RAN Base Station Control Using Predicted Network State

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing control systems for Radio Access Networks (RAN) struggle to achieve real-time, centralized, and reliable control of multiple base stations, often failing to match the actual state of the network due to timing discrepancies and lack of adaptability.

Innovation Solution

A control system that includes a central control device predicting the RAN state and generating control messages based on machine learning models, allowing base stations to adjust their operations in real-time to match the actual network state, ensuring high reliability and performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If control messages are generated based on predicted RAN state, then centralized control capability is improved, but control accuracy deteriorates due to time delay between prediction and actual state

Engineering Contradiction:
Improvecentralized control capabilityVSAvoidcontrol accuracy
Core Design Contradiction:
Extent of automationVSMeasurement precision

Solution Approach 1:

The control device generates control messages based on predicted RAN states at future time points before actually transmitting control instructions to base stations. This preliminary action allows the system to proactively prepare control decisions, improving centralized control capability while accounting for transmission delays and processing times in the control loop.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements a feedback mechanism where base stations report actual RAN states to the control device, which then compares predicted states with actual states. This feedback loop enables the control device to refine its prediction accuracy over time, resolving the contradiction between centralized control capability and control accuracy by continuously improving the precision of state predictions.

Inventive Principle:
Principle #23Feedback

2Speed

If control messages are transmitted without modification, then transmission speed is improved, but reliability deteriorates when predicted state does not match actual state

Engineering Contradiction:
Improvetransmission speedVSAvoidcontrol reliability
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system transmits control messages with partial modification capability - base stations first attempt to execute control messages as received, and only modify them locally if the predicted state does not match the actual state. This partial action approach maintains high transmission speed for most control scenarios while ensuring reliability when state mismatches occur, without requiring full modification of every control message.

Inventive Principle:
Principle #16Partial or excessive action

3Adaptability or versatility

If real-time RAN state acquisition is implemented at base stations, then control adaptability is improved, but system complexity increases

Engineering Contradiction:
Improvecontrol adaptabilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system segments the control functionality between the centralized control device and distributed base stations. The control device handles high-level decision-making based on predicted RAN states, while base stations handle local state acquisition and minor control adjustments. This segmentation allows real-time adaptability at the network level without requiring complex real-time processing at every base station, thus improving control adaptability while managing system complexity.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4730911A1Control system, control device, control unit, and program
Publication Date: 2026.04.22 SOFTBANK CORPORATION
  • EP4730911A1 patent drawingFigure 1
  • EP4730911A1 patent drawingFigure 2
  • EP4730911A1 patent drawingFigure 3

AI summary

Provided is a control system including: a control device which controls a plurality of base stations arranged in a RAN; and a control unit which is mounted on one base station among the plurality of base stations and which controls the one base station, wherein the control device includes a prediction unit which predicts a state of the RAN at a prediction timing after elapse of a transmission time from a generation start timing at which generation of a control message for controlling the one base station is started until data arrives at the one base station from the control device, a generation unit which generates the control message based on the state of the RAN predicted, and a transmission unit which transmits the control message to the one base station, and wherein the control unit includes a reception unit which receives the control message from the control device, an acquisition unit which acquires a state of the RAN at a reception timing at which the reception unit receives the control message, and a processing unit which executes processing related to control of the one base station based on the control message and the state of the RAN acquired by the acquisition unit.