High-speed refresh positioning method and system based on wireless signal processing

Abstract

The application discloses a high-speed refreshing positioning method and system based on wireless signal processing, relates to the technical field of high-speed refreshing positioning, and specifically comprises the following steps: analyzing the period number of a signal data unit in a current refreshing period, judging whether there is a signal data unit with a period number not equal to the current period number, and starting a cross-period data synchronization state judgment process when the signal data unit exists and the number ratio exceeds a preset threshold; based on the time tag distribution, historical data ratio and source period number of the signal data unit in the current refreshing period, feature information for representing time consistency and period structure is extracted, and the feature information is input into a classification judgment function to determine the cross-period data synchronization state corresponding to the current refreshing period. The application solves the positioning mis-refreshing problem caused by cross-period data mispositioning, realizes refreshing control based on synchronization state judgment, and effectively improves positioning accuracy and system stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of high-speed refresh positioning technology, and in particular to a high-speed refresh positioning method and system based on wireless signal processing. BACKGROUND

[0002] In scenarios such as the Internet of Things, intelligent transportation, intelligent manufacturing, and augmented reality that require high spatial perception capabilities, real-time and high-precision positioning capabilities have become key supporting technologies. Wireless signals have the advantages of flexible deployment, stable transmission, and wide perception range, and are widely used in indoor and outdoor positioning systems. By analyzing signal characteristic parameters such as received signal strength (RSSI), time of arrival (ToA), time difference of arrival (TDoA), phase information, or spectral features from multiple wireless transmitting sources, the position of a mobile target can be calculated. In order to adapt to high-speed mobile targets, continuous tracking scenarios, and tasks that require high-frequency updates of position data, positioning systems need to have high-speed acquisition, rapid processing, and high-frequency refresh capabilities. High-speed refresh positioning technology based on wireless signal processing is gradually becoming an important direction for high-dynamic application environments. Such technology usually needs to build a method framework with multi-source signal fusion, low-latency computing, and dynamic prediction capabilities, and combine with the corresponding system architecture to meet the needs of stable, continuous, and high-precision positioning in complex environments.

[0003] Existing high-speed refresh positioning methods based on wireless signal processing usually deploy multiple wireless signal transmitting sources (such as Wi-Fi routers, Bluetooth beacons, UWB base stations, or 5G small base stations), continuously collect wireless signal data sent by each transmitting source at the target device receiver end, and preprocess the signal parameters using time synchronization mechanisms or known coordinate references. The positioning method generally includes the following core links: first, the signal acquisition and preprocessing module, which filters, normalizes, and analyzes the time / frequency domain of received RSSI, ToA, TDoA, phase, or CSI data; second, the positioning engine part, which performs position estimation calculation based on geometric positioning models (such as trilateration, multilateration) or statistical models (such as Kalman filtering, particle filtering, deep learning models); third, the data refresh and update mechanism, which usually combines sliding windows, prediction algorithms, or edge computing frameworks to achieve high-frequency output updates to meet real-time requirements; finally, the positioning results can be transmitted to a remote system for navigation control or visualization services. The entire process involves signal decoding, feature extraction, positioning calculation, error correction, and result feedback, forming an integrated wireless positioning processing framework.

[0004] The existing technology has the following deficiencies:

[0005] In a high-speed refresh positioning method that uses a cross-cycle distributed sampling strategy, the system merges and processes signal data received in multiple consecutive refresh cycles to improve anti-interference capabilities and result stability. However, in the event of signal source congestion, data packet delays, or short-term loss, data from some cycles may not enter the processing flow of the current cycle in a timely manner, but instead are added to the queue in subsequent cycles. This causes the system to mix the lagged signals of the previous two cycles with some new data of the current cycle within one refresh cycle, forming a specific state of "cross-cycle input time misalignment." Because this misalignment occurs in the underlying data reception and queue scheduling links, the refresh logic cannot sense whether the input data comes from the same timing reference. As a result, the data with time offset is incorrectly used for current position estimation during processing, ultimately outputting an "outdated positioning result" that does not match the current actual position. The existing high-speed refresh positioning method based on wireless signal processing cannot determine whether the current positioning refresh should be executed based on the "cross-cycle data synchronization status" when the "distributed sampling input is delayed and misaligned at the cycle boundary". As a result, the incorrect refresh not only causes the system output trajectory to "jump back" or "drift", but also causes incorrect judgment of the forward direction and distorted path prediction. In severe cases, it may even affect the normal decision-making of the navigation control and dynamic obstacle avoidance modules, destroying the responsiveness and accuracy of the entire system.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0007] The purpose of the present invention is to provide a high-speed refresh positioning method and system based on wireless signal processing to solve the problems in the above-mentioned background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a high-speed refresh positioning method based on wireless signal processing, specifically comprising the following steps:

[0009] Processing the wireless signal data received in each refresh cycle, adding timestamp information and cycle number information, generating signal data units containing time information and cycle information, and storing all signal data units in a data buffer;

[0010] Analyze the cycle numbers of the signal data units in the current refresh cycle to determine whether there are signal data units with cycle numbers that are not equal to the current cycle number, and if such signal data units exist and their proportion exceeds a preset threshold, start the cross-cycle data synchronization status determination process;

[0011] Based on the time tag distribution of the signal data units in the current refresh cycle, the historical data proportion, and the number of source cycles, feature information representing time consistency and cycle structure is extracted, and the feature information is input into a classification discriminant function to determine the cross-cycle data synchronization state corresponding to the current refresh cycle, which is synchronization consistent, synchronization uncertain, or synchronization failure.

[0012] According to the cross-cycle data synchronization state, it is determined whether the current positioning refresh should be performed, and the corresponding refresh control operation is performed.

[0013] In the continuous refresh cycles after the cross-cycle data synchronization state is synchronization failure, only the wireless signal data with the cycle number equal to the current cycle number is accessed and used for positioning calculation, and the synchronization failure state is recorded for sampling control strategy adjustment.

[0014] Preferably, the process of processing the received wireless signal data in each refresh cycle includes:

[0015] Each wireless signal data is attached with timestamp information and cycle number information, the timestamp information records the receiving time of the wireless signal data, and the cycle number information is an identifier generated according to the refresh cycle number;

[0016] Based on the timestamp information and the cycle number information, a signal data unit is generated, which includes a signal content field, a time tag field, and a cycle number field;

[0017] The signal data unit generated in each refresh cycle is stored in a data buffer area divided according to the refresh cycle, and the data buffer area indexes and manages the signal data units according to the refresh cycle number.

[0018] Preferably, the process of analyzing the cycle number of the signal data units in the current refresh cycle and starting the cross-cycle data synchronization state discrimination process when a certain condition is met includes:

[0019] The current refresh cycle number is obtained as the current cycle number, and a current cycle data set is constructed, which contains all the signal data units received in the current refresh cycle;

[0020] Each signal data unit in the current cycle data set is traversed, the cycle number field contained in each signal data unit is extracted, the signal data unit with a cycle number different from the current cycle number is marked as a lag data unit, and a lag data set is constructed;

[0021] The ratio between the number of the lag data set and the total number of the current cycle data set is calculated as a lag proportion, and when the lag proportion is higher than a preset threshold, it is determined that there is a cycle number misplacement risk in the current refresh cycle;

[0022] When it is determined that there is a risk of period number misplacement in two consecutive refresh cycles, a cross-period data synchronization state discrimination process is started;

[0023] Period offset distribution information is constructed based on the period number offset degree of each signal data unit in the lag data set, which is used as the input basis for subsequent synchronization state classification discrimination.

[0024] Preferably, the feature information representing time consistency includes:

[0025] The reception time label of each signal data unit is extracted to construct a time label sequence;

[0026] The range, standard deviation, skewness, kurtosis and time window uniformity are calculated based on the time label sequence to depict the time distribution characteristics;

[0027] The refresh cycle time axis is divided into equal length time segments, the number of signal data units in each time segment is counted, and the Shannon entropy is calculated as the time distribution entropy value;

[0028] All time distribution characteristics are uniformly constructed into a time consistency feature set for subsequent synchronization state recognition.

[0029] Preferably, the feature information representing historical data structure includes:

[0030] The period number of each signal data unit in the current refresh cycle is obtained and compared with the current period number to identify historical data units;

[0031] The signal data units with period numbers less than the current period number are determined as historical data, and the proportion between the number of historical data and the total number of data in the current period is calculated to construct the historical data proportion;

[0032] A period source distribution vector is constructed according to the number of signals from each period in the historical data, and the maximum source period concentration and the historical period span are calculated;

[0033] The number of unique period numbers appearing in the historical data is counted to determine the diversity of historical period sources;

[0034] The historical data proportion, source concentration, period span and period source diversity are combined into a feature information set representing the historical data structure.

[0035] Preferably, the process of extracting source period number related feature information includes:

[0036] The period numbers of all signal data units in the current refresh cycle are extracted;

[0037] constructing a set containing unique cycle numbers, determining the number of different cycle numbers in the set as the source cycle number;

[0038] calculating the source cycle redundancy based on the source cycle number, the source cycle redundancy being the source cycle number minus one;

[0039] taking the source cycle number and the source cycle redundancy as characteristic indicators for measuring the complexity of the cycle structure;

[0040] combining the source cycle redundancy and the historical data proportion to construct a composite feature vector to enhance the cycle structure recognition capability;

[0041] taking the cycle source structure characteristics as input to participate in the synchronization state discrimination.

[0042] Preferably, the process of determining the cross-cycle data synchronization state based on the feature information comprises:

[0043] performing normalization processing on each feature information extracted in the current cycle, and mapping all feature values to a unified numerical interval;

[0044] constructing a classification discrimination function, and inputting the normalized feature information into the classification discrimination function;

[0045] determining the cross-cycle data synchronization state of the current cycle based on the result output by the classification discrimination function, the state type including synchronization consistency, synchronization uncertainty or synchronization failure;

[0046] when the confidence level output by the classification discrimination function is lower than a threshold, marking the current discrimination as an uncertain state, and delaying the final refresh judgment to the next cycle;

[0047] performing a sliding average processing on the discrimination results of multiple consecutive refresh cycles to suppress the judgment jitter caused by short-term fluctuations in the synchronization state.

[0048] Preferably, the process of constructing the classification discrimination function comprises:

[0049] normalizing the feature information representing time consistency, historical data composition and cycle source structure, and unifying all feature values to a preset numerical range;

[0050] combining the normalized feature information into a multi-dimensional input vector as the input data of the classification discrimination function;

[0051] constructing the classification discrimination function based on a three-class support vector machine, wherein the support vector machine adopts a nonlinear kernel function to realize feature space mapping and boundary fitting;

[0052] setting the output result of the classification discrimination function to correspond to three synchronization state types of synchronization consistency, synchronization uncertainty and synchronization failure;

[0053] The labeled historical period sample data is used to train the classification discriminant function, and the cross-validation method is used to optimize the model parameters to improve the generalization ability of the classification discriminant function.

[0054] Preferably, according to the cross-period data synchronization state, it is judged whether the current positioning refresh should be performed, and the corresponding refresh control operation is performed, specifically:

[0055] The cross-period data synchronization state is matched with the pre-defined synchronization consistent, synchronization uncertain, and synchronization invalid three state types, and it is judged whether the condition for performing positioning refresh is met according to the matching result;

[0056] When it is judged to be synchronization consistent, it is considered that the wireless signal data in the current period has time continuity and period consistency, and the positioning refresh operation of the current position is performed, and the refresh result is output;

[0057] When it is judged to be synchronization uncertain, it is considered that there is a certain proportion of historical data mixed or period source inconsistency in the current period, and the positioning refresh is suspended, the current period wireless signal data is stored to the delay data buffer area, and the synchronization state is re-discriminated in the next period combined with the new received data;

[0058] When it is judged to be synchronization invalid, it is considered that the wireless signal data in the current period has serious time offset or period source redundancy, and the positioning refresh operation of the current period is cancelled, and the current period wireless signal data is cleared;

[0059] When the synchronization state is synchronization uncertain or synchronization invalid, the current position refresh state is marked as incomplete, and the trajectory continuity estimation strategy is called to generate a substitute trajectory point in the path output for navigation continuity maintenance.

[0060] Preferably, the high-speed refresh positioning system based on wireless signal processing includes a unit construction module, a number screening module, a state discrimination module, a refresh decision module, and an access isolation module;

[0061] The unit construction module processes the received wireless signal data in each refresh period, adds timestamp information and period number information, generates signal data units containing time information and period information, and stores all signal data units in the data buffer area;

[0062] The number screening module analyzes the period number of the signal data unit in the current refresh period, judges whether there is a signal data unit with period number not equal to the current period number, and starts the cross-period data synchronization state discrimination process when the signal data unit exists and the number ratio exceeds the preset threshold.

[0063] The state discrimination module extracts feature information for representing time consistency and cycle structure based on the time tag distribution of the signal data unit in the current refresh cycle, the historical data proportion, and the number of source cycles, and inputs the feature information into a classification discrimination function to determine the cross-cycle data synchronization state corresponding to the current refresh cycle, the cross-cycle data synchronization state being synchronization consistent, synchronization uncertain, or synchronization failure.

[0064] The refresh decision module determines whether the current positioning refresh should be performed according to the cross-cycle data synchronization state, and performs a corresponding refresh control operation.

[0065] The access isolation module only accesses and uses wireless signal data with a cycle number equal to the current cycle number for positioning calculation in the continuous refresh cycle after the cross-cycle data synchronization state is synchronization failure, and records the synchronization failure state for use in sampling control strategy adjustment.

[0066] In the above technical solution, the technical effects and advantages provided by the present application are as follows:

[0067] 1. The present application realizes the periodical structured management of wireless signal data by constructing a timestamp and cycle number double labeling mechanism, significantly enhancing the perceptibility of the system to the cross-cycle data mixing phenomenon. In the high-speed refresh scenario, wireless signals often have transmission delays caused by network congestion or channel jitter. This scheme accurately marks the reception time and cycle attribution of each piece of data, enabling the system to finely identify whether there is "historical lag data" interference within the refresh cycle, thereby avoiding the use of time reference distorted data for current positioning calculation. This mechanism effectively improves the time sequence consistency of the data processing flow, laying a high-reliability data foundation for subsequent judgment and control.

[0068] 2. The present application establishes a cross-cycle synchronization state discrimination model based on multi-dimensional feature extraction and classification function recognition by introducing three quantifiable indexes: "time tag distribution feature", "historical data proportion", and "source cycle redundancy", realizing intelligent evaluation of the time sequence structure health degree of signal data. In this model, the system can dynamically output whether the current cycle is suitable for refresh according to the distribution skewness of the data, the historical mixing intensity, and the cycle source complexity. Compared with the traditional periodic blind refresh or static threshold judgment algorithm, this method has higher dynamic adaptability and anti-interference ability, can effectively suppress positioning refresh false triggering, positioning back jump, and trajectory drift, and ensure that the system always makes high-reliability refresh decisions based on structure consistent data.

[0069] 3、The application uses the identified synchronization state for refresh control strategy hierarchical execution: refresh immediately when synchronization is consistent, ensuring timeliness; delay buffering when synchronization is uncertain, retaining potential available information; refuse refresh when synchronization is invalid, and tighten data access standards in subsequent periods, realizing rapid isolation and recovery control of the misaligned period. Such refresh control strategy not only guarantees the continuity and accuracy of positioning results, but also effectively prevents the chain destruction of periodic interference on the overall decision chain of the system (such as path prediction and navigation planning). In summary, this scheme has significant technical advantages in dealing with complex wireless data environments such as high-frequency refresh, period mixing and data lag, and is especially suitable for scenes with high coupling of multi-period fusion, distributed sampling and real-time control. BRIEF DESCRIPTION OF DRAWINGS

[0070] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description only represent some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0071] Figure 1 The flowchart of the high-speed refresh positioning method and system based on wireless signal processing of the present application.

[0072] Figure 2 The module diagram of the high-speed refresh positioning method and system based on wireless signal processing of the present application. DETAILED DESCRIPTION

[0073] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these example implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art.

[0074] The present application provides a high-speed refresh positioning method based on wireless signal processing as shown in Figure 1 The high-speed refresh positioning method based on wireless signal processing specifically includes the following processes:

[0075] Process the received wireless signal data in each refresh period, add timestamp information and period number information, generate signal data units containing time information and period information, and store all signal data units in the data buffer area;

[0076] A high-speed refresh positioning method based on wireless signal processing is applicable to positioning terminal devices containing multiple wireless signal receivers, such as smart wearable devices, vehicle-mounted positioning modules, and mobile robotic platforms. This method relies on precise periodic structuring of wireless signal data received within continuous refresh cycles. By introducing time tags and a period numbering mechanism, it converts the raw received signals into structured signal data units, providing the foundation for refresh-level data buffering control and subsequent synchronization state modeling.

[0077] In its implementation, the system periodically samples external wireless signal sources (such as Wi-Fi routers, UWB base stations, Bluetooth beacons, and 5G small cells). The sampling frequency can be dynamically adjusted based on the refresh rate requirements of the application scenario. At the beginning of each refresh cycle, the system master clock initiates a sampling task, which runs continuously from the current cycle until the next cycle is triggered, continuously receiving signal information from the wireless interface. Each piece of received wireless signal data typically includes, but is not limited to, parameters such as signal strength (RSSI), time of arrival (ToA), carrier phase, and subcarrier frequency response (such as CSI). Upon receipt, this raw signal data is immediately cached in the local raw receive buffer and marked as "pending."

[0078] To implement structured processing at the refresh cycle level, the system appends two key pieces of information to each received wireless signal: a timestamp and a cycle number. The timestamp records the actual moment the signal data was received, typically provided by a local high-precision timer with microsecond accuracy. The cycle number identifies the refresh cycle to which the data belongs and is generated by the refresh counter of the system's master clock. For example, if the system refresh cycle is Tms, the cycle number can be defined as floor(received timestamp / T).

[0079] After appending time and cycle number information, signal data is structured as a standard data unit. This signal data unit includes at least: a signal content field (used to store raw data such as RSSI, ToA, and CSI); a time tag field (used to store the absolute timestamp of reception); a cycle number field (used to store the refresh cycle number to which it is assigned); and an optional source identification field (used to record the transmitter or channel from which the data originated). This structure enables the system to classify data based on cycle number and perform timing consistency analysis based on time tags. It serves as the fundamental input format for the subsequent discrimination algorithm.

[0080] At the end of each refresh cycle, the system will transfer all the signal data units constructed in the current cycle from the original receiving buffer to the cycle data buffer. The cycle data buffer adopts a labeled structured ring storage structure, which realizes data caching and cleaning control by cycle. Each refresh cycle corresponds to a logical buffer segment, which contains all the completed structured data units of the cycle. In the embodiment, the system can set the cycle data buffer to save data of the last N refresh cycles (N is usually dynamically adjusted according to the algorithm window length), to realize memory control and fast data rollback ability under high-frequency refresh conditions.

[0081] In order to support the subsequent detection of the "cross-cycle input misplacement state", the system can also append the cycle-in statistics index in the data buffer, such as the number of data units corresponding to each cycle number, the earliest receiving time and the latest receiving time, the source distribution structure, etc. These index information can be used as one of the judgment bases for judging whether there is data from non-current cycle number and whether it exceeds the threshold in the subsequent steps.

[0082] In order to reduce the system offset of the time label, the generation of the time stamp should be executed immediately after the completion of the physical reception of the signal, and the time stamp needs to be converted by referring to the system master clock, to avoid the time drift problem caused by the use of local timers in multiple modules. In the distributed architecture system, the time synchronization mechanism can adopt NTP, PTP or high-precision synchronization protocol based on 5G NR, to ensure that the cycle number logic is consistent at each sampling end.

[0083] The system can also sort the signal data units in the same cycle number according to the receiving time, or group them according to the source ID, as one of the preprocessing steps for signal clustering, positioning fitting or consistency evaluation in the subsequent processing stage. All the data units that have completed structured processing in the current refresh cycle and are stored in the buffer will be used as the basic data set for the "cross-cycle data synchronization state" discrimination in the subsequent steps.

[0084] In summary, by attaching a unified time stamp and cycle number to the received wireless signal data in the refresh cycle, structuring it into a manageable and analyzable signal data unit, and classifying and storing it through the cycle data buffer, a data organization framework at the refresh cycle level is constructed. This mechanism not only guarantees the data access stability in the high-speed refresh positioning environment, but also provides a solid structural support for the subsequent synchronization state modeling, misplacement state identification and refresh strategy selection.

[0085] Analyzing the cycle number of the signal data units in the current refresh cycle, judging whether there is a signal data unit with a cycle number not equal to the current cycle number, and starting the cross-cycle data synchronization state discrimination process if such signal data unit exists and the number exceeds a preset threshold;

[0086] To ensure the positioning accuracy under high-speed refresh conditions, the system will uniformly analyze the period number fields of all signal data units in the refresh cycle after the signal data unit construction is completed in each refresh cycle, and then identify whether there are lag data units from non-current cycles. When the number of lag data exceeds a certain threshold proportion in a certain cycle, the system determines that there is a risk of cross-cycle data input time misalignment, and then triggers a special "cross-cycle data synchronization state judgment process" to determine whether to allow the positioning refresh of the current cycle to be executed.

[0087] In a specific implementation, the system main clock identifies the current cycle number as C_cur when entering each refresh cycle. Then, the system scans the data unit set D_cur belonging to the current cycle refresh window in the cycle data buffer. This set contains all signal data units that are received and structured in the current refresh cycle.

[0088] When traversing each signal data unit in the set D_cur, the system reads its bound cycle number field C_i. For each signal data unit, if C_i ≠ C_cur, it is marked as a lag data unit, i.e., the data does not come from the current refresh cycle, but from the previous cycle or earlier cycle, but due to transmission delay, scheduling blockage or buffer overflow, etc. It is delayed to be injected into the current refresh cycle for processing. All such lag data units are collected to form the set D_late.

[0089] After all traversals are completed, the system calculates the total number of data units N_all in D_cur and the number of lag data units N_late in the set D_late, and thus calculates the lag data proportion indicator R_late = N_late / N_all. This indicator is used to represent the severity of cross-cycle data injection in the current cycle.

[0090] To ensure the error tolerance of the system, the system introduces a preset lag proportion threshold T_sync, the value of which is initialized according to the requirements of the system's sensitivity to the synchronization state (for example, it can be set to a lower value in a positioning scene with greater environmental interference). When R_late > T_sync, it means that a large number of data units with non-current cycle numbers are mixed in the current cycle, and the system may have entered the cycle misalignment state.

[0091] To prevent accidental network jitter or individual signal fluctuations from causing false triggering, the system can also introduce a continuity condition mechanism, i.e., if R_late > T_sync is observed twice in a row in the current cycle and the previous cycle, the "cross-cycle data synchronization state judgment process" is triggered. This design can effectively suppress temporary misalignment false positives and improve the overall robustness of the algorithm.

[0092] In a preferred embodiment, the system can classify the lag data units according to the absolute value of the difference between their cycle number and the current cycle number, forming a cycle offset distribution vector. For example, there is a class of data C_i = C_cur-1, another class C_i = C_cur-2, etc. The distribution vector can be further used as one of the input features of the subsequent discrimination model to assist in determining the severity of the synchronization state.

[0093] In addition, in order to improve processing efficiency and discrimination accuracy, the system can also introduce a cycle number index structure in the construction phase of the signal data unit. The cycle data buffer will update a cycle number statistical index table at the same time when receiving each new data. The index table maintains the number of occurrences of all cycle numbers in the current cycle window, so that the cycle mixing situation can be judged in real time without the need for scanning each piece of data.

[0094] When the system determines that there is a risk of cycle number misplacement in the current refresh cycle according to the proportion of lag data, cycle offset characteristics or continuity conditions, it can activate a special synchronization state analysis process. This process will further analyze the time consistency, data stability and cycle structure integrity in subsequent steps to determine whether to perform the positioning refresh operation of the current cycle.

[0095] In addition, in order to avoid high processing overhead or waste of system resources, if R_late < T_sync and no lag data unit appears in consecutive multiple cycles, the system can enter a "cycle number consistency stable state", in which the synchronization state discrimination process is no longer triggered every cycle, but is set to trigger at intervals or only activated when abnormal traffic appears in the buffer, further reducing the system operation load.

[0096] In summary, the present embodiment quantifies the proportion of lag data mixing by structurally analyzing the cycle number field of the signal data unit in the current refresh cycle, and introduces threshold comparison, continuous judgment and offset distribution modeling mechanism to establish cycle misplacement detection capability at the cycle number level. As the pre-filtering logic of subsequent synchronization state modeling, this detection mechanism not only improves the capture ability of cross-cycle misrefresh problems, but also enhances the tolerance of the system to network jitter and signal delay, ensuring the stability and accuracy of the high-speed refresh positioning method in complex environments.

[0097] Based on the time tag distribution, historical data proportion and source cycle number of the signal data unit in the current refresh cycle, the feature information representing time consistency and cycle structure is extracted, and the feature information is input into a classification discrimination function to determine the cross-cycle data synchronization state corresponding to the current refresh cycle. The cross-cycle data synchronization state is synchronization consistent, synchronization uncertain or synchronization failure.

[0098] To achieve automatic discrimination and control of cross-period data input state, the system sets a processing mechanism based on feature information extraction and classification function recognition to identify the synchronization consistency of data structure in the refresh period. The key of this mechanism is to extract feature information from three specific dimensions according to the basic information such as time label and period number of the received signal data unit in each refresh period, which are time label distribution, historical data proportion, and source period number. These three factors are used to reflect the time consistency and period structure integrity of the signal data in the current period, and finally the feature information is sent as input into the classification discrimination function to determine the cross-period data synchronization state of the current period, which can be synchronous consistent, synchronous uncertain or synchronous failure.

[0099] Firstly, in the time label distribution dimension, the system establishes a time distribution model within the refresh period according to the time stamp bound to each signal data unit. All signal data units are assigned a receiving time label when received, which is generated by a unified clock source (such as a local high-precision timer, a synchronous network clock, etc.). For a refresh period T, the system first obtains the time stamp sequence of the signal data unit set D_T in the period. According to the time sequence, the system performs a series of statistical processing operations to extract the following distribution characteristics:

[0100] Range R_t: the difference between the maximum and minimum time, reflecting the time span covered by the signal. If R_t is too large, it means that part of the data is likely to be injected from the previous period.

[0101] Standard deviation σ_t: represents the dispersion degree of data time distribution. The larger the value, the worse the time concentration.

[0102] Skewness Sk_t and kurtosis Ku_t: used to judge whether the data distribution deviates from the center and whether there is abnormal concentration, such as obvious positive bias or high peak, which may indicate that the signal is batched and injected, and is not typical periodic sampling.

[0103] Time window uniformity U_t: the system divides the period into equal length intervals (such as 10ms), counts the number of data in each interval, and calculates the standard deviation to quantify the uniformity of the distribution.

[0104] Time label entropy H_t: calculate Shannon entropy according to the data quantity proportion in each time period. The larger the entropy value, the more uniform the distribution, and the smaller the data concentration.

[0105] The above five types of time characteristics are key indicators for measuring whether the data in the current refresh period is naturally continuous sampling. If the time label distribution shows skewness, high peak, local density or large range extension, it is highly suspected that there is historical data mixing or period boundary misplacement.

[0106] Secondly, the system analyzes the historical data proportion. Each signal data unit is also attached with its original sampling period number. In the current period T, the system extracts the period number sequence where p_i represents the period number of the i-th signal data unit. Comparing all p_i with the current period number T, if there exists p_i < T, these data units are determined as historical data. The system further calculates the following indicators:

[0107] Historical data proportion R_h: defined as the ratio of the number of historical data units Nh and the total number Nt, i.e. R_h = Nh / Nt. This value is used to reflect the intensity of historical mixing, and the larger the value represents the more serious the misplacement.

[0108] Maximum historical source period concentration C_h: among all historical data, the data from a certain specific historical period has the highest proportion, which reflects whether the historical interference comes from a single delayed period.

[0109] Historical data period span S_h: the difference between the farthest historical period number and the current period number, representing the "depth" of historical data delay injection.

[0110] Historical period source diversity D_h: the number of period types of historical data sources, which can reflect the complexity of data delay sources.

[0111] Through the quantification of the above parameters, the system can further characterize the composition structure of historical data in the current period. For example, if R_h exceeds a certain threshold and C_h is close to 1, it means that the current period may be controlled by the delayed data input of the previous period; if S_h is greater than 2, it means that the historical data is not simply lagged by one period, but there may be serious queue accumulation or data retransmission behavior.

[0112] The third dimension is the number of source periods. The system extracts the period numbers of all signal data units in the current period T, and constructs a unique number set P = {p e D_T | p is a unique period number}. The number of source periods S_p = |P| represents how many different signal source periods appear in this period. If S_p = 1, it means that all data comes from the current period, which is the most ideal state of synchronization consistency; if S_p > 1, it means that there is historical data mixing. The system further defines the source period redundancy R_d = S_p - 1 as an evaluation indicator of period structure complexity.

[0113] To enhance the collaborative representation effect of the indicators, the system also introduces a combined feature construction mechanism. For example, R_h and R_d are combined as a feature pair <R_h, R_d> to identify abnormal period structures with strong historical proportion and multiple sources, or H_t and C_h are combined to identify typical skewed batch delay injection phenomena.

[0114] After the above feature extraction is completed, the system normalizes all original and derived features to avoid different dimensions or value ranges from causing bias to the classification function. Normalization can use Z-score standardization or minimum-maximum mapping to map all feature values to the interval [0, 1] or [-1, 1].

[0115] The system inputs the standardized feature set generated in each refresh cycle into a classification discriminant function. In a preferred scheme, the classification function uses a three-class support vector machine (SVM) model to construct, and uses a nonlinear kernel function (such as a radial basis kernel) to complete the fitting of the boundary line in the high-dimensional feature space. The model accepts three synchronous states as output labels:

[0116] Synchronization consistent: the current cycle data time distribution is uniform, there is no obvious mixing of historical data, and the source cycle is single;

[0117] Synchronization uncertain: the time distribution is slightly deviated, the historical data proportion is slight but exists, and the source cycle is less;

[0118] Synchronization failure: time distribution is highly discrete or concentrated, historical data proportion is high, source cycle is diverse or span is large.

[0119] The above state labels are obtained after supervised labeling based on known samples in the training stage. The system selects cycle samples in the real running environment as training data, and constructs a state labeling data set combined with manual auditing results. Through cross-validation method, the model parameters are optimized to have strong generalization ability and low misjudgment rate.

[0120] In each refresh cycle, the system inputs the extracted features into the classification model, and determines whether to continue to perform positioning refresh operation according to the state label output by the model. The confidence of the classification result is also calculated and output, if the confidence is low, the system can mark the result as “delayed decision” state, and use the old data in the next cycle to make a judgment again to avoid critical value misjudgment.

[0121] Further, to ensure the stability of classification, the system also establishes a classification output cache queue to perform sliding average processing on the classification results of continuous multiple cycles to determine whether there is classification oscillation. If the synchronous state repeatedly changes for continuous multiple cycles, the system can temporarily adjust the classification threshold or activate the backup judgment strategy (such as rule template judgment) to enhance the robustness of the system.

[0122] The synchronization state judgment mechanism has the advantages that the design is based on clear and quantifiable feature dimensions, covers multiple dimensions of time, structure and historical interference, has implementation feasibility and algorithm innovation in feature selection, standardization processing, model construction and strategy response, and can effectively avoid the positioning refresh false triggering problem caused by chaotic input of cycle boundary.

[0123] According to the cross-cycle data synchronization state, it is judged whether the current positioning refresh should be performed, and the corresponding refresh control operation is performed.

[0124] In order to improve the judgment accuracy and refresh decision rationality of the high-speed refresh positioning method based on wireless signal processing when processing cross-cycle data, after completing the classification identification of the synchronization state of the cross-cycle data in the current refresh cycle, the system needs to further determine whether to perform the current position refresh operation based on the synchronization state, and accordingly take the corresponding refresh control strategy, so as to prevent the problems of refresh false triggering, trajectory deviation and false positioning caused by data misplacement, lag or inconsistency.

[0125] In order to achieve the above object, the embodiment first divides the synchronization state into three types, which are: synchronization consistency, synchronization uncertainty and synchronization failure. The three types of states are high-level abstract reflections of the stability of the signal data structure and the consistency of the time sequence received in the current cycle, and represent the executability level of the positioning refresh operation of the system in the current cycle. The system executes differentiated refresh decision processes according to the synchronization state type, so as to ensure that the system maintains high refresh rate while considering data quality and positioning accuracy.

[0126] For the case that the synchronization state is synchronization consistency, the state indicates that all signal data units in the current cycle are derived from the current cycle number, or contain a small proportion of adjacent cycle data but do not produce abnormal disturbance in time label distribution and source structure. In this state, the system judges that the signal data has high consistency in time and structure, and has the basic conditions required for effective positioning refresh.

[0127] Therefore, in the synchronization consistent state, the system directly inputs the data set in the current cycle into the positioning calculation module to execute the position information refresh operation of the current cycle. The positioning calculation module can execute algorithm processes such as trilateration, multipath fusion, Kalman filtering or deep model estimation based on real-time wireless signal features (such as RSSI, ToF, CSI, etc.), generate the current position estimation value, and update to the system positioning cache and output interface for calling by downstream navigation and control modules.

[0128] To avoid refresh interference or false triggering, in the synchronous consistent state, the system can also perform a "refresh confirmation" action, that is, review the key feature indicators (such as time distribution entropy and historical proportion) used to support the "synchronous consistent" judgment in the completed classification discrimination. When they are within the safe interval and the current refresh operation does not have significant position deviation (low jump variable) from the previous period refresh, it is considered as the end of a normal refresh period. Otherwise, the system can still issue a warning mark for subsequent algorithm self-checking calls.

[0129] For the case where the synchronization state is synchronous uncertainty, this state is usually triggered by the following conditions: there is a certain proportion of non-current period data units in the current period, the time distribution appears local uneven but does not reach the complete chaos degree, and the number of source periods is more than one but still within the tolerance range. At this time, the system considers that the current period input data has a certain time sequence and structure risk, and if directly refreshed, it may cause positioning deviation, but completely discarding refresh may lose continuity and cause trajectory rupture.

[0130] In the synchronous uncertainty state, the system will adopt a delayed refresh control strategy, that is, suspend the positioning refresh of the current period, and instead store all signal data units of the current period in the "delayed refresh buffer". The buffer is a high-priority storage area, which is specially used to retain data sets that may still be effectively utilized in subsequent periods.

[0131] In the next refresh period, the system will first check whether there is delayed buffer data, if there is, it will be input into the synchronization state judgment process together with the signal data units received in the new period, and the decision whether to perform refresh will be based on the second comprehensive judgment result. If the subsequent judgment state is synchronous consistent, the system will perform "fusion refresh" on the delayed period and the current period data; if it is still an uncertain state, the system can allow to continue buffering once and set the maximum buffering times to avoid infinite loop; if the state becomes invalid, the delayed data is cleared and the refresh is abandoned, and the recovery mechanism is entered.

[0132] In addition, the system will also perform a marking operation during the delay control process, that is, mark the current period as "unrefreshed" and insert a "smooth jump reference point" in the system path trajectory, which is used for subsequent path correction and input adjustment of the speed estimation model, to avoid navigation misjudgment or dynamic obstacle avoidance failure caused by refresh rupture.

[0133] For the case where the synchronization state is synchronous invalid, this state indicates that the signal data in the current period has serious structure misplacement or time anomaly, such as historical data proportion far exceeding the threshold, more than two source periods, time label highly concentrated or presenting double peak distribution, etc. It is extremely likely to be caused by data delay accumulation, communication congestion or data scheduling anomaly.

[0134] In this state, the system determines that the data structure in the current period no longer has the basic consistency condition to support the positioning refresh operation. Forcing refresh may cause serious track "jumping", path backtracking or positioning drift, thereby affecting the stability and accuracy of the entire navigation link. Therefore, in the synchronization failure state, the system adopts the strategy of terminating the refresh operation.

[0135] Specifically, the system immediately gives up the positioning refresh operation in the current period and performs the following three control measures:

[0136] Clearing the current period data buffer: preventing the current invalid data from being mis-transmitted into the subsequent period for use, affecting the discrimination and estimation of the next period;

[0137] Refresh failure label mounting: the system writes a refresh failure label for the current period to the state record interface and records the failure reason (such as high historical proportion, abnormal structure of redundant source, etc.), which is used for system log and subsequent analysis call;

[0138] State recovery mechanism starts: the system enters the "refresh suppression mode" for a limited period after the failure state, that is, in the future several periods, only data units with the same source period number and current period number are allowed to be received, and all lagging or inconsistent data will be directly discarded to restore the stability of the data structure.

[0139] It is worth noting that in the several periods after synchronization failure, the system will significantly tighten the data access criteria and enhance the fault tolerance lower limit of the discriminator to prevent continuous positioning failure due to misidentification. At the same time, the system will perform short-term prediction interpolation based on the speed vector and direction estimation results of the last valid period to fill the refresh gap and maintain the continuity of the track.

[0140] This refresh control mechanism is based on hierarchical judgment and differential control based on synchronization state, which can not only achieve the dual goals of high-frequency refresh and data quality guarantee, but also effectively identify and isolate pseudo-updates caused by period misplacement, improving the dynamic response and robustness of the entire wireless positioning system. In actual deployment, this mechanism can adapt to different refresh period settings (such as 10ms, 20ms, etc.) and data density scenarios, and has good system compatibility and engineering landing nature.

[0141] In summary, the present embodiment provides a refresh judgment and control mechanism based on synchronization state identification results, which is based on state identification results to perform three kinds of refresh responses: direct refresh (synchronization consistent), delayed refresh (synchronization uncertain), and terminated refresh (synchronization failure), corresponding to normal, safety protection and abnormal isolation respectively. This mechanism can run independently in the algorithm link, or can be linked with the positioning estimation engine and path smoothing module to realize reliable refresh decision-making in complex dynamic environments.

[0142] In the continuous refresh cycle after the cross-cycle data synchronization state is invalid, only the wireless signal data with the cycle number equal to the current cycle number is accessed and used for positioning calculation, and the invalid state is recorded for sampling control strategy adjustment.

[0143] To improve the recovery ability and decision stability of the system after processing cross-cycle data abnormal state, when the cross-cycle data synchronization state is identified as invalid, the system will actively enter a continuous refresh cycle window with tightened control of data access and calculation behavior. In this window, the system limits the access of signal data units with cycle numbers equal to the current refresh cycle number for positioning calculation, and records the invalid state through a state recording mechanism for subsequent adjustment of the sampling control strategy.

[0144] Specifically, when any refresh cycle is identified as invalid, the system immediately triggers the "refresh conservative window mechanism", which specifies that in the multiple consecutive refresh cycles after the invalid cycle, the refresh logic will only accept data units with the same cycle number as the current cycle number, and signal data units from other source cycles (i.e. with cycle numbers less than the current cycle number) will be directly discarded as delayed data even if they fall within the current cycle in terms of receiving time. This cycle number consistency determination behavior can isolate historical data accumulated or delayed in previous cycles within a certain time range, and forcibly ensure that subsequent calculations are based only on raw data obtained from current sampling, thereby creating a foundation for the reconstruction of the system's timing consistency.

[0145] To achieve the above purpose, the system performs cycle number screening operation before distributing the data to the processing module after the signal data receiving module in the current cycle completes data access. The specific method is as follows: the system extracts all received signal data units from the data buffer, and compares their attached cycle number fields one by one, and filters out all data units with cycle numbers not equal to the current cycle number. This process does not depend on the relative newness or oldness of the timestamp or the data source location, but uses the number field as the only judgment basis to avoid mixing historical lag data into the positioning estimation process from the root.

[0146] In the positioning calculation module, to prevent precision degradation due to reduced data volume, the system enables a redundancy enhancement strategy in the recovery cycle, such as dynamically adjusting the minimum data set size threshold required in the signal processing algorithm, or appropriately reducing the positioning output frequency to enhance the tolerance to data jitter. In addition, the system can temporarily call the historical trajectory vectors stored in the path smoothing module to participate in position information correction, maintaining the continuity and credibility of the navigation trajectory.

[0147] It's worth noting that to prevent valid data from being discarded due to cycle numbering, the system also incorporates a time redundancy tolerance mechanism. If received data meets the cycle number requirement but its time stamp is significantly out of date (e.g., more than a few milliseconds behind the current cycle window), the system will automatically disregard it, thus ensuring the freshness and currency of refreshed data.

[0148] During this conservative refresh window, the system not only restricts data access but also continuously records information such as the duration of this state, the amount of data discarded, and the number of consecutive synchronization failures. All status information is written to an internal "synchronization anomaly status log" and regularly transmitted to the system's sampling control strategy analysis unit for dynamic adjustment of subsequent signal sampling behavior.

[0149] The sampling control strategy can automatically implement the following strategy adjustments based on the log analysis results:

[0150] Extended refresh cycle: If continuous failure states occur frequently, the system can temporarily increase the length of the refresh cycle to alleviate the problem of data queue accumulation;

[0151] Pre-cache expansion: Dynamically expand the data receiving buffer capacity to improve the historical data diversion capability;

[0152] The signal source rotation mechanism is triggered: an attempt is made to switch some data source base stations or channel frequency bands to avoid interference on specific channels;

[0153] Sampling rate adjustment: If redundant data causes frequent failures, the system can appropriately reduce the sampling rate to balance data density and processing delay.

[0154] This strategy forms a closed-loop feedback process: identification of failure states → refreshing behavior restrictions → state recording → strategy adjustment → data structure optimization → state recovery, ensuring that the system can achieve robust and efficient continuous positioning capabilities in dynamic and complex wireless communication environments.

[0155] The recovery and refresh mechanism after synchronization failure provided in this embodiment not only avoids the risk of "contamination" of subsequent cycles by data from the failure period, but also realizes the adaptive evolution of the sampling strategy while ensuring system recovery stability through a parallel design of control and strategy. This is of great significance for improving the fault tolerance and long-term stability of the high-speed refresh positioning algorithm under non-ideal network conditions.

[0156] like Figure 2 The high-speed refresh positioning system based on wireless signal processing shown includes a unit construction module, a number screening module, a state determination module, a refresh decision module and an access isolation module;

[0157] The unit construction module processes the received wireless signal data in each refresh cycle, adds timestamp information and cycle number information, generates signal data units containing time information and cycle information, and stores all signal data units in a data buffer;

[0158] The number screening module analyzes the cycle numbers of the signal data units in the current refresh cycle, determines whether there are signal data units with cycle numbers not equal to the current cycle number, and starts the cross-cycle data synchronization state judgment process when the signal data units exist and the number ratio exceeds a preset threshold.

[0159] The state judgment module extracts feature information representing time consistency and cycle structure based on the time tag distribution, historical data ratio, and source cycle number of the signal data units in the current refresh cycle, and inputs the feature information into a classification judgment function to determine the cross-cycle data synchronization state of the current refresh cycle, which is synchronization consistent, synchronization uncertain, or synchronization invalid.

[0160] The refresh decision module determines whether the current positioning refresh should be performed according to the cross-cycle data synchronization state, and performs the corresponding refresh control operation.

[0161] The access isolation module only accesses and uses wireless signal data with a cycle number equal to the current cycle number for positioning calculation in consecutive refresh cycles after the cross-cycle data synchronization state is synchronization invalid, and records the synchronization invalid state for sampling control strategy adjustment.

[0162] The above formulas are dimensionless numerical calculations, and the formulas are obtained by software simulation of a large amount of data to represent the current real situation. The preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0163] The above-described embodiments can be implemented in part or in whole through software, hardware, firmware or any combination thereof. When implemented in software, the above-described embodiments can be implemented using one or more computer programs written in any suitable programming language. Such programs can be stored in one or more storage media or memory devices (e.g., a hard disk, a floppy disk, a CD, an optical disk, a DVD, a Blu-ray disk, a RAM, a ROM, a flash memory, a cache, etc.) of the computer and loaded into the computer. Such programs can also be loaded into the computer from one or more external sources through a network or a communication channel. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus.

[0164] It should be understood that the sequence of the above processes is not intended to mean the execution order, and the execution order of the processes should be determined according to their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0165] Those skilled in the art can understand that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0166] In several embodiments provided in the present application, it should be understood that the disclosed system and method can be implemented in other ways. For example, the above-described embodiments are merely illustrative, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0167] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiment of the present application according to actual needs.

[0168] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0169] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A high-speed refresh positioning method based on wireless signal processing, characterized in that: The specific steps include: Processing the wireless signal data received in each refresh cycle, adding timestamp information and cycle number information, generating signal data units containing time information and cycle information, and storing all signal data units in a data buffer; Analyze the cycle numbers of the signal data units in the current refresh cycle to determine whether there are signal data units with cycle numbers that are not equal to the current cycle number, and if such signal data units exist and their proportion exceeds a preset threshold, start the cross-cycle data synchronization status determination process; Based on the time tag distribution of signal data units in the current refresh cycle, the proportion of historical data, and the number of source cycles, feature information used to characterize time consistency and cycle structure is extracted, and the feature information is input into the classification discriminant function to determine the cross-cycle data synchronization status corresponding to the current refresh cycle. The cross-cycle data synchronization status is synchronization consistency, synchronization uncertainty, or synchronization failure; The feature information extracted to represent temporal consistency includes: Extract the receiving time tag of each signal data unit and construct a time tag sequence; Calculate the range, standard deviation, skewness, kurtosis and time window uniformity based on the time tag sequence to characterize the time distribution characteristics; Divide the refresh cycle time axis into equal time segments, count the number of signal data units in each time segment, and calculate the Shannon entropy as the time distribution entropy value; All time distribution features are uniformly constructed into a time consistency feature set for subsequent synchronization state identification; Extracting characteristic information that characterizes the historical data structure includes: Obtaining the cycle number of each signal data unit in the current refresh cycle and comparing it with the current cycle number to identify the historical data unit; Determine that the signal data units with a cycle number smaller than the current cycle number are historical data, calculate the ratio between the number of historical data and the total number of data in the current cycle, and construct the historical data ratio; Construct a cycle source distribution vector based on the number of signals from each cycle in the historical data, and calculate the maximum source cycle concentration and historical cycle span; Count the number of unique cycle numbers that appear in historical data to determine the diversity of historical cycle sources; The historical data proportion, source concentration, period span and period source diversity are combined into a feature information set that characterizes the historical data structure; The process of extracting characteristic information related to the number of source cycles includes: Extract the cycle numbers of all signal data units in the current refresh cycle; Construct a set containing unique period numbers, and determine the number of different period numbers in the set as the number of source periods; Calculating source cycle redundancy based on the number of source cycles, the source cycle redundancy being the number of source cycles minus one; The number of source cycles and the redundancy of source cycles are used as characteristic indicators to measure the complexity of the periodic structure; Combining the source cycle redundancy with the historical data proportion to construct a composite feature vector to enhance the ability to identify periodic structures; The periodic source structural features are used as input to participate in the synchronization state discrimination; According to the cross-cycle data synchronization status, determine whether the current positioning refresh should be executed and perform the corresponding refresh control operation; During the continuous refresh cycle after the cross-cycle data synchronization status is synchronization failure, only the wireless signal data with a cycle number equal to the current cycle number is accessed and used for positioning calculation, and the synchronization failure status is recorded for use in sampling control strategy adjustment.

2. The high-speed refresh positioning method based on wireless signal processing according to claim 1, characterized in that: The process of processing the wireless signal data received in each refresh cycle includes: Adding timestamp information and cycle number information to each wireless signal data, where the timestamp information records the reception time of the wireless signal data, and the cycle number information is an identifier generated according to the refresh cycle number; Generate a signal data unit based on the timestamp information and the cycle number information, the signal data unit including a signal content field, a timestamp field and a cycle number field; The signal data units generated in each refresh cycle are stored in a data buffer divided according to the refresh cycle, and the data buffer classifies and manages the signal data units using the refresh cycle number as an index.

3. The high-speed refresh positioning method based on wireless signal processing according to claim 2, characterized in that: The process of analyzing the cycle number of the signal data unit in the current refresh cycle and starting the cross-cycle data synchronization state determination process when specific conditions are met includes: Obtaining the current refresh cycle number as the current cycle number, and constructing a current cycle data set, where the current cycle data set includes all signal data units received in the current refresh cycle; Traverse each signal data unit in the current cycle data set, extract the cycle number field contained in each signal data unit, mark the signal data unit whose cycle number is not equal to the current cycle number as a lagging data unit, and construct a lagging data set; The ratio between the number of hysteresis data sets and the total number of data sets in the current cycle is calculated as the hysteresis ratio. When the hysteresis ratio is higher than the preset threshold, it is determined that there is a risk of cycle number misalignment in the current refresh cycle. When it is determined that there is a risk of cycle number misalignment in two consecutive refresh cycles, the cross-cycle data synchronization status judgment process is started; Based on the period number deviation degree of each signal data unit in the lag data set, period deviation distribution information is constructed as the input basis for subsequent synchronization state classification and judgment.

4. The high-speed refresh positioning method based on wireless signal processing according to claim 3, characterized in that: The process of determining the cross-cycle data synchronization status based on feature information includes: Normalize the feature information extracted in the current cycle and map all feature values ​​to a unified numerical range; Construct a classification discriminant function and input the normalized feature information into the classification discriminant function; Determine the cross-cycle data synchronization status of the current cycle based on the output of the classification discriminant function. The status types include synchronization consistency, synchronization uncertainty, or synchronization failure. When the confidence level of the classification discriminant function output is lower than the threshold, the current judgment is marked as uncertain and the final refresh judgment is delayed to the next cycle; A sliding average process is performed on the judgment results of multiple consecutive refresh cycles to suppress the judgment jitter caused by short-term fluctuations in the synchronization state.

5. The high-speed refresh positioning method based on wireless signal processing according to claim 4, characterized in that: The process of constructing a classification discriminant function includes: Normalize the feature information used to characterize time consistency, historical data composition, and periodic source structure, unifying all feature values ​​into a preset value range; The normalized feature information is combined into a multi-dimensional input vector as input data for the classification discriminant function; A classification discriminant function is constructed based on a three-class support vector machine, where the support vector machine uses a nonlinear kernel function to achieve feature space mapping and boundary fitting; The output results of the classification discriminant function are set to correspond to three synchronization status types: synchronization consistency, synchronization uncertainty and synchronization failure; The classification discriminant function is trained using labeled historical period sample data, and the cross-validation method is used to optimize the model parameters to improve the generalization ability of the classification discriminant function.

6. The high-speed refresh positioning method based on wireless signal processing according to claim 5, characterized in that: According to the cross-cycle data synchronization status, determine whether the current positioning refresh should be executed and perform the corresponding refresh control operation, specifically: Match the cross-cycle data synchronization status with the predefined synchronization consistency, synchronization uncertainty, and synchronization failure status types, and determine whether the conditions for performing positioning refresh are met based on the matching results; When it is judged to be synchronous and consistent, the wireless signal data in the current period is considered to have time continuity and period source consistency, and the positioning refresh operation of the current position is performed and the refresh result is output; When synchronization is determined to be uncertain, it is considered that a certain proportion of historical data is mixed in the current cycle or the cycle source is not completely consistent. Positioning refresh is suspended, the wireless signal data of the current cycle is stored in the delayed data buffer area, and the synchronization status is re-determined in the next cycle in combination with the newly received data; When synchronization failure is determined, it is considered that there is a serious time offset or period source redundancy in the wireless signal data in the current cycle, and the positioning refresh operation of the current cycle is canceled and the wireless signal data of the current cycle is cleared; When the synchronization status is synchronization uncertain or synchronization failure, the current position refresh status is marked as incomplete, and the trajectory continuity estimation strategy is called to generate alternative trajectory points in the path output for navigation continuity maintenance.

7. A high-speed refresh positioning system based on wireless signal processing, used to implement the high-speed refresh positioning method based on wireless signal processing according to any one of claims 1 to 6, characterized in that: It includes unit construction module, number screening module, status identification module, refresh decision module and access isolation module; a unit construction module, processing the wireless signal data received in each refresh cycle, adding timestamp information and cycle number information, generating a signal data unit containing time information and cycle information, and storing all signal data units in a data buffer; The number screening module analyzes the cycle numbers of the signal data units in the current refresh cycle, determines whether there are signal data units with cycle numbers that are not equal to the current cycle number, and starts the cross-cycle data synchronization status judgment process if such signal data units exist and the proportion of the number exceeds a preset threshold; The state discrimination module extracts characteristic information used to characterize time consistency and cycle structure based on the time tag distribution of signal data units in the current refresh cycle, the proportion of historical data, and the number of source cycles. The characteristic information is input into the classification discrimination function to determine the cross-cycle data synchronization state corresponding to the current refresh cycle. The cross-cycle data synchronization state is synchronized, uncertain, or failed. The refresh decision module determines whether the current positioning refresh should be executed based on the cross-cycle data synchronization status and performs the corresponding refresh control operation; The access isolation module only accesses and uses the wireless signal data with a cycle number equal to the current cycle number for positioning calculation during the continuous refresh cycle after the cross-cycle data synchronization status is synchronization failure, and records the synchronization failure status for use in sampling control strategy adjustment.

Images