Distributed Location Computing System in 5G and Beyond Networks

Abstract

This invention relates to a system (1) that performs location estimation and connection prediction optimization of user equipment (2) between 5G base stations in 5G, advanced 5G and 6G mobile telecommunication networks, in particular - Radio Access Network. System (1) proactively allocates network resources in the form of network slice, quality of service, security switches, power levels by estimating the location, direction of movement, speed and connection probability of user equipment (2) with artificial intelligence / machine learning based analysis.

Description

1 TARIFF Distributed Location Computing System in 5G and Beyond Networks Technical Area This invention has implications for 5G, advanced 5G, and 6G mobile telecommunications networks, particularly in RAN – Radia applications. Access Network - Radio Access Network between 5G base stations UE - User 5 Equipment - location estimation and connection estimation for user equipment. It is related to a system that performs optimization. State of the Art Today, location estimation in 5G networks is typically achieved through Location Management in the core network. The function is managed by the Location Management Function (LMF) component. Last 10 User equipment receives power from nearby base stations – gNodeBs, the power It measures parameters such as level and noise ratio and uses these parameters on a source basis with various triggers. reports to the station. Measurement reports indicate the signal strength of the neighboring cell to the service. It is triggered when the cell's performance exceeds a certain threshold value. In other words... The current handover mechanism is reactive, not proactive. Today, 15 A predictive, proactive approach is needed. Handover processes are initiated by the source base station and transferred to the target base station. Session information such as QoS, security, and carrier is shared via the Xn interface. However, Raw measurement data in RSSI / ToA / AoA format is directly transmitted between base stations. It is not shared. Instead, the source base station receives 20 from the user equipment. It uses the measurements to make the handover decision and communicates this decision to the target base station. When location calculations are required, there is a reliance on the core network. Open RAN specifications enable AI / ML integration on the Intelligent Controller. Although Ericsson suggests it, data sharing between base stations is not standard. Ericsson's AI / ML RAN in optimization white papers, for example "Leveraging AI for RAN optimization in 5G", 25 In 2020, AI is recommended for load balancing and mobility, but not for predictive positioning. Sharing measurements between base stations is not recommended. In current systems, location information is not shared between base stations, predicting the potential connection probability of user equipment and the network proactively optimizing resource allocation such as slice, service quality, security, and power, 30 2 In general, this makes predictive connection of user equipment impossible. Location The prediction is that core network dependency, as is currently the case, will remain low, especially for URLLC and V2X. This leads to inefficiencies in 5G / 6G applications that require low latency and high reliability. Including these measurements in base station resource planning will allow for a shorter timeframe. After a while, you can instantly see how many people will connect and where they are currently located. It is necessary to know this. The increasing positional accuracy and near real-time capability of low-end technologies like urLLC and V2X. It reduces latency and inefficiency in applications that require latency. For example, in 5G networks. It reduces handover delays. Traditional location estimation techniques are dependent on the core network and require real-time, distributed 10 It cannot be used by base stations in some way; this results in delay and inaccuracy. It causes problems. Patent application number US11109283B1, which is included in the prior art. The document describes using machine learning to estimate handover success rate. It is explained. The system described in the relevant application document outlines the handover procedures. 15 It makes predictions by clustering, but it also shares measurements between base stations. It does not perform. Patent application number US10841853B1, which is included in the prior art. The document describes AI-based load balancing in 5G. The relevant application... The system described in the document uses artificial intelligence to balance the load, but 20 It does not make location-based predictions. Patent application number US20220272538A1, which is included in the prior art. The document describes classifier-based message routing in telecommunications networks. The system that performs this function is explained. The system described in the relevant application document. It performs classifier-based routing, but requires AI / ML integration and NTN 25. It falls short in terms of support. In conclusion, solutions that address the needs described above are relevant to the subject. Due to its shortcomings, it has become necessary to make improvements in the relevant technical field. Brief Description of the Invention 3 The invention was created by drawing inspiration from existing situations and overcoming the aforementioned drawbacks. It aims to solve the problem. The aim of this invention is to provide solutions for 5G, advanced 5G, and 6G mobile telecommunications networks, particularly in RAN (Range-to-Range) systems. – Radio Access Network - UE between 5G base stations within the Radio Access Network - User Equipment - location estimation and connection estimation of user equipment. It is about developing a system that performs optimization. The system uses distributed and centralized components to support devices like mobile phones. User equipment Received Signal Strength Indicator – RSSI, Measurements such as Time of Arrival (ToA) and Angle of Arrival (AoA) It shares via the Xn interface. The system uses artificial intelligence / machine learning-based analysis. 10 with the user equipment's location, direction of movement, speed, and connection possibility. by estimating the network segment, quality of service, security switches, and power levels. It proactively allocates its resources. The system, Vehicle to Everything - V2X – Vehicle to Everything, communication, smart cities, factories, Autonomous driving and Integrated Sensing and Communication - ISAC - Integrated Sensing and 15 It can be used in communication scenarios. The system is particularly Ultra Reliable. Low Latency Communication - For applications requiring uRLLC and high reliability, RAN-based distributed location computing by reducing reliance on traditional core networks. This enables increased processing power (edge ​​computing) and Artificial Intelligence (AI). Thanks to its intelligence capabilities, it enables real-time processing at edge nodes. It is recognized. This creates a bridge to the expected ISAC integration in 6G. The structural and characteristic features and all the advantages of the invention are given in the figures below. This becomes clearer thanks to the detailed explanation written with references to these figures. This will be understood, and therefore the evaluation should also take these forms and detailed explanations into consideration. It needs to be done by taking precautions. 25 Figures that will help understand the invention. Figure 1 is a schematic representation of the general structure of the system that is the subject of the invention. Figure 2 is a schematic representation of the detailed structure of the system that is the subject of the invention. Explanation of Part References 1. System 30 4 2. User equipment 3. Source: gNodeB 4. Target gNodeB 5. Analysis module 6. Vector database 5 7. In-memory database 8. Permanent database 9. Core network A. DU - Distributed Unit B. CU - Centralized Unit - Central Unit 10 Detailed Description of the Invention In this detailed description, the preferred configurations of the system (1) that is the subject of the invention are only This is explained to facilitate a better understanding of the subject. This invention has implications for 5G, advanced 5G, and 6G mobile telecommunications networks, particularly in RAN – Radia applications. Access Network - Radio Access Network between 5G base stations UE - User 15 Equipment - location estimation and connection estimation of user equipment (2). It is related to a system (1) that performs optimization. The system (1) uses distributed unit and central unit components, such as mobile phones, etc. (2) Received Signal Strength – RSSI - Received Signal Strength of user equipment Indicator, Time of Arrival – ToA, Angle of Arrival – AoA, etc. 20 It shares its measurements via the Xn interface. The system (1) uses artificial intelligence / machine learning. with base analysis the position, direction of movement, speed and (2) of the user equipment. By estimating the likelihood of connection, consider network slice, quality of service, security switches, and power. It proactively allocates network resources in the form of levels. The system subject to the invention (1); 25  mobile devices such as phones, IoT devices, and autonomous vehicles, the signal received Collecting data, Received Signal Strength – RSSI Indicator and / or Time of Arrival – ToA - Time of Arrival and / or Angle of Arrival – AoA - Angle of Arrival measurements are performed on the Non-Terrestrial Network (NTN). The 5 fundamental principles of position estimation that are critical for terrestrial network Doppler shift. User equipment that provides data (2),  a base station that takes the measurements performed by user equipment (2), The handover decision was initiated by local prediction with edge AI. performing, measuring and base station location data via the new Xn interface Shared via XnAP message, the new message 10 is critical for ensuring low latency. gNodeB (3), which adds Doppler data with type,  Shared measurement and neighboring base station data from source base gNodeB (3) Xn a base station that receives the signal via its interface, calculates the probability of connection Network slice in case of high connection probability, Quality of Service - QoS, security, Target gNodeB (4), 15 which provides resource allocation in the form of power.  Analyzing shared data at DU (A) / CU (B) level, based on the data In the analysis it performs, it estimates the position / velocity / Doppler, and in this process... Predicting the possibility of connecting user equipment (2) and the direction of movement proactively allocating resources in the target gNodeB (4) 20 that provides optimization and is updated with federated learning. artificial intelligence / machine learning - AI / ML analysis module (5),  For storing measurement vectors, fast querying, and machine learning training. vector database used, which increases accuracy with historical data (6),  Provides real-time data access, reduces latency, and is Ultra Reliable Low Delayed Communication - providing real-time prediction for uRLLC - in-memory data 25 base (7),  Long-term storage, historical analysis and legal monitoring - Lawful Interception - LI for GSM - Global System for Mobile Communications - Mobile Data protection compliant with Global System for Communications standards. permanent database (8) and 30  Traditional Location Management Function LMF (Landfill Multi-Terrestrial Network) or Non-Terrestrial Network Gateway AMF - Aether Management Platform - Aether, which provides integration with the gateway. Management Platform / UPF - User Plane Function core network (9) 35 6 It includes. In the system (1) measurement collection process, user equipment (2), RSS and / or ToA and / or performs AoA measurements and reports them to the source gNodeB (3). Source gNodeB (3) measures and neighboring gNodeB locations via Xn interface Shares with target gNodeB (4). 5 The AI / ML analysis module (5) extracts the shared data from the vector database (6). This data by analyzing the possibility of connecting user equipment (2) and predicting the direction of movement does. Target gNodeB (4) allocates resources and data in case of high connection probability. It writes to the in-memory database (7). Thus, the security key and power / energy optimization 10 It is applied. Vector database (6), in-memory database (7) and persistent databases (8) read and Writing operations are performed. In this way, measurement data and prediction results are stored. Integration with the core network (9) is provided. System (1), Vehicle to Everything - V2X – Vehicle to Everything communication, smart cities, 15 factories, autonomous driving and Integrated Sensing and Communication - ISAC - Integrated Sensing and Communication It can be used in communication scenarios. The system (1) is especially Ultra Reliable. Low Latency Communication - For applications requiring uRLLC and high reliability, RAN-based distributed location computing by reducing dependence on traditional core network (9) This enables increased processing power (edge ​​computing) and Artificial Intelligence - AI - Artificial 20 Thanks to its intelligence capabilities, it enables real-time processing at edge nodes. It is recognized. This creates a bridge to the expected ISAC integration in 6G. The working principle of the system (1) is shown in general in Figure-1 and in detail in Figure-2. It has been shown. The invention relates to the positioning structures and network (9) in the existing 5G core network. It can work in a supportive manner with slicing capabilities. In the system (1) close 25 In line with planned future NTN and Open RAN capabilities. A functional structure has been proposed. The invention provides an AI-based predictive connectivity and resource allocation system between base stations. The allocation system (1) provides each resource gNodeB (3), the exact location of neighboring base stations. It knows the coordinates and the user 30 who is connected or potentially able to connect. 7 They share the measurements of their equipment (2) with the target gNodeBs (4) via the Xn interface. The analysis module (5) analyzes this data through AI / ML models and enables the user The analysis estimates the connection probability, direction of movement and speed of the equipment (2). module (5) pre-allocates resources on the target gNodeB (4) as a result of this analysis. This method eliminates the need for traditional position estimation techniques. and is used in applications such as smart cities, autonomous vehicles, and traffic management. available. Solution Characteristics: Data Sharing Between Base Stations:  Base stations send a new XnAP message via the Xn interface. 10 Measurements (time, RSS, ToA) by type (e.g., PredictiveAttachmentData). and / or AoA) shares. gNode1, time t1: RSSI: -70, ToA: 100 ns, AoA: 45°, position1: {x: 10, y: 5} o gNode1, moment t2: RSSI: -65, ToA: 95 ns, AoA: 50°, pos2: {x: 12, y: 6} o gNode2, time t1: RSSI: -90, ToA: 1000 ns, AoA: 45°, position1: {x: 10, y: 5} 15 o gNode2, moment t2: RSSI: -35, ToA: 50 ns, AoA: 50°, pos2: {x: 12, y: 6} The time to connect to the second base station – T3 – is estimated. All base stations are connected at this T3. They can calculate their own estimated loads. AI / ML Based Prediction:  Each base station has 20 operating systems running at DU or CU level depending on CPU / GPU power. a distributed AI / ML model, shared time, analyzes RSS / ToA / AoA data. by determining the direction of UE's movement, its speed, and its connection to the target base station. It calculates the probability. For example, the result would be an 85% probability of connecting to gNode2. available. Prospective Resource Allocation: 25  When a high probability of connection is detected, the target base station is pre-selected. It allocates resources. These resources are, for example, used for preparing the relevant Network Slice. beamforming, power control, safety switch adjustment, It is in the form of sharing. 8  This reduces delays in the handover process and facilitates processes like URLC and V2X. This provides suitable conditions for applications requiring low latency. Eliminating the Requirement for Location Estimation:  The location of each device is directly linked to data shared between base stations. It is estimated that 5G Core and Location Management Function will be included. Dependence on the Management Function decreases.  In addition, Lawful Interception (LI) is used for traffic management in smart cities and It provides fast and accurate position estimation for low-latency communication in NTNs. 6G, Smart Cities:  Proposed method for ISAC, position and motion information detection in 6G networks 10 By combining these capabilities, it provides more accurate predictions.  In smart cities, predicting the movements of autonomous vehicles will help manage network resources. It optimizes things like traffic light coordination and vehicle routing.

Claims

9 REQUESTS 1. Mobile devices such as phones, IoT devices, and autonomous vehicles receive signal data. collecting, received signal strength and / or arrival time and / or arrival angle measurements performing position estimation, which is critical for Doppler shift in non-terrestrial networks. 5G, advanced 5G and 6G mobile 5 having user equipment (2) that provides basic data 5G base stations in telecommunication networks, particularly within the Radio Access Network. among user equipment (2) location estimation and connection estimation It is a system (1) that performs optimization, and its feature is;  a base station that takes the measurements performed by user equipment (2), initiating the handover decision, using cutting-edge artificial intelligence to make local predictions, measurements and 10 Base stations transmit location data via the Xn interface using a new XnAP message. Sharing, critical for ensuring low latency, new message type with Doppler gNodeB (3), the source that adds the data,  Shared measurement and neighboring base station data from source base gNodeB (3) Xn a base station that receives through its interface, calculates the probability of connection and 15 Network slice, quality of service, security, power in situations with high connection probability. Target gNodeB (4) which provides resource allocation in this way,  Analyzing shared data at DU (A) / CU (B) level, based on the data In the analysis it performs, it estimates the position / velocity / Doppler, and in this process... Predict the connection probability and direction of movement of user equipment (2) 20 proactively allocating resources in the target gNodeB (4) artificial intelligence / machine that provides optimization and is updated with federated learning learning analysis module (5),  For storing measurement vectors, fast querying, and machine learning training. vector database used, which increases accuracy with historical data (6), 25  Provides real-time data access, reduces latency, and is ultra-reliable low-frequency. In-memory database providing instant prediction for delayed communication (7),  Mobile for long-term storage, historical analysis and legal monitoring. Data protection compliant with Global System for Communications standards. permanent database (8) and 30  complements the traditional Location Management Function or non-terrestrial network Aether management platform / user plane that provides integration with the gateway. Core network (9) which is a functional feature It includes. 35

Images