A data reporting method based on narrowband internet of things retransmission mechanism optimization

Abstract

This invention relates to the field of narrowband Internet of Things (NIoT) and discloses a data reporting method optimized based on the NIoT retransmission mechanism. The method includes introducing a continuous timing characterization mechanism on the NIoT terminal side to splice and reassemble multiple transmission actions; extracting temporal disturbance information related to cell coverage fluctuations based on the transmission discontinuity and failure clustering characteristics presented in the transmission occupancy evolution structure; cross-mapping the failure backtracking path in the transmission occupancy evolution structure with the current uplink resource availability to derive the reachability probability distribution of each retransmitted data in the time slot; introducing wake-up cycle information in the terminal's energy-saving mode around the expected transmission time of each data unit in the retransmission scheduling set to correct the time offset between the retransmission trigger point and the terminal's active window; and mapping the successful convergence segment and the continuously failing diffusion segment back to the corresponding trigger time and alignment offset. This invention has the advantage of improving the data reporting success rate.

Description

Technical Field

[0001] This invention relates to the field of narrowband Internet of Things (IoT), specifically to a data reporting method optimized based on the narrowband IoT retransmission mechanism. Background Technology

[0002] In narrowband Internet of Things (NB-IoT) communication systems, terminal devices typically rely on base station scheduling to report small amounts of data at low rates, triggering retransmission mechanisms to ensure data reliability when link quality is unstable or transmission fails. However, current NB-IoT retransmission strategies often employ fixed-number triggering or simple acknowledgment feedback, lacking fine-grained characterization of the timeliness of actual business data and the dynamic characteristics of the link. In specific scenarios, such as underground utility tunnel monitoring, meter readings in densely built-up urban areas, or areas with weak signal coverage from power equipment, terminal devices often experience intermittent deep fading or uplink interference fluctuations during reporting. In such cases, even if a single transmission fails, the probability of subsequent successful retransmissions is unstable. With multiple terminals accessing concurrently, retransmitted data competes with newly transmitted data in the random access channel, easily leading to time slot congestion and repeated backoff, causing some low-priority but time-sensitive data to be delayed for extended periods. Especially in terminals operating in energy-saving mode, if the retransmission trigger timing is misaligned with the terminal wake-up window, multiple rounds of retransmission may not be executed in a timely manner, resulting in a "false queuing" phenomenon, where data is marked as pending retransmission in the protocol stack but does not actually receive a valid transmission opportunity. Therefore, it is essential to design a data reporting method based on the narrowband IoT retransmission mechanism optimization to improve the data reporting success rate. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a data reporting method optimized based on the narrowband Internet of Things retransmission mechanism, which has the advantage of improving the data reporting success rate and solves the problems mentioned in the background technology.

[0004] To achieve the aforementioned goal of improving data reporting success rate, this invention provides the following technical solution: a data reporting method optimized based on narrowband IoT retransmission mechanism, comprising the following steps: A continuous timing characterization mechanism is introduced on the narrowband IoT terminal side. Fine-grained time marking is performed on each stage from the data to be reported from queuing to the completion of transmission. Combined with the repeated transmission identifier and the confirmation feedback result, multiple transmission behaviors are spliced ​​and recombined to form a transmission occupancy evolution structure. Based on the characteristics of transmission discontinuity and failure clustering presented in the transmission occupancy evolution structure, the temporal disturbance information of cell coverage fluctuation is extracted, the original fixed retransmission interval is deconstructed and rearranged, and a non-uniform retransmission triggering scheme is generated. Under the constraint of the non-uniform retransmission triggering scheme, the failure backtracking path in the transmission occupancy evolution structure is cross-mapped with the current uplink resource availability, the reachability probability distribution of each data to be retransmitted in the time slot is derived, and a retransmission scheduling set with hierarchical constraints is constructed. Around the expected transmission time of each data unit in the retransmission scheduling set, the wake-up cycle information in the terminal power-saving mode is introduced to correct the time offset between the retransmission trigger point and the terminal active window, and the correction result is written back to the retransmission scheduling set to form a candidate transmission arrangement constrained by time alignment. Based on the candidate transmission arrangement, the successfully transmitted convergence segments and the continuously failing diffusion segments are mapped back to their corresponding trigger times and alignment offsets to construct a link response metric.

[0005] Preferably, the process of performing fine-grained time stamping on each stage of the data to be reported, from queuing to completion of transmission, is as follows: A time acquisition unit is set up inside the protocol stack of the narrowband IoT terminal to work in coordination with the access layer scheduling, and to calibrate the data unit entering the buffer queue, obtaining uplink authorization, and the end of radio frequency transmission one by one; The time intervals between each node are sampled in segments, and multi-dimensional time slice features are formed by combining the transmission power adjustment and the number of repeated transmissions. To address processing delay anomalies during continuous transmission, a time offset detection mechanism is established to mark and classify abnormal segments.

[0006] Preferably, the process of forming the transmission occupancy evolution structure is as follows: Extract the repeated transmission number and corresponding confirmation feedback status of each data unit during the uplink process of narrowband IoT, and associate the successful confirmation and failed retransmission records one by one; Using the time slice features corresponding to a single round of transmission as the basic unit, multiple rounds of transmission are spliced ​​on the time axis, and consecutive failure segments are aggregated. The correlation between sending interval and feedback delay is introduced during the splicing process to provide a structured representation of resource usage at different stages; Based on changes in occupancy intensity and the degree of failure concentration, a transmission occupancy evolution structure reflecting the evolution characteristics of the transmission process is constructed.

[0007] Preferably, the process of extracting time-domain disturbance information related to cell coverage fluctuations is as follows: A continuous analysis of the time distribution in the transmission occupancy evolution structure is performed to identify the transmission interruption location and its adjacent intervals; Within the interruption interval, the correlation between the number of repeated failures and signal quality indicators is statistically analyzed to extract link fluctuation characteristics; By constructing a failure density distribution function within a time window, the degree of transmission instability in different time segments is quantified; By correlating and fitting the failure density distribution with coverage intensity changes, temporal perturbation information representing the coverage fluctuations of narrowband IoT cells is generated.

[0008] Preferably, the process of generating a non-uniform retransmission triggering scheme is as follows: The fixed retransmission interval is divided into multiple candidate time intervals based on the time-domain perturbation information, and the stability of each interval is evaluated. For high failure density intervals, retransmission triggering is delayed; for low interference intervals, triggering is shifted forward. Based on the utilization of random access resources, conflict detection and avoidance are performed on the adjusted trigger time; Through multiple rounds of screening and correction, a non-uniform retransmission triggering scheme adapted to the link fluctuation characteristics was formed.

[0009] Preferably, the process of cross-mapping the failure backtracking path in the transmission occupancy evolution structure with the current uplink resource availability is as follows: Extract the time position and resource occupancy status of each failed segment in the transmission occupancy evolution structure; Obtain the distribution of uplink resource blocks within the current scheduling period and establish a mapping relationship between time and resources; Project the failed backtracking path to the available resource area to identify executable retransmission locations; For paths that cannot be directly matched, perform delay extension and path reconstruction to form an updated executable retransmission path.

[0010] Preferably, the process of constructing a retransmission schedule set with hierarchical constraints is as follows: Based on the executable retransmission path, the historical success rate of each data unit in different candidate time slots is statistically analyzed. By combining link quality indicators and resource contention intensity, the success probability of each time slot is estimated and calculated; Based on the success probability and the timeliness requirements of data services, the data to be retransmitted is prioritized. Data is grouped and scheduled according to priority results to construct a retransmission scheduling set with hierarchical constraints.

[0011] Preferably, the process of correcting the time offset between the retransmission trigger point and the terminal active window is as follows: Obtain the periodic wake-up parameters and active time window of the narrowband IoT terminal in power-saving mode; Map the expected transmission time of each data unit to the corresponding wake-up period and calculate the time deviation from the active window; Perform forward or backward adjustments for trigger times that exceed the active window range.

[0012] Preferably, the process of forming a candidate transmission arrangement constrained by timing alignment is as follows: Update the transmission time after time correction is completed to the scheduling record of the corresponding data unit; Perform consistency checks on the updated scheduling records to eliminate potential resource conflicts; The scheduling order is reorganized based on the corrected time distribution, and a candidate transmission arrangement with time alignment constraints is formed through the reconstruction of the scheduling structure.

[0013] Preferably, the process of constructing link response metrics is as follows: Continuously track the actual transmission results in the candidate transmission schedule and distinguish between successfully completed segments and repeatedly failed segments; By inversely correlating the trigger times and alignment offsets corresponding to different segments, and by statistically analyzing the concentration of successful segments and the expansion range of failed segments, link response characteristic parameters are extracted. A link response metric is constructed based on link response characteristic parameters to represent the stability of narrowband IoT links and the effectiveness of retransmission mechanisms.

[0014] Compared with existing technologies, this invention provides a data reporting method optimized based on the narrowband IoT retransmission mechanism, which has the following beneficial effects: This invention, by providing fine-grained time-series characterization of the entire data reporting process, reconstructs the originally discrete multi-round transmission behavior into a transmission occupancy evolution structure with continuous evolution characteristics, thus making the link status traceable and analyzable. By combining transmission intermittency and failure clustering characteristics to extract temporal disturbance information corresponding to cell coverage fluctuations, the retransmission triggering mode is transformed from a fixed-period mode to a non-uniform mechanism dynamically adjusted according to the link environment, effectively avoiding high-failure sections. By cross-mapping the failure backtracking path with uplink resource availability, the retransmission location and resource distribution are matched collaboratively, reducing resource conflicts and invalid occupancy. Simultaneously, the wake-up cycle constraint in the terminal energy-saving mode is introduced to align and correct the retransmission triggering time, avoiding transmission mismatch problems caused by terminal sleep. By constructing a link response metric, the execution effect of the retransmission strategy is quantitatively fed back and continuously corrected. In complex narrowband IoT scenarios, this invention significantly improves the data reporting success rate, reduces resource waste and latency accumulation caused by repeated retransmissions, and enhances the system's adaptability to link fluctuations and resource competition. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1: Please refer to Figure 1 An embodiment of the present invention provides a data reporting method optimized based on a narrowband IoT retransmission mechanism, comprising the following steps: S1: Introduce a continuous timing characterization mechanism on the narrowband IoT terminal side, perform fine-grained time marking on each stage from the data to be reported from enqueuing to the completion of transmission, and combine the repeated transmission identifier and the confirmation feedback result to splice and reassemble multiple transmission behaviors to form a transmission occupancy evolution structure.

[0018] The process of performing fine-grained time stamping on each stage of the data to be reported, from enqueuing to completion of transmission, in S1 is as follows: A time acquisition unit is set up inside the protocol stack of the narrowband IoT terminal to work in coordination with the access layer scheduling, and to calibrate the data unit entering the buffer queue, obtaining uplink authorization, and the end of radio frequency transmission one by one; The time intervals between each node are sampled in segments, and multi-dimensional time slice features are formed by combining the transmission power adjustment and the number of repeated transmissions. To address processing delay anomalies during continuous transmission, a time offset detection mechanism is established to mark and classify abnormal segments.

[0019] The process of forming the transmission occupancy evolution structure in S1 is as follows: Extract the repeated transmission number and corresponding acknowledgment feedback status of each data unit during the uplink process of narrowband IoT, and sequentially associate successful acknowledgments with failed retransmission records; read the retransmission counter and HARQ process identifier of each data unit on the narrowband IoT terminal side, and synchronously parse the ACK / NACK feedback bits returned by the base station to match the identifiers of the same data unit in different transmission rounds; construct a mapping table indexed by "data unit identifier + transmission round" to concatenate the transmission results of each round in chronological order, and set feedback consistency verification rules. When the interval between two consecutive feedbacks exceeds the preset scheduling period or feedback is missing, the interval is marked as a link abnormal segment, thus forming a complete transmission-feedback association chain. By explicitly establishing the sequential correspondence between transmission and feedback, the retransmission behavior becomes traceable, avoiding the state fragmentation problem under traditional statistical methods. Using the time slice features corresponding to a single round of transmission as the basic unit, multiple rounds of transmission are spliced ​​on the time axis, and consecutive failure segments are aggregated. Each transmission behavior is divided into a scheduling waiting stage, a resource occupation stage, and a feedback return stage, and the duration of each stage is recorded to form a standardized time slice description. Each transmission round is sorted using a unified time base, and adjacent time slices are spliced ​​using a time alignment algorithm. Failure judgment logic is set during the splicing process. When the feedback status of multiple consecutive time slices is NACK, it is judged as a consecutive failure segment. The failure intensity index is calculated by accumulating the number of failures and the duration. At the same time, interval compression processing is performed on the failure segment to reduce redundancy, transforming discrete transmission events into continuous time expressions, highlighting the failure concentration area, and facilitating the identification of unstable link intervals. During the splicing process, a correlation between transmission interval and feedback delay is introduced to structurally represent resource occupancy at different stages. The time interval between adjacent transmission actions is calculated, and the return delay of the corresponding feedback is obtained synchronously. The correlation coefficient between the two is calculated using a sliding window method. When the correlation coefficient is higher than a set threshold, a strong coupling relationship is determined. Furthermore, combined with uplink resource allocation information (including the number of subcarriers occupied and the scheduling cycle position), the resource usage within this coupling interval is encoded to form a ternary feature vector containing time interval, feedback delay, and resource occupancy intensity. By performing unified encoding on all time intervals, a structured resource occupancy expression result is obtained, establishing a clear correlation between time behavior and resource usage, which helps to reveal the intrinsic link between retransmission failure and resource contention. Based on changes in occupancy intensity and the concentration of failures, a transmission occupancy evolution structure reflecting the evolution characteristics of the transmission process is constructed. Continuity analysis is performed on the ternary feature vector along the time axis, and abrupt change locations are identified by calculating the gradient of occupancy intensity changes. High-risk transmission intervals are then screened in conjunction with failure intensity indicators. The time interval is segmented and labeled according to occupancy intensity and failure density, dividing the link state into stable, transitional, and unstable regions. The transition paths and frequencies between different states are recorded, forming a hierarchical transmission occupancy evolution structure that intuitively reflects the evolution law of the link state during narrowband IoT data reporting.

[0020] S2: Based on the transmission intermittency and failure clustering characteristics presented in the transmission occupancy evolution structure, extract the temporal disturbance information related to cell coverage fluctuations, deconstruct and rearrange the original fixed retransmission interval, and generate a non-uniform retransmission triggering scheme.

[0021] The process of extracting time-domain disturbance information related to cell coverage fluctuations in S2 is as follows: A continuity analysis is performed on the temporal distribution in the transmission occupancy evolution structure to identify the transmission interruption location and its adjacent intervals. For the constructed transmission occupancy evolution structure, the time nodes of each transmission behavior are sorted according to a unified time base, and time breakpoints are detected by setting a continuity judgment threshold (such as the time interval between adjacent transmissions exceeding a multiple of a preset scheduling period). When a sudden increase in interval is detected, the location is marked as a transmission interruption point, and a certain time window is extended before and after to form an adjacent analysis interval. Within this interval, the transmission frequency change rate and occupancy intensity change trend are further extracted as interruption feature criteria, which can accurately locate the transmission discontinuity area in the narrowband IoT link, avoid misjudging occasional delays as link anomalies, and thus improve the accuracy of interruption identification. Within each interruption interval, the correlation between the number of repeated failures and signal quality indicators is statistically analyzed to extract link fluctuation characteristics. Within each interruption interval, the number of retransmissions of the corresponding data unit is statistically analyzed, and signal quality indicators (including RSRP, RSRQ, and signal-to-noise ratio) are simultaneously collected to construct a "failure count - signal quality" data pair. By performing segmented statistical analysis on this data pair, when a monotonically increasing trend in the number of failures is detected as a decrease in signal quality, it is determined to be coverage fading fluctuation; when the number of failures and signal quality show a weak correlation or irregular changes, it is determined to be interference-dominated fluctuation. Simultaneously, a threshold grading mechanism is introduced to classify and label different types of fluctuations. By establishing a direct correlation between failure behavior and physical layer indicators, the source of link fluctuations can be distinguished, improving the ability to identify multi-factor interference in complex narrowband IoT environments. By constructing a failure density distribution function within a time window, the transmission instability of different time segments is quantified. Using a sliding time window as the unit, the number of failure events within each window is statistically analyzed, and the failure density value is calculated in conjunction with the total number of transmissions within the window, forming a discrete time series. This discrete series is then smoothed using kernel density estimation or piecewise linear fitting methods to obtain a continuous failure density distribution function. Threshold division points are set on the function curve to divide the time segments into low-failure, medium-fluctuation, and high-failure regions, and the duration and rate of change of each region are recorded. This functional expression enables a quantitative description of the transmission instability, transforming discrete failure events into continuous analyzable signals and improving the ability to characterize the trend of link state changes. The failure density distribution is correlated with coverage intensity changes to generate temporal disturbance information representing coverage fluctuations in narrowband IoT cells. The failure density distribution function is time-aligned with synchronously collected coverage intensity time series (such as RSRP change curves), and a mapping relationship between the two is established using least squares fitting or correlation analysis. When the fitting residual is lower than a preset threshold, the failure density change is considered to be mainly caused by coverage fluctuations, and the corresponding time segment is extracted as an effective disturbance area. For segments with large residuals, supplementary corrections are made in conjunction with interference indicators. The start and end times, fluctuation amplitudes, and change frequencies of each effective disturbance area are encoded to form temporal disturbance information output. By fusing failure statistics and physical layer coverage information, an accurate characterization of coverage fluctuations in narrowband IoT cells can be achieved.

[0022] The process of generating a non-uniform retransmission triggering scheme in S2 is as follows: Based on time-domain perturbation information, the fixed retransmission interval is divided into multiple candidate time intervals, and the stability of each interval is evaluated. Using the standard retransmission cycle currently used by narrowband IoT terminals as a benchmark, it is segmented according to the fluctuation boundary points in the time-domain perturbation information to obtain several candidate time intervals. Within each interval, the corresponding average failure density, signal quality fluctuation amplitude, and interruption frequency are extracted, and a stability evaluation index function is constructed. When the index function value is below a preset stability threshold, it is marked as a stable interval; when it is above the threshold, it is marked as an unstable interval. Simultaneously, the interval boundaries are smoothed to avoid scheduling oscillations caused by overly fine segmentation. By introducing interval division and quantitative evaluation based on actual link status, the retransmission timing is transformed from a fixed mode to an environment-driven mode, improving the adaptability of retransmission triggering to link changes. For high failure density intervals, retransmission triggering is delayed; for low interference intervals, triggering is moved forward. For intervals marked as unstable, a delay weight is set according to the failure density, shifting the original retransmission trigger time backward by one or more time units, with the maximum delay limited to the service latency constraint. For stable intervals, a forward weight is set according to their stability level, appropriately moving the trigger time forward to an earlier available time point. Monotonicity constraints are introduced during the adjustment process to ensure that the retransmission time adjustment does not disrupt the original time sequence. Through differentiated processing of forward and delay strategies, retransmission behavior actively avoids high-risk areas and utilizes low-interference windows, reducing the probability of repeated failures and improving the success rate of single retransmissions. Based on the utilization of random access resources, conflict detection and avoidance are performed on the adjusted trigger times. The resource occupancy of the narrowband IoT random access channel in each time interval is obtained, including the access preamble usage rate and scheduling load level. The adjusted retransmission trigger times are mapped to the corresponding resource intervals, and the resource conflict probability at that time is calculated. When the conflict probability is higher than a preset threshold, the trigger time is marked as a conflict point. Local time offset search is performed for the conflict point, and time points with lower resource occupancy in the adjacent intervals before and after it are selected as alternative trigger times, while maintaining the interval constraint with the retransmission times before and after. By introducing a resource-dimensional conflict detection and dynamic avoidance mechanism, access failures caused by multi-terminal competition are effectively reduced, the overall access success rate is improved, and invalid retransmissions are reduced. Through multiple rounds of screening and correction, a non-uniform retransmission triggering scheme adapted to link fluctuation characteristics is formed. All candidate triggering times after stability adjustment and conflict avoidance are uniformly sorted and screened according to business priority and latency constraints, eliminating triggering points that do not meet the conditions. The triggering times are globally adjusted through iterative optimization to achieve a balance between success probability and latency, outputting a set of non-uniform retransmission triggering schemes that dynamically change over time. Through multiple rounds of screening and global correction, the retransmission strategy is converged and optimized, simultaneously taking into account transmission success rate and timeliness in complex narrowband IoT environments, thereby improving the overall performance of data reporting.

[0023] S3: Under the constraint of the non-uniform retransmission triggering scheme, the failure backtracking path in the transmission occupancy evolution structure is cross-mapped with the current uplink resource availability, the reachability probability distribution of each data to be retransmitted in the time slot is derived, and a retransmission scheduling set with hierarchical constraints is constructed.

[0024] The process of cross-mapping the failed backtracking paths in the transmission occupancy evolution structure with the current uplink resource availability in S3 is as follows: Extract the time position and resource occupancy status of each failed segment in the transmission occupancy evolution structure; parse out the segments marked as failed from the transmission occupancy evolution structure, read their start and end times, duration, and corresponding transmission round information, and simultaneously extract the uplink resource parameters already occupied within the segment, including the number of subcarriers, time slot position, and scheduling cycle number; establish a "time segment - resource occupancy" association table to encode the resource usage patterns of different failed segments, and detect resource occupancy repetition. When the same resource repeatedly appears in consecutive failed segments, it is marked as a high-conflict resource; by refining the correspondence between failed segments and resource occupancy, the cause of failure becomes locatable. The system acquires the distribution of uplink resource blocks within the current scheduling period and establishes a mapping relationship between time and resources. It obtains uplink resource allocation information for the current and subsequent scheduling periods from the narrowband IoT network side, including the number of available resource blocks and their occupancy status in each time slot. The time axis is divided into discrete scheduling units, and a resource availability identifier is established for each unit, forming a "time unit - resource status" mapping matrix. Resource idleness, conflict probability, and historical usage frequency are marked in this matrix, and priority resource intervals are determined through threshold filtering. By constructing a unified time-resource mapping relationship, a visual representation of uplink resources is achieved, providing a clear resource selection space for retransmission scheduling and improving matching efficiency. The failed backtracking paths are projected onto available resource areas to identify executable retransmission locations. The time position of each failed backtracking path is mapped to the time-resource mapping matrix to find available resource blocks within its corresponding time unit. When the resource corresponding to the original time position is in a high-conflict or unavailable state, a local search is performed within its neighboring time units to select time points with resource idleness higher than a set threshold as candidate retransmission locations. At the same time, a minimum time offset constraint is introduced to ensure that the adjustment range does not affect the service latency requirements. After performing unified projection processing on all paths, a preliminary set of executable retransmission locations is formed. By directly connecting failed paths with resource distribution, the selection of retransmission locations becomes targeted, reducing the probability of resource conflicts and increasing the likelihood of successful retransmission. For paths that cannot be directly matched, delay extension and path reconstruction are performed to form updated executable retransmission paths. For failed paths that cannot find resources that meet the conditions within the current scheduling cycle, time delay extension is performed to extend them to subsequent scheduling cycles, and resource availability is continuously monitored during the extension process. When a resource interval that meets the conditions is detected, the time node sequence of the path is reconstructed, and its corresponding resource allocation information is updated. For paths that still cannot be matched after multiple extensions, a path reconstruction mechanism is introduced to regenerate the path structure by adjusting the retransmission interval or changing the sending priority. Finally, all updated paths are aggregated to form a set of executable retransmission paths. By combining delay extension and structural reconstruction, the problem of retransmission unreachability under resource constraints is solved, the coverage and executability of retransmission paths are improved, and the success rate of narrowband IoT data reporting is further improved.

[0025] The process of constructing a hierarchical retransmission schedule set in S3 is as follows: Based on executable retransmission paths, the historical success rate of each data unit in different candidate time slots is statistically analyzed. For the set of executable retransmission paths, the historical transmission records of each data unit in the corresponding candidate time slot are extracted one by one, and the success and failure flags are read from the narrowband IoT terminal-side cache or network-side logs. A three-dimensional statistical table of "data unit - time slot - result" is established with time slot as the index. The number of successful transmissions and the total number of transmissions in each candidate time slot are accumulated and the corresponding success ratio is calculated. At the same time, a time decay factor is introduced to give lower weight to earlier historical data to highlight the impact of recent link status on the statistical results. By performing structured statistics on historical behavior, a quantitative assessment of the feasibility of different time slots is achieved. By combining link quality indicators and resource contention intensity, the success probability of each time slot is estimated and calculated. Based on the success ratio, real-time link quality indicators (including RSRP and signal-to-noise ratio) and resource contention intensity parameters (such as access load and collision probability) are introduced to construct a multi-parameter fusion success probability estimation model. Through weighted calculation or regression analysis, historical success ratios and real-time indicators are fused to obtain the comprehensive success probability of each candidate time slot. A probability correction mechanism is set to dynamically adjust the estimation results when link quality changes abruptly. By fusing historical statistics and real-time environmental information, the success probability has dynamic response capabilities, improving the adaptability of retransmission decisions to complex narrowband IoT environments. Based on the success probability and data service timeliness requirements, the data to be retransmitted is prioritized. The highest success probability value corresponding to each data unit is jointly evaluated with its business attributes (such as latency tolerance and importance level) to construct a priority determination function. When a data unit has a high success probability and high timeliness requirements, it is assigned a high priority label. For data with a low success probability but high timeliness requirements, a compensation weight is introduced to increase its priority. At the same time, a priority stratification threshold is set to divide the data into an urgent layer, a priority layer, and a normal layer. By introducing joint constraints of business attributes and success probability, differentiated processing of scheduling decisions is achieved, ensuring timely reporting of key data while taking into account the overall success rate. Data is grouped and scheduled according to priority, constructing a retransmission scheduling set with hierarchical constraints. Based on the priority division, each data unit is assigned to its corresponding scheduling level, and within each level, they are sorted from high to low success probability. Different resource access weights and scheduling frequencies are assigned to different levels, ensuring that high-priority levels have priority access to high-quality time slots. Simultaneously, a scheduling isolation mechanism is established between levels to prevent low-priority data from preempting high-priority resources. Through unified orchestration and constraints on data at each level, a retransmission scheduling set with a clear hierarchical structure is formed. This hierarchical scheduling strategy achieves orderly resource allocation, reduces the risk of conflicts caused by multiple data competition, and improves the overall success rate and scheduling efficiency of narrowband IoT data reporting.

[0026] S4: Based on the expected transmission time of each data unit in the retransmission scheduling set, the wake-up cycle information in the terminal power-saving mode is introduced to correct the time offset between the retransmission trigger point and the terminal active window, and the correction result is written back to the retransmission scheduling set to form a candidate transmission arrangement constrained by timing alignment.

[0027] The process of correcting the time offset between the retransmission trigger point and the terminal active window in S4 is as follows: The system acquires the periodic wake-up parameters and active time windows of narrowband IoT terminals in energy-saving mode; it reads energy-saving operation configuration parameters, including PSM cycle, eDRX cycle, and Paging time window, from the narrowband IoT terminal side, and parses the current wake-up cycle length and the start and end positions of the active window in conjunction with system information blocks sent from the network side; by establishing a unified time reference, the periodic wake-up time axis is discretized into multiple wake-up cycle units, and the active time interval for executable transmission is marked in each cycle; at the same time, a parameter validity detection mechanism is introduced, and resampling correction is performed when abnormal cycle parameters or update lag are detected; by accurately acquiring the active time characteristics of the terminal, invalid transmission requests are avoided during the terminal's sleep period; The expected transmission time of each data unit is mapped to the corresponding wake-up cycle, and the time deviation from the active window is calculated. The expected transmission time of each data unit in the retransmission scheduling set is mapped to the corresponding wake-up cycle unit according to the time axis, and the time difference between this time and the start and end points of the active window is calculated. When the expected transmission time is within the active window, the deviation is marked as a valid deviation. When it is outside the window, its deviation direction (early or late) and offset are recorded. At the same time, a deviation threshold judgment rule is introduced to mark data units whose offset exceeds the set upper limit for subsequent key adjustments. By quantifying the relationship between the transmission time and the terminal's active state, the time mismatch problem can be expressed in a calculable way, improving the ability to identify invalid scheduling behavior. For trigger times outside the active window range, forward or backward adjustments are performed. For transmission times marked outside the window, an adjustment strategy is selected based on the direction of deviation: when the transmission time is earlier than the active window, it is moved forward to the nearest window start point; when the transmission time is later than the active window, it is delayed to the window start point of the next wake-up cycle. During the adjustment process, maximum offset limits and minimum interval constraints are set to ensure that the adjusted time satisfies the terminal's active conditions without disrupting the original retransmission rhythm. At the same time, conflict detection is performed on the adjusted time to avoid multiple data units concentrating in the same time slot. By directionally correcting the time offset, the retransmission trigger is aligned with the terminal's active window, reducing transmission failures caused by sleep mode and improving the effective execution rate of narrowband IoT data reporting.

[0028] The process of forming a candidate transmission arrangement with time alignment constraints in S4 is as follows: The transmission time after time correction is completed is updated in the scheduling record of the corresponding data unit; the transmission time after the wake-up window alignment is completed is written into the retransmission scheduling table on the narrowband IoT terminal side, and the corresponding time field is updated with the data unit identifier as the index, and the associated scheduling cycle number and resource occupancy flag are updated synchronously; a write verification mechanism is set during the update process to compare the time offset before and after the update. When the offset exceeds the preset threshold, it is recorded as a high change item for subsequent processing; at the same time, the update order of the scheduling record is ensured by version identifier or timestamp mechanism; by uniformly updating the scheduling information, the time correction result can directly participate in subsequent scheduling. The updated scheduling records undergo consistency checks to eliminate potential resource conflicts. Based on the updated transmission time, the resource occupancy of all data units within the same scheduling cycle is compared one by one to detect whether multiple data units are mapped to the same uplink resource block or the same time slot. When a conflict is detected, the resource occupancy of high-priority data units is preserved according to preset priority rules, while time fine-tuning or resource reallocation is performed on low-priority data. A conflict probability assessment mechanism is also introduced to pre-mark and pre-process potential conflict points. Continuous detection and correction are performed until all conflicts are eliminated or reduced to an acceptable range. Through strict consistency checks and conflict resolution mechanisms, the executability of the scheduling scheme is guaranteed, reducing the likelihood of further failures due to resource conflicts. The scheduling order is reorganized based on the corrected time distribution. Through restructuring the scheduling structure, a candidate transmission arrangement with time alignment constraints is formed. Scheduling records that have completed consistency checks are reordered according to their chronological order, and the scheduling order within the same time interval is optimized by combining the priority and success probability of data units. Time interval constraints are introduced during the restructuring process to ensure that adjacent transmissions meet the minimum interval requirement while avoiding concentrated transmissions within a short period. Furthermore, a hierarchical scheduling strategy is used to distribute data of different priorities to different time segments to achieve load balancing. A set of candidate transmission arrangements that meet the requirements of time alignment, resource conflict-free operation, and optimized order is output. Through the overall restructuring of the scheduling structure, coordinated optimization of time and resources is achieved.

[0029] S5: Based on the candidate transmission arrangement, map the successfully transmitted convergence segments and the continuously failing diffusion segments back to their corresponding trigger times and alignment offsets to construct a link response metric.

[0030] The process of constructing link response metrics in S5 is as follows: The system continuously tracks the actual transmission results in the candidate transmission schedule, distinguishing between successfully completed segments and repeatedly failed segments. A transmission result monitoring module is established on the narrowband IoT terminal side to record the actual execution status of each data unit in the candidate transmission schedule, collecting its transmission time, corresponding feedback result (ACK / NACK), and retransmission count. A sliding time window is used to scan the continuous transmission results; when multiple consecutive data units receive ACK feedback, the interval is marked as a successfully completed segment; when consecutive NACKs occur or the maximum retransmission count is reached without success, the corresponding interval is marked as a repeatedly failed segment. A minimum segment length threshold is set to filter out occasional errors. Through continuous tracking of transmission results and segment division, a structured expression of the link's operating status is achieved, intuitively identifying stable and unstable areas of the link. By inversely correlating the trigger times and alignment offsets corresponding to different segments, and extracting link response characteristic parameters by statistically analyzing the concentration of successful segments and the expansion range of failed segments, the successful and failed segments obtained above are traced back to their corresponding original retransmission trigger times and time alignment offsets to establish a ternary correlation relationship of "segment-trigger time-offset". The concentration index (such as variance or clustering) of the time distribution of successful segments is calculated, and the time expansion range and duration of failed segments are calculated. At the same time, the success rate changes within different offset intervals are statistically analyzed to extract the influence pattern of alignment offset on the transmission results. The statistical results are summarized to form a set of link response characteristic parameters. The inverse correlation mechanism reveals the intrinsic relationship between triggering strategies and results, and quantifies the degree of influence of different scheduling strategies on link performance. A link response metric is constructed based on link response characteristic parameters to represent the stability of narrowband IoT links and the effectiveness of retransmission mechanisms. The link response characteristic parameters are introduced into a unified metric calculation framework, and success concentration, failure spread, and offset sensitivity are weighted and fused to construct a comprehensive link response index. Normalization and threshold grading mechanisms are set in the model to classify link states into high stability, medium volatility, and low stability levels. Simultaneously, the weight coefficients are dynamically adjusted based on the metric results to adapt to different narrowband IoT application scenarios. The output link response metric value serves as the evaluation basis for the current link state and the effectiveness of the retransmission strategy. Through multi-parameter fusion modeling, a quantitative assessment of link performance is achieved.

[0031] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0032] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A data reporting method optimized based on the narrowband Internet of Things retransmission mechanism, characterized in that, Includes the following steps: A continuous timing characterization mechanism is introduced on the narrowband IoT terminal side. Fine-grained time marking is performed on each stage from the data to be reported from queuing to the completion of transmission. Combined with the repeated transmission identifier and the confirmation feedback result, multiple transmission behaviors are spliced ​​and recombined to form a transmission occupancy evolution structure. Based on the characteristics of transmission discontinuity and failure clustering presented in the transmission occupancy evolution structure, the temporal disturbance information of cell coverage fluctuation is extracted, the original fixed retransmission interval is deconstructed and rearranged, and a non-uniform retransmission triggering scheme is generated. Under the constraint of the non-uniform retransmission triggering scheme, the failure backtracking path in the transmission occupancy evolution structure is cross-mapped with the current uplink resource availability, the reachability probability distribution of each data to be retransmitted in the time slot is derived, and a retransmission scheduling set with hierarchical constraints is constructed. Around the expected transmission time of each data unit in the retransmission scheduling set, the wake-up cycle information in the terminal power-saving mode is introduced to correct the time offset between the retransmission trigger point and the terminal active window, and the correction result is written back to the retransmission scheduling set to form a candidate transmission arrangement constrained by time alignment. Based on the candidate transmission arrangement, the successfully transmitted convergence segments and the continuously failing diffusion segments are mapped back to their corresponding trigger times and alignment offsets to construct a link response metric.

2. The data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 1, characterized in that, The process of performing fine-grained time stamping on each stage of the reported data from enqueueing to completion of transmission is as follows: A time acquisition unit is set up inside the protocol stack of the narrowband IoT terminal to work in coordination with the access layer scheduling, and to calibrate the data unit entering the buffer queue, obtaining uplink authorization, and the end of radio frequency transmission one by one; The time intervals between each node are sampled in segments, and multi-dimensional time slice features are formed by combining the transmission power adjustment and the number of repeated transmissions. To address processing delay anomalies during continuous transmission, a time offset detection mechanism is established to mark and classify abnormal segments.

3. The data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 2, characterized in that, The process of forming the transmission occupancy evolution structure is as follows: Extract the repeated transmission number and corresponding confirmation feedback status of each data unit during the uplink process of narrowband IoT, and associate the successful confirmation and failed retransmission records one by one; Using the time slice features corresponding to a single round of transmission as the basic unit, multiple rounds of transmission are spliced ​​on the time axis, and consecutive failure segments are aggregated. The correlation between sending interval and feedback delay is introduced during the splicing process to provide a structured representation of resource usage at different stages; Based on changes in occupancy intensity and the degree of failure concentration, a transmission occupancy evolution structure reflecting the evolution characteristics of the transmission process is constructed.

4. The data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 3, characterized in that, The process of extracting time-domain disturbance information related to cell coverage fluctuations is as follows: A continuous analysis of the time distribution in the transmission occupancy evolution structure is performed to identify the transmission interruption location and its adjacent intervals; Within the interruption interval, the correlation between the number of repeated failures and signal quality indicators is statistically analyzed to extract link fluctuation characteristics; By constructing a failure density distribution function within a time window, the degree of transmission instability in different time segments is quantified; By correlating and fitting the failure density distribution with coverage intensity changes, temporal perturbation information representing the coverage fluctuations of narrowband IoT cells is generated.

5. The data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 4, characterized in that, The process of generating a non-uniform retransmission triggering scheme is as follows: The fixed retransmission interval is divided into multiple candidate time intervals based on the time-domain perturbation information, and the stability of each interval is evaluated. For high failure density intervals, retransmission triggering is delayed; for low interference intervals, triggering is shifted forward. Based on the utilization of random access resources, conflict detection and avoidance are performed on the adjusted trigger time; Through multiple rounds of screening and correction, a non-uniform retransmission triggering scheme adapted to the link fluctuation characteristics was formed.

6. The data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 5, characterized in that, The process of cross-mapping the failed backtracking paths in the transmission occupancy evolution structure with the current uplink resource availability is as follows: Extract the time position and resource occupancy status of each failed segment in the transmission occupancy evolution structure; Obtain the distribution of uplink resource blocks within the current scheduling period and establish a mapping relationship between time and resources; Project the failed backtracking path to the available resource area to identify executable retransmission locations; For paths that cannot be directly matched, perform delay extension and path reconstruction to form an updated executable retransmission path.

7. A data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 6, characterized in that, The process of constructing a retransmission schedule set with hierarchical constraints is as follows: Based on the executable retransmission path, the historical success rate of each data unit in different candidate time slots is statistically analyzed. By combining link quality indicators and resource contention intensity, the success probability of each time slot is estimated and calculated; Based on the success probability and the timeliness requirements of data services, the data to be retransmitted is prioritized. Data is grouped and scheduled according to priority results to construct a retransmission scheduling set with hierarchical constraints.

8. The data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 7, characterized in that, The process of correcting the time offset between the retransmission trigger point and the terminal active window is as follows: Obtain the periodic wake-up parameters and active time window of the narrowband IoT terminal in power-saving mode; Map the expected transmission time of each data unit to the corresponding wake-up period and calculate the time deviation from the active window; Perform forward or backward adjustments for trigger times that exceed the active window range.

9. A data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 8, characterized in that, The process of forming a candidate transmission arrangement constrained by time alignment is as follows: Update the transmission time after time correction is completed to the scheduling record of the corresponding data unit; Perform consistency checks on the updated scheduling records to eliminate potential resource conflicts; The scheduling order is reorganized based on the corrected time distribution, and a candidate transmission arrangement with time alignment constraints is formed through the reconstruction of the scheduling structure.

10. A data reporting method based on narrowband IoT retransmission mechanism optimization according to claim 9, characterized in that, The process of constructing link response metrics is as follows: Continuously track the actual transmission results in the candidate transmission schedule and distinguish between successfully completed segments and repeatedly failed segments; By inversely correlating the trigger times and alignment offsets corresponding to different segments, and by statistically analyzing the concentration of successful segments and the expansion range of failed segments, link response characteristic parameters are extracted. A link response metric is constructed based on link response characteristic parameters to represent the stability of narrowband IoT links and the effectiveness of retransmission mechanisms.

Images