System and method for user equipment mobility optimization in new radio dual connectivity networks
An AI/ML-based system at the MN in NR-DC networks predicts performance gains to optimize SN configurations, addressing dynamic network challenges and enhancing UE performance by aligning PSCell selection and traffic distribution.
Patent Information
- Application Number
- PCT/IN2026/050787
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-08
- Filing Date
- 2026-05-07
- Publication Date
- 2026-11-12
AI Technical Summary
Conventional rule-based mobility control mechanisms in 5G new radio dual connectivity (NR-DC) networks fail to adapt to dynamic network conditions, lacking predictive intelligence and contextual granularity for optimal UE performance, leading to suboptimal handovers and resource allocation.
Implementing an AI/ML-based inference model at the master node (MN) to predict performance gains, enabling real-time adjustments to secondary node (SN) configurations, such as addition or release, based on received performance data and traffic expectations, without altering existing RRC signaling flows.
Enhances mobility robustness and resource allocation, reducing unnecessary handovers and improving QoS by aligning PSCell selection and traffic distribution with AI/ML-driven predictions, thus optimizing UE performance in heterogeneous networks.
Smart Images

Figure IN2026050787_12112026_PF_FP_ABST
Abstract
Description
SYSTEM AND METHOD FOR USER EQUIPMENT MOBILITY OPTIMIZATION IN NEW RADIO DUAL CONNECTIVITY NETWORKSRESERVATION OF RIGHTS
[0001] A portion of the disclosure of this patent document contains material, which is subject to intellectual property rights such as, but are not limited to, copyright, design, trademark, Integrated Circuit (IC) layout design, and / or trade dress protection, belonging to Jio Platforms Limited (JPL) or its affiliates (hereinafter referred as owner). The owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent files or records, but otherwise reserves all rights. All rights to such intellectual property are fully reserved by the owner.TECHNICAL FIELD
[0002] The present disclosure relates generally to the field of communication systems. More particularly, the present disclosure relates to a system and a method for optimizing mobility of user equipment (UE) in a new radio-dual connectivity (NR-DC) network by using artificial intelligence / machine learning (AI / ML)-based techniques.DEFINITION
[0003] As used in the present disclosure, the following terms are generally intended to have the meaning as set forth below, except to the extent that the context in which they are used to indicate otherwise.
[0004] The term ‘NG-RAN’ as used herein in the specification refers to the Next Generation Radio Access Network. The NG-RAN is the radio access network component of a 5G communication system that provides wireless connectivity between a User Equipment and the core network. It typically includes base stations such as gNBs and is responsible for radio resource management, mobility handling, and forwarding signaling and data between the UE and core network functions.
[0005] The term “New Radio Dual Connectivity (NR-DC)” as used herein in the specification refers to a feature in 5G systems where the UE is simultaneously connected to two 5G NR base stations, namely a Master Node (MN) and a Secondary Node (SN), to enhance data throughput, reliability, and overall user experience.
[0006] The term “User Equipment (UE)”, as used herein, refers to a device utilized by an end user to communicate within a communication network. The UE may include, but is not limited to, a smartphone, tablet, modem, Internet of Things (loT) device, or any wireless communication device capable of performing channel measurements and reporting channel state information.
[0007] The term “Dual Connectivity (DC),” as used herein, refers to a radio access technology feature defined in 3 GPP specifications that allows user equipment (UE) to simultaneously connect to and communicate with two distinct network nodes, such as a master node (MN) and a secondary node (SN).
[0008] The term “Master Node (MN),” as used herein, refers to a primary radio access network (RAN) node responsible for managing a control plane connection with UE. The MN initiates and coordinates the establishment of dual connectivity by maintaining a main radio resource control (RRC) connection and interacting with a core network.
[0009] The term “Secondary Node (SN),” as used herein, refers to a radio access network (RAN) node that provides additional user plane resources to UE under the coordination of the MN. The SN is responsible for enhancing data throughput, coverage, or reliability by supplementary radio links that operate in parallel with the MN’s primary connection.
[0010] The term “internal radio access network (RAN) signaling,” as used herein, refers to communication or signaling that occurs within a RAN between network elements that make up the RAN, such as between the MN and SN.
[0011] The term “Radio Resource Control (RRC) signaling,” as used herein, refers to a protocol that manages communication between the UE and the network, handling tasks such as connection setup, mobility management (e.g., handovers), paging, and bearer configuration.
[0012] The term “bearer,” as used herein, refers to a logical data path used to carry user data between the UE and the core network, with specific quality of service (QoS) parameters to ensure optimal data transfer based on the type of service (e.g., voice, video, or general internet traffic).
[0013] The term ‘inter- node data sharing,” as used herein, refers to the exchange of information, intelligence, or control signals between two or more network nodes, such as the MN and the SN, to support coordinated decision-making, resource optimization, and seamless user experience in a distributed RAN.
[0014] The term “Primary Secondary Cell (PScell),” as used herein, refers to a main serving cell of the SN in a dual connectivity setup. It manages data transmission and control functions for the SN side.
[0015] The term “UE-associated signaling context,” as used herein, refers to signaling that is directly associated with a specific UE and its session or connection. The signaling messages in this context carry an identifier that links them to the UE.
[0016] The term “non-UE-associated signaling context,” as used herein, refers to signaling that is not tied to a specific UE. It typically involves control-plane procedures at the network level, such as configuration, capability exchange, or general status reporting between nodes.
[0017] The term “Reference Signal Received Power (RSRP)”, as used herein, refers to the linear average power level of a specific reference signal received by a UserEquipment (UE) from a serving cell, used as a key metric to evaluate signal quality, network coverage, and to identify coverage holes within the network.BACKGROUND
[0018] The following description of related art is intended to provide background information pertaining to the field of the disclosure. This section may include certain aspects of the art that may be related to various features of the present disclosure. However, it may be appreciated that this section be used only to enhance the understanding of the reader with respect to the present disclosure, and not as admissions of prior art.
[0019] The development of 5G new radio (NR) has introduced a dual connectivity (DC) framework, which allows user equipment (UE) to simultaneously connect to a master node (MN) and a secondary node (SN). This framework improves mobility robustness, enhances data throughput, and ensures better service continuity. In this framework, the MN manages radio resource control (RRC) signaling and bearer coordination, while the SN supplements user-plane resources to offload traffic and improve network performance.
[0020] As mobile networks become increasingly dynamic, especially in heterogeneous deployments, dense urban environments, and high-mobility scenarios, conventional rule-based mobility control mechanisms often fall short of sustaining optimal performance. These mechanisms rely on static thresholds and reactive handover triggers that may not adapt well to fluctuating radio conditions or traffic loads. In recognition of these challenges, 3 GPP previous standards identify artificial intelligence and machine learning (AI / ML) as promising technologies to enhance mobility decision-making within NR-DC.
[0021] However, NR-DC current standards offer limited procedural support for incorporating predictive intelligence or learning-based mechanisms into dualconnectivity operations. Data exchange between the MN and SN is primarily procedural and lacks the contextual granularity needed for intelligent coordination. For example, UE performance metrics collected at the SN are reported to the MN using a data collection reporting procedure via a data collection update message. Yet, a specification of measurement parameters and inclusion of contextual information such as actual or predicted primary secondary cell (PS Cell) identifiers remain insufficiently defined.
[0022] The procedures for collecting the UE performance metrics from the SN are further constrained by narrow exit conditions. For example, the collection of the UE performance metrics ceases either when a configured time duration elapses after NR-DC setup or when an NR-DC connection is released. These conditions may fail to capture transient performance degradations or short-term variations in link quality, especially in cases involving intermittent secondary cell group (SCG) activity or abrupt mobility events.
[0023] Additionally, the absence of explicit support for enhanced assistance information, such as predicted PSCell IDs from the MN or refined performance indicators in the UE performance information element, restricts the system’s ability to make proactive adjustments based on expected UE behavior. Hence, the existing NR-DC framework does not fully accommodate the complexity of modern network environments.
[0024] There is, therefore, a need in art to overcome the deficiencies of the prior arts.SUMMARY OF THE DISCLOSURE
[0025] In an exemplary embodiment, a method for optimizing mobility of user equipment (UE) in a new radio-dual connectivity (NR-DC) network is disclosed. The method includes steps of receiving, by a master node (MN), performance data associated with the UE from a secondary node (SN), predicting, by the MN and usingan inference model, performance gains expected to be achieved for the UE based on the received performance data and a configuration of the SN for handling at least a portion of a UE traffic, determining, by the MN, an action to be taken for a dual connectivity operation based on the predicted performance gains for the UE and transmitting, by the MN, a control information to the SN in response to the determined action for enabling the SN to configure the dual connectivity in accordance with the predicted performance gains for the UE.
[0026] In some embodiments, the performance data comprises at least one of radio link quality metrics, resource availability metrics, traffic load context, UE-specific mobility and history reports, latency and throughput reports, and interference and neighboring cell load.
[0027] In some embodiments, the method includes a step of enabling inter-node data sharing between the MN and the SN within the NR-DC network to enhance prediction of the performance gains by the inference model, where the inference model corresponds to a Machine Learning (ML) model.
[0028] In some embodiments, for predicting the performance gains, the method further includes a step of estimating, by the MN, a capability of the SN to handle the portion of the UE traffic based on the received performance data.
[0029] In some embodiments, for predicting the performance gains, the method further includes steps of generating, by the MN, a feature vector corresponding to the UE based on the received performance data, where the feature vector represents at least one of radio link conditions, resource availability, traffic load, and mobility characteristics associated with the UE and applying, by the MN, the inference model to the generated feature vector to obtain the predicted performance gains.
[0030] In some embodiments, the predicted performance gains include one or more of an expected throughput gain, an expected latency reduction, a reliabilityimprovement, a probability of quality of service (QoS) satisfaction, a mobility robustness improvement, and an energy efficiency improvement.
[0031] In some embodiments, for determining the action to be taken for the dual connectivity operation, the method includes steps of comparing, by the MN, the predicted performance gains with one or more predefined threshold conditions and determining, by the MN, whether the predicted performance gains satisfy the one or more predefined threshold conditions based on the comparison.
[0032] In some embodiments, the action to be taken for the dual connectivity operation includes at least one of: initiating, by the MN, an addition of the SN for the UE in association with the dual connectivity operation when the predicted performance gains satisfy the one or more predefined threshold conditions and maintaining, by the MN, a single connectivity when the predicted performance gains fail to satisfy the one or more predefined threshold conditions.
[0033] In some embodiments, upon adding the SN, the MN transmits control information to the SN for configuring one or more bearers and allocating radio resources for the dual connectivity operation.
[0034] In some embodiments, the control information includes an identifier (ID) of a predicted primary secondary cell (PSCell), an estimate of expected uplink and downlink traffic volume, and one or more quality of service (QoS) indicators.
[0035] In some embodiments, the method includes a steps of aligning, by the MN, a selected PSCell and an actual traffic volume handled by the SN with the corresponding predicted PSCell and expected traffic volume, to obtain the predicted performance gams.
[0036] In some embodiments, the method includes steps of monitoring, by the MN, actual performance gains achieved for the UE after the dual connectivity operation and updating, by the MN, the inference model based on the monitored performance gains
[0037] In some embodiments, the dual connectivity operation is performed using the inference model in accordance with an existing radio resource control (RRC) signaling procedure.
[0038] In another exemplary embodiment, a system for optimizing mobility of user equipment (UE) in a new radio-dual connectivity (NR-DC) network is disclosed. The system includes a master node (MN) configured to receive performance data associated with the UE from a secondary node (SN), predict, using an inference model, performance gains expected to be achieved for the UE based on the received performance data and a configuration of the SN for handling at least a portion of a UE traffic, determine an action to be taken for a dual connectivity operation based on the predicted performance gains for the UE and transmit a control information to the SN in response to the determined action for enabling the SN to configure the dual connectivity in accordance with the predicted performance gains for the UE.
[0039] In yet another exemplary embodiment, a computer program product comprising a non-transitory computer-readable medium is disclosed. The medium includes instructions that, when executed by one or more processors, cause the one or more processors to execute a method for optimizing mobility of user equipment (UE) in a new radio-dual connectivity (NR-DC) network is disclosed. The method includes steps of receiving, by a master node (MN), performance data associated with the UE from a secondary node (SN), predicting, by the MN and using an inference model, performance gains expected to be achieved for the UE based on the received performance data and a configuration of the SN for handling at least a portion of a UE traffic, determining, by the MN, an action to be taken for a dual connectivity operation based on the predicted performance gains for the UE and transmitting, by the MN, acontrol information to the SN in response to the determined action for enabling the SN to configure the dual connectivity in accordance with the predicted performance gains for the UE.
[0040] The foregoing general description of the illustrative embodiments and the following detailed description thereof are merely exemplary aspects of the teachings of this disclosure and are not restrictive.OBJECTIVES OF THE PRESENT DISCLOSURE
[0041] Some of the objectives of the present disclosure, which at least one embodiment herein satisfies, are as follows:
[0042] An objective of the present disclosure is to provide a method and system for optimizing mobility of a user equipment (UE) decisions in new radio-dual connectivity (NR-DC) networks using artificial intelligence and machine learning (AI / ML) techniques, with inference anchored at a master node (MN) to improve UE performance outcomes.
[0043] Another objective of the present disclosure is to utilize MN’s role as a central radio resource control (RRC) and bearer control anchor to host and execute AI / ML-based inference models, enabling it to predict the potential performance benefits of adding or switching a secondary node (SN) in real-time.
[0044] Yet another objective of the present disclosure is to enable MN to predict whether SN effectively serves a portion of traffic of UE, including expected traffic volume, thereby ensuring that dual connectivity activation is aligned with anticipated service needs and actual traffic distribution.
[0045] Another objective of the present disclosure is to ensure that SN is provided with both a predicted primary secondary cell (PSCell) identity and an estimated trafficvolume, enabling the SN to pre-allocate appropriate resources and align its configurations with AI / ML inference.
[0046] Yet another objective of the present disclosure is to enhance baseline NR-DC operation through AI / ML without modifying existing standardized radio resource control (RRC) procedure flows, thereby maintaining compatibility with current network deployments.
[0047] Another objective of the present disclosure is to improve the accuracy and effectiveness of MN-hosted ML inference by facilitating inter-node data sharing between the MN and SN over an Xn interface.
[0048] Yet another objective of the present disclosure is to improve user experience by aligning PSCell selection and traffic distribution strategies with AI / ML-driven performance predictions, minimizing a mismatch between expected and actual traffic patterns.
[0049] Other objects and advantages of the present disclosure will be more apparent from the following description, which is not intended to limit the scope of the present disclosure.BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWING
[0050] The accompanying drawings, which are incorporated herein, and constitute a part of this disclosure, illustrate exemplary embodiments of the disclosed methods and systems in which like reference numerals refer to the same parts throughout the different drawings. Components in the drawings are not necessarily to scale, emphasis instead being placed upon clearly illustrating the principles of the present disclosure. Some drawings may indicate the components using block diagrams and may not represent the internal circuitry of each component. It will be appreciated by those skilled in the art that disclosure of such drawings includes disclosure of electricalcomponents, electronic components or circuitry commonly used to implement such components.
[0051] FIG. 1 illustrates an exemplary network architecture implementing a system for optimizing mobility of a user equipment (UE) in a new radio-dual connectivity (NR-DC) network, in accordance with an embodiment of the present disclosure.
[0052] FIG. 2 illustrates an exemplary high-level system architecture for the NR-DC network, in accordance with an embodiment of the present disclosure.
[0053] FIG. 3 illustrates an exemplary block diagram of the system for optimizing the mobility of the UE in the NR-DC network, in accordance with an embodiment of the present disclosure.
[0054] FIG. 4 illustrates an exemplary sequence diagram of a secondary network (SN) addition preparation procedure for dual connectivity (DC) within the NG-DC network, in accordance with an embodiment of the present disclosure.
[0055] FIG. 5 illustrates an exemplary sequence diagram of a data collection reporting initiation procedure, in accordance with an embodiment of the present disclosure.
[0056] FIG. 6 illustrates an exemplary flowchart of a method implemented by the system for optimizing the mobility of the UE in the NR-DC network, in accordance with an embodiment of the present disclosure.
[0057] FIG. 7 illustrates an exemplary block diagram of a computer system in which or with which embodiments of the present disclosure may be implemented.
[0058] The foregoing shall be more apparent from the following more detailed description of the disclosure.List of reference numerals100 - Network Architecture102 - System104 - Network106 - User Equipment (UEs)108 -User(s)200 - System Architecture202 - Master Node (MN)204 - Secondary Node (SN)206 - User Equipment (UE)208 - Inference Model300 - Block diagram301 - Processor(s)302 - Memory304 -Interface(s)306 - Processing Engine308 - Database400, 500 - Exemplary sequence diagram600 - Method Flowchart700 - Computer System710 - External Storage Device720 - Bus730 - Main Memory740 - Read Only Memory750 - Mass Storage Device760 - Communication Port770 - ProcessorDETAILED DESCRIPTION
[0059] In the following description, for the purposes of explanation, various specific details are set forth in order to provide a thorough understanding of embodiments of the present disclosure. It will be apparent, however, that embodiments of the present disclosure may be practiced without these specific details. Several features described hereafter can each be used independently of one another or with any combination of other features. An individual feature may not address any of the problems discussed above or might address only some of the problems discussed above. Some of the problems discussed above might not be fully addressed by any of the features described herein. Example embodiments of the present disclosure are described below, as illustrated in various drawings in which like reference numerals refer to the same parts throughout the different drawings.
[0060] The ensuing description provides exemplary embodiments only, and is not intended to limit the scope, applicability, or configuration of the disclosure. Rather, the ensuing description of the exemplary embodiments will provide those skilled in the art with an enabling description for implementing an exemplary embodiment. It may be understood that various changes may be made in the function and arrangement of elements without departing from the spirit and scope of the disclosure as set forth.
[0061] Specific details are given in the following description to provide a thorough understanding of the embodiments. However, it will be understood by one of ordinary skill in the art that the embodiments may be practiced without these specific details. For example, circuits, systems, networks, processes, and other components may be shown as components in block diagram form in order not to obscure the embodiments in unnecessary detail. In other instances, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the embodiments.
[0062] Also, it is noted that individual embodiments may be described as a process that is depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of the operations may be re-arranged. A process is terminated when its operations are completed but could have additional steps not included in a figure. A process may correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When a process corresponds to a function, its termination can correspond to a return of the function to the calling function or the main function.
[0063] The word “exemplary” and / or “demonstrative” is used herein to mean serving as an example, instance, or illustration. For the avoidance of doubt, the subject matter disclosed herein is not limited by such examples. In addition, any aspect or design described herein as “exemplary” and / or “demonstrative” is not necessarily to beconstrued as preferred or advantageous over other aspects or designs, nor is it meant to preclude equivalent exemplary structures and techniques known to those of ordinary skill in the art. Furthermore, to the extent that the terms “includes,” “has,” “contains,” and other similar words are used in either the detailed description or the claims, such terms are intended to be inclusive like the term “comprising” as an open transition word without precluding any additional or other elements.
[0064] Reference throughout this specification to “one embodiment” or “an embodiment” or “an instance” or “one instance” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, the appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
[0065] The terminology used herein is to describe particular embodiments only and is not intended to be limiting the disclosure. As used herein, the singular forms “a”, “an”, and “the” are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms “comprises” and / or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. As used herein, the term “and / or” includes any combinations of one or more of the associated listed items. It may be noted that the terms “mobile device”, “user equipment”, “user device”, “communication device”, “device” and similar terms are used interchangeably for the purpose of describing the invention. These terms are not intended to limit the scope of the invention or imply any specific functionality or limitations on the described embodiments. The use of theseterms is solely for convenience and clarity of description. The invention is not limited to any particular type of device or equipment, and it may be understood that other equivalent terms or variations thereof may be used interchangeably without departing from the scope of the invention as defined herein.
[0066] While considerable emphasis has been placed herein on the components and component parts of the preferred embodiments, it will be appreciated that many embodiments can be made and that many changes can be made in the preferred embodiments without departing from the principles of the disclosure. These and other changes in the preferred embodiment, as well as other embodiments of the disclosure, will be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter is to be interpreted merely as illustrative of the disclosure and not as a limitation.
[0067] Aspects of this disclosure are directed to a system and method for optimizing user equipment (UE) mobility and dual connectivity (DC) operations in new radio (NR) networks (e.g., 5GNR-DC architectures) by leveraging artificial intelligence / machine learning (AIZML)-based inference at a master node (MN) to dynamically predict performance gains for UE. The system utilizes real-time performance data received from a secondary node (SN), along with contextual UE information, to infer whether adjustments to SN configuration, such as addition, removal, or modification, enhance network performance. Conventional approaches rely on static thresholds or predefined rule-based mobility triggers, which often fail to adapt effectively to complex and dynamic network conditions.
[0068] To address these challenges, the present disclosure introduces an intelligent decision-making framework where the MN applies a trained AI / ML model to estimate expected improvements (e.g., uplink / downlink throughput, latency reduction) resulting from a dual connectivity change. The MN then determines an appropriate DC action and issues control information to the SN accordingly, without altering standardizedradio resource control (RRC) signaling flows. This framework enables adaptive, data-driven mobility optimization, reduces reliance on rigid mobility rules, and improves resource allocation across heterogeneous network deployments. Unlike conventional approaches that lack predictive adaptation, the disclosed system reduces unnecessary handovers, enhances dual connectivity efficiency, and delivers a better quality of service (QoS) and user experience.
[0069] Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the accompanying drawings.
[0070] FIG. 1 illustrates an exemplary network architecture (100) implementing a system (102) for optimizing mobility of a user equipment (UE) (106) in a new radiodual connectivity (NR-DC) network, in accordance with an embodiment of the present disclosure.
[0071] As illustrated in FIG. 1, the network architecture (100) may include one or more user equipments (UEs) (106-1, 106-2, .... 106-N) associated with one or more users (108-1, 108-2, ...., 108-N) in an environment. A person of ordinary skill in the art will understand that one or more users (108-1, 108-2, ... 108-N) may be individually referred to as the user (108) and collectively referred to as the users (108). Similarly, a person of ordinary skill in the art will understand that one or more UEs (106-1, 106-2, ....106-N) may be individually referred to as the UE (106) and collectively referred to as the UEs (106). Although three UEs (106) are depicted in FIG. 1, however, any number of the user equipments (106) may be included without departing from the scope of the ongoing description. In an embodiment, each UE (106) may have a unique identifier attribute associated therewith. In an embodiment, the unique identifier attribute may be indicative of at least one of a Mobile Station International Subscriber Directory Number (MSISDN), International Mobile Equipment Identity (IMEI) number, an International Mobile Subscriber Identity (IMSI), a Subscriber Permanent Identifier (SUPI), and the like.
[0072] In an embodiment, the UE (106) may include smart devices operating in a smart environment, such as an Internet of Things (loT) system. In such an embodiment, the UE (106) may include, but is not limited to, smartphones, smart watches, smart sensors (e.g., mechanical, thermal, electrical, magnetic, etc.), networked appliances, networked peripheral devices, networked lighting system, communication devices, networked vehicle accessories, networked vehicular devices, smart accessories, tablets, smart television (TV), computers, a smart security system, a smart home system, other devices for monitoring or interacting with or for the users (108) and / or entities, or any combination thereof. A person of ordinary skill in the art will appreciate that the UE (106) may include, but is not limited to, intelligent, multi-sensing, network- connected devices that can integrate seamlessly with each other and / or with a central server or a cloud-computing system or any other device that is network-connected.
[0073] In an embodiment, the UE (106) may include, but is not limited to, a handheld wireless communication device (e.g., a mobile phone, a smart phone, a phablet device, and so on), a wearable computer device (e.g., a head-mounted display computer device, a head-mounted camera device, a wristwatch computer device, and so on), a Global Positioning System (GPS) device, a laptop computer, a tablet computer, or another type of portable computer, a media playing device, a portable gaming system, and / or any other type of computer device with a wireless communication capabilities, and the like. In an embodiment, the UE (106) may include, but is not limited to, any electrical, electronic, electro-mechanical, or an equipment, or a combination of one or more of the above devices, such as virtual reality (VR) devices, augmented reality (AR) devices, a laptop, a general-purpose computer, a desktop, a personal digital assistant, a tablet computer, a mainframe computer, or any other computing device. In addition, the UE (106) may include one or more in-built or externally coupled accessories including, but not limited to, a visual aid device such as a camera, an audio aid, a microphone, a keyboard, and input devices for receiving input from the user (108) or an entity such as touch pad, a touch enabled screen, an electronic pen, and the like. Aperson of ordinary skill in the art will appreciate that the UE (106) may not be restricted to the mentioned devices and various other devices may be used.
[0074] In FIG. 1, the UE (106) may communicate with the system (102) via a network (104) to send data to the system (102). In an embodiment, the network (104) may include at least one of a Fifth Generation (5G) network, a Sixth Generation (6G) network, or the like. In an embodiment, the network (104) is the NR-DC network that leverages dual connectivity (DC) to provide enhanced mobile network performance. The network (104) may enable the UEs (106) to communicate with other devices in the network architecture (100) and / or with the system (102). The network (104) may include a wireless card or some other transceiver connection to facilitate this communication. In another embodiment, the network (104) may be implemented as or include any of a variety of different communication technologies, such as a wide area network (WAN), a local area network (LAN), a wireless network, a mobile network, a Virtual Private Network (VPN), the Internet, the Public Switched Telephone Network (PSTN), or the like. In an embodiment, the network (104) may include, by way of example but not limitation, at least a portion of one or more networks having one or more nodes that transmit, receive, forward, generate, buffer, store, route, switch, process, or a combination thereof, etc. one or more messages, packets, signals, waves, voltage or current levels, some combination thereof, or so forth.
[0075] In an embodiment, the UE (106) may be communicatively coupled with the network (104). The system (102) may receive a connection request from the UE (106). The system (102) may send an acknowledgment of the connection request to the UE (106). The UE (106) may transmit a plurality of signals in response to the connection request. Once the connection is established, the system (102) may be configured to optimize the UE mobility process. A process for optimizing the UE mobility in the NR-DC network is explained in greater detail in conjunction with FIGS. 2-7.
[0076] Although FIG. 1 shows exemplary components of the network architecture (100), in other embodiments, the network architecture (100) may include fewer components, different components, differently arranged components, or additional functional components than depicted in FIG. 1. Additionally, or alternatively, one or more components of the network architecture (100) may perform functions described as being performed by one or more other components of the network architecture (100).
[0077] FIG. 2 illustrates an exemplary high-level system architecture (200) for the NR-DC network, in accordance with an embodiment of the present disclosure.
[0078] FIG. 2, with reference to FIG. 1, illustrates the high-level system architecture (200) that includes a master node (MN) (202), a secondary node (SN) (204) and the UE (206). In an example, the MN (202) may be a 5G NR gNB (Next Generation Node B) operating as a primary serving cell, whereas the SN (204) may be another gNB that operates on a higher-frequency band and is configured to provide additional data capacity to the UE (206) under NR-DC.
[0079] In an embodiment, the UE (206) may correspond to the UE (106) of FIG. 1. The MN (202) is configured to manage network connectivity and coordinate the DC for the UE (206). The MN (202) may serve as a centralized inference engine within a distributed radio access network (RAN) architecture that may be configured to support intelligent mobility optimization. The MN (202) is configured to perform a wide range of data collection, processing, and decision-making functions by aggregating real-time and historical mobility data received from the SN (204) and the UE (206).
[0080] In an exemplary embodiment, the MN (202) may be configured to collect and aggregate measurement reports from the UE (206), including signal strength indicators such as reference signal received power (RSRP), reference signal received quality (RSRQ), signal-to-interference-plus-noise ratio (SINR), and the like. In another exemplary embodiment, the MN (202) may be configured to ingest and analyze thehistorical mobility data, including UE handover trajectories, mobility state transitions, past dual connectivity activation / deactivation patterns, and the like. The MN (202) may also be configured to monitor network-level performance indicators, such as, but not limited to, handover success / failure rates, SN resource utilization, radio link reliability metrics, and the like. The MN (202) may also be configured to track quality of service (QoS) compliance for ongoing sessions and correlate with the impact of SN addition or release.
[0081] In an embodiment, as an anchor for radio resource control (RRC) signaling and bearer management in the NR-DC network, the MN (202) is uniquely positioned to host and execute an inference model (208) that guides mobility-related decisions. In such an embodiment, the MN (202) is configured to execute the inference model (208) (e.g., artificial intelligence / machine learning (AI / ML) model) that is trained to predict expected performance gains or risks associated with the SN addition, such as throughput improvement, latency reduction, or reduced handover failures.
[0082] By hosting the inference model (208) at the MN (202), latency in mobility-related decision-making is minimized, and decisions are harmonized across cells and interfaces managed by the MN (202). In an embodiment, an outcome of the inference model (208) may directly influence triggering conditions and thresholds for initiating the DC (e.g., SN addition based on predicted throughput gain or mobility robustness improvement), as well as the selection of optimal SN candidates. This predictive capability is beneficial in scenarios involving the UE (206) with high velocity, irregular mobility patterns, or in environments with dynamic interference conditions (e.g., urban macro deployments or cross-frequency DC).
[0083] In an embodiment, the inference model (208) is trained using a combination of the real-time and historical mobility data from multiple sources, including the UE (206), the MN (202), and the SN (204). The training process involves collecting the real-time and historical mobility data and pre-processing the data to ensure consistency,normalize values, and filter out noise. In an embodiment, various supervised learning techniques, such as, but not limited to, gradient boosting machines (GBMs), deep neural networks (DNNs), time series long short-term memory (LSTM), and so forth may be employed to train the inference model (208), where the inference model (208) learns to map input features (e.g., RSRP, traffic load, UE mobility history) to specific output variables (e.g., predicted throughput gain, latency reduction, or QoS satisfaction). During training, the inference model (208) may be optimized by minimizing a loss function that quantifies a difference between predicted outcomes and actual observed results. Once the inference model (208) is trained, the inference model (208) is deployed on the MN (202) to predict and optimize mobility-related decisions, such as when to trigger the DC (SN addition or release) based on the predicted performance gains or risks. Additionally, the inference model (208) may be periodically retrained with new data to refine its predictions and adapt to evolving network conditions.
[0084] In an embodiment, the SN (204) may be configured to provide supplementary radio resources to the UE (206), enhancing network throughput and coverage. The SN (204) operates in parallel with the MN (202), supporting the DC by establishing a secondary radio link with the UE (206). In an embodiment, the SN (204) may report performance metrics back to the MN (202) to assist in mobility optimization.
[0085] In an embodiment, the UE (206) is an end-user device that communicates with both MN (202) and SN (204). The UE (206) maintains a dual connectivity link, with one connection to the MN (202) and another to the SN (204), thereby benefiting from improved data throughput, coverage, and reliability. The detailed explanation of the interaction and message flows between the MN (202) and the SN (204) is provided in conjunction with FIGS. 3, 4 and 5.
[0086] FIG. 3 illustrates an exemplary block diagram (300) of the system (102) for optimizing the mobility of the UE (206) in the NR-DC network, in accordance with an embodiment of the present disclosure.
[0087] With reference to FIG. 2 and FIG. 1, FIG. 3, illustrates the system (102) that may include one or more processor(s) (301). The one or more processor(s) (301) may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and / or any devices that process data based on operational instructions. Among other capabilities, the one or more processor(s) (301) may be configured to fetch and execute computer-readable instructions stored in a memory (302) of the system (102). The memory (302) may be configured to store the computer-readable instructions or routines in a non-transitory computer-readable storage medium, which may be fetched and executed to create or share data packets over a network service. The memory (302) may include any non-transitory storage device including, for example, volatile memory such as randomaccess memory (RAM), or non-volatile memory such as erasable programmable readonly memory (EPROM), flash memory, and the like.
[0088] In an embodiment, the system (102) may include an interface(s) (304). The interface(s) (304) may include a variety of interfaces, for example, interfaces for data input and output devices (I / O), storage devices, and the like. The interface(s) (304) may facilitate communication through the system (102). The interface(s) (304) may also provide a communication pathway for one or more components of the system (102). Examples of such components include, but are not limited to, a processing engine (306) and a database (308).
[0089] In an embodiment, the processing engine (306) may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement one or more functionalities of the processing engine (306). In the examples described herein, such combinations of hardware and programming may beimplemented in several different ways. For example, the programming for the processing engine (306) may be processor-executable instructions stored on a non-transitory machine-readable storage medium, and the hardware for the processing engine (306) may include a processing resource (for example, one or more processors), to execute such instructions. In the present examples, the machine-readable storage medium may store instructions that, when executed by the processing resource, implement the processing engine (306). In such examples, the system (102) may include the machine-readable storage medium storing the instructions and the processing resource to execute the instructions, or the machine-readable storage medium may be separate but accessible to the system (102) and the processing resource. In other examples, the processing engine (306) may be implemented by electronic circuitry.
[0090] In an embodiment, the database (308) may store data that may be generated as a result of functionalities implemented by any of the components of the processor (301) or the processing engine (306). In an embodiment, the database (308) may be indicative of including, but not limited to, a relational database, a distributed database, a cloud-based database, or the like. In an embodiment, the database (308) may be configured to store the real-time and historical mobility data, radio-link metrics, traffic profiles, QoS indicators, and inference model outcomes to support master node-based mobility optimization in the NR-DC network.
[0091] In an exemplary embodiment, the processing engine (306) may include one or more units having functions that may include, but are not limited to, testing, storage, and peripheral functions, such as a wireless communication unit for remote operation and the like.
[0092] In an embodiment, the processing engine (306) may be configured to optimize the UE mobility in the NR-DC network. In an embodiment, optimizing the UE mobility in the NR-DC network refers to managing handover-related procedures for UEtransitioning between the MN (202) and the SN (204) in the DC architecture. The processing engine (306) may be configured to enable the coordination of handover signaling, context transfer (i.e., transfer of UE-specific state information), and measurement reporting across the MN (202) and SN (204). This allows the system (102) to dynamically evaluate and select optimal node associations, adjust mobility parameters (e.g., handover thresholds, timers, and priorities), and ensure continuity of service with reduced latency and radio link failure probability.
[0093] In one example, the processing engine (306) may be implemented as a mobility coordination module integrated within the MN (202). In implementation, the mobility coordination module may process mobility-related context and measurement reports received from the UE (206) or SN (204), identify optimal mobility strategy, and initiate mobility control signaling (e.g., dual connectivity control). By using the mobility coordination module, the MN (202) minimizes handover latency and signaling load by preemptively managing UE mobility decisions and by streamlining coordination with the SN (204). The mobility control signaling refers to message exchanges that enable seamless UE mobility within the NR-DC network. The message exchanges include RRC signaling, Xn interface messages, dual connectivity control procedures, and the like between the MN (202) and SN (204).
[0094] In another example, the processing engine (306) may be implemented as a mobility assistance module integrated within the SN (204). In implementation, the mobility assistance module within the SN (204) may evaluate UE measurement reports, monitor radio link quality metrics such as RSRP and SINR, and generate mobility-related data that is transmitted to the MN (202). By embedding the processing engine (306) within the SN (204), the SN (204) may locally process real-time radio conditions and contribute to the mobility decision-making. This integration reduces a computational burden on the MN (202), minimizes handover delays, and enhances responsiveness to dynamic radio environments.
[0095] In order to optimize the UE mobility in the NR-DC network, the MN (202) may first receive performance data associated with the UE (206) from the SN (204). In an embodiment, the MN (202) may receive the performance data through a communication interface, such as, but not limited to, a next-generation core interface (NG-C interface), X2 interface, Fl interface, N1 interface, and so forth. In an embodiment, the communication interface may be Xn interface. In another embodiment, the MN (202) may receive the performance data through an internal radio access network (RAN) signaling.
[0096] In an embodiment, the performance data may include, but is not limited to, radio link quality metrics, resource availability metrics, traffic load context, UE-specific mobility and history reports, latency and throughput reports, interference and neighboring cell load, and the like. In an embodiment, the MN (202) may also receive reports including results from prior DC activations, to support predictive and context-aware mobility decisions.
[0097] In an embodiment, the radio link quality metrics may be collected on a per-UE basis. The radio link quality metrics may include RSRP, RSRQ, and SINR measurements at the SN (204) for candidate primary secondary cell (PSCell), along with channel state information (CSI) feedback or channel quality indicator (CQI) reports from the UE (206) as received by the SN (204). The radio link quality metrics assist the MN (202) in estimating a potential data rate and link reliability.
[0098] In an embodiment, the resource availability metrics may include a current physical resource block (PRB) load on SN cells, scheduling capacity, buffer occupancy, and so forth. The resource availability metrics indicate whether the SN (204) has sufficient resources to support new traffic.
[0099] The traffic load context may include metrics, such as an active UE count and traffic load distribution at the SN (204), along with QoS profiles being supported.1These metrics help the MN (202) determine whether the SN (204) serves new flows without affecting the QoS. Furthermore, the UE-specific mobility and history reports may include data about past success or failure of SN addition / release procedures for the specific UE (206) or similar UEs, and dwell time statistics at SN cells. These metrics help predict session stability and guide future mobility actions.
[0100] The latency and throughput reports may include recent uplink and downlink throughput data for the UE (206) with similar radio conditions, as well as packet delay and jitter trends. These metrics provide insight into real-time performance and network responsiveness. Further, the interference and neighboring cell load metrics may include inter-cell interference levels and load balancing indicators from neighboring SN cells. These metrics help the MN (202) make informed decisions on interference avoidance and efficient load distribution.
[0101] In an embodiment, the optimization of UE mobility and DC configuration is further facilitated by inter-node data sharing of intelligence and data insights between the MN (202) and the SN (204). The inter-node data sharing may involve real-time exchange of the radio metrics, historical performance logs, context-aware mobility analytics, and so forth across the Xn interface or other suitable interfaces. This collaborative data sharing allows each node to contribute to the mobility prediction, resource planning, and DC decision-making. For example, the SN (204) may periodically transmit enriched UE measurement reports, historical throughput trends, local cell load forecasts, and the like to the MN (202), enabling more accurate inference of the expected performance gains. Likewise, the MN (202) may share mobility decision thresholds, model-generated feature vector templates, feedback results, and the like to synchronize learning strategies across the nodes. Such inter-node cooperation enhances consistency in decision-making, reduces redundant data collection efforts, and enables distributed intelligence across the RAN nodes for a more robust and adaptive NR-DC environment.
[0102] In an embodiment, the MN (202) applies the inference model (208) to predict the performance gains expected to be achieved for the UE (206), based on the received performance data and a configuration of the SN (204) for handling at least a portion of a UE traffic.
[0103] In an exemplary embodiment, the MN (202) may estimate a capability of the SN (204) to handle the portion of the UE traffic. Herein the term “capability” of the SN (204) refers to the ability of the SN (204) to effectively accommodate and handle the portion of the UE traffic while meeting desired performance objectives such as throughput, latency, and reliability. This capability is not limited to a single parameter but represents a composite measure derived from multiple factors, including radio link quality (e.g., Channel Quality Indicator (CQI), RSRP, etc.), available radio resources (e.g., Physical Resource Blocks), current load conditions, scheduling efficiency, and backhaul constraints. For example, an SN with high signal quality and low resource utilization may exhibit high capability and may offload a significant portion of traffic, whereas an SN experiencing high congestion or poor radio conditions may have limited capability, thus traffic offloading in these condition is less beneficial.
[0104] In an embodiment, the estimation of the capability of the SN (204) may be performed using received performance data. Initially, the MN (202) may collect relevant metrics such as CQI reports, throughput measurements, block error rates, resource utilization levels, and latency indicators. These inputs are then processed by the MN (202) using the inference model (208) to derive meaningful features, such as estimated spectral efficiency, available capacity, and traffic load trends. Subsequently, the inference model (208) applies either analytical computations or AI / ML-based prediction techniques to estimate the expected performance gains if the portion of UE traffic is offloaded to the SN (204). Based on this estimation, the MN (202) determines whether the SN (204) has sufficient capacity to handle the portion of the UE traffic.
[0105] In another exemplary embodiment, the MN (202) may apply the inference model (208) to generate or construct a feature vector for the UE (206) in order to predict the expected performance gains of the SN (204). The feature vector represents at least one of radio link conditions, resource availability, traffic load, and mobility characteristics associated with the UE (206). In an embodiment, the MN (202) feeds the feature vector “xu” into the inference model (208) to generate an inference of the expected performance gain, mobility robustness, or QoS enhancement resulting from the SN addition or release. Further, the MN (202) may construct the feature vector for each UE (206) by aggregating the real-time and historical mobility-related data. In an exemplary embodiment, the feature vector may include metrics represented as: xu = [RSPR SN, SINR_SN, PRB_SN, Load_SN, QoS_u, Traffic_u, Speed_u, PastThroughput u] .
[0106] The parameters RSPR_SN, SINR_SN, PRB SN, Load_SN, QoS_u, Traffic_u, Speed_u, and PastThroughput_u represent key inputs used by the MN (202) to evaluate the suitability and capability of the SN (204) for serving UE traffic.
[0107] RSPR SN represents to the measured received power level of the reference signals transmitted by the SN (202) at the UE (206). RSPR_SN indicates the strength of the radio signal from the SN (204) and is used to assess coverage quality, higher values that mply better signal conditions.
[0108] SINR SN represents the quality of the received signal at the UE (206) from the SN (204) by comparing the useful signal power against interference and noise.
[0109] PRB SN represents the amount of radio resources available at the SN (204). PRB SN reflects how much bandwidth can be assigned to the UE (206) for data transmission.
[0110] Load_SN indicates the current utilization level of the SN (204), typically expressed as a percentage of occupied resources (e g., PRB usage or active users). High load suggests congestion, which may limit the SN’s ability to serve additional traffic.
[0111] QoS u defines the service requirements associated with the UE’s traffic, such as latency, reliability, and priority. QoS u determines how traffic should be handled and scheduled.
[0112] Traffic u represents the current or predicted traffic demand of the UE (206), including data rate requirements, burst characteristics, and application type (e.g., video streaming, file download).
[0113] Speed u denotes the mobility speed of the UE (206). Higher speeds may lead to rapid channel variations and reduced stability of the SN link, affecting the effectiveness of dual connectivity.
[0114] PastThroughput u captures the historical throughput experienced by the UE (206) over a recent time window. PastThroughput u provides insight into past network performance and helps in predicting future achievable rates.
[0115] Based on the generated inference, the inference model (208) may output a set of predictive performance gains that help the MN (202) evaluate the expected performance gain of adding the SN (204). The predictive performance gains may include, but not be limited to, an expected predicted throughput gain (T gain), an expected latency reduction (L reduction), higher reliability (P robability), probability of QoS satisfaction (P qos), improved mobility robustness (M robustness), energy efficiency (E_gain), and the like.
[0116] The T gain represents an anticipated improvement in aggregated downlink and uplink throughput when the SN (204) is added to assist the MN (202). In anembodiment, the T gain is modeled using historical link budget data, traffic profiles, observed scheduling behavior, and so forth.
[0117] The L reduction reflects a projected decrease in transmission delay and jitter, especially in an uplink direction, if the SN (204) is geographically or topologically closer to the UE (206). This prediction may take into account differences in transmission time intervals (TTI) and hybrid automatic repeat request (HARQ) timings between MN (202) only and dual connectivity scenarios (i.e., MN (202) and SN (204) together).
[0118] The P robability estimates the probability of maintaining service continuity in challenging radio conditions, such as high mobility or fading environments. The P qos represents an estimate of how well UE’s QoS requirements can be met under dual connectivity compared to single connectivity.
[0119] The P qos helps decide whether important QoS traffic needs to be offloaded to the SN (204) for better service quality. The M_robustness estimates an expected reduction in radio link failures (RLF) or unsuccessful handovers, which may result from proactive SN anchoring based on historical handover trends, UE velocity, and prior RLF events.
[0120] The E gain captures a potential reduction in UE transmission power and signaling overhead achieved by offloading part of the traffic to the SN (204). This is useful for scenarios focused on optimizing power consumption and promoting sustainable network operations.
[0121] The MN (202) further determines an action to be taken for a DC operation based on the predicted performance gains for the UE (206). In an exemplary embodiment, the MN (202) may determine whether to initiate, defer, or skip the addition of the SN (204) for the DC operation based on the predicted performance gains generated by the inference model (208). The determination may involve evaluating thepredicted performance gain metrics against one or more decision threshold evaluation rules.
[0122] In one implementation, the MN (202) may compare the predicted performance gains with one or more predefined threshold conditions to assess whether the predicted performance gains of adding the SN (204) is justified. Based on the comparison the MN (202) determines whether the predicted performance gains satisfy the one or more predefined threshold conditions. In an embodiment, the action to be taken for the dual connectivity operation may include initiating an addition of the SN (204) for the UE (206) in association with the dual connectivity operation when the predicted performance gains satisfy the one or more predefined threshold conditions and maintaining a single connectivity when the predicted performance gains fails to satisfy the one or more predefined threshold conditions.
[0123] For example, if the T gain exceeds a first threshold value 0_T, or the P qos exceeds a second threshold value 0_Q, then the MN (202) may decide that initiating the SN addition is beneficial for the UE (206). On the other hand, if these predicted performance gain fall below the respective thresholds, the MN (202) may conclude that maintaining the single connectivity mode is preferable in order to conserve network and UE resources. The threshold-based decision logic ensures that the DC is selectively enabled only when it is likely to result in significant performance improvements, thus aligning with network efficiency and user experience goals. In an embodiment, the one or more predefined thresholds (e.g., 0_T and 0_Q) may be dynamically updated based on network policies, learned models, or historical performance outcomes to further refine the decision-making process.
[0124] Further, the MN (202) transmits a control information to the SN (204) in response to the determined action for enabling the SN (204) to configure the DC in accordance with the predicted performance gains for the UE (206). In an exemplary embodiment, upon determining that the DC is to be established based on the predictedperformance gains, the MN (202) may transmit the control information to the SN (204) to assist in configuring the DC setup tailored to the specific UE (206). The control information may include an identifier (ID) of a predicted primary secondary cell (PSCell) at the SN (204), an estimate of expected uplink and downlink traffic volume, one or more indicators associated with the QoS requirements of the UE (206), and the like. In an embodiment, the control information enables the SN (204) to perform configuration activities that align with the model-inferred traffic and QoS expectations for the UE (206).
[0125] In an embodiment, upon adding the SN (204), the MN (202) transmits the control information to the SN (204) for configuring one or more bearers and allocating radio resources for the DC operation. Based on the control information, the SN (204) may configure appropriate bearers, assign radio resources (e.g., PRBs), and align a selected PSCell and an actual traffic volume handled by the SN (204) with the corresponding predicted PSCell and expected traffic volume to obtain the predicted performance gains. Further, the SN (204) may initiate scheduling optimizations to ensure that the predicted performance gains (e.g., throughput increase, lower latency, improved reliability) are executed in practice. This coordinated configuration process may reduce latency in DC establishment and align network behavior with predictive analytics for improved user experience.
[0126] The MN (202) is further configured to monitor actual performance gains achieved for the UE (206) following the DC operation by the SN (204). The monitoring process may involve evaluating the extent to which the predicted performance gains have been realized in an operational environment. To facilitate this, the MN (202) may collect post-activation performance data from the SN (204) and the UE (206), including, but not limited to, actual uplink and downlink throughput, latency measurements, jitter levels, radio link stability, bearer-level QoS satisfaction, and so forth. The post-activation performance data may be compared with predictedperformance gain generated by the inference model (208) during the decision-making phase.
[0127] In one implementation, the MN (202) may log an outcome of each DC activation instance, classifying each DC activation instance as a success or failure based on one or more predefined performance thresholds. The logged outcomes may then be used as feedback data to refine or retrain the inference model (208), either in real-time (online learning) or as part of a scheduled batch process (offline learning). By continuously validating and updating the inference model (208) with real-world outcomes, the system (102) may improve its prediction accuracy over time, thereby enhancing the efficiency and effectiveness of future DC activation decisions for other UEs in similar contexts.
[0128] FIG. 4 illustrates an exemplary sequence diagram (400) of the SN addition preparation procedure for the DC within the NG-DC network, in accordance with an embodiment of the present disclosure.
[0129] With reference to FIG. 1, FIG. 2, and FIG. 3, FIG.4 illustrates the SN addition preparation procedure.
[0130] In an embodiment, the SN addition preparation procedure enables the MN (202) to initiate the addition of the SN (204) for the specific UE (206), by requesting resource allocation and context preparation.
[0131] In the inference model-assisted mobility optimization framework, the SN addition preparation procedure is enhanced by predictive analytics and data-driven decision-making, allowing the MN (202) to proactively identify beneficial dual connectivity opportunities.
[0132] In an embodiment, this enhancement is achieved through AI / ML-assisted analysis without altering standardized RRC procedure flows, thereby maintaining compliance with baseline NR-DC operation while improving decision accuracy.
[0133] The SN addition preparation procedure is essential for establishing the DC where the UE (206) simultaneously connects to two different next-generation node Bs (gNBs) to improve performance, reliability, and mobility handling.
[0134] In the illustrated sequence diagram (400), the SN addition preparation procedure is triggered by the MN (202), which determines, based on the predicted performance gain, that initiating the DC may benefit the UE (206). Accordingly, the MN (202) sends an “SN addition request” message (402) to the SN (204). This message (402) is UE-associated signaling and contains the control information for the SN (204) to allocate the resources and prepare for the DC setup.
[0135] Upon receiving the message (402), the SN (204) performs resource preparation, including the configuration of radio bearers, allocation of the PRBs, internal context setup for the UE (206), and so forth. The SN (204) responds with the “SN addition request acknowledge” message (404). This message (404) confirms a successful setup of the DC operation.
[0136] In an embodiment, the MN (202), upon receiving the message (404) may send an “RRC reconfiguration” message to the UE (206). The RRC configuration message enables the UE (206) to configure a secondary cell group (SCG) and establish a connection to the SN (204). Consequently, the UE (206) applies the configuration and confirms via an “RRC reconfiguration complete” message to the MN (202) about establishing the DC.
[0137] FIG. 5 illustrates another exemplary sequence diagram (500) of a data collection reporting initiation procedure, in accordance with an embodiment of the present disclosure.
[0138] With reference to FIG. 1, FIG. 2, FIG. 3, and FIG. 4, FIG.5 illustrates the data collection reporting initiation procedure. The data collection reporting initiation procedure facilitates the exchange of data reporting configurations between the MN (202) and SN (204) to support functions such as inference model-based analysis and network optimization.
[0139] The data collection reporting initiation procedure is initiated by the MN (202), which seeks specific data from the SN (204) in a non-UE-associated signaling context. The MN (202) sends a “data collection reporting initiation request” message (502), which specifies the type of data to be reported, such as radio performance statistics, traffic patterns, or measurement logs relevant to mobility, interference, or load conditions.
[0140] Upon receiving the message (502), the SN (204) responds with a “data collection reporting initiation response” message (504). The “data collection reporting initiation response” message (504) includes confirmation of acceptance or failure, along with optional parameters like reporting intervals or supported data formats. The data collected through this mechanism may be periodically transmitted to the MN (202), where it may be consumed by the inference model (208) for predictive mobility management.
[0141] FIG. 6 illustrates an exemplary flowchart of a method (600) implemented by the system (102) for optimizing the mobility of the UE in the NR-DC network, in accordance with an embodiment of the present disclosure.
[0142] FIG. 6, with reference to FIG. 1, FIG. 2, FIG. 3, FIG. 4 and FIG. 5 illustrates the method (600) for optimizing the UE mobility by using the MN (202) and the SN (204) of the system (108).
[0143] At step (602), the method (600) includes receiving, by the MN (202), the performance data associated with the UE (206) from the SN (204). In an aspect, theperformance data may include the performance data the radio link quality metrics, the resource availability metrics, the traffic load context, the UE-specific mobility and history reports, the latency and throughput reports, and the interference and neighboring cell load as described earlier in FIG.3
[0144] In an aspect, the method (600) enables the inter-node data sharing between the MN (202) and the SN (204) within the NR-DC network to enhance the prediction of the performance gains by the inference model (208), where the inference model (208) corresponds to the ML model.
[0145] At step (604), the method (600) includes predicting, by the MN (202) using the inference model (208), the performance gains expected to be achieved for the UE (206), based on the received performance data and the configuration of the SN (204) for handling at least the portion of the UE traffic. In an aspect, predicting the performance gains includes estimating, by the MN (202), the capability of the SN (204) to handle the portion of the UE traffic based on the received performance data.
[0146] In another embodiment, for predicting the performance gains, the method (600) further includes generating, by the MN (202), the feature vector corresponding to the UE (206) based on the received performance data where the feature vector represents the at least one of radio link conditions, resource availability, traffic load, and mobility characteristics associated with the UE (206), and applying, by the MN (202), the inference model (208) to the generated feature vector to obtain the predicted performance gains. The MN (202) feeds the generated feature vector into the inference model (208) to generate the inference of the expected performance gain. Based on the generated inference, the inference model (208) may output a set of predictive performance metrics such as the expected throughput gain, the expected latency reduction, the reliability improvement, the probability of quality of service (QoS) satisfaction, the mobility robustness improvement, and the energy efficiencyimprovement that help the MN (202) to evaluate the expected performance gains of adding the SN (204).
[0147] At step (606), the method (600) includes determining, by the MN (202), the action to be taken for the DC operation based on the predicted performance gains for the UE (206). In an aspect, the DC operation is performed using the inference model (208) in accordance with the existing RRC signaling procedure. In an aspect, for determining the action to be taken for the dual connectivity operation, the method (600) includes steps of comparing, by the MN (202), the predicted performance gains with the one or more predefined threshold conditions and based on the comparison, determining, by the MN (202), whether the predicted performance gains satisfy the one or more predefined threshold conditions.
[0148] In an aspect, the action to be taken for the dual connectivity operation includes the one of: initiating, by the MN (202), an addition of the SN (204) for the UE (206) in association with the DC operation when the predicted performance gains satisfy the one or more predefined threshold conditions or maintaining, by the MN (202), the single connectivity when the predicted performance gains fail to satisfy the one or more predefined threshold conditions.
[0149] At step (608), the method (600) includes transmitting, by the MN (202), the control information to the SN (204) in response to the determined action for enabling the SN (204) to configure the DC in accordance with the predicted performance gains for the UE (206). The control information includes at least one of the predicted PSCell, the expected traffic volume and the QoS indicators.
[0150] In an aspect, upon adding the SN (204), the MN (202) transmits the control information to the SN (204) for configuring the one or more bearers and allocating the radio resources for the DC operation.
[0151] In an aspect, the method (600) further includes aligning, by the MN (202), the selected PSCell and the actual traffic volume handled by the SN (204) with the corresponding predicted PSCell and expected traffic volume, to obtain the predicted performance gains.
[0152] In an aspect, the method (600) further includes steps of monitoring, by the MN (202), actual performance gains achieved for the UE (206) after the DC operation and updating, by the MN (202), the inference model (206) based on the monitored performance gains.
[0001] FIG. 7 illustrates an exemplary computer system (700) in which or with which embodiments of the present disclosure may be implemented.
[0002] As shown in FIG. 7, the computer system (700) may include an external storage device (710), a bus (720), a main memory (730), a read only memory (740), a mass storage device (750), a communication port (760), and a processor (770). A person skilled in the art will appreciate that the computer system (700) may include more than one processor (770) and communication ports (760). Processor (770) may include various modules associated with embodiments of the present disclosure.
[0003] In an embodiment, the communication port (760) may be any of an RS-232 port for use with a modem- based dialup connection, a 10 / 100 Ethernet port, a Gigabit or 10 Gigabit port using copper or fiber, a serial port, a parallel port, or other existing or future ports. The communication port (760) may be chosen depending on a network (104), such a Local Area Network (LAN), Wide Area Network (WAN), or any network to which the computer system (700) connects.
[0004] In an embodiment, the memory (730) may be Random Access Memory (RAM), or any other dynamic storage device commonly known in the art. Read-only memory (740) may be any static storage device(s) e.g., but not limited to, a Programmable ReadOnly Memory (PROM) chips for storing static information e.g., start-up or Basic Input / Output System (BIOS) instructions for the processor (770).
[0005] In an embodiment, the mass storage device (750) may be any current or future mass storage solution, which may be used to store information and / or instructions. Exemplary mass storage solutions include, but are not limited to, Parallel Advanced Technology Attachment (PATA) or Serial Advanced Technology Attachment (SATA) hard disk drives or solid-state drives (internal or external, e.g., having Universal Serial Bus (USB) and / or Firewire interfaces), one or more optical discs, Redundant Array of Independent Disks (RAID) storage, e.g., an array of disks (e.g., SATA arrays).
[0006] In an embodiment, the bus (720) communicatively coupled to the processor(s) (770) with the other memory, storage and communication blocks. The bus (720) may be, e.g., a Peripheral Component Interconnect (PCI) / PCI Extended (PCI-X) bus, Small Computer System Interface (SCSI), Universal Serial Bus (USB) or the like, for connecting expansion cards, drives and other subsystems as well as other buses, such a front side bus (FSB), which connects the processor (770) to the computer system (700).
[0007] Optionally, operator and administrative interfaces, e.g., a display, keyboard, joystick, and a cursor control device, may also be coupled to the bus (720) to support direct operator interaction with the computer system (700). Other operator and administrative interfaces may be provided through network connections connected through the communication port (760). Components described above are meant only to exemplify various possibilities. In no way should the aforementioned exemplary computer system (700) limit the scope of the present disclosure.
[0153] In an exemplary embodiment, the processor (770) is configured to execute the method for optimizing mobility of the UE in the NR-DC network. The method includes steps of receiving, by the MN, performance data associated with the UE from the SN,predicting, by the MN and using the inference model, the performance gains expected to be achieved for the UE based on the received performance data and the configuration of the SN for handling at least a portion of a UE traffic, determining, by the MN, the action to be taken for the dual connectivity operation based on the predicted performance gains for the UE and transmitting, by the MN, the control information to the SN in response to the determined action for enabling the SN to configure the dual connectivity in accordance with the predicted performance gains for the UE.
[0154] The present disclosure provides significant technical advancements in wireless communication systems by introducing an Artificial Intelligence / Machine Learning (AIZML)-driven inference framework anchored at a Master Node (MN) within a New Radio Dual Connectivity (NR-DC) architecture. Unlike conventional approaches that rely primarily on static signaling triggers such as PSCell configuration for Secondary Node (SN) addition, the present disclosure enables the MN serving as the Radio Resource Control (RRC) and bearer control anchor to perform predictive inference of UE performance metrics prior to DC activation. This inference-driven approach allows the MN to proactively determine the feasibility and expected performance gain of offloading traffic to the SN, thereby reducing unnecessary SN additions, minimizing signaling overhead, and improving overall spectral efficiency.
[0155] The present disclosure further enhances baseline NR-DC operation without introducing modifications to standardized RRC signaling procedures, ensuring backward compatibility while embedding intelligence into existing control mechanisms.
[0156] The present disclosure further improves system performance and resource utilization by enabling coordinated decision-making between the MN and SN through AI / ML-informed data exchange. Specifically, the MN generates and communicates both the predicted Primary Secondary Cell (PSCell) and the estimated UE traffic volume to be served by the SN, thereby allowing the SN to perform optimized resourceallocation and configuration aligned with inferred traffic demands. This coordinated mechanism overcomes limitations of conventional systems where reliance solely on PSCell signaling leads to suboptimal traffic distribution and degraded Quality of Experience (QoE).
[0157] Additionally, the present disclosure introduces inter-node data sharing capabilities to enhance inference accuracy at the MN, enabling continuous learning and adaptation based on network conditions and historical performance. By ensuring alignment between predicted and actual traffic handling, the system mitigates performance inconsistencies and enhances reliability. Furthermore, the integration of AI / ML-based inference with dynamic traffic steering significantly improves DC activation efficiency, reduces latency, and enables scalable deployment of intelligent radio access network optimization..
[0158] The method and system of the present disclosure may be implemented in a number of ways. For example, the methods and systems of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order for the steps of the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless specifically stated otherwise. Further, in some embodiments, the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
[0159] While considerable emphasis has been placed herein on the preferred embodiments, it will be appreciated that many embodiments can be made and that many changes can be made in the preferred embodiments without departing from the principles of the disclosure. These and other changes in the preferred embodiments ofthe disclosure will be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter is to be implemented merely as illustrative of the disclosure and not as a limitation.ADVANTAGES OF THE PRESENT DISCLOSURE
[0160] The present disclosure provides a system and a method for optimizing mobility decisions in new radio-dual connectivity (NR-DC) by utilizing artificial intelligence / machine learning (AI / ML) inference anchored at a master node (MN). This enables intelligent and predictive decision-making for an addition of s secondary node (SN), Primary Secondary Cell (PSCell) selection, and traffic offloading.
[0161] The present disclosure provides a system and a method for aligning SN addition and traffic offloading decisions with AI / ML-predicted user equipment (UE) performance gains. By estimating both the appropriate PSCell and expected traffic volume, the system ensures that dual connectivity is activated only when beneficial, thereby enhancing user experience and network efficiency.
[0162] The present disclosure enhances dual connectivity operation without requiring changes to standardized radio resource control (RRC) procedure flows. The inference-driven architecture retains compatibility with existing signaling protocols while incorporating AI / ML predictions to improve performance outcomes.
[0163] The present disclosure improves dual connectivity effectiveness by ensuring synchronization between predicted and actual traffic distribution across the MN and SN. This alignment prevents performance degradation that may arise from mismatched traffic handling.
[0164] The present disclosure enhances SN resource planning by providing SN with both a predicted PSCell ID and estimated UE traffic volume. This enables the SN toproactively allocate radio and scheduling resources, improving responsiveness and capacity utilization.
[0165] The present disclosure supports accurate and context-aware AI / ML inference at MN by enabling inter-node data sharing from the SN. This shared data allows an AI / ML model to learn from real-time and historical UE behavior, thereby improving prediction accuracy over time.
[0166] The present disclosure reduces signaling overhead and processing burden associated with frequent, rule-triggered DC activations. By enabling MN to infer SN addition or change based on anticipated performance benefits to UE, the system ensures that DC operations are executed only when aligned with actual traffic distribution and service needs.
Claims
We Claim:
1. A method (600) for optimizing mobility of user equipment (UE) (206) in a new radio-dual connectivity (NR-DC) network, the method (600) comprising:receiving (602), by a master node (MN) (202), performance data associated with the UE from a secondary node (SN) (204);predicting (604), by the MN (202) and using an inference model (208), performance gains expected to be achieved for the UE (206) based on the received performance data and a configuration of the SN (204) for handling at least a portion of a UE traffic;determining (606), by the MN (202), an action to be taken for a dual connectivity operation based on the predicted performance gains for the UE (206); andtransmitting (608), by the MN (202), a control information to the SN (204) in response to the determined action for enabling the SN (204) to configure the dual connectivity in accordance with the predicted performance gains for the UE (206).
2. The method (600) as claimed in claim 1, wherein the performance data comprises at least one of radio link quality metrics, resource availability metrics, traffic load context, UE-specific mobility and history reports, latency and throughput reports, and interference and neighboring cell load.
3. The method (600) as claimed in claim 1, comprising enabling inter-node data sharing between the MN (202) and the SN (204) within the NR-DC network to enhance prediction of the performance gains by the inference model (208), wherein the inference model (208) corresponds to a Machine Learning (ML) model.
4. The method (600) as claimed in claim 1, wherein predicting the performance gains comprises:estimating, by the MN (202), a capability of the SN (204) to handle the portion of the UE (206) traffic based on the received performance data.
5. The method (600) as claimed in claim 1, wherein predicting the performance gains further comprises:generating, by the MN (202), a feature vector corresponding to the UE (206) based on the received performance data, wherein the feature vector represents at least one of radio link conditions, resource availability, traffic load, and mobility characteristics associated with the UE (206); andapplying, by the MN (202), the inference model (208) to the generated feature vector to obtain the predicted performance gains.
6. The method (600) as claimed in claim 1, wherein the predicted performance gains comprise one or more of: an expected throughput gain, an expected latency reduction, a reliability improvement, a probability of quality of service (QoS) satisfaction, a mobility robustness improvement, and an energy efficiency improvement.
7. The method (600) as claimed in claim 1, wherein determining the action to be taken for the dual connectivity operation comprises:comparing, by the MN (202), the predicted performance gains with one or more predefined threshold conditions; anddetermining, by the MN (202), whether the predicted performance gains satisfy the one or more predefined threshold conditions based on the comparison.
8. The method (600) as claimed in claim 7, wherein the action to be taken for the dual connectivity operation comprises at least one of:initiating, by the MN (202), an addition of the SN (204) for the UE (206) in association with the dual connectivity operation when the predicted performance gains satisfy the one or more predefined threshold conditions; and maintaining, by the MN (202), a single connectivity when the predicted performance gains fail to satisfy the one or more predefined threshold conditions.
9. The method (600) as claimed in claim 8, wherein upon adding the SN (204), the MN (202) transmits control information to the SN (204) for configuring one or more bearers and allocating radio resources for the dual connectivity operation.
10. The method (600) as claimed in claim 1, wherein the control information comprises an identifier (ID) of a predicted primary secondary cell (PSCell), an estimate of expected uplink and downlink traffic volume, and one or more quality of service (QoS) indicators.
11. The method (600) as claimed in claim 10, further comprising:aligning, by the MN (202), a selected PSCell and an actual traffic volume handled by the SN with the corresponding predicted PSCell and expected traffic volume, to obtain the predicted performance gains.
12. The method (600) as claimed in claim 1, further comprising:monitoring, by the MN (202), actual performance gains achieved for the UE (206) after the dual connectivity operation; andupdating, by the MN (202), the inference model (208) based on the monitored performance gains.
13. The method (600) as claimed in claim 1, wherein the dual connectivity operation is performed using the inference model (208) in accordance with an existing radio resource control (RRC) signaling procedure.
14. A system (108) for optimizing mobility of user equipment (UE) (206) in a new radio-dual connectivity (NR-DC) network, the system (108) comprising:a master node (MN) (202) configured to:receive performance data associated with the UE (206) from a secondary node (SN) (204);predict, using an inference model (208), performance gains expected to be achieved for the UE (206) based on the received performance data and a configuration of the SN (204) for handling at least a portion of a UE traffic; determine an action to be taken for a dual connectivity operation based on the predicted performance gains for the UE (206); andtransmit a control information to the SN (204) in response to the determined action for enabling the SN (204) to configure the dual connectivity in accordance with the predicted performance gains for the UE (206).
15. The system (108) as claimed in claim 14, wherein the performance data comprises at least one of radio link quality metrics, resource availability metrics, traffic load context, UE-specific mobility and history reports, latency and throughput reports, and interference and neighboring cell load.
16. The system (108) as claimed in claim 14, comprising enabling inter-node data sharing between the MN (202) and the SN (204) within the NR-DC network to enhance prediction of the performance gains by the inference model (208), wherein the inference model (208) corresponds to a Machine Learning (ML) model.
17. The system (108) as claimed in claim 14, wherein to predict the performance gains, the MN (202) is configured to estimate a capability of the SN (204) to handle the portion of the UE traffic based on the received performance data.
18. The system (108) as claimed in claim 14, wherein to predict the performance gains, the MN (202) is configured to:generate a feature vector corresponding to the UE (206) based on the received performance data, wherein the feature vector represents at least one of radio link conditions, resource availability, traffic load, and mobility characteristics associated with the UE (206); andapply the inference model (208) to the generated feature vector to obtain the predicted performance gains, wherein the predicted performance gains comprise one or more of: an expected throughput gain, an expected latency reduction, a reliability improvement, a probability of quality of service (QoS) satisfaction, a mobility robustness improvement, and an energy efficiency improvement.
19. The system (108) as claimed in claim 14, wherein to determine the action to be taken for the dual connectivity operation, the MN (202) is confirmed to:compare the predicted performance gains with one or more predefined threshold conditions; anddetermine whether the predicted performance gains satisfy the one or more predefined threshold conditions based on the comparison,wherein the action to be taken for the dual connectivity operation comprises at least one of:initiate an addition of the SN (204) for the UE (206) in association with the dual connectivity operation when the predicted performance gains satisfy the one or more predefined threshold conditions; andmaintain a single connectivity when the predicted performance gains fail to satisfy the one or more predefined threshold conditions.
20. The system (108) as claimed in claim 19, wherein upon adding the SN (204), the MN (202) transmits control information to the SN (204) for configuring one or more bearers and allocating radio resources for the dual connectivity operation, and wherein the control information comprises an identifier (ID) of a predicted primary secondary cell (PSCell), an estimate of expected uplink and downlink traffic volume, and one or more quality of service (QoS) indicators.
21. The system (108) as claimed in claim 20, wherein the MN (202) is further configured to:align a selected PSCell and an actual traffic volume handled by the SN (204) with the corresponding predicted PSCell and expected traffic volume, to obtain the predicted performance gains;monitor actual performance gains achieved for the UE (206) after the dual connectivity operation; andupdate the inference model (208) based on the monitored performance gains, wherein the dual connectivity operation is performed using the inference model (208) in accordance with an existing radio resource control (RRC) signaling procedure.
22. A computer program product comprising a non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform a method (600) for optimizing mobility of user equipment (UE) (206) in a new radio-dual connectivity (NR-DC) network, the method (600) comprising:receiving (602), by a master node (MN) (202), performance data associated with the UE from a secondary node (SN) (204);predicting (604), by the MN (202) and using an inference model (208), performance gains expected to be achieved for the UE (206) based on the receivedperformance data and a configuration of the SN (204) for handling at least a portion of a UE traffic;determining (606), by the MN (202), an action to be taken for a dual connectivity operation based on the predicted performance gains for the UE (206); andtransmitting (608), by the MN (202), a control information to the SN (204) in response to the determined action for enabling the SN (204) to configure the dual connectivity in accordance with the predicted performance gains for the UE (206).
