AI-powered XAPP system that selects adaptive waveform based on channel conditions in O-RAN based access networks.

Abstract

The invention is a system that performs adaptive waveform selection according to channel conditions in O-RAN-based access networks. The system includes a Channel Condition Monitoring Module (1), Waveform Decision Mechanism (2), RIC Control Interface (3), O-DU (4), O-RU (5) and Model Update Unit (6). The system analyzes channel condition information to determine the most suitable waveform and transmits this decision to the O-DU (4) unit, thereby updating the physical layer parameters. Thanks to the Model Update Unit (6), the system continuously optimizes the artificial intelligence model using performance data. In this way, connection continuity and spectral efficiency are increased under high mobility and variable channel conditions.

Description

1 TARIFF Adaptive Wave in O-RAN Based Access Networks According to Channel Conditions THE FORM IS SELECTED BY AN AI-POWERED XAPP SYSTEM. Technical Area The invention relates to radio access networks used in O-RAN (Open Radio Access Network) based access networks. It relates to resource management systems. Specifically, AI-powered xApp applications. a control mechanism that selects the dynamic waveform according to channel conditions via 10 It is related to. State of the Art In current O-RAN architectures, radio source management is typically based on static parameters or 15 This is done through predefined simple rule sets. In these systems, Even if channel conditions change momentarily, the connection between the base station and user equipment remains stable. Waveform selection is managed by a fixed algorithm. The use of a fixed waveform provides high efficiency. Signal loss for users with mobility issues or devices under heavy interference, and This leads to low data rate problems. Current AI applications generally use 20 focusing solely on traffic forecasting or energy saving, the physical layer It does not offer a decision-making mechanism that instantly optimizes its parameters. This situation, This leads to a decrease in spectral efficiency and connection drops, especially in extreme case scenarios. This is causing it to increase. Purpose of the Invention The invention describes adaptive waveform selection based on channel conditions in O-RAN-based access networks. The aim is to create a system that achieves this. The main purpose of the invention is to improve the signal-to-noise ratio. Using metrics such as SNR, latency, and Doppler shift, the optimal waveform can be determined. The aim is to determine CP-OFDM and DFT-s-OFDM by analyzing real-time data streams. or allows switching between specific modulation techniques. This enables high-performance modulation. Waveforms that minimize the Doppler effect are selected for users moving at high speed. Connection continuity is increased. The invention operates on a RIC (Radio Intelligent Controller). By processing channel state information (CSI) data via an xApp, the optimal waveform is determined. 35 It generates commands. Unlike existing technology, the invention increases not only the data speed, 2 By analyzing the physical characteristics of the channel with artificial intelligence, the physical layer parameters It changes dynamically. Figures that will help understand the invention. Figure 1 shows a general representation of the system that is the subject of the invention. Explanation of Part References 1: Channel Status Monitoring Module 10 2: Waveform Decision Mechanism 3: RIC Control Interface 4: O-DU (Distributed Unit) 5: O-RU (Radio Unit) 6: Model Update Unit 15 Detailed Description of the Invention The invention describes adaptive waveform selection based on channel conditions in O-RAN-based access networks. It is an AI-powered xApp system, Channel Status Monitoring Module (1), Waveform 20 Decision Mechanism (2), RIC Control Interface (3), O-DU (4), O-RU (5) and Model Update It includes the unit (6). Channel Status Monitoring Module (1), instant channel from O-RU (5) and O-DU (4) units It is the unit that collects and processes the signal-to-noise ratio (SNR). It continuously monitors metrics such as Doppler shift, delay propagation, and interference level. 25 The collected data were normalized in time series format to form Waveform Decision. The mechanism is transferred to unit (2). The Waveform Decision Mechanism (2) is the core decision unit of the system and is trained within it. It contains an artificial intelligence model. This module is the Channel Status Monitoring Module (1) It takes the metrics transmitted as input data. The decision-making mechanism, for example, high 30 When Doppler shift is detected, it corresponds to DFT-s-OFDM waveform, with low latency and high In the case of SNR, it makes a decision about transitioning to the CP-OFDM waveform. This decision is made using a scoring system. It is supported by an algorithm that provides a performance score for each waveform depending on the channel conditions. The waveform is calculated and the waveform with the highest score is selected. The decisions generated by the RIC Control Interface (3), Waveform Decision Mechanism (2) are O-35 This unit converts control commands into RAN standard-compliant commands. This interface uses the E2 protocol. 3 It transmits waveform change commands to the O-DU (4) unit via which user. The commands are transmitted to which user It includes specific parameters that determine which waveform to use for the equipment (UE). O-DU (4) is called a distributed unit and provides real-time radio resources. It is the unit responsible for its management. Commands received via the RIC Control Interface (3) By doing so, it updates the physical layer processing processes. O-DU (4) 5 suitable for the selected waveform. It performs modulation and encoding operations and transmits the data to the O-RU (5) unit. O-RU (5) is called a radio unit and converts digital signals into RF signals. It is a unit. Antennas receive the data in waveform format transmitted by O-DU (4). It enables transmission through the air. O-RU (5) also has a feedback mechanism. It sends the channel status information back to the Channel Status Monitoring Module (1). 10 The Model Update Unit (6) is the unit that enables the system's adaptive learning capability. The unit compares the actual performance of the selected waveform with the expected performance. If If the selected waveform cannot meet the target metrics (e.g., block error rate), this dataset will be rejected. The artificial intelligence model is retrained using the updated model Waveform Decision. The mechanism is loaded into unit (2). 15 The system operating scenario is as follows: The O-RU (5) unit receives from the user equipment. It analyzes the signals and transmits the channel status information to the Channel Status Monitoring Module (1). The Channel Condition Monitoring Module (1) extracts metrics such as SNR and Doppler shift from Waveform. The form is sent to the Decision Mechanism (2) unit. The Waveform Decision Mechanism (2) sends this It determines the most suitable waveform using metrics and makes this decision in the RIC Control Interface (3) 20 It transmits the decision to the unit. The RIC Control Interface (3) transmits the decision to the O-DU (4) unit via the E2 protocol. By transmitting it, it triggers a waveform change. O-DU (4) processes the signal with the new waveform and O- It sends to the RU (5) unit and O-RU (5) transmits this signal over the air. Model Update Unit (6), By analyzing performance data throughout this cycle, the artificial intelligence model is continuously optimized. It equals 25. 35

Claims

4 REQUESTS 1. Adaptive waveform selection based on channel conditions in O-RAN based access networks. It is an AI-powered xApp system; its features include: - Collecting real-time channel status information from O-RU (5) and O-DU (4) units and 5 Operating Channel Status Monitoring Module (1), - Optimal waveform using metrics transmitted by Channel Status Monitoring Module (1) Waveform Decision Mechanism (2) which determines the form, - Decisions generated by the Waveform Decision Mechanism (2) are standardized to O-RAN standards. RIC Control Interface (3), which converts into appropriate control commands, 10 - Physical layer processing by receiving commands from the RIC Control Interface (3) O-DU (4) updating its processes, - Receiving the data in waveform format transmitted by O-DU (4) and through the antennas O-RU (5) which enables transmission through the air and - 15 that updates the artificial intelligence model by analyzing the performance of the selected waveform. Model Update Unit (6) It includes.

2. According to Claim 1, adaptive waveforming in O-RAN based access networks according to channel conditions. It is an AI-powered xApp system that selects the form, and its feature is a signal-to-noise ratio of 20. Using metrics such as SNR (Surface-to-Noise Ratio), Doppler shift, delay propagation, and interference level, the most It includes the Waveform Decision Mechanism (2) which determines the appropriate waveform.

3. According to Claim 1, adaptive waveforming in O-RAN based access networks according to channel conditions. It is an AI-powered xApp system that selects the form, and its feature is high Doppler 25. When wavelength shift is detected, DFT-s-OFDM waveform, low latency and high SNR In this case, the Waveform Decision Switching Unit makes the decision to switch to the CP-OFDM waveform. The mechanism is that it includes (2).

4. According to claim 1, adaptive waveform 30 in O-RAN based access networks according to channel conditions. It is an AI-powered xApp system that selects the form, and its feature is the E2 protocol. RIC Control Interface transmitting waveform change commands to the O-DU (4) unit via (3) is included.

5. According to claim 1, adaptive waveform 35 in O-RAN based access networks according to channel conditions. It is an AI-powered xApp system that selects the form; its feature is that the selected wave artificial intelligence compares the actual performance of the form with the expected performance. It includes a Model Update Unit (6) that retrains the model.

6. According to Claim 1, adaptive waveforming in O-RAN based access networks according to channel conditions. It is an AI-powered xApp system that selects the waveform, and its feature is; 5 for each waveform. a performance score is calculated based on channel conditions and the wave with the highest score The form selected is the Waveform Decision Mechanism (2) which includes.

7. According to Claim 1, adaptive waveforming in O-RAN based access networks according to channel conditions. It is an AI-powered xApp system that selects the form and its feature is; 10 from O-RU (5) unit. Channel Status Monitoring Module receives channel status information via a feedback mechanism. (1) is included.

8. According to Claim 1, adaptive waveforming in O-RAN based access networks depends on channel conditions. It is an AI-powered xApp system that selects the waveform, and its feature is that the selected waveform is 15. When the target metrics are not met, the AI ​​model uses this dataset. It includes the Model Update Unit (6) which updates.

Images