBS. The number of types is not limited to two, and the TBS may be further divided into a plurality of stages, and a transmission error model for each of the plurality of stages may be used.

[0097] 22, the Near-RT RIC 100 determines an MCS based on the estimated BLER for each MCS (S205). The MCS determiner 160 determines an MCS index that satisfies the target BLER based on the BLER for each MCS estimated by the BLER estimator 150. Furthermore, when the BLER estimator 150 estimates the BLER for each number of retransmissions (number of transmissions), the MCS determiner 160 may determine an MCS index from the BLER for the number of retransmissions that satisfies the allowable delay.

[0098] The MCS determiner 160 may determine the MCS taking into account the time interval between CSI reports from the UE. Channel fluctuations can reduce the accuracy of the PM (Precoding Matrix) used in beamforming and MIMO, resulting in a decrease in SINR. When acquiring CQI and BLER from a base station (O-DU or O-CU), the MCS determiner 160 simultaneously collects information on the frequency and time of CSI reports for each UE, and determines the MCS according to the frequency and interval of the CSI reports. For example, if the interval between CSI reports is long, wireless quality may be reduced, so the MCS Index may be lowered. Alternatively, if the interval between CSI reports is short, the MCS Index may be increased.

[0099] Next, the Near-RT RIC 100 controls the RAN using the determined MCS (S206). The control information transmitter 170 transmits the control information (MCS) determined by the MCS determiner 160 to the E2 node 300 including either or both of the O-DU and the O-CU, thereby controlling the RAN.

[0100] As described above, in this embodiment, during learning, an estimation model is generated that calculates the BLER for each MCS at the next moment from the time-series data of radio wave quality using time-series data of past radio wave quality and BLER for each MCS. Also, during inference, the generated estimation model is used to calculate the BLER for each MCS at each moment from the time-series data of radio wave quality.

[0101] By setting two types of parameters in the estimation model, it becomes possible to estimate the next BLER for each MCS from the past wireless quality and the BLER for each MCS. The first parameter is a parameter of the time-varying part and is a parameter that identifies the wireless quality that varies over time. The second parameter is a parameter of the time-invariant part and is a parameter that identifies the relationship between the instantaneous wireless quality and the BLER. In order to identify the relationship between the instantaneous wireless quality and the BLER, data accumulation is required. By estimating the time-varying part of the first parameter and removing its influence, data used to calculate the second parameter can be accumulated within a certain period of time. Once the second parameter is known, it is possible to calculate the BLER for each MCS from the first parameter. In this embodiment, the first parameter is estimated using a radio wave fluctuation model, and the second parameter is estimated using a transmission error model.

[0102] This makes it possible to accurately estimate the BLER for each MCS even when the wireless quality fluctuates and the target BLER is low. Furthermore, the MCS to be set in the RAN can be determined based on the estimated BLER for each MC.

[0103] (Embodiment 2) Next, embodiment 2 will be described. In this embodiment, an example will be described in which a learning model is used as an estimation model for estimating BLER for each MCS. Note that this embodiment can be implemented in combination with embodiment 1, and the configurations shown in embodiment 1 may be used as appropriate.

[0104] FIG. 25 shows an example configuration of the Near-RT RIC 100 and the E2 node 300 in the RAN system 1 according to this embodiment. As shown in FIG. 25, in this embodiment, an estimation model used to estimate BLER for each MCS is stored in the model storage unit 130 of the Near-RT RIC 100. The estimation model is a model that predicts future (one period ahead) wireless quality information (BLER for each CQI and BLER for each MCS) from time-series data of past wireless quality information (BLER for each CQI and BLER for each MCS). The model storage unit 130 stores, as estimation models, a radio wave fluctuation learning model M3 and a transmission error learning model M4, which are machine learning models. Note that the radio wave fluctuation learning model M3 and the transmission error learning model M4 may be combined into a single learning model.

[0105] The radio wave fluctuation learning model M3 is a machine learning model that estimates fluctuations in radio quality (radio wave quality) from time-series data of radio quality information, similar to the radio wave fluctuation model M1 of the first embodiment. The radio wave fluctuation learning model M3 is a model that can analyze and predict time-series data. For example, the radio wave fluctuation learning model M3 may be a recurrent neural network (RNN) or a long-short term model (LSTM), or may be another neural network (DNN: deep neural network). The radio wave fluctuation learning model M3 is not limited to a neural network, and may be another learning model.

[0106] The transmission error learning model M4, like the transmission error model M2 in the first embodiment, is a machine learning model that estimates the BLER for each MCS from the wireless quality (radio wave quality) estimated by the radio wave fluctuation model M1. The transmission error learning model M4 is a model that can analyze and predict time-series data. For example, like the radio wave fluctuation learning model M3, the transmission error learning model M4 may be an RNN or LSTM, or may be another neural network (DNN). The transmission error learning model M4 is not limited to a neural network, and may be another learning model.

[0107] In this embodiment, the Near-RT RIC 100 includes a model learning unit 180 instead of the model calculation unit 140 of the first embodiment. The model learning unit 180 performs machine learning such as deep learning using time-series data of radio quality information received from the E2 node 300 to generate a trained estimation model. The model learning unit 180 stores the trained estimation model in the model storage unit 130. The model learning unit 180 includes a radio wave fluctuation model learning unit 181 and a transmission error model learning unit 182. The radio wave fluctuation model learning unit 181 trains a radio wave fluctuation learning model M3 using time-series data of radio quality information (such as wideband CQI) to generate a trained radio wave fluctuation learning model M3. The transmission error model learning unit 182 trains a transmission error learning model M4 using the estimation result by the trained radio wave fluctuation learning model M3 and radio quality information (error rate for each MCS) to generate a trained transmission error learning model M4. The other configurations are the same as those of the first embodiment.

[0108] 26 shows an example of the operation of the learning phase process in the RAN system 1 according to the present embodiment. As shown in FIG. 26, similar to the first embodiment, the Near-RT RIC 100 acquires wireless quality information from the E2 node 300, stores the acquired wireless quality information (S101), and chronologically serializes the stored wireless quality information (S102).

[0109] Next, the Near-RT RIC 100 performs a learning process for the estimation model using the time-series data of wireless quality information stored in the wireless information recording unit 120, and generates a learned estimation model (S111-S112). For example, the model learning unit 180 acquires time-series data of BLER for each CQI and MCS, divides the time-series data of BLER for each CQI and MCS into past data and future data, and trains the estimation model to generate a model that predicts future BLER for each CQI and MCS from past BLER for each CQI and MCS.

[0110] Specifically, the radio wave fluctuation model learning unit 181 performs a learning process for the radio wave fluctuation learning model M3 (S111), and the transmission error model learning unit 182 performs a learning process for the transmission error learning model M4 (S112). The radio wave fluctuation model learning unit 181 and the transmission error model learning unit 182 store the generated, trained radio wave fluctuation learning model M3 and transmission error learning model M4 in the model storage unit 130.

[0111] 27 shows the input and output of the radio wave fluctuation learning model M3 in S111 (the first step of the learning process). The radio wave fluctuation model learning unit 181 inputs time-series data of radio wave quality, such as wideband CQI, into the radio wave fluctuation learning model M3 and trains the radio wave fluctuation learning model M3. For example, the time-series data of radio wave quality from the radio information recording unit 120 is divided into past and future data, and the radio wave quality of the future part of the time-series data is set as the correct answer. As shown in FIG. 27, the radio wave quality of the past part of the time-series data is input into the radio wave fluctuation learning model M3, and the radio wave fluctuation learning model M3 is trained so that the error between the predicted radio wave quality result output from the radio wave fluctuation learning model M3 and the correct answer for the radio wave quality of the future part of the time-series data is reduced.

[0112] 28 shows the input and output of the radio wave fluctuation learning model M3 and the transmission error learning model M4 in S112 (the second step of the learning process). The transmission error model learning unit 182 inputs the prediction results of the learned radio wave fluctuation learning model M3 and time-series data on the number of block errors for each MCS into the transmission error learning model M4 to train the transmission error learning model M4. For example, the time-series data on the number of block errors for each MCS from the radio information recording unit 120 is divided into past and future data, and the number of block errors for each MCS in the future part of the time-series data is set as the correct answer. As shown in FIG. 28, time-series data on radio wave quality, such as wideband CQI, from the radio information recording unit 120 is input into the radio wave fluctuation learning model M3 that was learned in S111 to predict future radio wave quality. Next, the radio wave quality prediction result of the radio wave fluctuation learning model M3 and the number of block errors for each MCS of the past time series data are input into the transmission error learning model M4, and the transmission error learning model M4 is trained so that the error between the BLER prediction result for each MCS output from the transmission error learning model M4 and the number of block errors (BLER) for each MCS of the future time series data is small.

[0113] 29 shows an example of the operation of the inference phase process in the RAN system 1 according to the present embodiment. As shown in FIG. 29, similar to the first embodiment, the Near-RT RCI 100 acquires wireless quality information from the E2 node 300, stores the acquired wireless quality information (S201), and chronologically serializes the stored wireless quality information (S202).

[0114] Next, the Near-RT RIC 100 uses the trained estimation model in the model storage unit 130 to predict BLER for each MCS from the time-series data of radio quality time information stored in the radio information recording unit 120 (S211-S212). For example, the BLER estimation unit 150 acquires time-series data of CQI and BLER, and uses the trained estimation model to predict BLER for each MCS one period ahead from the acquired time-series data of CQI and BLER. Specifically, the BLER estimation unit 150 predicts radio wave quality using the trained radio wave fluctuation learning model M3 (S211), and predicts BLER for each MCS using the trained transmission error learning model M4 (S212).

[0115] 30 shows the input and output of the radio wave fluctuation learning model M3 and the transmission error learning model M4 in S211 and S212. As shown in Fig. 30, time-series data on radio wave quality, such as Wideband CQI, from the radio information recording unit 120 is input to the trained radio wave fluctuation learning model M3, and future radio wave quality is predicted. Next, the predicted radio wave quality result of the radio wave fluctuation model M1 is input to the trained transmission error learning model M4, and future BLER for each MCS is predicted.

[0116] Thereafter, similarly to the first embodiment, the Near-RT RIC 100 determines an MCS based on the predicted BLER for each MCS (S205), and controls the RAN using the determined MCS (S206).

[0117] Note that the estimation model of embodiment 1 and the estimation model of embodiment 2 may be provided, and either model may be selected to estimate the BLER for each MCS. FIG. 31 shows another example configuration of the Near-RT RIC 100 according to this embodiment. As shown in FIG. 31, the model storage unit 130 of the Near-RT RIC 100 may store the radio wave fluctuation model M1 (radio wave fluctuation calculation model) and transmission error model M2 (error rate calculation model) of embodiment 1, and the radio wave fluctuation learning model M3 and transmission error learning model M4 of embodiment 2. The Near-RT RIC 100 may also include the model calculation unit 140 and model learning unit 180 of embodiment 1. Either or both of the model calculation unit 140 and the model learning unit 180 may be located in the Non-RT RIC 200. Some or all of the radio wave fluctuation model M1, transmission error model M2, radio wave fluctuation learning model M3, and transmission error learning model M4 may be stored in the Non-RT RIC 200.

[0118] The estimation model (calculation model) of the first embodiment can learn quickly because it includes approximations in the estimation, but the estimation accuracy may be limited to a certain extent. On the other hand, the estimation model (learning model) of the second embodiment takes time to learn, but may be superior to the estimation model of the first embodiment in terms of estimation accuracy. For this reason, estimation using the estimation model of the first embodiment and estimation using the estimation model of the second embodiment may be switched depending on the learning status. Note that time-series data of wireless quality information is accumulated during operation of the Near-RT RIC 100, and the model learning unit 180 performs learning processing of the radio wave fluctuation learning model M3 and the transmission error learning model M4 at regular intervals. The BLER estimation unit 150 acquires time series data of CQI and time series data of BLER for each MCS from the E2 node 300, and if the amount of training data is equal to or less than a certain value, it estimates the BLER for each MCS using the estimation model of embodiment 1 (radio wave fluctuation model M1 and transmission error model M2), and if the amount of training data is equal to or more than a certain value, it estimates the BLER for each MCS using the estimation model of embodiment 2 (radio wave fluctuation learning model M3 and transmission error learning model M4).

[0119] The selection criterion for the estimation model may be the total amount of learning data, or the amount of data similar to the behavior of the current observation value (CQI). If the accumulated time-series data of radio quality information contains a large amount of data similar to the current observation value, the estimation model (learning model) of the second embodiment may be selected, and if the accumulated time-series data of radio quality information contains a small amount of data similar to the current observation value, the estimation model (calculation model) of the first embodiment may be selected.

[0120] As described above, an estimation model may be generated by machine learning the future BLER for each MCS according to past wireless quality information, and the trained estimation model may be used to estimate the BLER for each MCS according to the acquired wireless quality information. As in the first embodiment, this allows the BLER for each MCS to be estimated with high accuracy even when the target BLER is low. Using a learning model that has undergone machine learning makes it possible to further improve the estimation accuracy.

[0121] (Embodiment 3) Next, embodiment 3 will be described. In this embodiment, an example will be described in which BLER is measured using a measurement bearer at the E2 node. Note that this embodiment can be implemented in combination with embodiment 1 or 2, and the configurations shown in embodiment 1 or 2 may be used as appropriate.

[0122] Fig. 32 shows a configuration example of the Near-RT RIC 100 and the E2 node 300 in the RAN system 1 according to the present embodiment. As shown in Fig. 32, in the present embodiment, the E2 node 300 includes a bearer creation unit 350 in addition to the configuration of the first embodiment.

[0123] The bearer creation unit 350 creates multiple bearers for BLER measurement for each MCS. The bearer creation unit 350 sets up a bearer for BLER measurement between the O-DU or O-CU and the UE. The bearer creation unit 350 selects one 5QI (5G QoS Indicator) for BLER measurement and creates a bearer for the selected 5QI. The bearer creation unit 350 selects multiple 5QIs of different classes and creates multiple bearers for each 5QI.

[0124] The radio information acquisition unit 310 transmits measurement data through the created multiple bearers and measures the BLER for each MCS. Data may be transmitted from the E2 node 300 to the UE, and the UE may measure the downlink BLER. Alternatively, data may be transmitted from the UE to the E2 node 300, and the E2 node 300 may measure the uplink BLER. The radio information acquisition unit 310 transmits data of a different MCS for each bearer and measures the BLER for each MCS. For example, the Near-RT RIC 100 may specify multiple MCS indices to be measured on multiple measurement bearers to the E2 node 300 in order to acquire the BLER (number of block errors) for each MCS as radio quality information. The radio information acquisition unit 310 may periodically transmit transport blocks, frozen using the coding and modulation methods of the specified multiple MCS indices, via multiple measurement bearers, to measure the BLER for each MCS. Other configurations are the same as those in the first embodiment.

[0125] In this way, a bearer for BLER measurement may be created and the BLER may be measured using the created bearer. By creating multiple measurement bearers with different 5QI classes and transmitting data with different MCSs on each measurement bearer to measure the BLER, BLER measurements for multiple MCSs can be performed simultaneously and efficiently. This can improve the learning speed and inference accuracy.

[0126] (Fourth Embodiment) Next, a fourth embodiment will be described. In this embodiment, an example in which location information is acquired from an external application server will be described. Note that this embodiment can be implemented in combination with any of the first to third embodiments, and each configuration shown in any of the first to third embodiments may be used as appropriate.

[0127] Fig. 33 shows a configuration example of the Near-RT RIC 100 and the E2 node 300 in the RAN system 1 according to the present embodiment. As shown in Fig. 33, the RAN system 1 includes an application server 400 in addition to the configuration of the first embodiment. For example, the application server 400 is connected to the Near-RT RIC 100 so as to be able to communicate with it, but may also be connected to the Non-RT RIC 200 so as to be able to communicate with it.

[0128] The application server 400 is a server external to the RAN including the E2 node 300. The application server 400 may be a physical server or a virtual server on the cloud, for example, a server on the Internet. The application server 400 is a device that provides application services in cooperation with the RAN and manages data related to applications. For example, the application server 400 may be a management server for an automated guided vehicle (AGV) system or an autonomous mobile robot (AMR) system, a management server for a vehicle platooning system or an autonomous driving system, or a management server for an automated construction system. For example, the application server 400 provides UE location information to the Near-RT RIC 100. Note that the application server 400 may provide other application data in addition to UE location information.

[0129] Furthermore, in this embodiment, Near-RT RIC 10 includes a location information acquisition unit 190 in addition to the configuration of embodiment 1. Location information acquisition unit 190 acquires location information of UEs from application server 400. Location information acquisition unit 190 may periodically acquire location information of UEs from application server 400 and store time-series data of the location information in radio information recording unit 120.

[0130] The model calculation unit 140 calculates parameters of the estimation model based on the time-series data of the radio quality information acquired from the E2 node 300 and the time-series data of the UE location information acquired from the application server 400. When using the learning model of the second embodiment, learning may be performed using the radio quality information and the location information to generate the learning model.

[0131] The radio wave fluctuation model calculation unit 141 may calculate parameters of the radio wave fluctuation model M1 based on location information of the UE. For example, it may estimate obstruction from the location information of the UE and calculate parameters of the radio wave fluctuation model M1 based on the estimated obstruction. The BLER estimation unit 150 may estimate obstruction from the location information of the UE, and when obstruction is estimated, it may estimate the BLER for each MCS using the radio wave fluctuation model M1 generated taking the obstruction into consideration. Other configurations are the same as those in the first embodiment.

[0132] In this way, location information may be acquired from the application server, and the acquired location information may be used to perform learning and inference of the estimation model. By additionally inputting location information into the input of the estimation model, it is possible to improve the accuracy of calculating the BLER.

[0133] The present disclosure is not limited to the above-described embodiment, and can be modified as appropriate within the scope of the present disclosure.

[0134] Each component in the above-described embodiments may be configured with hardware, software, or both, and may be configured with a single piece of hardware or software, or may be configured with multiple pieces of hardware or software. Each device, including the Non-RT RIC and Near-RT RIC, and each function (processing) may be implemented by a computer 70 having a network interface 71, a processor 72 such as a CPU (Central Processing Unit), and a memory 73 serving as a storage device, as shown in FIG. 34 . The network interface 71 may include a network interface card (NIC) that communicates with devices including network nodes. For example, a program for performing the method in the embodiment may be stored in the memory 73, and each function may be implemented by executing the program stored in the memory 73 by the processor 72.

[0135] These programs include instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The programs may be stored on a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable media or tangible storage media include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disc (DVD), Blu-ray disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The programs may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals.

[0136] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

[0137] Some or all of the above embodiments can be described as, but are not limited to, the following supplementary notes. (Supplementary Note 1) A system comprising: an acquisition means for acquiring radio quality information from a wireless network; and a generation means for generating, based on the acquired radio quality information, a radio wave fluctuation model for estimating fluctuations in radio wave quality in accordance with the radio quality information, and for generating an error rate model for estimating an error rate for each modulation and coding scheme in accordance with the radio wave quality. (Supplementary Note 2) The system according to Supplementary Note 1, wherein the radio wave fluctuation model is a calculation model for calculating fluctuations in the radio wave quality using a predetermined calculation method, and the generation means sets radio wave fluctuation parameters used for calculating fluctuations in the radio wave quality based on the acquired radio quality information. (Supplementary Note 3) The system according to Supplementary Note 2, wherein the radio wave fluctuation model is a calculation model for calculating a probability distribution of the radio wave quality, and the generation means sets at least one of a slope of the probability distribution and a position of a peak of the probability distribution as the radio wave fluctuation parameter. (Supplementary Note 4) The system according to Supplementary Note 3, wherein the generation means averages the radio wave quality for each resource block obtained from the probability distribution of the radio wave quality, and sets the radio wave fluctuation parameter based on the averaged radio wave quality and the acquired radio quality information. (Supplementary Note 5) The system according to any one of Supplements 2 to 4, wherein the generation means sets the radio wave fluctuation parameter according to radio wave shielding or interference in the wireless network. (Supplementary Note 6) The system according to any one of Supplements 2 to 5, wherein the generation means sets the radio wave fluctuation parameter according to spatial correlation of base stations in the wireless network. (Supplementary Note 7) The system according to any one of Supplements 2 to 6, wherein the generation means sets the radio wave fluctuation parameter based on location information of a terminal device in the wireless network acquired from an external server.(Supplementary Note 8) The system according to any one of Supplements 1 to 7, wherein the error rate model is a calculation model that calculates an error rate for each of the modulation and coding schemes using a predetermined calculation method, the acquiring means acquires error rate information for each of the modulation and coding schemes from the wireless network, and the generating means sets an error rate parameter used to calculate the error rate for each of the modulation and coding schemes based on the estimation result of the radio wave fluctuation model and the acquired error rate information for each of the modulation and coding schemes. (Supplementary Note 9) The system according to Supplementary Note 8, wherein the acquiring means acquires the error rate information measured using a measurement bearer for each of the modulation and coding schemes. (Supplementary Note 10) The system according to Supplementary Note 8 or 9, wherein the error rate model is a calculation model that calculates an error rate characteristic with respect to radio wave quality for each of the modulation and coding schemes, and the generating means sets at least one of a slope of the error rate characteristic and a peak position of the error rate characteristic as the error rate parameter. (Supplementary Note 11) The system according to Supplementary Note 10, wherein the generation means calculates a block error rate based on an error rate for each resource block calculated from the estimation result of the radio wave fluctuation model and the error rate characteristics, and sets the error rate parameter based on the calculated block error rate and the acquired error rate information for each modulation and coding scheme. (Supplementary Note 12) The system according to Supplementary Note 1, wherein the radio wave fluctuation model is a learning model that predicts fluctuations in the radio wave quality according to the radio quality information, and the generation means uses the acquired radio quality information to train the radio wave fluctuation model to learn fluctuations in the radio wave quality according to the radio quality information. (Supplementary Note 13) The system according to Supplementary Note 1 or 12, wherein the error rate model is a learning model that predicts an error rate for each modulation and coding scheme according to the radio wave quality, and the acquisition means acquires error rate information for each modulation and coding scheme from the wireless network, and the generation means uses the estimation result of the radio wave fluctuation model and the acquired error rate information for each modulation and coding scheme to train the error rate model to learn the error rate for each modulation and coding scheme according to the radio wave quality.(Supplementary Note 14) The system according to any one of Supplements 1 to 13, comprising: estimation means for estimating an error rate for each modulation and coding scheme according to the acquired radio quality information, using the generated radio wave fluctuation model and error rate model. (Supplementary Note 15) The system according to Supplementary Note 14, comprising determination means for determining a modulation and coding scheme to be set in the wireless network, based on the estimated error rate for each modulation and coding scheme. (Supplementary Note 16) The system according to Supplementary Note 15, wherein the determination means determines the modulation and coding scheme based on a transmission interval of a radio quality report including the radio quality information transmitted from a terminal device. (Supplementary Note 17) The system according to Supplementary Note 15 or 16, wherein the error rate model estimates an error rate for each number of retransmissions in the modulation and coding scheme, and the determination means determines the modulation and coding scheme based on the error rate for each number of retransmissions. (Supplementary Note 18) The system according to any one of Supplements 1 to 17, comprising: a RAN Intelligent Controller (RIC) for controlling a RAN (Radio Access Network). (Supplementary Note 19) A system comprising: an acquisition means for acquiring radio quality information from a wireless network; and an estimation means for estimating an error rate for each modulation and coding scheme based on the acquired radio quality information, using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the radio quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality. (Supplementary Note 20) The system according to Supplementary Note 19, wherein the radio wave fluctuation model is a calculation model in which parameters are set to calculate fluctuations in the radio wave quality using a predetermined calculation method, and the error rate model is a calculation model in which parameters are set to calculate an error rate for each modulation and coding scheme using a predetermined calculation method. (Supplementary Note 21) The system according to Supplementary Note 19, wherein the radio wave fluctuation model is a learning model that has learned fluctuations in the radio wave quality according to the radio quality information, and the error rate model is a learning model that has learned an error rate for each modulation and coding scheme according to the radio wave quality.(Supplementary Note 22) The system according to Supplementary Note 19, wherein the radio wave fluctuation model includes a radio wave fluctuation calculation model in which parameters are set to calculate fluctuations in the radio wave quality using a predetermined calculation method, and a radio wave fluctuation learning model that has learned fluctuations in the radio wave quality according to the radio quality information, the error rate model includes an error rate calculation model in which parameters are set to calculate an error rate for each modulation and coding scheme using a predetermined calculation method, and an error rate learning model that has learned an error rate for each modulation and coding scheme according to the radio wave quality, and the estimation means selects the radio wave fluctuation calculation model or the radio wave fluctuation learning model and estimates fluctuations in the radio wave quality using the selected model, and selects the error rate calculation model or the error rate learning model and estimates the error rate for each modulation and coding scheme using the selected model. (Supplementary Note 23) The system according to Supplementary Note 22, wherein the estimation means selects the radio wave fluctuation calculation model or the radio wave fluctuation learning model based on the amount of data learned by the radio wave fluctuation learning model, and selects the error rate calculation model or the error rate learning model based on the amount of data learned by the error rate learning model. (Supplementary Note 24) An apparatus comprising: an acquisition means for acquiring radio quality information from a wireless network; and a generation means for generating, based on the acquired radio quality information, a radio wave fluctuation model for estimating fluctuations in radio wave quality in accordance with the radio quality information, and generating an error rate model for estimating an error rate for each modulation and coding scheme in accordance with the radio wave quality. (Supplementary Note 25) An apparatus comprising: an acquisition means for acquiring radio quality information from a wireless network; and an estimation means for estimating an error rate for each modulation and coding scheme based on the acquired radio quality information, using the radio wave fluctuation model generated to estimate fluctuations in radio wave quality in accordance with the radio quality information and the error rate model generated to estimate the error rate for each modulation and coding scheme in accordance with the radio wave quality.(Supplementary Note 26) A method comprising: acquiring radio quality information from a wireless network; and generating, based on the acquired radio quality information, a radio wave fluctuation model for estimating fluctuations in radio wave quality in accordance with the radio quality information, and generating an error rate model for estimating an error rate for each modulation and coding scheme in accordance with the radio wave quality. (Supplementary Note 27) A method comprising: acquiring radio quality information from a wireless network; and estimating an error rate for each modulation and coding scheme based on the acquired radio quality information, using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality in accordance with the radio quality information and an error rate model generated to estimate the error rate for each modulation and coding scheme in accordance with the radio wave quality. (Supplementary Note 28) A non-transitory computer-readable medium having stored thereon a program for causing a computer to execute processes of: acquiring radio quality information from a wireless network; and generating, based on the acquired radio quality information, a radio wave fluctuation model for estimating fluctuations in radio wave quality in accordance with the radio quality information, and generating an error rate model for estimating the error rate for each modulation and coding scheme in accordance with the radio wave quality. (Supplementary Note 29) A non-transitory computer-readable medium storing a program for causing a computer to execute a process of acquiring radio quality information from a wireless network, and estimating an error rate for each modulation and coding scheme based on the acquired radio quality information using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the radio quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality.

[0138] 1 RAN system 10 First system 11 Acquisition unit 12 Generation unit 20 Second system 21 Acquisition unit 22 Estimation unit 30 System 40 First device 50 Second device 60 Device 70 Computer 71 Network interface 72 Processor 73 Memory 100 Near-RT RIC 110 Radio information acquisition unit 120 Radio information recording unit 130 Model storage unit 140 Model calculation unit 141 Radio wave fluctuation model calculation unit 142 Transmission error model calculation unit 150 BLER estimation unit 160 MCS determination unit 170 Control information transmission unit 180 Model learning unit 181 Radio wave fluctuation model learning unit 182 Transmission error model learning unit 190 Location information acquisition unit 200 Non-RT RIC 210 Radio information acquisition unit 220 Radio information recording unit 300 E2 node 310 Radio information acquisition unit 320 Radio information transmission unit 330 Control information reception unit 340 RAN control unit 350 Bearer creation unit 400 Application server M1 Radio wave fluctuation model M2 Transmission error model M3 Radio wave fluctuation learning model M4 Transmission error learning model

Claims

1. acquiring means for acquiring wireless quality information from a wireless network; a generating means for generating a radio wave fluctuation model for estimating fluctuations in radio wave quality according to the acquired radio quality information, and generating an error rate model for estimating an error rate for each modulation and coding scheme according to the radio wave quality; A system comprising:

2. the radio wave fluctuation model is a calculation model for calculating fluctuations in the radio wave quality by a predetermined calculation method, the generating means sets radio wave fluctuation parameters to be used for calculating fluctuations in the radio wave quality based on the acquired radio quality information. The system of claim 1 .

3. the radio wave fluctuation model is a calculation model for calculating a probability distribution of the radio wave quality, the generating means sets at least one of a slope of the probability distribution and a position of a peak of the probability distribution as the radio wave fluctuation parameter. The system of claim 2 .

4. the generating means averages the radio wave quality for each resource block obtained from the probability distribution of the radio wave quality, and sets the radio wave fluctuation parameter based on the averaged radio wave quality and the acquired radio quality information. The system of claim 3.

5. the error rate model is a calculation model for calculating an error rate for each modulation and coding scheme by a predetermined calculation method, the acquiring means acquires error rate information for each modulation and coding scheme from the wireless network; the generating means sets an error rate parameter to be used for calculating the error rate for each modulation and coding scheme based on the estimation result of the radio wave fluctuation model and the acquired error rate information for each modulation and coding scheme. A system according to any one of claims 1 to 4.

6. acquiring means for acquiring wireless quality information from a wireless network; an estimation means for estimating an error rate for each modulation and coding scheme based on the acquired wireless quality information, using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the wireless quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality; A system comprising:

7. Obtaining wireless quality information from a wireless network; generating a radio wave fluctuation model for estimating fluctuations in radio wave quality according to the acquired radio quality information, and generating an error rate model for estimating an error rate for each modulation and coding scheme according to the radio wave quality; method.

8. Obtaining wireless quality information from a wireless network; estimating an error rate for each modulation and coding scheme based on the acquired wireless quality information, using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the wireless quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality; method.

9. Obtaining wireless quality information from a wireless network; generating a radio wave fluctuation model for estimating fluctuations in radio wave quality according to the acquired radio quality information, and generating an error rate model for estimating an error rate for each modulation and coding scheme according to the radio wave quality; A program that causes a computer to execute a process.

10. Obtaining wireless quality information from a wireless network; estimating an error rate for each modulation and coding scheme based on the acquired wireless quality information, using a radio wave fluctuation model generated to estimate fluctuations in radio wave quality according to the wireless quality information and an error rate model generated to estimate an error rate for each modulation and coding scheme according to the radio wave quality; A program that causes a computer to execute a process.