An adaptive handover method for wireless networks

Abstract

The invention is related to a novel mechanism in clustering the transmission / reception points into two regions including a sensing region and non-sensing region.

Description

[0001] DESCRIPTION

[0002] AN ADAPTIVE HANDOVER METHOD FOR WIRELESS NETWORKS Technical Field

[0003] The invention is related to a novel mechanism in clustering the transmission / reception point into two regions including a sensing region and non-sensing region.

[0004] Prior Art

[0005] Handover (or handoff) in wireless communication refers to the process of transferring an ongoing call, data session, or network connection from one base station, access point, or cell tower to another without interruption. This is a critical aspect of mobility in wireless networks, ensuring seamless connectivity as users move between different network coverage areas.

[0006] The ETSI TS 136 300 V17.0.0 (2022-05) [1] document specifies two primary handover (HO) mechanisms utilized in LTE (Long Term Evolution) and NR (New Radio) systems:

[0007] Baseline Handover (HO): A traditional handover mechanism where the base station (BS) initiates the preparation and execution of the HO after deciding on the most suitable target cell. While effective, it can be prone to failures under dynamic scenarios, such as high-speed mobility or abrupt blockages, due to its reactive nature [2],

[0008] Conditional Handover (CHO): In the handover preparation phase, the CHO creates multiple targets cells and chooses the most appropriate cell. Besides, the user carries out the execution of the CHO when one or more conditions of the handover executions are satisfied. The evaluation of the execution condition(s) for candidate cells of the CHO begins when the UE receives the CHO configuration. After obtaining the CHO command, the user does not disconnect from the serving gNB, as is the case in the standard handover procedure. Instead, the UE persists in the exchange of uplink and downlink data with the serving gNB until the CHO execution condition is satisfied. However, the UE performs CHO when the specified execution condition(s) are satisfied for a CHO candidate cell. Whenever the execution is met, the UE terminate the execution condition(s) [3], However, CHO improves reliability by reducing handover failures (HOFs) associated with delayed preparation or mismatched target selection.

[0009] Despite these advancements, existing handover mechanisms face significant challenges in the context of 5G and beyond, particularly in heterogeneous networks featuring diverse BSs and access points (APs). These challenges are amplified by the reliance on high frequency mmWave bands, which are more susceptible to blockages, rapid signal degradation, and limited coverage.

[0010] Key issues are including [2]: handover failures (HOFs) which Caused by long handover time taken during the handover procedure or improper selection of the optimal target base station and handover delays and interference which is especially critical in dense network deployments and ping-pong handovers (HOPP) results wastage of network resources due to the contribution to the signaling overhead and unbalanced network loads which can lead to suboptimal user experiences and degraded Quality of Experience.

[0011] The paper of Ge, et al [4] focuses on addressing the limitations of existing ISAC systems in large-scale, distributed architectures for 6G networks. Traditional ISAC frameworks face challenges in maintaining target tracking across wide areas due to the limited field of view (FoV) of base stations (BSs) and inefficient mechanisms for information sharing during target handover. This leads to disruptions in trajectory tracking and impacts sensing accuracy. The proposed handover mechanism relies on real-time tracking but does not account for scenarios where neighboring base stations (BSs) are unavailable or face latency issues. Target handover is restricted to nearby BSs, assuming the target leaving one BS’s field of view (FoV) will enter the FoV of an adjacent BS. Additionally, at cell edges, interference levels are higher, which can degrade sensing performance, especially if unexpected blockages occur near BS areas. Such blockages can disrupt both sensing and communication processes. Furthermore, while the Trajectory Poisson Multi -Bernoulli Mixture filters improve trajectory tracking, they may introduce computational complexity and latency, potentially impacting real-time applications

[0012] Study of Zhang et al [5] discusses strategies to improve mobility management and resource allocation, particularly focusing on how handover processes can be optimized to protect against failures. This includes the use of advanced sensing and communication frameworks, as well as the integration of perceptive technologies that allow networks to anticipate mobility needs and adapt resource allocations dynamically. For instance, predictive handover mechanisms that account for real-time sensing data (e.g., from vehicle radar or environmental sensors) help mitigate issues like handover failures or delays.

[0013] The authors [8] presented the mmalert system, a predictive system designed to detect potential mmWave link blockages caused by mobile blockers, such as people moving in the Line-Of- Sight (LoS) path. The system utilizes passive sensing with two receive beams at the data communication receiver. This approach offers a low false alarm probability and provides enough warning time to handle potential link blockages in real-time communication scenarios. The proposed system comes with several challenges. First, it necessitates increased hardware requirements, as it relies on two beamforming paths and separate RF chains at the receiver. This addition significantly increases the system's complexity and cost. Second, the receiver is tasked with both sensing and trajectory estimation, which can impose a considerable computational burden. This may result in latency or processing bottlenecks, particularly if the receiver has limited capacity. Finally, while the integration of radar-aided proactive blockage prediction enhances the system's ability to forecast link degradation due to obstacles and enables preemptive measures, its implementation in real-world millimeter-wave systems introduces further challenges, such as ensuring robust performance in diverse environments.

[0014] System of Zhang et al [5] dynamically adjusts the handover and time-to-trigger (TTT) parameters to optimize handover efficiency. The integration of sensing with handover planning improves overall mobility management and reduces unnecessary handovers. Employs sensing parameters (e.g., time delay, Doppler effect, and angle of arrival) for accurate mobility predictions and optimized resource allocation. In their method they didn’t specified a certain region for sensing they assume that the base station can get a sensing measurement report interference at cell Edges. At the cell edge, the system may suffer from high interference from neighboring base stations or cross-link interference, which can degrade the performance of sensing and mobility management.

[0015] Several approaches have been explored for mobility trajectory prediction using Al and machine learning techniques. One such method is Trajectory -Based Handover Optimization, which analyzes historical mobility patterns to forecast user equipment (UE) movement and enhance handover decision-making. For instance, a study in [9] demonstrated that Long Short-Term Memory (LSTM) networks could effectively learn from historical signal data to predict future handover events, thereby reducing the likelihood of missed or unnecessary handovers. Another approach, Mobility Prediction for Seamless Connectivity, integrates predictive models with real-time sensing data to anticipate UE trajectory changes. This ensures reduced handover latency and improved throughput, as highlighted in

[0010] , Additionally, Al-Driven Mobility Management for 5G and Beyond

[0012] employs machine learning techniques for trajectory prediction, enabling adaptive network configurations in response to user mobility. However, despite these advancements, existing approaches rely solely on deep learning models, which often depend on centralized data collection and require large datasets, presenting challenges in scalability and real-time adaptability.

[0016] As a result, all of the problem mentioned above has made it necessary to provide a novelty in the related field.

[0017] Brief Description and Objects of the Invention

[0018] The evolution of next-generation mobile networks introduces significant challenges due to their unique characteristics and requirements. Technologies such as millimeter-wave (mmWave) communication, ultra-dense small BS deployments, exponential growth in mobile connection traffic, and an increasing number of connected applications present distinct hurdles for efficient mobility management.

[0019] The purpose of this invention is to address the inefficiencies in handover control within mobility environments, particularly in 5G and beyond networks. By introducing a sensing cell clustering along with dynamic auto-tuning mechanism for handover control parameters such as Time-to- Trigger (TTT) and Handover Margin (HOM), the invention optimizes handover timing and improves overall mobility management. This prevents issues like handover failures (HOFs), signaling overhead and handover ping-pong (HOPP), ensuring smoother transitions between cells and minimizing network disruptions.

[0020] The invention contributes to reducing the mobility issues in the presence of high mobility / low mobility, blockage, and changing network conditions. By incorporating real-time data for mobility prediction, the invention provides a solution to reduce the mobility issues including radio link failures (RLF) and unnecessary handovers.

[0021] Furthermore, it addresses the issue of interference between communication and sensing, as discussed in ISAC standards. Utilizing sensing-reflection in areas near the cell edge can increase interference between communication and sensing. The invention proposes methods to mitigate such interference, ensuring effective communication and sensing coexistence.

[0022] The frequent connections of the UE between the serving base station and the target base station are what causes the ping-pong effect due to signal fluctuations. The more often this happens, the more HOs will require processing, which will cause network delays and wastage of network resources. To achieve above-mentioned advantages, the proposed solution involves an auto-tuning mechanism that dynamically adjusts HOM and TTT based on real-time mobility data. This ensures that handovers occur at the optimal time, reducing the likelihood of failures and unnecessary handovers, and enhancing overall network reliability. The method includes determining whether the wireless communication entity is in the sensing or non-sensing region of the base station. A blockage prediction algorithm is applied if the entity is in the sensing region, or a handover prediction algorithm is applied if the entity is in the non-sensing region. The method further determines if the mobility of the wireless communication entity is highspeed or low-speed, applying conditional handover for high-speed mobility and baseline handover for low-speed mobility.

[0023] In accordance with other embodiments, the method involves repeating the blockage prediction algorithm using sensing measurements as input until a predetermined condition indicating that handover is required is met by the output of the blockage prediction algorithm.

[0024] In yet other embodiments, the handover prediction is carried out using a blockage prediction algorithm with sensing measurement reports collected from the base station of the non-sensing region as input.

[0025] Further embodiments involve repeating the handover prediction until a predetermined condition, which indicates that handover is required, is met by the output of the blockage prediction algorithm.

[0026] In accordance with additional embodiments, the predetermined condition is defined as the reference signal received power of the target being higher than that of the serving station.

[0027] In other embodiments, the handover prediction algorithm is executed using Al-driven or machine learning models to forecast user mobility patterns.

[0028] Additional embodiments include steps such as transmitting Synchronization Signal Blocks in burst mode, determining the number of beams in the bursts based on network configurations, detecting echoes from reflected beams, estimating blockage probabilities, calculating sensing metrics, defining a sensing region, and utilizing sensing measurements to identify potential blockages and prepare for handover operations.

[0029] In accordance with further embodiments, the prediction of blockage for sensing region involves employing beam-sweeping techniques to determine the necessity of handover by optimizing parameters such as Time-to-Trigger and Handover Margin. Other embodiments involve identifying the wireless communication entity within a sensing region defined by proximity to the base station, receiving sensing measurement reports, determining the device's transition from the sensing region to a non-sensing region, ceasing requests for sensing measurement reports when the device transitions beyond the sensing region, and dynamically adjusting parameters based on the device's movement patterns.

[0030] In additional embodiments, the sensing region is defined as a region within a certain distance from the base station, while the non-sensing region is beyond that distance.

[0031] Further embodiments determine the transition of the wireless communication entity from the sensing region to the non-sensing region based on predictive analysis of the entity's trajectory.

[0032] In accordance with other embodiments, the method includes dynamically adapting the sensing region based on network measurements or decisions to optimize sensing accuracy.

[0033] Additional embodiments account for various scenarios such as fixed sensing regions with the device moving in fixed or random directions, and adaptive sensing regions with the device moving in fixed or random directions.

[0034] In yet other embodiments, for fixed sensing regions with the wireless communication entity moving in a fixed direction, a constant value is set, and requests for sensing measurement reports cease when a threshold is reached.

[0035] In other embodiments, for fixed sensing regions with the wireless communication entity moving in a random direction, parameters are dynamically adjusted based on variations in the entity's trajectory.

[0036] Further embodiments involve adaptive sensing regions with the wireless communication entity moving in a fixed direction, where the sensing region dynamically expands or contracts to accommodate the entity's trajectory.

[0037] In accordance with additional embodiments, for adaptive sensing regions with the wireless communication entity moving in a random direction, the sensing region adapts dynamically to align with the device's unpredictable trajectory, enhancing robustness and flexibility.

[0038] In other embodiments, a processor device is provided comprising means for carrying out the steps of the method of any of the preceding claims.

[0039] The proposed sensing-based handover mechanism leverages integrated sensing and communication (ISAC) functionalities. It employs passive reflections and echoes from communication signals to estimate sensing parameters. This approach enhances accuracy and reduces signaling overhead. By avoiding excessive transmitter power for sensing, it mitigates interference between BS-to-BS, BS-to-UE, and sensing-to-communication operations, ensuring optimized spectrum utilization and improved network reliability.

[0040] The method adopts a two-region strategy. In inner regions, it uses real-time sensing parameters such as Doppler estimation, Angle of Arrival (AoA), Time of Arrival (TOA), Signal-to- Interference-plus-Noise Ratio (SINR), and reflection beam indices derived from Synchronization Signal Block. This improves handover decision precision, predicts mobility patterns, and proactively addresses potential blockages. In outer regions, the method focuses on trajectory prediction using AI / ML models, incorporating factors like cell size and BS type (i.e., micro, macro, gNB). This enables early detection of handover candidates, reduces latency, and improves scalability in dense networks.

[0041] The proposed method dynamically adjusts handover control parameters (HCPs) such as Time- to-Trigger (TTT) and Handover Margin (HOM) based on user mobility and network requirements. This reduces unnecessary handovers, such as ping-pong effects, and ensures seamless transitions between cells. The clustering of sensing regions further enhances handover efficiency by mitigating Radio Link Failure (RLF) caused by blockages and ensuring accurate trajectory -based decisions near the cell edge.

[0042] This approach ensures dynamic adaptation to varying mobility scenarios, whether users are stationary, slow-moving, or high-speed. By combining sensing-based techniques for inner regions with trajectory -based methods for outer regions, it achieves a balance between accuracy and efficiency. It is scalable and flexible, making it applicable across different network types and deployment scenarios, including microcells, macro-cells, and mm-Wave-based gNBs.

[0043] The sensing-based approach reduces signaling overhead compared to existing methods and improves mobility prediction by increasing trajectory prediction resolution. It minimizes interference between sensing and communication operations while optimizing spectrum utilization. This method provides a robust framework for addressing mobility challenges, enhancing overall network reliability and handover performance in next-generation wireless systems. The setting values of the handover control parameters will be higher in low-speed scenarios compared to high-speed scenarios. This ensures adaptability to varying mobility scenarios while maintaining efficient and reliable handover performance.

[0044] The frequent connections of the UE between the serving base station and the target base station are what causes the ping-pong effect due to signal fluctuations. The more often this happens, the more HOs will require processing, which will cause network delays and wastage of network resources.

[0045] Description of the Figures of the Invention

[0046] The figures and related descriptions necessary for the subject matter of the invention to be understood better are given below.

[0047] Figure 1. A flow chart of the implementation of the handover types (baseline HO and CHO) based on user experience (mobility and blockage)

[0048] Figure 2. Illustration of a handover and blockage-awareness clustering mechanism based on two regions: the sensing region and the non-sensing region, base station is equipped with sensing capabilities and utilizes sensing measurement reports in the inner region, while relying on mobility trajectory predictions in the outer region.

[0049] Figure 3. A flow chart illustrating the handover and blockage prediction process.

[0050] Figure 4. An illustration of the proposed handover mechanism with multiple handover positionalities and multi-base station nodes.

[0051] Figure 5. An illustration of the proposed approach for handover enhancement and blockage awareness into inner area (sensing region).

[0052] Figure 6. An illustration of the proposed approach for handover enhancement and blockageawareness in outage area regions.

[0053] Figure 7. An illustration of utilizing integrating sensing capabilities with SYNCHRONIZATION SIGNAL BLOCK-based prediction, cells can proactively prepare for potential handover scenarios, ensuring a seamless and immediate response when signal conditions change.

[0054] Figure 8. An illustration of sensing regions used for estimation within the Conditional Handover (CHO) process. Figure 9. An illustration of the concept of Tmax, which defines the maximum number of measurement reporting cycles (steps) in sensing region.

[0055] Figure 10. The flowchart presents the possible scenarios for adaptive Tmax, adjustments based on the status of the sensing region for the given base station.

[0056] Reference Numbers

[0057] The parts and components are given in the figures are referenced for the subject matter of the invention to be understood better.

[0058] 1. Transmission / reception point

[0059] 2. Wireless communication entity

[0060] 10. Sensing region

[0061] 20. Non-sensing region

[0062] 30. Blockage

[0063] 40. Synchronization Signal Block

[0064] 40a. Synchronization Signal Block X

[0065] 40b. Synchronization Signal Block Y

[0066] 50. Echo

[0067] Detailed Description of the Invention

[0068] The invention is related to a novel mechanism in clustering the base station into two regions including a sensing region (10) and non-sensing region (20).

[0069] Figure 1 illustrates a flowchart outlining the fundamental operations of the proposed approach, introducing a novel concept of clustering based on two distinct regions, each with unique operational statuses. These regions are dynamically classified based on specific detection criteria and user mobility characteristics. The proposed approach integrates various strategies to address the specific requirements of each scenario, enabling differentiated handling of mobility events. The clustering concept adapts to user conditions, where the status can be defined by High-Speed and Low-Speed scenarios. Status of High-Speed and Low-Speed is determined according to the wireless communication entity’s (2) (for example a user device) speed which determines the type of handover, whether conditional handover for high-speed user or baseline handover.

[0070] In the low-speed scenarios, key performance indicators (KPI), such as radio link failures (RLF) and Ping-Pong Handovers (HOPP), that can be utilized to evaluate the system's performance.

[0071] Sensing regions (10): Handover decision is influenced by the measurement sensing report as already specified the sensing region to sense more accurately. The system may utilize measurements to predict the speed of wireless communication entity (2) and trajectory based on mobility patterns.

[0072] Non-Sensing region (20): The handover (HO) will be initiated if handover decision is satisfied. The handover decision may be satisfied either by the blockage (30) or by the degradation of the signal quality of the serving transmission / reception points (1), such as base stations.

[0073] FIG. 2 illustrates this distinction for an individual the transmission / reception point. Two regions named as a sensing region (10) (inner region) and a non-sensing region (20) (outer region). The diameter of each region can be estimated based on network and environmental factors, and it may be fixed, pre-selected, or dynamically updated based on specific metrics or multi-metrics such as channel, number of users, scattering in the environment, or channel state information (CSI) while considering both sensing and communication applications. Consequently, the areas of sensing region and non-sensing region for transmission / reception point (1) (which may be a base station or user device) can be adjusted based on specific parameters derived from network estimations.

[0074] Notably, transmission / reception point (1) may operate the sensing measurement to make predictions either independently in a decentralized manner or as part of a larger network, such as heterogeneous or cell-free networks.

[0075] For the sensing region (10), handover decisions are primarily influenced by blockage / shadowing predictions, which are determined using sensing estimations. This region should ideally be managed by a large base station, such as one on a heterogeneous network or another transmission / reception point (1) (it may be a UAV, HAPS...Etc.), capable of providing service without adversely affecting other users within the region. These sensing approaches may rely on reflections from communication signals (ISAC signals), or radar signals. Additionally, the trajectory of both users and devices within this region is estimated using sensing data. The sensing methodology can be monostatic or bistatic, utilizing either active devices or passive reflections (collected echoes / reflection sensing signal). An advantage of this approach is that echoes in the inner sensing region are still detectable, providing valuable data for reliable handover decisions.

[0076] In the outer region, known as the non-sensing region (20), the approach shifts towards mobility trajectory-AI based predictions. Here, blockage (30) and the specific targets or users requiring service are predicted using advanced techniques, which may include machine learning (ML) or artificial intelligence (Al). The methods implemented in this region are designed to enhance the accuracy and efficiency of mobility management while addressing the inherent limitations of sensing in areas farther from the transmission / reception point (1). By focusing on a selected region, the approach reduces network overhead associated with data collection and training processes that would otherwise span the entire area. This targeted strategy minimizes backhaul load, decreases processing delays, and improves the accuracy and timeliness of the results.

[0077] In a preferred embodiment, the AI / ML model is trained by a training data which comprises channel measurements obtained by the UE (User Equipment) and transmitted to the APs (Access Points). The features of the training data used for training the model include at least Channel State Information (CSI) and Power Delay Profile (PDP). The known models are sufficient for prediction. However, Convolutional Neural Networks (CNNs) such as VGG16 or ResNet can be used for specifically. This models’ accuracy depends on the diversity and quantity of data collected, particularly at varying Signal-to-Interference-plus-Noise Ratio (SINR) levels. The model's output accuracy improves with more comprehensive data at different SINR conditions, ensuring the model's robustness and reliability.

[0078] As can be seen in FIG. 3a, AI / ML models in the evaluation use either: wireless communication entity (2) (i.e, User Equipment) measurement data and gNB (gNodeB) measurement as data as input. The evaluation includes the following AI / ML positioning methods:

[0079] Direct AI / ML positioning involves the straightforward application of artificial intelligence and machine learning (AI / ML) techniques to directly estimate position using available measurement data. In contrast, assisted AI / ML positioning employs collaborative techniques with multiple configurations to enhance accuracy. One such approach is multi- transmission / reception point (1) construction, where AI / ML models utilize measurements from multiple transmission / reception points (1) to assist in positioning. Another configuration, single- transmission / reception point (1) construction with one model for N transmission / reception points (1), involves a single AI / ML model processing measurement collectively for multiple transmission / reception points (1). Lastly, single- transmission / reception point (1) construction with N models for N transmission / reception points (1) uses separate AI / ML models for each transmission / reception point (1) independently.

[0080] FIG. 4 is illustration of the proposed handover mechanism with multi- transmission / reception point nodes and positionalities. This figure demonstrates the handover mechanism employing multiple transmission / reception points (1), utilizing a clustering approach based on two distinct regions — sensing and non-sensing — defined for each transmission / reception point (1). The clustering enables efficient management of handover processes and blockage prediction, particularly in dynamic network conditions.

[0081] The sensing region focuses on leveraging real-time measurements, such as angle of arrival (AoA) and beam sweeping with synchronization signal block (40) index reflections, to detect and predict blockages (30). These blockages (30) may include both fixed and dynamic obstacles. This sensing capability is crucial for environments like dense urban areas, where multipath propagation and high user mobility are common. In this region, the transmission / reception point (1) can proactively estimate blockage (30) positions and adapt the handover strategy to maintain seamless connectivity.

[0082] The non-sensing region operates beyond the direct sensing capability of the transmission / reception point (1) and relies on AI / ML-based trajectory prediction for mobility management. This approach ensures that even when direct sensing is not possible, the system can anticipate user movement and blockages (30). For instance, in sparse rural or suburban areas, this prediction minimizes signaling overhead, reduces processing delays, and optimizes key parameters such as time-to-trigger (TTT) and handover margin (HOM).

[0083] This mechanism enables versatile configurations. In heterogeneous networks, transmission / reception points (1) with broader coverage can manage handovers to smaller cells or other transmission / reception points, balancing load and maintaining service quality. In dense network deployments, transmission / reception points also collaborate efficiently to handle high user density and maintain uninterrupted service, as demonstrated in handovers between transmission / reception points. FIG. 3 provides a flow diagram illustrating the handover and blockage prediction process, delineating two distinct regions: the sensing region (10) and the non-sensing region (20). FIG. 5 and FIG. 6 depict the detailed steps of the prediction and handover process for moved terminals (high speed or low speed), spanning from the first step to the last step of FIG. 3.

[0084] FIG. 5 focuses on the inner sensing region (10) where the proposed handover mechanism begins. In the first step, the wireless communication entity (2), which moves, is identified within the sensing region (10). As the wireless communication entity (1) resides in the sensing region (20), it proceeds to step where measurement reporting is initiated. Measurement reporting is conducted based on sensing data, which transitions to next step which sensing measurements, including parameters such as trajectory prediction for both the wireless communication entity and the blockage (30) (as on FIG. 5), are utilized. Predictions are generated based on sensing topologies (monostatic, bistatic) and detection algorithms.

[0085] The sensing estimation can leverage the reflection of beam sweeping, using parameters like Angle of Arrival (AoA) and the communication signal beam Synchronization Signal Block (40) index, where the primary synchronization signal (PSS) and secondary synchronization signal (SSS) as defined in TS 38.21 l(more details in figure 6 and figure 7). Additionally, sensing can also involve direct measurements between the wireless communication entity and the transmission / reception point.

[0086] Based on the sensing report, the system detects both the trajectory of blockage (30) and the speed of the wireless communication entity (2).

[0087] If the wireless communication entity (2) exhibits high-speed mobility, it advances to next step for a conditional handover. If the wireless communication entity (2) exhibits low-speed mobility, it follows next step for a baseline handover.

[0088] Two different handover types are initiated bases on the user mobility. However, since the fast degradation of the signal quality in high-mobility scenarios, CHO handover will be initiated since the preparation phase is done while the user connection to the serving transmission / reception point (1) is still strong. Thereby reducing the HOFs. Furthermore, low triggering value will be assigned during the automatic adjustments of the TTT in CHO. On the other side, during low-speed scenarios, baseline handover will be initiated due to reduce the signaling loads caused by applying the CHO. However, higher triggering values will be assigned during low scenarios compared to high mobility users.

[0089] It is ensured that the system achieves enhanced reliability and efficiency by evaluation of key performance indicators (KPIs), such as Handover Ping-Pong (HOPP), Radio Link Failures (RLF), radio link failures (RLFs), and signaling overhead,

[0090] Noted that the measurement reports are calculated in sensing region (10) only if the blockage (30) is detected since the user is near to the serving base station. Thereby, signaling load to the serving transmission / reception point (1) is reduced. If blockage (30) is not detected in sensing region, the measurement reports will be calculated when the user enters the non-sensing region (20) of the serving transmission / reception point (1).

[0091] FIG. 6 focuses on the non-sensing region (20), where the handover mechanism begins. It is started with starting with the wireless communication entity (2) is identified within the outage coverage area. The measurement reports are calculated in the non-sensing region (20) since the wireless communication entity (2) is far from the serving base station and may experience signal degradation.

[0092] Trajectory prediction is utilized for both the wireless communication entity and the blockage (30) (as shown in FIG. 6). These predictions are generated based on AI / ML trajectory prediction algorithms.

[0093] Depending on the scenario: if the wireless communication entity (2) exhibits high-speed mobility, it proceeds to a step for a conditional handover. If the wireless communication entity (2) exhibits low-speed mobility, it follows step for a baseline handover. In cases where no blockage (30) in non-sensing region (20), the HO can be decided based on one criterion or more such as RSRP, SINR, load, cost, speed scenarios, HO history.

[0094] A decision is made by regarding that if blockage (30) is predicted in non-sensing region (20), the wireless communication entity (2) proceeds through the same steps based on the speed scenario.

[0095] FIG. 7 and FIG. 8 illustrate the utilization of sensing processes that may be utilized for handover prediction and blockage (30) awareness in sensing regions. This approach leverages communication signals for sensing capabilities to predict blockage (30) positioning and adapt accordingly to maintain seamless connectivity.

[0096] In this method, synchronization signal block (40) blocks, standardized by 3 GPP, are exploited for network operations. For example, as outlined in 3GPP Rel-19, an synchronization signal block (40) occupies a grid of 240 subcarriers in frequency (20 Resource Blocks, RBs) and 4 OFDM symbols in time, fitting into a 5 ms within a specific frequency domain position. These synchronization signal block (40) bursts are periodically repeated, with the number of beams depending on: Frequency range (e.g., FR1 vs. FR2), synchronization signal block (40) periodicity, and Network configuration, including subcarrier spacing and beamwidth.

[0097] The reflection of synchronization signal block (40) beams can be used for sensing. For instance, FIG. 7 illustrates the estimation of blockage (30) possibilities from the synchronization signal block Y-index (40b) echo (50) and the calculation of metrics such as angle of arrival (AoA) and threshold power using access beams from synchronization signal block X (40a) index. The length of the sensing region determines the threshold value for detection. These sensing measurements help detect potential blockages (30) and prepare for the handover process.

[0098] In FIG. 8, the sensing-based prediction of blockage is further elaborated. By employing beamsweeping techniques, the system determines when handover (HO) is necessary, optimizing parameters such as Time-to-Trigger (TTT) and Handover Margin (HOM). This ensures minimal failures or delays in the handover process.

[0099] When operating in sensing regions (10), such as those in heterogeneous networks, the handover can occur between a smaller serving transmission / reception point (1) and a larger transmission / reception point (1). Similarly, in dense networks, handover transitions may involve the wireless communication entities (2), depending on the configuration and sensed conditions.

[0100] This sensing-based approach improves handover reliability and reduces signaling overhead, particularly in challenging network scenarios involving high mobility or dense environments.

[0101] FIG. 9 illustrates the handover process and measurement reporting when a user transitions from a sensing region (10) to a non-sensing region (20). The measurement report provides an indication of the user's movement across these regions, highlighting the need for dynamic adaptation of network parameters. In the context of FIG 9, the scenario involves the wireless communication entity (2) within sensing region (10). Initially, the wireless communication entity (2) requires sensing measurement reports while it remains close to the wireless communication entity (2) (e.g., within less than distance dl). These measurements ensure accurate tracking and prediction of the wireless communication entity’s (2) trajectory.

[0102] As the wireless communication entity (2) transitions further from the wireless communication entity (e.g., at a distance dl< the wireless communication entity positioning ), predictions indicate that sensing reports are no longer necessary for the next Tsimcycle. This adaptation, based on Tmax, allows the network to cease requesting unnecessary sensing data, optimizing energy usage while maintaining performance. The condition Tsim< Tmaxcontrols this process, ensuring the system balances energy efficiency with operational accuracy. This approach dynamically adjusts the reporting requirements based on the wireless communication entity’s status and trajectory, showcasing the adaptive and energy-efficient capabilities of the system.

[0103] FIG 10 illustrates the adaptive Tmaxscenarios, highlighting transitions between sensing regions (10) and non-sensing regions (20). Here, Tmaxrefers to the maximum measurement cycles (i.e., the total measurement time). Within sensing regions (10), wireless communication entities (2) may exhibit varying movement patterns across different scenarios. These dynamic adaptations reflect the network's ability to adjust Tmaxto maintain efficiency.

[0104] As sensing regions can change dynamically or adapt based on various factors, including measurements or network decisions, the expected scenarios are shaped by conditions such as fixed sensing region (10) and wireless communication entity (2) moving in a fixed direction, fixed sensing region (10) and wireless communication entity (2) moving in a random direction, adaptive sensing region (10) and wireless communication entity (2) moving in a fixed direction and adaptive sensing region (10) and wireless communication entity (2) moving in a random direction.

[0105] In the scenario of sensing region (10) is fixed and wireless communication entity (2) moving in a fixed direction, the sensing region (109 is static, and the wireless communication entity (2) follows a predefined trajectory.

[0106] In the scenario of sensing region (10) is fixed and wireless communication entity (2) moving in a random direction, the sensing region (10) remains static, but the wireless communication entity’s (2) trajectory is unpredictable, requiring adjustments in monitoring and decisionmaking.

[0107] In the scenario of adaptive sensing region (10) and wireless communication entity (2) moving in a fixed direction, the sensing region (10 9dynamically adapts, accommodating the wireless communication entity’s (2) fixed trajectory to optimize sensing accuracy.

[0108] In the scenario of adaptive sensing region (10) and wireless communication entity (2) moving in a random direction, the sensing region (10) adapts to match the unpredictable trajectory of the wireless communication entity (2), enhancing flexibility and robustness.

[0109] For example, In the fixed direction scenario, Tmaxcan be set to a constant value, such as 4000 measurement cycles (steps), which mean when Tsimreach that Tmaxsteps no need for asked for sensing measurement reports as next step it expected to be in non-sensing region (20).

[0110] For a random direction (trajectory), Tmaxmust adapt dynamically, with Tsimadjusting accordingly.

[0111] Similarly, the sensing region's fixed or adaptive nature influences the system's ability to maintain efficiency and accuracy across various mobility patterns. This ensures the optimization of resources and sensing reliability.

[0112] When the wireless communication entity (2) moves from the sensing region (10) to the nonsensing region (20), the network must adjust Tmaxthe maximum time allowed for maintaining connectivity or making a handover decision. The adaptation of Tmaxinvolves steps of analyzing the measurement report, dynamic adjustment and coordination across regions.

[0113] Analyzing the Measurement Report: Based on the wireless communication entity ’ s (2) mobility trajectory and sensed conditions, the network evaluates the rate of change in signal strength, blockage (30) risks, and speed.

[0114] Dynamic Adjustment: If the user is moving quickly or towards an area with higher interference or blockage (30), Tmaxmay be reduced to ensure timely handover. Conversely, for slower movements or stable conditions, Tmaxcan be extended to reduce unnecessary handovers. Coordination Across Regions: In the non-sensing region (20), predictions rely on historical data, mobility patterns, and AI / ML algorithms to ensure smooth transitions, compensating for the lack of real-time sensing.

[0115] The invention finds extensive applications in advanced wireless communication scenarios. These include channel estimation and detection, enabling precise signal processing and identification, and enhanced throughput and reliability mechanisms to optimize data transmission. It supports MAC layer management for efficient resource allocation and multicell MIMO technologies, including coordinated multi-point (CoMP), which improves network performance through collaborative base station operations.

[0116] Additionally, the invention is applicable to Cloud-RAN (C-RAN) and fog-RAN (F-RAN) architectures, enhancing centralized and distributed network deployments. It facilitates cooperative user relaying networks, enabling better connectivity in challenging environments, and supports efficient precoding designs to improve signal transmission quality. Joint Radar and Communication (JRC) applications also benefit from this invention by integrating communication with radar functionalities.

[0117] Further applications include massive MIMO networks, which enhance capacity and reliability, and ultra-reliable low-latency communications (URLLC) for mission-critical operations. The invention also plays a key role in mmWave communication, enabling high-frequency network deployment, and unmanned aerial vehicle-aided communications (UAV), expanding connectivity to remote areas.

[0118] Emerging use cases like joint communication and sensing (JCAS) and Perceptive Mobile Networks (PMN) leverage its capabilities for seamless integration of sensing and communication tasks. Lastly, it is instrumental in massive machine-type communication (mMTC) and Integrated Sensing and Communication (ISAC) technologies, addressing the unique demands of loT and future network paradigms. These applications demonstrate the invention’s versatility in overcoming mobility challenges, managing shadowing and blockage (30), and improving handover performance. REFERENCES:

[0119] [1] 3GPP. (2024). LTE; Evolved Universal Terrestrial Radio Access (E-UTRA) and Evolved

[0120] Universal Terrestrial Radio Access Network (E-UTRAN); Overall description; Stage 2 (3 GPP TS 36.300 version 17.0.0 Release 17). https: / / www.etsi.org / deliver / etsi_ts / 136300_136399 / 136300 / 17.00.00_60 / ts_136300vl70000 p.pdf

[0121] [2] Alraih, Saddam, et al. "Effectiveness of handover control parameters on handover performance in 5G and beyond mobile networks." Wireless Communications and Mobile Computing 2022.1 2266282, (2022).

[0122] [3] Rehman, A. U., Roslee, M. B., & Jun Jiat, T. (2023). A survey of handover management in mobile HetNets: current challenges and future directions. Applied Sciences, 13(5), 3367.

[0123] [4] Ge, Y., Kaltiokallio, O., Chen, EL, Talvitie, J., Xia, Y., Madhusudan, G., ... & Wymeersch, H. (2024). Target Handover in Distributed Integrated Sensing and Communication. arXiv preprint arXiv:2411.01871.

[0124] [5] Zhang, H., Zhang, Y., Liu, X., Sun, K., & Zhang, Y. Resource Allocation and Mobility Management for Perceptive Mobile Networks in 6G. IEEE Wireless Communications, 2024.

[0125] [6] A Conditional Handover Strategy Based on Trajectory Prediction for High-Speed Terminals in LEO Satellite Networks

[0126] [7] Yu, C., Sun, Y., Luo, Y., & Wang, R. (2023). mmAlert: mmWave Link Blockage Prediction via Passive Sensing. IEEE Wireless Communications Letters.

[0127] [8] Zhang, Haijun, et al. "Resource Allocation and Mobility Management for Perceptive Mobile Networks in 6G." IEEE Wireless Communications (2024).

[0128] [9] Luo, Yi, et al. "Handover algorithm based on Bayesian-optimized LSTM and multi-attribute decision making for heterogeneous networks." Ad Hoc Networks 157 103454. (2024).

[0129]

[0010] Asad, Syed Muhammad. User mobility prediction and management using machine learning. Diss. University of Glasgow, 2022.

[0011] Jawad, Aqeel Thamer, Rihab Maaloul, and Lamia Chaari. "A comprehensive survey on 6G and beyond: Enabling technologies, opportunities of machine learning and challenges." Computer Networks (2023): 110085.

[0130]

[0012] Huang, Zhi-Hong, Yu-Shu Chen, and Ming-Jer Tsai. "Efficient conditional handover algorithm in 5G with blockages using recurrent neural network." 2023 IEEE 20th Consumer

[0131] Communications & Networking Conference (CCNC). IEEE, 2023.

Claims

CLAIMS1. An adaptive handover method for wireless networks having transmission / reception points (1) that carry out communication with wireless communication entities (2) in a predetermined sensing region (10) and non-sensing regions (20) characterized by comprises steps ofDetermining if the wireless communication entity (2) to be communicated in the sensing region (10) or non-sensing region (2) of the transmission / reception points (1);Applying blockage prediction algorithm in case that the wireless communication entity (2) in the sensing region (10) or applying handover prediction algorithm in case that the wireless communication entity (2) in the non-sensing region (20);Determining if mobility of the wireless communication entity (2) to be communicated is high-speed or low-speed mobility;Applying a conditional handover in case that the mobility of wireless communication entity (2) is high-speed or applying a base line handover in case that the mobility of wireless communication entity (2) is low-speed.

2. A method according to Claim 1, wherein repeating blockage prediction algorithm by sensing measurement as an input until a predetermined condition, which shows handover is required, is met by output of the blockage prediction algorithm.

3. A method according to Claim 1, wherein the handover prediction is carried out in the sensing region by a blockage prediction algorithm using sensing measurement reports collected from the transmission / reception point (1) of the non-sensing region (20) as an input.

4. A method according to Claim 3, characterized by repeating the handover prediction until a predetermined condition, which shows handover is required, is met by output of the blockage prediction algorithm.

5. A method according to Claim 4, wherein the predetermined condition is reference signal received power of target is higher than reference signal received power of serving.

6. A method according to Claim 1, wherein the handover prediction algorithm is carried out by AI-Driven or machine learning models for to forecast user mobility pattern.

7. A method according to Claim 1, characterized by further comprises steps of Transmitting synchronization signal blocks in burst mode;Determining index of beams in the synchronization signal blocks (40) bursts based on network configurations, including frequency range, synchronization signal blocks (409 periodicity, subcarrier spacing, and beamwidth;Detecting echoes (50) from reflected synchronization signal blocks beams (40);Estimating blockage (30) probabilities using the synchronization signal blocks Y-index (40b) and corresponding echo signals (50);Calculating sensing metrics, including angle of arrival (AoA) and threshold power of the sensing region, using access beams associated with the synchronization signal block X-index (40a);Defining a sensing region (10) to determine a threshold value for detection; and Utilizing the sensing measurements to identify potential blockages (30) and prepare for handover operations in the network.

8. A method according to Claim 1, wherein the prediction of blockage (30) in sensing region further comprises step of employing beam-sweeping techniques to determines necessity of handover is by optimizing parameters such as Time-to-Trigger and Handover Margin.

9. A method according to Claim 1, characterized by further comprises steps of identifying the wireless communication entity (2) within a sensing region (10) defined by proximity to the transmission / reception point (1); receiving sensing measurement reports from the wireless communication entity (2) when the wireless communication entity (2) is within a predetermined distance from the transmission / reception point (1); determining, based on the wireless communication entity’s (2) trajectory and position relative to the transmission / reception point (1), whether the wireless communication entity (2) is transitioning from the sensing region (10) to a non-sensing region (20);ceasing the request for sensing measurement reports when the wireless communication entity (2) transitions beyond the sensing region (10) for a duration controlled by a time parameter (Tsim) less than a maximum threshold (Tmax) dynamically adjusting Tmaxbased on the device's movement patterns to balance energy efficiency with operational accuracy.

10. A method according to Claim 9, characterized by further comprises of reducing size of the sensing measurement report of the sensing region in case that the no blockage (30) is detected.

11. A method according to Claim 9 or 10, wherein measurement of sensing continues in case that the no blockage (30) is detected.

12. A method according to Claim 9, wherein the sensing region (10) is defined as a region within a distance from the transmission / reception point (1) from serving transmission / reception point (1) to the maximum threshold (Tmax) and the non-sensing region (20) is defined as a region beyond the distance.

13. A method according to Claim 9, wherein the determination of the wireless communication entity’s (2) transition from the sensing region (10) to the non-sensing region (20) is based on predictive analysis of the wireless communication entity’s (2) trajectory.

14. A method according to Claim 9, characterized by further comprising dynamically adapting the sensing region (10) based on network measurements or decisions to optimize sensing accuracy.

15. A method according to Claim 14, wherein the sensing region adaptation accounts for: fixed sensing regions (10) with the wireless communication entity (2) moving in a fixed direction; fixed sensing regions (10) with the wireless communication entity (2) moving in a random direction;adaptive sensing regions (10) with the wireless communication entity (2) moving in a fixed direction; or adaptive sensing regions (10) with the wireless communication entity (2) moving in a random direction.

16. A method according to Claim 14, wherein for fixed sensing regions (10) with the wireless communication entity (2) moving in a fixed direction, Tmaxis set to a constant value, and the request for sensing measurement reports ceases when the time parameter (Tsim) reaches the maximum threshold (Tmax).

17. A method according to Claim 14, wherein for fixed sensing regions (10) with the wireless communication entity (2) moving in a random direction, Tmaxis dynamically adjusted based on variations in the wireless communication entity’s (2) trajectory.

18. A method according to Claim 14, wherein for adaptive sensing regions (10) with the wireless communication entity (2) moving in a fixed direction, the sensing region (10) dynamically expands or contracts to accommodate the wireless communication entity’s (2) trajectory.

19. A method according to Claim 14, wherein for adaptive sensing regions (10) with the wireless communication entity (2) moving in a random direction, the sensing region (10) adapts dynamically to align with the wireless communication entity’s (2) unpredictable trajectory, enhancing robustness and flexibility.

20. A processor device comprising means for carrying out the steps of the method of any of preceding claims.

21. A computer program comprising instructions which, when the program is executed by the processor device of claim 20, cause the computer to carry out the method of claim 1 to 19.

22. A computer-readable data carrier having stored thereon the computer program of claim21.

Images