Method for mobility-optimized handover

Abstract

A method and a system for handover, particularly for random access channel-less (RACH-less) handover, in a wireless communication system, comprising: configuring a user equipment (UE) for performing movement estimation and / or trajectory prediction with respect to one or more candidate or target cells or beams; defining measurement and / or prediction events and associated thresholds performing, by the UE, movement estimation and / or trajectory prediction based on the received configuration; calculating, by the UE, a probability of the UE entering at least one of the candidate or target cells or beams; comparing, by the UE, the calculated probability to a configured threshold; and performing, by the UE, a timing advance (TA) estimation for the respective candidate or target cell(s) or beam(s) if the calculated probability exceeds the configured threshold.

Description

[0001] 202404977

[0002] 1

[0003] Description

[0004] Method for mobility-optimized Handover

[0005] The invention relates to wireless communication systems, and more specifically to methods, systems, and devices for improving mobility management and enabling optimized handover, particularly random access channel-less (RACH-less) handover, procedures, particularly under high-mobility conditions, using user equipment (UE)-side movement estimation, trajectory prediction, and timing advance (TA) estimation.

[0006] In legacy cellular handover schemes, such as those standardized in LTE and 5G NR, the UE performs radio measurements on downlink reference signals received from the network, such as reference signal received power (RSRP) and reference signal received quality (RSRQ). These measurements are reported to the network, specifically to the serving base station (source gNB), e.g., depending on configured triggering conditions or time intervals.

[0007] Based on the reported measurement results, the source gNB determines whether handover criteria are fulfilled. If so, it initiates a handover procedure by transmitting a handover request to a target gNB.

[0008] However, this process introduces inherent temporal misalignment: the UE’s measurements are taken at the time the reference signals are received, but the source gNB makes the final handover decision later, after processing the reports and coordinating with the target gNB. This gap can be critical, particularly in mobile UE scenarios where the UE position and radio conditions can change significantly within milliseconds, such as in vehicular or high-speed train scenarios.

[0009] In general, three types of AI / ML models that can improve network management and mobility handling: Network-side models, UE-side models, and Joint UE and gNB-side models.

[0010] Importantly, since Layer 3 (L3) mobility management relies on measurements made by the UE, it is increasingly recognized that UE-side AI / ML models deserve 202404977

[0011] 2 focused study and analysis for potential performance gains in mobility management, particularly for L3 enhancements. The mobility management at L3 can be divided into two main types: beam-level mobility and cell-level mobility.

[0012] Beam-level mobility is handled entirely within the lower protocol layers, specifically in the medium access control (MAC) layer and the physical layer. It is essentially identical to beam management, where the device switches between beams under the control of the same cell, without requiring higher-layer signalling.

[0013] In contrast, cell-level mobility requires higher-layer signalling, particularly at the radio resource control (RRC) layer, because it involves changing the serving cell. This type of mobility triggers a full handover procedure, where the network transfers the device’s connection from one base station or cell to another.

[0014] To support this, the device is configured with specific measurement tasks. It is instructed to perform measurements on reference signals from candidate cells, apply filtering to these measurements, and report the results to the network based on event-triggered conditions. Because the network controls the handover decisions, it maintains awareness of the device’s location at the cell level. This allows the network to decide when the device should be handed over to a new cell, based on the reported measurement data and defined handover criteria.

[0015] In the context of 5G New Radio (NR), the reporting of measurement results by the user equipment is governed by the event-triggered mechanisms defined in the 3GPP RRC specifications. These so-called NR measurement events serve as conditions under which the UE transmits measurement reports to the network, allowing the network to make informed mobility and handover decisions. The defined events include Event A1 , where the serving cell becomes better than a defined threshold; Event A2, where the serving cell becomes worse than a threshold; Event A3, where a neighbouring cell becomes offset better than the serving cell (SpCell); Event A4, where a neighbour becomes better than a threshold; and Event A5, where the serving cell becomes worse than one threshold while a neighbour becomes better than another. Additionally, Event A6 relates to situations where a secondary neighbour cell (SCell) becomes offset better than another. 202404977

[0016] 3

[0017] For inter-RAT (Radio Access Technology) scenarios, two events are defined: Event B1 , which occurs when the measured signal from a neighbouring inter-RAT cell (such as a 4G cell) exceeds a threshold after applying an inter-RAT offset, and Event B2, which is triggered when the primary cell (PCell) measurement drops below a threshold while the inter-RAT measurements exceed a second threshold. These event-based mechanisms ensure that the UE only sends measurement reports when meaningful changes are detected, minimizing signalling overhead and enabling the network to maintain efficient and timely control over mobility decisions.

[0018] Handover procedures in mobile networks are typically triggered by changing radio conditions as the UE moves through the network. As the UE’s connection with the current (source) cell weakens, the network monitors whether specific thresholds are reached that indicate degraded service quality. Once such thresholds are exceeded, the network can decide to hand over the UE to another cell that offers better radio conditions.

[0019] The network’s awareness of these conditions relies on the continuous measurement reports provided by the UE. For example, in the case of a 5G NR Event A3 scenario, the handover decision is based on the condition that a neighbouring (target) cell has become better than the source cell by a defined offset value, and that this condition has been sustained for at least a certain duration, known as the time-to-trigger (TTT). This helps ensure that transient fluctuations in signal strength do not lead to unnecessary handovers, maintaining a balance between connection quality and network stability.

[0020] In typical mobility scenarios, the network triggers the handover procedure based on measurement events reported by the UE. For example, events such as A3 and A5 are commonly used to trigger coverage-based mobility, while A4 events can be employed to support load balance-based mobility decisions. Traditionally, these procedures rely on reactive mechanisms where the network acts once specific event conditions are fulfilled and reported.

[0021] With the evolution of 5G NR, however, there is growing interest in shifting toward predictive mobility management, particularly through the use of artificial intelligence (Al) and machine learning (ML). The intention behind measurement event prediction 202404977

[0022] 4 is to anticipate when and which measurement event will likely be fulfilled, introducing a temporal dimension into mobility control. This predictive approach enables the network and the UE to act proactively, reducing latency and improving the reliability of handovers.

[0023] Timing Advance (TA) is a critical mechanism in 5G NR networks that enables the UE to adjust the timing of its uplink transmissions so that they align precisely with the network’s expected reception window. This adjustment ensures that uplink transmissions from different UEs arrive at the base station (gNB) in a synchronized manner, avoiding interference and maintaining uplink efficiency. The TA value is calculated based on multiple components, including the measured propagation delay between the UE and the network, as well as predefined offsets such as NTA, offset, or derived from system information (e.g., SIB19).

[0024] In particular, non-terrestrial networks (NTN), such as those involving GEO or LEO satellite links, introduce significantly longer propagation delays due to the extended round-trip times over satellite links. To compensate for this, the network applies additional offsets - including common TA adjustments and UE-specific adjustments - to ensure proper frame alignment between downlink (DL) and uplink (UL) at the synchronization reference point. The total TA value, expressed as incorporates these various factors, where Tcis the basic time unit, and the specific offsets account for both cell and system configuration aspects.

[0025] In the context of RACH-less handovers, accurate and timely estimation of the expected TA at the target cell becomes even more critical, particularly under high-mobility or NTN scenarios. Without proper pre-compensation, the target cell may allocate uplink resources that the UE cannot use effectively, leading to resource waste or handover failures. Integrating AI / ML-based mechanisms at the UE to predict TA values proactively - alongside movement estimation and trajectory prediction - offers significant potential to optimize handover performance and improve overall service continuity. 202404977

[0026] 5

[0027] The evolution of non-terrestrial networks (NTNs) plays an increasingly important role in extending 5G New Radio (NR) coverage beyond traditional terrestrial infrastructures. NTNs involve the integration of satellite systems or high-altitude platform stations (HAPS) with terrestrial networks, enabling connectivity over wide geographic areas, including remote and underserved regions. The 3GPP specifications define a “transparent” or “bent pipe” architecture, where the satellite or HAPS serves as a relay, simply converting the frequency of the uplink radio signal, filtering, and amplifying it before transmitting it back on the downlink. In this architecture, the intelligence and radio access network (RAN) functions are handled entirely on the ground, typically at a gateway linked to the 5G RAN and 5G core network (CN).

[0028] In contrast, a “non-transparent” or “regenerative” architecture is characterized by a satellite or HAPS that is equipped with onboard processing capabilities. Here, the satellite itself performs RAN functions, including decoding, re-encoding, and amplifying uplink radio signals before forwarding them to the ground network. This enables a more flexible and autonomous network architecture, reducing dependency on the ground gateway and improving latency and resource management. Both architectures present specific challenges for mobility management, especially regarding timing advance calculations, beam management, and RACH-less handovers, as the network must account for the unique propagation delays and dynamic characteristics of the NTN environment.

[0029] The 3GPP Releases 17 and 18 baselines for NTN (non-terrestrial network) connectivity, as specified in TS38.331 , introduces a range of system information broadcast elements (SIB19) designed to support mobility and synchronization under NTN-specific conditions. These include ephemeris data and common TA parameters, the reference location of the serving cell, distance thresholds for measurement triggering, and service timing information indicating when a satellite will start or stop serving a given area. Notably, the specification also defines mechanisms such as the moving reference location for Earth-moving systems (e.g., low Earth orbit satellites), which informs the evaluation of handover events and location-based measurement initiation. The specification accounts for both fixed and moving NTN systems and introduces neighbour cell configurations and 202404977

[0030] 6 synchronization time offsets between beams or satellites. Together, these elements allow the network and UE to manage mobility intelligently, even in dynamically changing environments, by knowing the relation between the UE’s location, the cell’s reference location (which typically coincides with the subsatellite point), and the planned satellite movements.

[0031] In Release 17, the 3GPP specification defines mobility procedures for UEs in the RRC_CONNECTED state, including handovers between non-terrestrial networks (NTN) and terrestrial networks (TN). Notably, during NTN-TN mobility, the UE is not required to maintain simultaneous connections to both network types; handovers can occur in either direction - from NTN to TN (hand-in) or from TN to NTN (hand-out). While dual-active protocol stack (DAPS) handovers are not supported for NTN in this release, the UE may support mobility between gNBs using NTN payloads in different orbits, such as between geostationary (GSO) and non-geostationary (NGSO) systems.

[0032] Importantly, RACH-less handover, as specified in TS 38.321 and TS 38.331 , is supported in NTNs. In this procedure, the RRCReconfiguration message used to trigger the handover includes a timing adjustment indication along with either a configured uplink grant or a beam indication for accessing the target cell. The UE applies the timing adjustment, synchronizes to the target cell, and transmits the RRCReconfigurationComplete message using the configured uplink grant if provided. If no valid configured grant is included, the UE can fall back to random access (RACH) or obtain an uplink grant by monitoring the PDCCH as indicated by the beam information.

[0033] The 3GPP specification further introduces enhancements related to conditional handover (CHO) and satellite switching in NTN environments. Time-based conditional handover can be performed using a RACH-less approach, improving efficiency by avoiding the need for random access procedures when predefined timing conditions are met.

[0034] Additionally, the specification defines a satellite switch with re-synchronization procedure for quasi-Earth-fixed NTN scenarios where the same synchronization signal block (SSB) frequency and the same gNB are used. This procedure enables 202404977

[0035] 7 both hard and soft satellite switching while avoiding Layer 3 mobility operations, by maintaining the same physical cell identity (PCI) across the geographical area covered by the satellite beam. CHO can be configured alongside the satellite switch with re-sync, allowing flexible mobility management.

[0036] For soft satellite switching, the UE can begin synchronization with the target satellite before the source satellite stops serving the cell, eliminating the need for simultaneous connections. In contrast, for hard satellite switching, the UE initiates synchronization with the target satellite only after the switch has been triggered, ensuring orderly transition between satellite links.

[0037] According to the 3GPP specification, upon reception of an “RRCReconfiguration" message, the UE is required to carry out specific actions depending on the configuration and mobility scenario. In the context of conditional reconfiguration procedures - such as conditional handover (CHO), conditional primary cell addition (CPA), conditional primary cell change (CPC), or subsequent conditional primary cell addition / change (CPAC) - or during an LTM (L1 / L2 Triggered Mobility) cell switch, the UE follows detailed handling steps.

[0038] Specifically, if a RACH-less handover is included in the jecon figuration With Sync" field within the „spCellConfig“ of a master cell group (MCG), and the lower layers have indicated that the RACH-less handover has been successfully completed, the UE proceeds with the post-handover procedures. Furthermore, if the RACH-less handover was configured and an NTN-specific configuration („cg-NTN-Configuration“) was included, the UE is required to release the uplink grant that was initially configured for the RACH-less handover procedure. This ensures that radio resources are efficiently released and reallocated following the completion of the handover process.

[0039] A key challenge in current RACH-less handover procedures arises from uncertainty regarding the precise timing of the UE’s first uplink (UL) transmission after switching to the target cell or gNB. When the network lacks accurate knowledge of when the UE will begin transmitting in the target cell, it may preemptively allocate UL resources - such as uplink grants or scheduling opportunities - that the UE cannot 202404977

[0040] 8 yet use. This leads to inefficient resource utilization, with scheduled UL capacity being wasted while the UE is still completing the handover process.

[0041] This problem becomes particularly severe in scenarios involving high-speed UE mobility, where the time window for handover is short and unpredictable, and in transitions between terrestrial networks (TN) and non-terrestrial networks (NTN), or vice versa. In these cases, the increased propagation delays, dynamic movement patterns, and complex synchronization requirements significantly amplify the risk of misaligned scheduling and resource waste. Addressing this issue requires enhanced mechanisms that allow the network to anticipate the UE’s mobility and accurately predict the handover completion moment, enabling optimized RACH-less handover preparation and execution.

[0042] From the above, the problem arises of how to enable the network to accurately anticipate the timing of the UE’s first uplink transmission during RACH-less handover particularly under high-mobility and TN-to-NTN (and vice versa) handover conditions in order to avoid resource waste, improve synchronization, and ensure reliable handover execution.

[0043] This problem is solved by the method of claim 1 and the system of claim 11. Preferred embodiments are defined in the respective dependent sub-claims.

[0044] The present invention provides a method for performing a mobility-optimized RACH-less handover in a wireless communication system, designed to address the challenges of resource inefficiency and synchronization uncertainty, particularly under high-mobility and terrestrial-to-non-terrestrial network (TN-NTN) handover conditions.

[0045] According to the method, the serving network node, respectively the serving base station, particularly the gNB, configures the UE with a configuration that enables the UE to perform movement estimation and / or trajectory prediction with respect to one or more candidate or target cells or beams. This configuration defines specific measurement and / or prediction events and associated thresholds, including, for example, thresholds for prediction probabilities, thresholds for motion changes or reporting conditions. In this regard the configuration may be provided via a 202404977

[0046] 9

[0047] ..measurement configuration", a radio resource control (re-)configuration”. a ..mobility estimation / prediction" configuration. or „AI / ML configuration".

[0048] Based on the received configuration, the UE performs movement estimation and / or trajectory prediction, determining the likelihood of entering at least one of the configured candidate or target cells or beams. The UE calculates a probability value and compares it against a predefined and / or configured threshold. If the calculated probability exceeds the threshold, the UE proactively performs TA estimation for the respective candidate or target cell or beam.

[0049] Further, the network adjusts the configured prediction parameters based on received UE reports. For example, the network configures prediction time periods and / or prediction time windows depending on UE mobility, e.g., in case of low UE mobility longer prediction time periods, and in case of high UE mobility shorter prediction time periods are configured.

[0050] This proactive mechanism allows the network to anticipate the UE’s mobility state and synchronize resources more accurately, reducing the risk of unsuccessful uplink (UL) scheduling and improving overall handover performance. By integrating AI / ML-based movement prediction with TA estimation at the UE side, the invention enhances the efficiency, reliability, and responsiveness of RACH-less handovers, particularly under the demanding conditions of high-speed mobility and cross-domain TN-NTN transitions.

[0051] As claimed, the invention provides a method for RACH-less handover in a wireless communication system. The method comprises configuring a UE for performing movement estimation and / or trajectory prediction with respect to one or more candidate or target cells or beams. The method further includes defining measurement and / or prediction events and associated thresholds, performing, by the UE. movement estimation and / or trajectory prediction based on the configuration, calculating, by the UE. a probability of the UE entering at least one of the candidate or target cells or beams, comparing, by the UE. the calculated probability to a configured threshold, and performing, by the UE. a TA estimation for the respective candidate or target cell or beam if the calculated probability exceeds the configured threshold. 202404977

[0052] 10

[0053] The advantage of this two-step predictive approach is that the UE can proactively assess its mobility path and prepare synchronization parameters before completing the handover. This reduces the uncertainty in handovertiming, allows the network to avoid unnecessary or premature uplink resource allocations, and improves handover reliability, particularly in scenarios with high-speed movement or transitions between terrestrial and non-terrestrial networks.

[0054] In a preferred embodiment, the configuration provided to the UEs determined and supplied by the serving element, particularly the gNB. This ensures that the UE operates under network-controlled parameters, allowing the gNB to specify which candidate or target cells or beams should be evaluated, as well as how movement estimation and trajectory prediction should be carried out. By centralizing this configuration, the system achieves harmonized operation, allowing the network to flexibly adapt measurement and prediction strategies in response to varying mobility scenarios, such as urban high-density areas, rural coverage, or NTN environments.

[0055] In another preferred embodiment, the measurement and / or prediction events and associated thresholds used by the UE are explicitly defined by the network, in particular by the serving gNB. This allows the network to dictate under what conditions the UE should trigger movement estimation, trajectory prediction, or reporting. For example, the network can adjust thresholds to be more sensitive in high-mobility conditions (e.g., fast-moving trains) or less sensitive in static environments. This network-driven control improves adaptability and optimizes the balance between signalling overhead and prediction accuracy.

[0056] In a further preferred embodiment, the UE reports, to the serving element, particularly the gNB, detailed information including one or more candidate or target cell or beam identifiers, the calculated handover probabilities, and the corresponding estimated TA values. This reporting allows the network to obtain a precise picture of not only which target cells or beams are under consideration but also how likely the UE is to enter them and what timing corrections will be needed. This enhances the network’s ability to validate predictions, preconfigure uplink resources, and prepare efficient RACH-less handover execution. 202404977

[0057] 11

[0058] In another embodiment, the configuration provided to the UE comprises at least one or more parameters, such as the time period over which movement estimation or trajectory prediction should be performed, thresholds that define significant motion changes, specific motion change events to monitor, and reporting conditions, under which the UE should inform the gNB. This rich configuration allows the network to tailor the UE’s behaviour to the specific mobility context, ensuring fine-grained control of prediction intervals and reporting sensitivity, which improves prediction reliability while keeping signalling efficient.

[0059] In a further preferred embodiment, the UE reports additional information to the gNB, specifically including the estimated movement vectors or predicted trajectory paths, in addition to cell / beam identifiers and probabilities. This extra layer of reporting allows the network to understand not just where the UE might go, but also how it is moving, for example, detecting directional trends, velocity, or anticipated path changes. Such predictive insight enables the network to plan ahead, optimize handover decisions, and even proactively adjust beamforming or resource scheduling.

[0060] In another embodiment, the serving gNB processes the reported UE information together with internal radio access network (RAN) data, feeding these combined inputs into an artificial intelligence or machine learning (AI / ML) model. The AI / ML model is configured to determine and / or validate the optimal candidate or target cell or beam for handover, update the UE’s conditional handover (CHO) configuration, send a HO command message including optimal candidate or target cell or beam, corresponding TA values, and configured grant for UL transmissions to initiate a RACH-less handover, and / or prepare for a mobility-optimized RACH-less handover. This network-side intelligence enables advanced decision-making, improving accuracy and reliability over rule-based systems and adapting handover planning to real-time conditions, historical patterns, and multi-cell coordination.

[0061] In a preferred embodiment, the AI / ML model used at the gNB is trained using historical mobility data of the UE and / or additional context information associated with the network, such as NTN constellation characteristics (e.g., ephemeris data), typical movement patterns in the area, beam configurations, or user profiles. This 202404977

[0062] 12 learning-based approach allows the system to go beyond instantaneous measurements, using statistical trends and past observations to predict likely future movements, improve handover timing precision, and reduce the risk of failures, particularly in environments with complex or rapidly changing mobility profiles.

[0063] In another embodiment, the UE performs the probability calculation and the timing advance estimation not just for a single candidate, but for a plurality of k candidate or target cells or beams. This multi-candidate assessment increases robustness by ensuring that even if one target cell becomes unavailable or degrades, alternative handover options are already evaluated and ready. This is particularly advantageous in dense urban deployments, multi-beam NTN scenarios, or situations where user mobility is highly dynamic and unpredictable.

[0064] Finally, in a preferred embodiment, the method further comprises that the UE reports a motion change event to the gNB upon detecting a change in its motion state that exceeds a configured threshold. This event-driven reporting ensures that significant mobility changes - for example, sudden acceleration, deceleration, or a sharp change in direction - are immediately communicated to the network, allowing it to reassess handover configurations and maintain service quality even under abrupt movement conditions.

[0065] In a further aspect, the invention provides a system for performing RACH-less handover in a wireless communication system. The system comprises a UE and at least two gNBs, including a serving gNB and a target gNB. The serving gNB is configured to provide the UE with a configuration that enables the UE to perform movement estimation and / or trajectory prediction with respect to one or more candidate or target cells or beams. It also defines measurement and / or prediction events and thresholds that the UE uses to determine when significant movement or trajectory changes occur. The UE, in turn, is configured to carry out movement estimation and / or trajectory prediction based on the provided configuration, calculate the probability of entering at least one candidate or target cell or beam, compare that probability against a configured threshold, and, if the threshold is exceeded, perform a TA estimation for the respective candidate or target cell or beam. 202404977

[0066] 13

[0067] The advantage of this system design is that it enables predictive, coordinated mobility management between the UE and the network, improving handover timing, reducing uplink resource waste, and supporting high-speed and NTN mobility scenarios.

[0068] In a preferred embodiment, the system is further configured so that the UE reports, to the serving gNB, detailed information including one or more candidate or target cell or beam identifiers, the estimated handover probabilities, and the corresponding estimated TA values. This reporting gives the gNB precise predictive input, allowing it to validate the UE’s estimates, prepare resources proactively, and optimize handover execution.

[0069] In another preferred embodiment, the serving gNB in the system is further configured to process the reported information, together with radio access network (RAN)-internal data, using an artificial intelligence or machine learning (AI / ML) model. This model determines and / or validates the best handover target, updates the UE’s conditional handover (CHO) configuration, sends a HO command message including optimal candidate or target cell or beam, corresponding TA values, and configured grant for UL transmissions to initiate a RACH-less handover, and / or prepares for a mobility-optimized RACH-less handover. This Al-enhanced design improves decision accuracy and adapts to real-time mobility conditions.

[0070] In another aspect, the invention provides a UE configured for RACH-less handover. The UE comprises a processor and memory storing instructions that, when executed, enable the UE to perform several tasks. Specifically, the UE can carry out movement estimation and / or trajectory prediction for one or more candidate or target cells or beams, using a measurement configuration received from the serving gNB. Based on this, the UE calculates the probability of entering at least one candidate or target, compares this probability to a configured threshold, and, if the threshold is exceeded, performs a TA estimation for the corresponding cell or beam.

[0071] This intelligent UE-side behaviour enables the device to play an active role in predicting mobility transitions, improving the precision and success rate of RACH-less handovers. 202404977

[0072] 14

[0073] In another aspect, the invention provides a serving gNB configured for RACH-less handover. The gNB includes a processor and memory storing instructions that allow it to provide a UE with a measurement configuration for movement estimation and / or trajectory prediction. Additionally, the gNB is designed to receive detailed measurement reports from the UE, including candidate or target cell or beam identifiers, estimated handover probabilities, and estimated TA values. It processes this information, together with RAN-internal data, using an artificial intelligence or machine learning (AI / ML) model to determine the optimal handover target and prepare a mobility-optimized RACH-less handover.

[0074] This allows the network side to make faster, more accurate, and predictive handover decisions, reducing handover failures and improving resource efficiency.

[0075] In yet another aspect, the invention provides a target gNB configured to support RACH-less handover. This gNB includes a processor and memory storing instructions that enable it to receive a handover request from a serving gNB, including validated or updated TA values for a UE. Based on the received TA values, the target gNB schedules uplink resources (such as configured uplink grants) for the UE. It then sends a handover acknowledgment back to the serving gNB, indicating that the UE can access the target cell.

[0076] This design ensures that the target gNB is ready to serve the UE immediately upon handover, minimizing downtime and improving handover reliability, especially in fast-changing environments or across TN-NTN boundaries.

[0077] In some embodiments, a more general term “network node” may be used and may correspond to any type of radio network node or any network node, which communicates with a UE (directly or via another node) and / or with another network node. Examples of network nodes are NodeB, MeNB, ENB, Integrated Access and Backhaul (IAB) node, a network node belonging to MCG or SCG, base station (BS), multi-standard radio (MSR) radio node such as MSR BS, eNodeB (eNB), gNodeB (gNB), network controller, radio network controller (RNC), base station controller (BSC), relay, donor node controlling relay, base transceiver station (BTS), access point (AP), transmission points, transmission nodes, RRU, RRH, nodes in distributed antenna system (DAS), core network node (e.g. Mobile Switching Center 202404977

[0078] 15

[0079] (MSC), Mobility Management Entity (MME), etc.), Operations & Maintenance (O&M), Operations Support System (OSS), Self Optimized Network (SON), positioning node (e.g. Evolved- Serving Mobile Location Centre (E-SMLC)), Minimization of Drive Tests (MDT), test equipment (physical node or software), etc.

[0080] The non-limiting term user equipment (UE) or wireless device may be used and may refer to any type of wireless device communicating with a network node and / or with another UE in a cellular or mobile communication system. Examples of UE are target device, device to device (D2D) UE, machine type UE or UE capable of machine to machine (M2M) communication, PDA, PAD, Tablet, mobile terminals, smart phone, laptop embedded equipped (LEE), laptop mounted equipment (LME), USB dongles, UE category Ml, UE category M2, ProSe UE, V2V UE, V2X UE, etc.

[0081] In some embodiments, the non-limiting term base station (BS) or network node may be used and may refer to any type of entity in a mobile communication system supporting any wireless communication technology, e.g., 4G, 5G, 6G, etc.

[0082] As will be appreciated by one skilled in the art, aspects of the embodiments may be embodied as a system, apparatus, method, or program product. Accordingly, embodiments may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects.

[0083] For example, the disclosed embodiments may be implemented as a hardware circuit comprising custom very-large-scale integration (“VLSI”) circuits or gate arrays, off-the-shelf semiconductors such as logic chips, transistors, or other discrete components. The disclosed embodiments may also be implemented in programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices, or the like. As another example, the disclosed embodiments may include one or more physical or logical blocks of executable code which may, for instance, be organized as an object, procedure, or function.

[0084] Furthermore, embodiments may take the form of a program product embodied in one or more computer readable storage devices storing machine readable code, 202404977

[0085] 16 computer readable code, and / or program code, referred hereafter as code. The storage devices may be tangible, non- transitory, and / or non-transmission. The storage devices may not embody signals. In a certain embodiment, the storage devices only employ signals for accessing code.

[0086] For a better understanding of the principle of the present invention, embodiments of the invention will be explained in more detail below with reference to the figures. Like reference numerals are used in the figures for the same or equivalent elements and are not necessarily described again for each figure.

[0087] The invention is explained in more detail below with reference to the figures. The figures show:

[0088] Figure 1 a problem scenario

[0089] Figure 2 a procedural flow carried out at the UE,

[0090] Figure 3 a procedural flow carried out at the serving gNB,

[0091] Figure 4 a procedural flow carried out at the target gNB,

[0092] Figure 5 an exemplary scenario illustrating UE motion changes in NTN coverage area.

[0093] Figure 1 illustrates a problem scenario involving a handover from a terrestrial network (TN) to a non-terrestrial network (NTN), specifically highlighting the challenges arising from altitude differences between the two systems.

[0094] In this setup, the UE 1 is initially connected to a terrestrial base station 2 (indicated as PCI #1 ), operating on the ground network. As the UE 1 moves, it autonomously detects a candidate NTN cell for handover - either a geostationary satellite 3 (GSO) or a non-geostationary satellite 4 (NGSO) - marked with their coverage areas PCI #2 and PCI #3 circles, respectively. The UE performs handover preparation by adjusting its timing and synchronizing to the target cell, following the RACH-less handover principles. 202404977

[0095] 17

[0096] However, a critical issue arises due to the significant altitude differences between the terrestrial base stations 2 and the satellite-based gNBs 3 and 4. The propagation distance 5 from the UE 1 to the satellite (and back) is much greater than the distance 6 in the terrestrial case, resulting in non-negligible time shifts in the uplink and downlink communication. Even though the UE 1 autonomously adjusts its timing, these altitude-induced time shifts can cause misalignment, if not correctly predicted and compensated.

[0097] The figure 1 further shows the Xn interface connections 7 between the terrestrial and satellite gNBs, which are essential for coordination during handover. However, without accurate advance knowledge of the altitude differences and the resulting propagation delays, the target gNB (e.g., on the satellite) may attempt to schedule the UE 1 prematurely - before the UE has actually completed the handover and become uplink-synchronized. This can lead to resource waste, failed uplink attempts, or interrupted service continuity.

[0098] Overall, the scenario demonstrates the need for intelligent, predictive timing adjustment taking into account not only horizontal movement and cell IDs but also vertical distance and satellite ephemeris data - to enable successful and efficient TN-to-NTN handovers, especially under RACH-less conditions.

[0099] As described above the invention provides that the serving gNB configures the UE 1 to carry out movement estimation and / or trajectory prediction as well as TA estimation for one or more candidate or target cells or beams. To enable this, the gNB provides the UE 1 with a configuration, e.g., as part of a measurement configuration, which may define new AI / ML-based measurement and / or prediction events and associated thresholds.

[0100] This measurement configuration can include, for example, a defined time period for performing movement estimation or trajectory prediction, including a corresponding prediction time window, threshold values for detecting significant motion changes, specified motion change events, and reporting conditions under which the UE 1 is expected to send results back to the gNB. 202404977

[0101] 18

[0102] Once the UE 1 receives this configuration, it monitors its mobility context. If the measurement and / or prediction criteria are met — for instance, if motion change thresholds are exceeded or periodic prediction is required — the UE determines, based on AI / ML-supported movement estimation and / or trajectory prediction, the probability that it will enter a given candidate or target cell or beam. If the calculated probability exceeds the configured threshold, the UE performs TA estimation for the respective candidate or target cell(s) or beam(s).

[0103] If the reporting conditions are fulfilled, the UE transmits a report to the gNB. Further, the network adjusts the configured prediction parameters based on received UE reports. For example, the network configures prediction time periods and / or prediction time windows depending on UE mobility, e.g., in case of low UE mobility longer prediction time periods, and in case of high UE mobility shorter prediction time periods are configured. In one variant, the report includes the identifiers (IDs) of the measured cells or beams and the corresponding handover probabilities, as well as the estimated TA values for each candidate. In another variant, the report additionally contains the UE’s estimated movement vectors or predicted trajectory paths, offering the network a richer dataset for handover preparation.

[0104] Upon receiving the report, the gNB may combine the provided UE data (cell / beam IDs, probabilities, TA values, and optionally movements or trajectories) with RAN-internal data and input this combined dataset into its own AI / ML model. This model can be used to check and validate the most suitable target cell or beam, update the UE’s conditional handover (CHO) configuration, send a HO command message (incl. optimal candidate or target cell or beam, corresponding TA values, and configured grant for UL transmissions) to initiate a RACH-less handover and / or prepare the necessary steps for executing a mobility-optimized RACH-less handover.

[0105] This approach is particularly beneficial in challenging scenarios, such as TN to NTN handovers, where altitude differences between terrestrial and satellite cells cause significant propagation time shifts. By enabling proactive estimation and Al-supported prediction of both movement and TA, the invention reduces handover failures and radio link failures (RLFs), improves mobility robustness, and ensures 202404977

[0106] 19 seamless service continuity, even under high-mobility or cross-domain network transitions.

[0107] For example, the UE 1 may be configured to perform movement estimation and trajectory prediction periodically or specifically in response to detected motion changes that exceed configured thresholds. Such motion changes can be identified based on internal UE mechanisms or based on defined measurement and / or prediction events configured and / or updated by the network. The applied AI / ML algorithms support the UE in evaluating whether it is likely to move into another cell or beam, and if the calculated probabilities exceed the defined limits, the UE proactively performs TA estimation for the affected candidates, ensuring the target cell or beam can prepare its uplink scheduling accurately.

[0108] Figure 2 illustrates the procedural flow carried out at the UE side during a mobility-optimized RACH-less handover. The flow begins when the UE receives a measurement configuration from the serving gNB. This configuration includes instructions for movement estimation and / or trajectory prediction, target timing advance (NTA) estimation, and the related measurement, prediction, and reporting conditions.

[0109] The UE first evaluates whether the conditions for movement estimation and / or trajectory prediction are met. If yes, it proceeds to perform the movement estimation and / or trajectory prediction as per the configuration. Next, it checks whether the conditions for NTA estimation are satisfied. If they are, the UE carries out the target NTA estimation, determining the required timing advance adjustments toward the candidate or target cell or beam.

[0110] Once these steps are complete, the UE evaluates if the reporting conditions are met. If so, it generates and sends a measurement report to the serving gNB. This report includes the calculated target NTA estimation(s) along with other relevant information such as cell or beam identifiers and probabilities.

[0111] Upon receiving the report, the serving gNB provides the UE with a radio resource control (RRC) reconfiguration message, which includes updated target NTA parameters and a configured grant (CG) for uplink transmission. Based on this 202404977

[0112] 20 configuration, the UE determines the appropriate moment to detach from the serving gNB and initiates re-synchronization with the target gNB.

[0113] Finally, the UE performs the RACH-less handover procedure, applying the provided configured grant (CG) to begin uplink transmissions toward the target cell. This flow ensures that the UE proactively prepares for the handover, reducing the risk of timing misalignments, uplink resource waste, or service interruption.

[0114] Figure 3 illustrates the procedural flow executed at the serving gNB side as part of a mobility-optimized RACH-less handover.

[0115] The process begins with the serving gNB configuring the UE to perform movement estimation and / or trajectory prediction, as well as target timing advance (NTA) estimation and reporting. This configuration equips the UE with the necessary instructions and thresholds to anticipate mobility changes and calculate predictive timing values.

[0116] Afterward, the serving gNB receives the measurement report(s) from the UE, including predicted target NTA values. Based on the received data, the gNB evaluates whether a handover is necessary and decides whether to switch the UE from the serving gNB to the target gNB.

[0117] Once the handover decision is made, the serving gNB checks, validates, or updates the target NTA value reported by the UE to ensure accuracy. It then sends a handover request to the target gNB, including the validated or updated NTA values for the UE.

[0118] Upon receiving an acknowledgment (ACK) from the target gNB, which also includes a configured uplink (UL) grant, the serving gNB provides the UE with an updated radio resource control (RRC) reconfiguration message. This message includes the final target NTA value and the configured UL grant, enabling the UE to synchronize and access the target cell.

[0119] Next, the serving gNB forwards the user data to the target gNB, ensuring continuity of data flow during the handover. Finally, once the target gNB indicates that the UE 202404977

[0120] 21 has successfully completed the switch, the serving gNB releases the UE context, completing the handover procedure.

[0121] This flow allows for proactive and coordinated handover management, reducing handover delays, minimizing uplink resource waste, and improving mobility support, particularly under challenging high-mobility or TN-to-NTN transition scenarios.

[0122] Figure 4 illustrates the procedural flow carried out at the target gNB during a mobility-optimized RACH-less handover.

[0123] The process begins when the target gNB receives a handover request from the serving gNB. This request includes validated or updated target timing advance (NTA) values for the UE, ensuring the target gNB has accurate timing information to prepare for uplink synchronization.

[0124] Based on the received NTA values, the target gNB schedules uplink (UL) resources for the UE, preparing the necessary configured uplink grants or dynamic scheduling resources that the UE will use after synchronization. Once the uplink resources are prepared, the target gNB sends a handover request acknowledgment (ACK) back to the serving gNB, which includes the configured UL grant information.

[0125] While the UE completes the handover process, the target gNB buffers incoming user data that is being forwarded by the serving gNB, ensuring that no data is lost during the transition. After the UE has successfully synchronized and the handover is finalized, the target gNB signals back to the serving gNB that the switching process is complete.

[0126] Finally, the target gNB indicates the path switch to the access and mobility management function (AMF) or user plane function (UPF), enabling the network core to update the data path accordingly and finalize the handover procedure.

[0127] This process ensures smooth and coordinated handover execution, minimizing latency and data loss, and is particularly critical for high-mobility scenarios or terrestrial-to-non-terrestrial network (TN-to-NTN) transitions. 202404977

[0128] 22

[0129] Figure 5 illustrates a scenario where the UE 10 moves within the coverage area 11 of a satellite-based cell (e.g., from a Low Earth orbit, LEO, satellite) and undergoes a change 12 in movement direction. The figure 5 shows the UE 10 initially moving along direction v, then changing to a new direction u, forming an angle a that indicates the shift in movement trajectory.

[0130] The serving gNB can be configured to instruct the UE to monitor and report such motion changes, particularly when they exceed configured thresholds. There are several ways to define and detect a “motion change,” including:

[0131] - an absolute travel distance measured in meters,

[0132] - a proportion of travelled distance relative to the cell diameter (e.g., in NTN systems, the constellation type information can be used to infer the beam footprint or cell size on the ground, with LEO satellite cells typically spanning ~50 km in diameter),

[0133] - an orientation change, defined as a change in movement direction relative to a reference direction,

[0134] - or a velocity change, defined as a change in the UE’s speed, either in absolute terms or relative to the velocity of the NTN cell.

[0135] Furthermore, the gNB can configure the UE to report specific events such as entering a cell edge area, shown in the figure as the outer boundary region of the satellite footprint. A combination of conditions can also be used: for example, reporting may be triggered only when the UE enters the edge area and is moving at a certain velocity or in a particular orientation that exceeds a configured threshold.

[0136] By enabling the detection and reporting of such motion or edge-related events, the network gains early insight into the UE’s movement dynamics, which can be used to improve handover preparation, beam management, and resource allocation, particularly in mobility-challenging NTN environments.

Claims

20240497723Patent claims1 . A method for handover, particularly for random access channel-less (RACH-less) handover, in a wireless communication system, comprising: configuring a user equipment (UE) for performing movement estimation and / or trajectory prediction with respect to one or more candidate or target cells or beams; defining measurement and / or prediction events and associated thresholds performing, by the UE, movement estimation and / or trajectory prediction based on the received configuration; calculating, by the UE, a probability of the UE entering at least one of the candidate or target cells or beams; comparing, by the UE, the calculated probability to a configured threshold; and performing, by the UE, a timing advance (TA) estimation for the respective candidate or target cell(s) or beam(s) if the calculated probability exceeds the configured threshold.

2. The method of claim 1 , wherein the configuration is provided by a network node which may be a base station, particularly by a gNB, to the UE via a measurement configuration, a radio resource control (re-)configuration, a mobility estimation / prediction configuration, or an AI / ML configuration.

3. The method of claim 1 or 2, wherein the measurement and / or prediction events and the associated thresholds and further parameters, particularly prediction time period, are defined and configured by the network, in particular by the gNB.

4. The method of any one of the preceding claims, wherein the UE reports, to the serving element, information comprising:- one or more cell or beam identifiers,20240497724- estimated handover probabilities, and- corresponding estimated timing advance (TA) values.

5. The method of any one of the preceding claims, wherein the configuration comprises at least one of:- a time period for movement estimation and / or trajectory prediction,- one or more thresholds for motion changes,- one or more motion change events, and- one or more reporting conditions.

6. The method of any one of the preceding claims, wherein the UE reports, to the serving element, information further comprising estimated movements and / or predicted trajectories.

7. The method of any one of the preceding claims, the serving element processes the reported information, together with radio access network (RAN)-internal data, as input to an AI / ML model configured to:- determine and / or validate an optimal candidate or target cell or beam for handover, particularly including corresponding timing advance (TA) values and configured grant for UL transmissions,- update the UE’s conditional handover (CHO) configuration, and / or- send a HO command message, particularly including optimal candidate or target cell or beam, corresponding timing advance (TA) values, and configured grant for UL transmissions) to initiate a mobility-optimized RACH-less handover.

8. The method of any one of the preceding claims, wherein the AI / ML algorithm is trained to predict handover probabilities based on historical mobility data of the UE and / or context information associated with the network.

9. The method of any one of the preceding claims, wherein the UE performs the probability calculation and the timing advance estimation for a plurality of (k) candidate or target cells or beams.2024049772510. The method of any one of the preceding claims, further comprising reporting, by the UE to the serving element, a motion change event upon detection of a change in motion exceeding a configured threshold.

11. A system for performing random access channel-less (RACH-less) handover in a wireless communication system, the system comprising: a user equipment (UE) and at least two network elements, particularly base stations (gNB), including a serving element, particularly serving gNB, and a target element, particularly target gNB; wherein the serving element is configured to:- provide the UE with a configuration for performing movement estimation and / or trajectory prediction with respect to one or more candidate or target cells or beams,- define and configure measurement and / or prediction events and associated thresholds, wherein the UE is configured to:- perform movement estimation and / or trajectory prediction based on the configuration,- calculate a probability of entering at least one candidate or target cell or beam,- compare the calculated probability to a configured threshold, and- perform timing advance (TA) estimation for the respective candidate or target cell or beam if the calculated probability exceeds the configured threshold.

12. The system of claim 11 ,20240497726 wherein the UE is further configured to report, to the serving element, information comprising:- one or more cell or beam identifiers,- estimated handover probabilities and- corresponding estimated TA values.

13. The system of claim 11 or 12, wherein the serving element is further configured to process the reported information, together with radio access network (RAN)-internal data, as input to an artificial intelligence or machine learning (AI / ML) model to:- determine or validate an optimal candidate or target cell or beam for handover, particularly including corresponding timing advance (TA) values and configured grant for UL transmissions,- update the UE’s conditional handover (CHO) configuration, and / or- send a HO command message (incl. optimal candidate or target cell or beam, corresponding timing advance (TA) values, and configured grant for UL transmissions) to initiate a mobility-optimized RACH-less handover.

14. A user equipment (UE) configured for random access channel-less (RACH-less) handover, comprising a processor and memory storing instructions that, when executed, cause the UE to:- perform movement estimation and / or trajectory prediction with respect to one or more candidate or target cells or beams, based on a configuration received from a serving element,- calculate a probability of entering at least one candidate or target cell or beam,- compare the calculated probability to a configured threshold, and20240497727- perform timing advance (TA) estimation for the respective candidate or target cell or beam if the calculated probability exceeds the configured threshold.

15. A serving element, particularly a serving base station (gNB), configured for random access channel-less (RACH-less) handover, comprising a processor and memory storing instructions that, when executed, cause the serving element to:- provide a user equipment (UE) with a configuration for performing movement estimation and / or trajectory prediction,- receive reports from the UE, including candidate or target cell or beam identifiers, estimated handover probabilities, and estimated TA values, and- process the received information, together with radio access network (RAN)-internal data, using an artificial intelligence or machine learning (AI / ML) model to determine and / or validate an optimal handover target and prepare and / or initiate a mobility-optimized RACH-less handover.

16. A target element, particularly a serving base station (gNB), configured for random access channel-less (RACH-less) handover, comprising a processor and memory storing instructions that, when executed, cause the target element to:- receive a handover request from a serving element, the request including validated or updated target TA values for a user equipment (UE),- schedule uplink resources for the UE based on the received target TA values, and- send a handover request acknowledgment including an uplink grant for the UE to the serving element.

Images