An adaptive waveform dynamic switching method based on channel state prediction

Abstract

This invention discloses an adaptive waveform dynamic switching method based on channel state prediction. The method includes: constructing a channel dynamic predictor, which predicts the trajectory of changes in key channel parameters of the serving link within a future time window based on real-time satellite ephemeris, beam pointing information, and location information; determining the target transmission configuration corresponding to the terminal within the future time window based on the trajectory of changes in key channel parameters; and pre-configuring and / or sending dynamic instructions to the terminal based on the target transmission configuration, so that the terminal adopts the target transmission configuration for communication at the effective time corresponding to the prediction result within the future time window. This invention enables the terminal to match the waveform and parameters with optimal spectral efficiency and transmission robustness under current and recent channel conditions, thereby improving overall throughput, coverage reliability, and resource utilization.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to an adaptive waveform dynamic switching method based on channel state prediction. Background Technology

[0002] Non-terrestrial networks (NTNs) are a crucial component of 5G and subsequent mobile communication systems. They provide communication capabilities to areas with insufficient terrestrial cellular network coverage via satellites or high-altitude platforms, holding practical significance in scenarios such as emergency communications, the Internet of Things (IoT), broadcasting, and broadband access. Since Release 17, 3GPP has gradually introduced NTN-related technologies, and in Releases 18 and 19, it has continuously expanded the scope of applications, covering IoT, enhanced mobile broadband, and broadcasting services. Among these, low Earth orbit (LEO) satellites, due to their relatively low propagation latency and flexible deployment, have become a key focus of current NTN research and engineering applications.

[0003] Compared to terrestrial networks, NTN communication scenarios based on LEO satellites exhibit significant differences in channel characteristics. The propagation distance between the satellite and the terminal continuously changes with orbital motion, resulting in not only a large absolute value of propagation delay but also pronounced dynamic characteristics. Under these conditions, timing adjustments relying on feedback mechanisms face high signaling overhead and response pressure. Simultaneously, the high-speed motion of LEO satellites relative to the ground introduces a significant Doppler shift, reaching tens of kilohertz, and changing rapidly, posing a challenge to maintaining orthogonality in multi-carrier systems. Furthermore, the signal propagation path may be affected by factors such as terrain, buildings, or vegetation, resulting in time-varying obstructions and shadow fading, making link conditions highly time-varying.

[0004] To adapt to different channel conditions, the 5G NR standard defines various physical layer waveforms. Cyclic Prefix Orthogonal Frequency Division Multiplexing (CFD) is widely used for uplink and downlink transmission due to its high spectral efficiency and good compatibility with multi-antenna technologies. Discrete Fourier Transform Extended OFDM (DFT), as a single-carrier waveform, has a low peak-to-average power ratio, which is beneficial for improving uplink transmission efficiency and reducing sensitivity to frequency offset to some extent. In 5G-Advanced research, the characteristics of single-carrier frequency domain equalization waveforms in resisting phase noise in high-frequency bands have also attracted attention.

[0005] Existing schemes have proposed dynamically instructing terminals to switch between different waveforms using downlink control information, but these schemes mostly rely on measurements of current or historical channel states for decision-making. In NTN scenarios, due to significant propagation and processing delays, the channel state often changes by the time the waveform switching command reaches the terminal, making it difficult to match the actual transmission time. Furthermore, selection is typically limited to a finite set of pre-configured parameters, with limited ability to adjust key parameters such as subcarrier spacing and cyclic prefixes, making it difficult to cover the large variations in Doppler spread and delay spread in NTN. In addition, existing signaling is primarily designed for low-latency terrestrial networks; directly applying it to NTN can easily lead to problems with scheduling timing alignment and signaling efficiency. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an adaptive waveform dynamic switching method based on channel state prediction. This method helps terminals match the waveforms and parameters that offer the best spectral efficiency and transmission robustness under current and recent channel conditions, thereby improving overall throughput, coverage reliability, and resource utilization.

[0007] To address the aforementioned technical problems, this invention discloses an adaptive waveform dynamic switching method based on channel state prediction, the method comprising: A channel dynamic predictor is constructed on the network-side node. The channel dynamic predictor predicts the trajectory of changes in key channel parameters of the service link within a future time window based on real-time satellite ephemeris, beam pointing information, and location information. Based on the channel key parameter change trajectory, the target transmission configuration of the terminal within the future time window is determined. The target transmission configuration includes at least: waveform switching strategy, target transmission waveform, and key parameter configuration corresponding to the target transmission waveform. Based on the target transmission configuration, the terminal is pre-configured and / or a dynamic indication is sent, so that the terminal uses the target transmission configuration to communicate at the effective time corresponding to the prediction result within the future time window.

[0008] In some implementations, the channel key parameter change trajectory is a time-ordered predicted channel state parameter sequence. The predicted channel state parameter sequence includes at least multiple prediction time points and corresponding channel state parameter sets. The channel state parameter sets include one or more of the following: Doppler shift, predicted signal-to-noise ratio / channel quality indication, link propagation delay variation, and shadow fading risk.

[0009] In some implementations, network-side nodes obtain the shadow fading state marker through the following steps: Obtain terminal location information and corresponding satellite location information; The line-of-sight propagation path between the terminal and the satellite is determined based on the terminal location information and the satellite location information. By combining the digital elevation model and / or building model, it is determined whether the line-of-sight propagation path is blocked by terrain or buildings; The shadow decay state risk is generated based on the occlusion judgment result.

[0010] In some implementations, determining the target transmission configuration of the terminal within the future time window based on the channel key parameter change trajectory includes: Based on the Doppler frequency shift, Doppler change rate, predicted channel quality indication, and shadow fading risk corresponding to different prediction time points, the channel state change characteristics within the future time window are analyzed, and the target transmission waveform is determined by selecting from multiple candidate transmission waveforms. Based on the predicted channel state parameter sequence, adjust the configuration of key parameters corresponding to the target transmission waveform.

[0011] In some implementations, the target transmission waveform includes an orthogonal frequency division multiplexing waveform and a single-carrier frequency domain equalization waveform; Waveform switching strategies include: When the Doppler frequency shift and / or Doppler rate of change in the predicted channel state parameter sequence meet the preset conditions, or when the predicted channel state parameter sequence indicates that the terminal will enter or be in a shadow fading state, the target transmission waveform is determined to be a single-carrier frequency domain equalized waveform. When the Doppler frequency shift and / or Doppler rate of change do not meet the preset conditions, and the predicted channel state parameter sequence indicates that the channel state is relatively stable, the target transmission waveform is determined to be an orthogonal frequency division multiplexing waveform.

[0012] In some implementations, the key parameter configuration includes at least the subcarrier spacing and the cyclic prefix length; The network-side node adjusts the subcarrier spacing based on the Doppler frequency shift and / or Doppler rate of change in the predicted channel state parameter sequence; and adjusts the cyclic prefix length based on the propagation delay change in the predicted channel state parameter sequence.

[0013] In some implementations, pre-configuration and / or sending dynamic indications to the terminal based on the target transmission configuration includes: The available waveform and parameter set is configured to the terminal through radio resource control reconfiguration messages and / or system information blocks; the available waveform and parameter set includes multiple uplink bandwidth part configurations, each uplink bandwidth part configuration is associated with a transmission waveform and a corresponding key parameter configuration, the key parameter configurations include subcarrier spacing and cyclic prefix length; In the radio resource control reconfiguration message and / or system information block, the terminal is enabled to perform adaptive waveform dynamic switching.

[0014] In some implementations, pre-configuring and / or sending dynamic indications to the terminal based on the target transmission configuration further includes: Based on the predicted channel state parameter sequence, the target effective time of the terminal within a future time window is determined. Based on the target effective time, scheduling offset parameters, downlink propagation delay, and terminal processing delay, the timing for sending downlink control information is determined. At the specified transmission timing, downlink control information containing waveform indication information and the scheduling offset parameter is sent to the terminal. The downlink control information is used to instruct the terminal to switch to the target transmission configuration from the available waveform and parameter set, so that the terminal starts to use the target transmission configuration for communication in the corresponding time slot according to the scheduling offset parameter at the time when the target takes effect.

[0015] In some implementations, it also includes: When the predicted channel state parameter sequence exhibits periodic or regular changes, a waveform switching time series table is generated based on the change trajectory of the key channel parameters. The waveform switching time sequence table is sent to the terminal so that the terminal can adaptively switch waveforms at a future preset time reference point; the waveform switching time sequence table includes the future preset time reference point and time offset, waveform configuration index and / or effective duration.

[0016] In some implementations, it also includes: The accuracy of channel prediction is verified based on the deviation between the predicted channel state parameter sequence and the actual measured channel state parameters at the terminal. When the deviation exceeds a preset threshold, compensation is provided based on the scheduling parameters determined by the prediction results or the channel dynamic predictor is updated. When the prediction error continues to exceed the allowable range, the prediction-based dynamic waveform switching is paused, and the terminal is instructed to fall back to the resource scheduling mode based on terminal feedback or the preset high robustness transmission configuration.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention constructs a channel dynamic predictor on the network side based on satellite ephemeris, beam pointing information, and terminal location information. This predicts changes in key channel parameters within a future time window and determines the target transmission configuration of the terminal. This allows the adjustment of the transmission waveform and its key parameters to maintain a better time match with the time of channel changes. By pre-configuring the terminal and combining it with dynamic indication, the terminal communicates using the target transmission configuration at the effective time corresponding to the prediction result. Under the conditions of long latency in non-terrestrial networks and rapid channel changes, this invention improves the adaptability of waveform switching and scheduling control and reduces the performance loss caused by switching lag and mismatch. Attached Figure Description

[0018] Figure 1 A schematic diagram of the system architecture of an adaptive waveform dynamic switching method based on channel state prediction provided by the present invention; Figure 2 This is a geometrical diagram illustrating the calculation of the intermediate point height between the UE and the satellite in an adaptive waveform dynamic switching method based on channel state prediction provided by the present invention. Figure 3 This is a flowchart illustrating the process of calculating shadow fading state risk in an adaptive waveform dynamic switching method based on channel state prediction provided by the present invention. Figure 4 This is a schematic diagram of the channel prediction, verification, and update process in an adaptive waveform dynamic switching method based on channel state prediction provided by the present invention. Figure 5 This is a schematic diagram of the adaptive waveform and parameter configuration in an adaptive waveform dynamic switching method based on channel state prediction provided by the present invention. Figure 6 The flowchart shows the waveform switching signaling interaction in the adaptive waveform dynamic switching method based on channel state prediction provided by this invention. Detailed Implementation

[0019] To better understand and implement this invention, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] The terms “comprising” and “having” and any variations thereof in this invention are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products or devices.

[0021] In non-terrestrial network (NTN) communication scenarios, due to the high-speed movement of satellites, the long distance between satellites and the ground, and the complex propagation environment, the propagation delay, Doppler shift, and rate of change of the link between the terminal and the network exhibit time-dependent and predictable characteristics. This invention discloses an adaptive waveform dynamic switching method based on channel state prediction. By introducing a channel dynamic prediction mechanism on the network side, the method predicts the channel state within a future time window and makes waveform switching and parameter adjustment decisions in advance based on the prediction results, achieving timely adaptive transmission of the terminal during channel state changes.

[0022] like Figure 1 As shown, this method includes: Step S1: Construct a channel dynamic predictor. The channel dynamic predictor predicts the trajectory of changes in key channel parameters of the service link within a future time window based on real-time satellite ephemeris, beam pointing information, and location information.

[0023] The input information for the channel dynamic predictor, in addition to real-time satellite ephemeris, beam pointing, and terminal location information, can also incorporate one or more of the following information to improve prediction accuracy: device capability information reported by the terminal (such as antenna gain, supported waveforms, and bandwidth capabilities); motion state determined by the terminal location sequence or direct reporting by the terminal (such as speed and direction of motion); and historical channel measurement information of the terminal or the region stored on the network side. Based on the fused multidimensional information, the predictor can generate more accurate trajectory changes in key channel parameters using geometric models, filtering algorithms, or machine learning models.

[0024] Network-side nodes are base stations (gNBs) in non-terrestrial networks, on-board processing units, or ground gateways, and are equipped with channel dynamic predictors for performing channel state prediction. The network side broadcasts or configures auxiliary information for channel state prediction to terminals via system information blocks or radio resource control signaling. This auxiliary information includes ephemeris data of serving satellites or equivalent orbital parameter models, cell reference point location information, and common timing advance (Common TA), providing a basic reference for subsequent link delay and Doppler calculations.

[0025] The terminal can selectively report its geographic location information obtained through the Global Navigation Satellite System (GNSS) via Radio Resource Control (RRC) signaling or Media Access Control Unit (MACCE). When the terminal cannot obtain GNSS positioning information, the network side obtains the terminal's location information based on existing network positioning technologies. The channel dynamic predictor deployed on the network side predicts the link status of the terminal or a group of terminals within a beam within a future time window based on the aforementioned ephemeris information, location information, and beam pointing information. This prediction includes link propagation delay, Doppler shift, and Doppler rate of change.

[0026] In one embodiment, the channel dynamic predictor outputs a structured trajectory of changes in key channel parameters, which is a time-ordered sequence of predicted channel state parameters. This output is not an instantaneous value but a continuous prediction of the channel state within a future configurable time window. The predicted channel state parameter sequence includes at least multiple prediction time points and corresponding sets of channel state parameters, which include one or more of the following: Doppler shift, predicted signal-to-noise ratio / channel quality indication, link propagation delay variation, and shadow fading risk.

[0027] In one embodiment, the predictor output is a discrete-time series, whose data structure can be represented as a set of vectors: {(t1, H(t1)), (t2, H(t2)), …, (tN, H(tN))}. Here, ti represents the i-th future prediction time point, and H(ti) represents the synthesized channel state vector predicted at time point ti. The prediction time point is aligned with the basic scheduling time unit of the new air interface system, preferably at the time slot or subframe level; for example, with a 15kHz subcarrier spacing, a time slot is 1 millisecond. This granularity supports the scheduler in performing time-slot-by-time or multi-slot-by-multi-time-slot resource and waveform assignment. The predicted coverage future time window length (tN-t1) is a configurable parameter, typically ranging from hundreds of milliseconds to several seconds, to balance prediction accuracy and scheduling foresight: shorter windows (e.g., 100–500 ms) are used for rapid response to channel changes; longer windows (e.g., 1–5 s) are used to leverage the strong regularity of LEO satellite orbits, supporting longer-term resource reservations and pattern switching, and can be implemented in conjunction with the planned switching mechanism issued by MAC CE. The length of the prediction time window is a configurable parameter, which can be set to hundreds of milliseconds to several seconds according to actual service requirements, achieving a balance between prediction accuracy and scheduling foresight, and is not limited in this application.

[0028] The core prediction terms included in the synthesized channel state vector are at least: link propagation delay (τ, in seconds or milliseconds), Doppler shift (fd, in Hz), and Doppler rate of change (dfd / dt, in Hz / s). Link propagation delay is used for precise timing advance compensation and scheduling alignment; Doppler shift is used to assess the severity of inter-carrier interference (ICI) and determine the transmission waveform (OFDM / SC-FDE) and subcarrier spacing (SCS); Doppler rate of change reflects the rapid change characteristics of Doppler and is related to the density configuration of the phase tracking reference signal (PTRS). The channel dynamic predictor can also selectively output derived or evaluation terms, such as the predicted signal-to-noise ratio / channel quality indication (SINR / CQI), which can be obtained based on the predicted path loss and interference model for modulation and coding strategy (MCS) pre-selection.

[0029] To predict non-frequency selective slow fading caused by terrain or ground cover, in some implementations the channel dynamic predictor also combines digital maps or terrain models to assess the possible shadow fading state at future times, and obtain the shadow fading state risk, also known as the shadow fading state flag (ShadowFlag), which is in the form of Boolean value or probability value to indicate whether the corresponding time is likely to be in a shadow occlusion state.

[0030] The evaluation process integrates ensemble computing and geographic information system data; the specific evaluation process is as follows: Figure 2 , 3 And the following examples: Step S11: Obtain terminal location information and corresponding satellite location information; Step S12: Determine the line-of-sight propagation path between the terminal and the satellite based on the terminal location information and the satellite location information; Step S13: Combine the digital elevation model and / or building model to determine whether the line-of-sight propagation path is blocked by terrain or buildings; Step S14: Generate the shadow fading state risk based on the occlusion judgment result.

[0031] The system acquires dynamic geometric relationship data and digital map data. The channel dynamic predictor calculates the geometric relationship between the terminal and the serving satellite in real time, including the line-of-sight propagation path between the terminal and the satellite. To this end, the predictor acquires terminal location information and satellite location information. The terminal location information includes longitude, latitude, and altitude, which can be obtained and reported by the terminal through the Global Navigation Satellite System, or obtained through network positioning technology in the absence of navigation satellite positioning capabilities; satellite location information is calculated based on real-time ephemeris data. Network-side nodes (e.g., ground gateways or onboard processing units) pre-store or acquire digital map data of the service area in real time. This digital map data includes a high-precision digital elevation model (DEM) and / or 3D building model data. The DEM describes the terrain elevation, while the 3D building model describes the outline and height information of fixed obstructions such as major buildings.

[0032] To ensure the accuracy of the geometric calculations, in this embodiment, the shadow fading risk assessment process is performed in a unified three-dimensional coordinate system. In one embodiment, the geometric calculations use the Earth-centered Earth-fixed (ECEF) coordinate system as the reference coordinate system. This coordinate system is a three-dimensional rectangular coordinate system, with its origin located at the Earth's center of mass, the Z-axis pointing towards the Earth's North Pole, the X-axis pointing towards the intersection of the Prime Meridian and the Equator, and the Y-axis forming a right-handed coordinate system together with the X and Z axes.

[0033] Based on the aforementioned coordinate system, the terminal coordinates Pue and the satellite coordinates Psat are determined. According to the terminal's longitude, latitude, and altitude, the terminal's position is converted from the geographic coordinate system to the terminal coordinates Pue (Xue, Yue, Zue) in the ECEF coordinate system. The altitude is the geodetic height relative to the Earth's reference ellipsoid (e.g., the WGS84 ellipsoid). Based on real-time ephemeris data, the satellite coordinates Psat (Xsat, Ysat, Zsat) in the ECEF coordinate system are obtained. In this embodiment, the Earth's reference ellipsoid (e.g., WGS84) serves as the unified reference datum for altitude calculation.

[0034] After determining the terminal and satellite coordinates, the channel dynamic predictor performs line-of-sight propagation path profile analysis between the terminal and the satellite, and makes a blocking judgment. In the ECEF coordinate system, the terminal coordinates Pue and the satellite coordinates Psat determine a three-dimensional straight line, representing the line-of-sight propagation path between the terminal and the satellite. This straight line can be represented parametrically as follows: P(λ) = Pue + λ·(Psat) Pue) Where λ is a scaling factor; when λ=0, it corresponds to the terminal coordinates; when λ=1, it corresponds to the satellite coordinates. The change of λ from 0 to 1 describes the entire propagation path from the terminal to the satellite.

[0035] In one implementation, only the path interval from the terminal upwards to the satellite orbit is analyzed; therefore, the sampling interval is set to λ∈[0,1]. Within this interval, sampling is performed at a fixed Δλ, which corresponds to a step size of 10 to 100 meters in the ground projection distance, to obtain a series of intermediate parameter values ​​λi. The corresponding three-dimensional spatial coordinates of the intermediate point Pi=(Xi,Yi,Zi) are then calculated. The number and distribution of the intermediate points can be configured according to accuracy requirements and are not limited to covering the entire interval.

[0036] For each intermediate point Pi, the channel dynamic predictor calculates and compares the line height Hlinki with the actual terrain height Hterraini. The ECEF coordinates of intermediate point Pi are converted to geographic coordinates, including longitude, latitude, and geodetic height. The resulting geodetic height is the height Hlinki of the line at that point relative to the reference ellipsoid. This height represents the theoretical height of the radio propagation path from the Earth's surface at that location under an ideal, smooth Earth model. The intermediate point Pi is projected onto the Earth's surface along the normal direction of the Earth's reference ellipsoid, obtaining the corresponding surface projection point. Based on the latitude and longitude indexed digital elevation model of this projection point, the corresponding terrain elevation Hterraini is obtained through interpolation. This terrain elevation also uses the reference ellipsoid as the height reference and can comprehensively consider the height information of terrain and fixed buildings. The actual terrain height Hterraini is compared with the line height Hlinki. The line-of-sight propagation path is determined to be obstructed at the intermediate point when the following conditions are met: Hterraini+Margin>Hlinki Margin is a protective margin used to compensate for the accuracy errors of the digital elevation model, the effects of unmodeled objects (such as trees and small buildings), and atmospheric refraction.

[0037] The channel dynamic predictor traverses all intermediate points Pi in λ from 0 to 1. Once any intermediate point satisfies the above-mentioned occlusion condition, it can be determined that the link between the terminal and the satellite is blocked at the corresponding time, and this intermediate point can be regarded as the key location where the occlusion occurs. After completing the occlusion judgment, the shadow fading risk assessment process outputs the assessment result. In one embodiment, the output result includes the shadow fading state risk, which can be a Boolean value to indicate whether line-of-sight occlusion exists; in another embodiment, the output result can also be a shadow fading probability parameter to characterize the probability of significant shadow fading. Combining factors such as the number of occlusion points, the degree of occlusion, and the accuracy of the digital elevation model, a predicted shadow fading depth parameter (e.g., in decibels) can also be output to represent the possible degree of slow fading.

[0038] The aforementioned shadow fading status markers, shadow fading probabilities, or predicted shadow fading depths are used as part of the channel state parameters output by the channel dynamic predictor, and are then used by the upper-layer waveform switching decision engine or parameter adjustment module. For example, when the predicted shadow fading probability exceeds a preset threshold at a future time slot, the network-side node can instruct the terminal to switch to a more robust transmission waveform several scheduling cycles in advance and adopt corresponding modulation and coding parameters to reduce the impact of shadow fading on communication continuity.

[0039] Step S2: Based on the channel key parameter change trajectory, determine the target transmission configuration of the terminal within the future time window. The target transmission configuration includes at least: waveform switching strategy, target transmission waveform, and key parameter configuration corresponding to the target transmission waveform.

[0040] Specifically, it includes the following steps: Step S21: Based on the Doppler frequency shift, Doppler change rate, predicted channel quality indication, and shadow fading state risk corresponding to different prediction time points, analyze the change characteristics of the channel state within the future time window, and select from multiple candidate transmission waveforms to determine the target transmission waveform; Step S22: Based on the predicted channel state parameter sequence, adjust the configuration of key parameters corresponding to the target transmission waveform.

[0041] The decision engine in the network-side node formulates a waveform switching strategy for the future time window based on the predicted channel state parameter sequence. This waveform switching strategy indicates the type of transmission waveform adopted by the terminal at different predicted time points or within different time periods. The decision engine analyzes the changing characteristics of the channel state within the future time window based on the Doppler shift, Doppler change rate, predicted channel quality indication, and shadow fading risk corresponding to different predicted time points, and compares multiple candidate transmission waveforms to determine the target transmission waveform.

[0042] In one embodiment, the waveform switching strategy includes: when the Doppler frequency shift and / or Doppler rate of change in the predicted channel state parameter sequence meets a preset condition, or when the predicted channel state parameter sequence indicates that the terminal will enter or be in a shadow fading state, the target transmission waveform is determined to be a single-carrier frequency domain equalized waveform. When the Doppler frequency shift and / or Doppler rate of change do not meet the preset conditions, and the predicted channel state parameter sequence indicates that the channel state is relatively stable, the network-side node determines that the target transmission waveform is an orthogonal frequency division multiplexing waveform.

[0043] For example, when the predicted channel state parameters indicate a large Doppler frequency shift or a high Doppler rate of change, such as when the terminal is located at the edge of the beam or near the moment a satellite passes overhead, a single-carrier frequency-domain equalized waveform is preferentially selected, taking advantage of the single-carrier characteristics and stronger robustness to frequency offset. When the predicted channel state is relatively stable, the Doppler frequency shift is small, and the terminal is in a good coverage area, an orthogonal frequency division multiplexing waveform is selected to maximize spectral efficiency and spatial multiplexing gain. When the predicted channel state parameters predict that the terminal will enter or leave a shadowed area, waveform switching is performed before or after the corresponding prediction time point. The power efficiency advantage is utilized to maintain the connection, switching to a single-carrier frequency-domain equalized waveform in advance, or switching back to an orthogonal frequency division multiplexing waveform upon leaving, increasing the data rate to cope with impending shadow fading or channel recovery. The specific judgment conditions and thresholds of the above waveform switching strategy can be adjusted according to the system configuration, and this invention is not limited to the above examples. After determining the waveform switching strategy, the network-side node selects a target transmission waveform from multiple candidate transmission waveforms based on the waveform switching strategy, and the terminal transmits uplink or downlink within the future time window. In one embodiment, the candidate transmission waveforms include at least orthogonal frequency division multiplexing waveforms and single-carrier frequency domain equalization waveforms, and the selection of the target transmission waveform is based on the Doppler characteristics, channel stability, and shadow fading risk reflected in the predicted channel state parameter sequence.

[0044] Based on the selected target transmission waveform, the network-side node further dynamically adjusts the configuration of key parameters corresponding to the target transmission waveform based on the predicted channel state parameter sequence.

[0045] In one embodiment, when the target transmission waveform is an orthogonal frequency division multiplexing (OFDM) waveform, the network-side node adjusts the subcarrier spacing based on the predicted Doppler frequency shift and / or Doppler rate of change; and configures or adjusts the cyclic prefix length based on the predicted link propagation delay and its variation. For example, when the predicted Doppler spread increases, a larger subcarrier spacing can be selected; when the predicted delay spread increases, a longer cyclic prefix or an extended cyclic prefix can be configured.

[0046] When the target transmission waveform is a single-carrier frequency-domain equalized waveform, the network-side node can adjust the physical layer parameters related to the single-carrier frequency-domain equalized waveform based on the predicted channel state, and can also adjust the configuration of the phase tracking reference signal in combination with the predicted Doppler variation characteristics and phase stability requirements. For example, during periods when the predicted phase noise is severe, the phase noise may be caused by high-frequency transmission or large Doppler frequency shift and its rate of change. The network-side node can instruct to increase the insertion density of the phase tracking reference signal to improve the phase tracking accuracy. At the same time, it can configure the corresponding subcarrier spacing and cyclic prefix length for the single-carrier frequency-domain equalized waveform according to the prediction results.

[0047] Step S3: Based on the target transmission configuration, pre-configure and / or send a dynamic indication to the terminal so that the terminal uses the target transmission configuration to communicate at the effective time corresponding to the prediction result within the future time window.

[0048] In one embodiment, the network-side node semi-statically configures an available waveform and parameter set to the terminal via a radio resource control reconfiguration message and / or system information block. The available waveform and parameter set includes multiple uplink bandwidth portion configurations, each associated with a transmission waveform and a set of key parameter configurations corresponding to that transmission waveform. The key parameter configurations include at least the subcarrier spacing and the cyclic prefix length.

[0049] Simultaneously, the network-side node enables the terminal to perform prediction-based dynamic waveform adaptive operation in the radio resource control reconfiguration message and / or system information block. This allows the terminal to switch to a target transmission configuration from the available waveform and parameter set based on the predictive decision of the network side when it subsequently receives a dynamic indication. Through this pre-configuration method, the network-side node pre-establishes multiple switchable transmission configuration bases for the terminal without increasing frequent higher-layer signaling overhead.

[0050] In another implementation, the network-side node determines the target effective time of the terminal within a future time window based on a predicted channel state parameter sequence. The target effective time corresponds to the point in time when the predicted channel state will change; this point in time can be represented based on absolute time and further mapped to a system frame number and a timeslot number. After determining the target effective time, the network-side node determines the timing for transmitting downlink control information based on the target effective time, scheduling offset parameters, downlink propagation delay, and terminal processing delay.

[0051] In one embodiment, the timing of sending the downlink control information is determined by working backward from the target effective time, so as to ensure that the terminal can complete the reception, parsing and target transmission configuration of the downlink control information before the target effective time.

[0052] At a predetermined transmission timing, the network-side node sends downlink control information to the terminal. This downlink control information includes waveform indication information and the scheduling offset parameters. The downlink control information instructs the terminal to switch to the target transmission configuration from the available waveform and parameter set.

[0053] After receiving the downlink control information, the terminal parses the waveform indication information to determine the corresponding target uplink bandwidth configuration. Based on the scheduling offset parameter, it determines at which time slot after receiving the downlink control information to begin communication using the target transmission configuration. The moment the terminal begins communication using the target transmission configuration is aligned with the target activation time determined by the network side based on the predicted channel state parameter sequence, achieving prediction-driven, timely waveform switching and parameter adjustment.

[0054] Specifically, to support the aforementioned prediction-based dynamic waveform adaptive operation, network-side nodes dynamically trigger terminal waveform switching via downlink control information (DCI). The downlink control information can employ a newly defined downlink control information format or be implemented by extending existing downlink control information formats used for scheduling, such as extending downlink control information formats used for uplink scheduling (e.g., DCI format 01) to introduce fields for waveform switching indication. The downlink control information may also include waveform indication information, used to instruct the terminal to switch to the corresponding target transmission configuration from a pre-configured set of available waveforms and parameters. This waveform indication information can take the form of a 1-2 bit indication field, used to distinguish different uplink bandwidth portion configurations (BWP) or different transmission waveform types.

[0055] In non-terrestrial network (NTN) scenarios, due to the significant propagation delay between the satellite and the terminal, the network side's decision to switch waveforms at future times based on predictions needs to be sent in advance via downlink control information to ensure that the terminal can perform waveform switching on time when the target takes effect. Therefore, this embodiment introduces a timing alignment mechanism based on scheduling offset parameters to align the scheduling timing of downlink control information with the predicted channel state change points.

[0056] Specifically, the network-side channel dynamic predictor determines, based on the predicted channel state parameter sequence at time t0, that the terminal's channel state will change significantly at a future absolute time t2, requiring a switch in the transmission waveform. The absolute time t2 can be further mapped to the System Frame Number (SFN) and the slot number.

[0057] After determining the target activation time t2, the network-side node sends downlink control information containing waveform indication information at time t1. Time t1 is determined by working backward from the target activation time t2, and takes into account the scheduling offset parameter K2, terminal processing delay (Δprocessing), and downlink propagation delay (Δpropagation) to ensure that the downlink control information can be correctly received and parsed before the terminal performs the handover operation.

[0058] The terminal receives the downlink control information at time t1 after adding the downlink propagation delay, and parses the waveform indication information therein to determine the target uplink bandwidth configuration to which it needs to switch. Based on this, the terminal determines, according to the scheduling offset parameter K2 indicated in the downlink control information, to start uplink transmission using the new uplink bandwidth configuration from the K2th time slot after receiving the downlink control information, thereby achieving prediction-driven, timely waveform switching.

[0059] In one implementation, the determination of the target activation time is based on an absolute time alignment mechanism. Network-side nodes and terminals share a unified time reference based on the global navigation satellite system or network synchronization time, enabling the predicted channel state change points to be accurately mapped to system frame number (SFN) and slot number. The scheduling offset parameter K2, as a relative time offset, implicitly indicates the specific slot position where the terminal performs waveform switching in the downlink control information.

[0060] When the predicted channel state parameter sequence exhibits periodic or regular changes, the network-side node no longer triggers waveform switching solely through successive downlink control information. Instead, it generates a waveform switching time sequence table based on the trajectory of the channel's key parameter changes and sends it to the terminal via the Medium Access Control Element (MAC CE) to reduce the signaling overhead caused by frequent dynamic indications. Specifically, this includes the following steps: Step S4: When the predicted channel state parameter sequence exhibits periodic or regular changes, generate a waveform switching time series table based on the change trajectory of the key channel parameters; The waveform switching time sequence table is sent to the terminal, enabling the terminal to adaptively switch waveforms at the specified time. The waveform switching time sequence table includes a future preset time reference point and time offset, waveform configuration index, and / or effective duration.

[0061] In one embodiment, the waveform switching time sequence table is a structured dataset containing at least one list entry, each indicating the waveform configuration to be used at a specific time point. The time reference point is an absolute starting time reference, serving as the basis for subsequent time calculations; for example, a specific System Frame Number (SFN) and Slot number. The time offset is the relative time offset relative to the time reference point, indicating the occurrence of subsequent waveform switching actions, for example, an offset of N slots. The waveform configuration index indicates the target configuration index to be activated at the corresponding time point, within the pre-configured set of available waveforms and parameters in the Radio Resource Control reconfiguration message or System Information Block, such as the Uplink Bandwidth Part Identifier (BWP-ID). The effective duration determines the duration of application of the waveform configuration after activation; after the duration expires, the terminal can either revert to the default configuration or be indicated by the next entry in the list for subsequent configuration.

[0062] The waveform switching time series table is generated based on the output of the channel dynamic predictor and serves as the specific execution plan for the prediction results. Specifically, the network-side nodes take the predicted channel state parameter sequence as input, which includes parameters such as Doppler frequency shift, Doppler rate of change, and shadow fading risk over a future period. The intelligent scheduler analyzes the predicted channel state parameter sequence to identify periodic or regular variation characteristics.

[0063] For example, the intelligent scheduler identifies peaks in Doppler shift every predetermined number of time slots, or increases the risk of shadow fading several time slots before the satellite enters a specific terrain-obstructed area. Based on the identified periodic patterns, the intelligent scheduler generates a mapping relationship between time and optimal waveform configuration, forming a waveform switching time series table.

[0064] In one exemplary embodiment, the waveform switching time sequence table may include the following entry: (SFN=100, Slot=0) → BWP-ID 2, which corresponds to the use of a single-carrier frequency domain equalized waveform to enhance robustness against Doppler. (Offset time offset = 200 Slots) → BWP-ID 1, corresponding to the use of orthogonal frequency division multiplexing waveform, to improve spectral efficiency; (Offset time offset = 280 Slots) → BWP-ID 2, corresponding to switching back to single-carrier frequency domain equalization waveform.

[0065] In one embodiment, the network-side node defines a new MAC CE type, the payload of which carries the waveform switching time sequence table. Upon receiving the MAC CE, the terminal stores the waveform switching time sequence table locally and, based on the locally maintained system frame number and timeslot timing information, automatically switches to the corresponding waveform configuration at the specific time indicated by the time sequence table. This reduces control signaling overhead and eliminates the need to wait for dynamic triggering of downlink control information during each switch.

[0066] By employing the above methods, in scenarios where channel state changes are highly periodic or regular, control signaling overhead is reduced, and the determinism and timing accuracy of waveform switching execution are improved. This approach is applicable to scenarios where predictable channel changes are caused by the orbital motion of low-Earth orbit satellites.

[0067] To ensure long-term effectiveness and overall system robustness during actual operation, this embodiment introduces the following... Figure 4 The diagram illustrates a closed-loop prediction-verification-update mechanism, providing corresponding backoff and protection measures in case of prediction failure. Network-side nodes continuously compare the predicted channel state with the actual measured channel state to evaluate the accuracy of the prediction model in real time. The comparison can be based on multiple verification metrics, including at least one or more of the following: Timing Advance Error (TAError), Doppler Frequency Offset Error (Doppler Error), and Channel Quality Error (CQError). Timing Advance Error is the deviation between the predicted link propagation delay τpred(t) and the actual timing advance value used to maintain terminal uplink synchronization. Doppler Frequency Offset Error is the difference between the predicted Doppler frequency shift fdpred(t) and the frequency offset value actually estimated using an uplink reference signal (e.g., demodulation reference signal DMRS). Channel Quality Error is the difference between the signal-to-noise ratio (SNR) or channel quality indication estimated based on the predicted shadow fading risk and the SNR or channel quality indication actually measured and reported by the terminal.

[0068] The network side sets dynamic thresholds for the aforementioned verification indicators. If the verification indicator remains below the corresponding threshold within a continuous scheduling period, the prediction model is considered to be in a valid state; if the verification indicator exceeds the threshold, the network-side node initiates a multi-level prediction model adjustment process to gradually restore prediction accuracy.

[0069] In one implementation, for short-term rapid compensation, the network applies a dynamic compensation offset to the scheduling parameters calculated based on the prediction results at the physical layer, quickly correcting for the next or a few subsequent transmissions. The scheduling parameters include, but are not limited to, the scheduling offset parameter K2 and pre-compensated Doppler frequency offset. For medium-term model calibration, the network analyzes the source type of the prediction error. For example, when the timing lead error shows a stable offset trend, it can be determined that the terminal location information may be inaccurate, triggering a higher-precision positioning process; when the Doppler frequency offset error shows a regular deviation, it can be determined that there is drift in the satellite ephemeris or orbital parameters, and backup ephemeris data can be called or the orbital parameters can be fine-tuned and calibrated. During long-term model adaptive updates, the network uses historical prediction data and corresponding actual measurement data as samples to periodically update the internal parameters of the prediction model. The update process can employ online learning or incremental learning methods, enabling the channel dynamic predictor to adapt to long-term environmental changes, such as satellite orbit perturbations or seasonal ground feature changes.

[0070] When the channel dynamic predictor experiences significant errors or prediction failures, the system must possess robust backoff and protection capabilities to prevent prediction failures from adversely affecting communication reliability. In one implementation, when the prediction error exceeds the maximum permissible threshold for multiple consecutive scheduling cycles, or when a terminal reports prediction indication failure information through the media access control unit, the network-side node suspends prediction-based dynamic waveform switching operations and stops sending downlink control information or waveform switching time sequence tables based on the prediction results. In this backoff mode, the network-side node instead relies on the channel state information and sounding reference signals periodically reported by the terminal to perform reactive resource scheduling and waveform switching based on real-time feedback.

[0071] In one implementation, if the link quality continues to deteriorate after falling back to the traditional feedback mode, for example, if the bit error rate increases significantly, the network-side node can command the terminal to switch to a preset high-robustness default transmission configuration. This high-robustness default configuration may include using a single-carrier frequency-domain equalization waveform, a larger subcarrier spacing, and a longer cyclic prefix length. This configuration prioritizes maintaining uninterrupted connection rather than optimal spectral efficiency.

[0072] In one implementation, the network-side node can send a fallback instruction to the terminal via explicit signaling, such as through a specific downlink control information field or an emergency media access control unit, explicitly commanding the terminal to immediately enable the specified fallback configuration.

[0073] In another implementation, the fallback rule can also be implicitly agreed upon through radio resource control pre-configuration information. For example, when the terminal does not receive any dynamic waveform switching indication within a preset number of consecutive time windows, the terminal automatically activates a predefined default robust uplink bandwidth portion configuration.

[0074] The invention will now be explained using a LEO satellite NTN uplink transmission scenario as an example: Assuming a low-Earth orbit satellite acts as a transparent relay payload, or carries a regenerative onboard base station, it provides 5G NR-NTN communication services to multiple ground terminals within its coverage area. The base station functional unit on the satellite side, or its associated ground gateway, operates a channel dynamic predictor and an intelligent scheduler. The channel dynamic predictor predicts the trajectory of key channel parameter changes between the terminal and the satellite within a future time window; the intelligent scheduler, based on the prediction results, performs waveform switching decisions, parameter configuration, and corresponding signaling scheduling operations.

[0075] After obtaining the above information, the channel dynamic predictor calculates the channel state within a preset time window based on the prediction model. For example, it predicts the channel changes within a future time window of T = 500ms. In an exemplary embodiment, the prediction calculation may include, but is not limited to, the following: propagation delay prediction, Doppler shift prediction, and shadow fading risk. Based on the satellite position S(t) and the UE position U, the Euclidean distance d(t) = |S(t) - U| is calculated, then the propagation delay τ(t) = d(t) / c (c is the speed of light). The relative radial velocity vr(t) between the satellite and the terminal UE is calculated as vr(t) = d(d(t)) / dt, then the Doppler shift fd(t) = fc * vr(t) / c (fc is the carrier frequency). Combining the UE position digital elevation model (DEM) and the satellite elevation angle θ(t), the existence of direct path obstruction is determined through geometric relationships, and the additional loss is estimated to obtain the shadow fading risk.

[0076] It should be noted that the above prediction calculation model is only an exemplary description, and the present invention does not limit the specific mathematical model or calculation method.

[0077] Assuming the channel dynamic predictor outputs the predicted trajectory shown in the attached figure, corresponding to the changes in the channel state over a future period, it can be divided into the following categories: Figure 5The three typical time periods shown are as follows: In time period A (t0-t1), the satellite elevation angle is low, the absolute value of the Doppler shift is large and changes rapidly, and a slight shadowing risk is predicted. Based on the above predictions, the intelligent scheduler instructs the terminal to switch to a single-carrier frequency domain equalized waveform, using a larger subcarrier spacing (e.g., 60kHz) to enhance robustness against Doppler spread; a standard cyclic prefix is ​​used; and the insertion density of the phase tracking reference signal is increased to enhance the tracking capability against potential phase noise. In time period B (t1-t2), the satellite is near overhead, with a high elevation angle. The Doppler shift first decreases and then increases, but the overall absolute value is small, and the predicted channel conditions are relatively good. The intelligent scheduler instructs the terminal to switch to an orthogonal frequency division multiplexing waveform, using a smaller subcarrier spacing (e.g., 15kHz) to improve spectrum utilization efficiency; a standard cyclic prefix is ​​used, and a higher modulation and coding level and spatial multiplexing layer are configured to improve the uplink transmission rate. In time period C (t2-t3), the satellite gradually leaves the coverage area, and the terminal is predicted to enter a densely built-up area, increasing the risk of shadowing fading. Before the risk of shadowing increases, the intelligent scheduler instructs the terminal to switch back to a single-carrier frequency domain equalization waveform, using a smaller subcarrier spacing (such as 15kHz) to obtain a more concentrated power spectral density and improve penetration capability. It also uses an extended cyclic prefix to deal with possible multipath effects.

[0078] To enhance the feasibility of this method in real-world non-terrestrial network scenarios, the contextual description used to describe channel state changes can be associated with specific physical quantities or communication parameters to form quantifiable criteria for waveform switching and parameter adjustment.

[0079] In low-Earth orbit (LEO) satellite communication scenarios, when the predicted satellite elevation angle is lower than a preset threshold (e.g., less than 30°), or less than 10° near the horizon, the terminal is considered to be in a low-elevation coverage area. This is usually accompanied by a large Doppler frequency shift and its rate of change. Accordingly, the absolute value of the predicted Doppler frequency shift and the rate of change can be combined for judgment. For example, when the absolute value of the predicted Doppler frequency shift exceeds a preset threshold, or its rate of change exceeds a preset threshold, the channel can be considered to be in a rapidly changing state. When the predicted satellite elevation angle is high, e.g., greater than 60°, and the predicted signal-to-noise ratio (SNR) is higher than a preset threshold, the channel conditions can be considered relatively good, and a transmission waveform and parameter configuration with high spectral efficiency are suitable. For scenarios with the risk of occlusion, a quantitative judgment can be made based on the predicted shadow fading probability. For example, when the predicted shadow probability exceeds a preset threshold (e.g., 0.7), or when the predicted SNR shows a significant downward trend, the terminal can be considered to be about to enter or be in a shadowed area.

[0080] The threshold parameters in the aforementioned quantification indicators can be configured based on the carrier frequency band, satellite orbit parameters, and system link budget, and can be calibrated and optimized through simulation analysis or field testing before system deployment. These quantification indicators can be used individually or in combination to support waveform switching decisions and parameter adjustments under different channel conditions.

[0081] In practical applications, waveform switching and parameter configuration are not based on a simple threshold judgment of a single prediction parameter, but rather on a decision-making process based on multiple parameters. The network-side intelligent scheduler uses the multi-dimensional prediction results output by the channel dynamic predictor as decision input, and comprehensively determines the target transmission configuration within each decision cycle (e.g., each time slot).

[0082] In one exemplary embodiment, the input to the intelligent scheduler can be represented as a comprehensive state vector consisting of multiple predicted channel state parameters, for example: State(t) = [θ(t), fd(t), dfd / dt(t), Pshadow(t), τ(t), Predicted SNR(t)] Where θ(t) represents the predicted satellite elevation angle, fd(t) represents the predicted Doppler frequency shift, dfd / dt(t) represents the Doppler rate of change, Pshadow(t) represents the predicted shadow fading probability, τ(t) represents the predicted link propagation delay, and Predicted SNR(t) represents the predicted channel quality index.

[0083] Based on the comprehensive state vector, the intelligent scheduler can use a predefined rule set, lookup table method, or scoring function to jointly evaluate multiple candidate waveforms and their parameter configurations. For example, in an exemplary rule set, the following decision logic can be defined: When the predicted shadow fading probability Pshadow(t) is greater than 0.7, or the predicted absolute value of Doppler frequency shift |fd(t)| is greater than 30kHz, a single-carrier frequency domain equalization waveform is selected and a larger subcarrier spacing (e.g., 60kHz) is configured to prioritize link robustness. When the predicted shadow fading probability Pshadow(t) is less than 0.2, the predicted absolute value of Doppler frequency shift |fd(t)| is less than 5kHz, and the predicted signal-to-noise ratio PredictedSNR(t) is greater than 20dB, an orthogonal frequency division multiplexing waveform is selected, and a smaller subcarrier spacing (e.g., 15kHz) is configured. At the same time, the modulation coding level or the number of spatial multiplexing layers is increased to pursue higher system capacity. When the predicted satellite elevation angle θ(t) is in the range of 10° to 30° and the predicted absolute value of the Doppler frequency shift is in the range of 5kHz to 20kHz, a single-carrier frequency domain equalization waveform is selected and a medium-sized subcarrier spacing (e.g., 30kHz) is configured to achieve a trade-off between robustness and spectral efficiency.

[0084] In an alternative implementation, the joint decision-making process described above can also be implemented using a machine learning model, for example, by using a lightweight neural network to directly map the integrated state vector to a target transmission configuration index.

[0085] To avoid frequent waveform and parameter switching caused by slight fluctuations in the predicted channel state near the handover threshold, a hysteresis mechanism is introduced in one embodiment to improve the stability of the handover process. This hysteresis mechanism forms a handover hysteresis zone by setting different thresholds for handover triggering and handover recovery. For example, the following hysteresis rule can be set for Doppler frequency shift: When the predicted absolute value of the Doppler frequency shift, |fd(t)|, rises from a lower value and exceeds a first threshold (e.g., 25kHz), a switch from the orthogonal frequency division multiplexing waveform to a single-carrier frequency domain equalized waveform is triggered. Switching back from a single-carrier frequency domain equalized waveform to an orthogonal frequency division multiplexed waveform is only permitted when the predicted absolute value of the Doppler frequency shift, |fd(t)|, decreases from a higher value and falls below the second threshold (e.g., 15 kHz). When the predicted absolute value of the Doppler frequency shift is between the two thresholds mentioned above, the current transmission configuration remains unchanged.

[0086] In practical systems, hysteresis mechanisms can also be applied to the joint decision results, rather than a single prediction parameter. For example, a handover can be triggered only when the comprehensive decision score corresponding to a candidate transmission waveform exceeds a first threshold, and a handover back to the original configuration can be allowed only when the score is below a second threshold, thus avoiding back-and-forth switching and reducing signaling overhead.

[0087] With reference to the figure, taking a typical LEO satellite NTN uplink transmission scenario as an example, the exemplary implementation process of the present invention is explained.

[0088] After the terminal accesses the NTN cell, it reads common auxiliary information such as satellite ephemeris and common timing advance from the system information block (e.g., SIB1 or other SIBs). Subsequently, the network-side base station configures multiple uplink bandwidth portions (BWPs) for the terminal through Radio Resource Control Reconfiguration messages (RRCReconfiguration). For example, it can be configured as follows: BWP-ID 1: OFDM waveform, 15 kHz subcarrier spacing; BWP-ID 2: SC-FDE waveform, 15 kHz subcarrier spacing; BWP-ID 3: SC-FDE waveform, 60 kHz subcarrier spacing.

[0089] The RRC reconfiguration message also includes a new information element, such as WaveformAdaptationConfig, which instructs the terminal to enable the channel prediction-based dynamic waveform adaptation function and identifies the available waveforms and parameter sets.

[0090] During operation, the network-side channel dynamic predictor predicts that in a time slot (e.g., slot #n) approximately 10 frames later, the terminal's channel state will evolve from "Slot A" to "Slot B". Based on this prediction, at the corresponding scheduling time, the network-side base station sends a downlink grant downlink control message to the terminal, for example, using DCI format01. In this downlink control message, in addition to the regular scheduling fields, the target BWP is indicated by reusing or extending the frequency domain resource allocation field, or by adding a waveform indication field. For example, 2-bit encoding can be used: "00" indicates BWP-ID 1. "01" indicates BWP-ID 2. "10" indicates BWP-ID 3.

[0091] The scheduling offset parameter K2 indicated in the downlink control information is set so that the corresponding uplink PUSCH transmission occurs exactly at the predicted start time of slot #n, which is aligned with the timing of the predicted channel state change point.

[0092] After receiving the downlink control information, the terminal parses the waveform indication field to identify the target uplink bandwidth configuration to which it needs to switch, such as BWP-ID1 (OFDM, 15 kHz). The terminal then automatically activates the BWP configuration in the corresponding slot #n according to the scheduling offset parameter K2 indicated in the downlink control information. This includes adopting the corresponding subcarrier spacing, cyclic prefix length, and corresponding reference signal configuration, and uses this configuration to perform uplink PUSCH transmission.

[0093] For channel scenarios with strong predictable changes caused by satellite orbital motion, the network side can also employ a MAC control unit-based approach for periodic waveform switching. The network-side base station sends a waveform switching pattern table to the terminal via a Waveform SwitchingPattern MAC CE, for example: {Start SFN = 100, Start Slot = 0, Pattern = [BWP-ID 2 lasts for 50 ms, BWP-ID 1 lasts for 200 ms, ...]}.

[0094] After receiving and storing the switching mode table, the terminal automatically performs waveform switching at a predetermined time point based on the locally maintained system frame number and timeslot number, without having to trigger it through downlink control information each time, thus reducing control signaling overhead.

[0095] The predictive waveform adaptation scheme proposed in this invention can be naturally integrated with the NTN enhancement working direction of 3GPP R19 and later versions (e.g., R20). In one embodiment, new information elements, such as WaveformAdaptationConfig, can be defined in the Radio Resource Control protocol (38.331 series) to pre-configure waveform and parameter sets and enable predictive adaptation functionality. In the physical layer protocol (38.212 series), new or extended downlink control information formats are specified, clarifying the field positions and interpretation methods used for dynamic waveform indication, and defining timing relationships in conjunction with the large scheduling offset parameter K2 in the NTN scenario. In the Media Access Control protocol (38.321 series), a new MAC control unit can be defined to carry the waveform switching mode table, and its triggering, transmission, and execution procedures are specified.

[0096] The aforementioned enhancement mechanism can work in conjunction with NTN’s existing ephemeris information broadcasting, terminal location reporting, and enhanced timing advance reporting mechanism. It can be implemented through software upgrades and limited protocol enhancements without changing the basic structure of the existing physical layer, and has good backward compatibility and product feasibility.

[0097] This invention incorporates satellite ephemeris and terminal location information into the channel state prediction on the network side, enabling proactive joint scheduling of the transmission waveform and its key parameters. This transforms waveform switching in NTN scenarios from a feedback-dependent reactive mechanism to a prediction-based coordinated control, improving the matching relationship between switching timing and actual channel changes under long propagation delay conditions. Simultaneously, by adaptively selecting between waveforms such as OFDM and SC-FDE, and dynamically adjusting the corresponding subcarrier spacing, cyclic prefix, and reference signal configuration, the transmission configuration better aligns with multi-dimensional channel characteristics such as Doppler variations, delay fluctuations, and obstruction. Combined with a signaling approach that integrates higher-layer pre-configuration and lower-layer dynamic triggering, this invention reduces control overhead while ensuring scheduling flexibility, improves spectrum utilization efficiency under favorable channel conditions, maintains link stability under deteriorating channel conditions, and balances overall coordination between terminal transmission efficiency and network resource utilization.

[0098] This invention constructs a channel dynamic predictor on the network side based on satellite ephemeris, beam pointing information, and terminal location information. This predicts changes in key channel parameters within a future time window and determines the target transmission configuration of the terminal. This allows the adjustment of the transmission waveform and its key parameters to maintain a better time match with the time of channel changes. By pre-configuring the terminal and combining it with dynamic indication, the terminal communicates using the target transmission configuration at the effective time corresponding to the prediction result. Under the conditions of long latency in non-terrestrial networks and rapid channel changes, this invention improves the adaptability of waveform switching and scheduling control and reduces the performance loss caused by switching lag and mismatch.

[0099] Based on the same inventive concept, the method of the present invention can be executed by a non-terrestrial network communication system, the system including network-side nodes and terminals, wherein the network-side nodes can be spaceborne base stations, ground gateways that work in conjunction with spaceborne base stations, or processing units with base station functions.

[0100] The network-side node includes a prediction processing module, a scheduling decision module, and a signaling control module. The prediction processing module is used to construct a channel dynamic predictor based on real-time satellite ephemeris, beam pointing information, and terminal location information, and outputs the channel key parameter change trajectory of the service link within a future time window; the channel key parameter change trajectory represents the evolution of the channel state over time in the form of a time series.

[0101] The scheduling decision module is connected to the prediction processing module and is used to determine the target transmission configuration of the terminal in the future time window based on the change trajectory of the channel key parameters. The target transmission configuration includes at least a waveform switching strategy, a target transmission waveform, and key parameter configurations corresponding to the target transmission waveform.

[0102] The signaling control module is connected to the scheduling decision module and is used to pre-configure and / or dynamically instruct the terminal on the transmission configuration according to the target transmission configuration, so that the terminal uses the target transmission configuration for communication at the effective time corresponding to the prediction result.

[0103] In one embodiment, the signaling control module sends an available set of waveforms and parameters to the terminal via radio resource control signaling, and instructs the terminal to switch to the target transmission configuration in the set via downlink control information or media access control unit.

[0104] The terminal includes a signaling receiving module and a waveform configuration module, which are used to receive pre-configured signaling and dynamic indication information from network-side nodes, and generate corresponding transmission waveforms and their parameter configurations according to the target transmission configuration at the corresponding effective time to complete uplink communication.

[0105] It should be noted that the above module division is only used to illustrate the functional structure of the system and does not limit the specific implementation method. The modules can be implemented by software, hardware or a combination of software and hardware, and can also be merged or split according to actual deployment needs.

[0106] Based on the same inventive concept, the present invention also provides a computer device, comprising: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and executed the steps of the above-described adaptive waveform dynamic switching method based on channel state prediction.

[0107] The processing methods for computer devices can be referred to the description of the methods above, and will not be repeated here.

[0108] This application also provides a non-transitory machine-readable storage medium storing an executable program, which, when run by a microprocessor, causes the processor to execute the method provided in the above embodiments.

[0109] This invention discloses a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to perform the described methods.

[0110] This invention discloses a computer program product including a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform the described method.

[0111] The embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0112] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0113] Finally, it should be noted that the embodiments disclosed in this invention are merely preferred embodiments of this invention and are only used to illustrate the technical solutions of this invention, not to limit it. Although this invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this invention.

Claims

1. An adaptive waveform dynamic switching method based on channel state prediction, applied to network-side nodes, characterized in that, The method includes: A channel dynamic predictor is constructed on the network-side node. The channel dynamic predictor predicts the trajectory of changes in key channel parameters of the service link within a future time window based on real-time satellite ephemeris, beam pointing information, and location information. Based on the channel key parameter change trajectory, the target transmission configuration of the terminal within the future time window is determined. The target transmission configuration includes at least: waveform switching strategy, target transmission waveform, and key parameter configuration corresponding to the target transmission waveform. Based on the target transmission configuration, the terminal is pre-configured and / or a dynamic indication is sent, so that the terminal uses the target transmission configuration to communicate at the effective time corresponding to the prediction result within the future time window.

2. The adaptive waveform dynamic switching method based on channel state prediction according to claim 1, characterized in that, The channel key parameter change trajectory is a predicted channel state parameter sequence ordered by time. The predicted channel state parameter sequence includes at least multiple prediction time points and corresponding channel state parameter sets. The channel state parameter sets include one or more of the following: Doppler frequency shift, predicted signal-to-noise ratio / channel quality indication, link propagation delay change, and shadow fading state risk.

3. The adaptive waveform dynamic switching method based on channel state prediction according to claim 2, characterized in that, Network-side nodes obtain shadow fading status markers through the following steps: Obtain terminal location information and corresponding satellite location information; The line-of-sight propagation path between the terminal and the satellite is determined based on the terminal location information and the satellite location information. By combining the digital elevation model and / or building model, it is determined whether the line-of-sight propagation path is blocked by terrain or buildings; The shadow decay state risk is generated based on the occlusion judgment result.

4. The adaptive waveform dynamic switching method based on channel state prediction according to claim 2, characterized in that, Based on the change trajectory of the key channel parameters, the target transmission configuration of the terminal within the future time window is determined, including: Based on the Doppler frequency shift, Doppler change rate, predicted channel quality indication, and shadow fading risk corresponding to different prediction time points, the channel state change characteristics within the future time window are analyzed, and the target transmission waveform is determined by selecting from multiple candidate transmission waveforms. Based on the predicted channel state parameter sequence, adjust the configuration of key parameters corresponding to the target transmission waveform.

5. The adaptive waveform dynamic switching method based on channel state prediction according to claim 4, characterized in that, The target transmission waveform includes an orthogonal frequency division multiplexing waveform and a single-carrier frequency domain equalization waveform; Waveform switching strategies include: When the Doppler frequency shift and / or Doppler rate of change in the predicted channel state parameter sequence meet the preset conditions, or when the predicted channel state parameter sequence indicates that the terminal will enter or be in a shadow fading state, the target transmission waveform is determined to be a single-carrier frequency domain equalized waveform. When the Doppler frequency shift and / or Doppler rate of change do not meet the preset conditions, and the predicted channel state parameter sequence indicates that the channel state is relatively stable, the target transmission waveform is determined to be an orthogonal frequency division multiplexing waveform.

6. The adaptive waveform dynamic switching method based on channel state prediction according to claim 4, characterized in that, The key parameter configuration includes at least the subcarrier spacing and the cyclic prefix length; The network-side node adjusts the subcarrier spacing based on the Doppler frequency shift and / or Doppler rate of change in the predicted channel state parameter sequence; and adjusts the cyclic prefix length based on the propagation delay change in the predicted channel state parameter sequence.

7. The adaptive waveform dynamic switching method based on channel state prediction according to claim 4 or 6, characterized in that, Based on the target transmission configuration, pre-configure and / or send dynamic indications to the terminal, including: The available waveform and parameter set is configured to the terminal through the radio resource control reconfiguration message and / or system information block; the available waveform and parameter set includes multiple uplink bandwidth part configurations, each uplink bandwidth part configuration is associated with a transmission waveform and the corresponding key parameter configuration; In the radio resource control reconfiguration message and / or system information block, the terminal is enabled to perform adaptive waveform dynamic switching.

8. The adaptive waveform dynamic switching method based on channel state prediction according to claim 7, characterized in that, Based on the target transmission configuration, pre-configuring and / or sending dynamic indications to the terminal further includes: Based on the predicted channel state parameter sequence, the target effective time of the terminal within a future time window is determined. Based on the target effective time, scheduling offset parameters, downlink propagation delay, and terminal processing delay, the timing for sending downlink control information is determined. At the specified transmission timing, downlink control information containing waveform indication information and the scheduling offset parameter is sent to the terminal. The downlink control information is used to instruct the terminal to switch to the target transmission configuration from the available waveform and parameter set, so that the terminal starts to use the target transmission configuration for communication in the corresponding time slot according to the scheduling offset parameter at the time when the target takes effect.

9. The adaptive waveform dynamic switching method based on channel state prediction according to claim 7, characterized in that, Also includes: When the predicted channel state parameter sequence exhibits periodic or regular changes, a waveform switching time series table is generated based on the change trajectory of the key channel parameters. The waveform switching time sequence table is sent to the terminal so that the terminal can adaptively switch waveforms at a future preset time reference point; the waveform switching time sequence table includes the future preset time reference point and time offset, waveform configuration index and / or effective duration.

10. The adaptive waveform dynamic switching method based on channel state prediction according to claim 7, characterized in that, Also includes: The accuracy of channel prediction is verified based on the deviation between the predicted channel state parameter sequence and the actual measured channel state parameters at the terminal. When the deviation exceeds a preset threshold, compensation is provided based on the scheduling parameters determined by the prediction results or the channel dynamic predictor is updated. When the prediction error continues to exceed the allowable range, the prediction-based dynamic waveform switching is paused, and the terminal is instructed to fall back to the resource scheduling mode based on terminal feedback or the preset high robustness transmission configuration.

Images