Methods, apparatus, equipment, media and products for cross-network resource scheduling
By building a local base station behavior database in the mobile terminal, predicting stable connection windows and centrally performing data synchronization, the problem of invalid communication in mobile terminals under frequent network switching scenarios is solved, and low-power and high-efficiency data synchronization is achieved.
Patent Information
- Application Number
- CN202611386164.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-08
- Publication Date
- 2026-11-17
AI Technical Summary
In scenarios with periodic mobility and frequent network switching, mobile terminals continuously retry establishing TCP connections and sending small data packets when in areas with weak signal coverage or when frequently switching cells due to high-speed movement. This leads to invalid communication behavior, increased RF power consumption, reduced battery life, and core network signaling congestion.
By predicting stable connection windows based on the historical mobility patterns of terminals, and constructing a local base station behavior database using the historical mobility feature data of the terminals themselves, stable connection windows are identified, and resource-intensive data synchronization operations are performed centrally during the window period to avoid invalid network connection attempts.
Significantly reduces the number of invalid network connection attempts, lowers the workload of the terminal RF module and overall power consumption, and improves the success rate of data synchronization and the terminal's battery life.
Smart Images

Figure CN123124040A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of network resource scheduling technology, and in particular to a cross-network resource scheduling method, apparatus, equipment, medium and product. Background Technology
[0002] In scenarios with periodic mobility and frequent network switching, such as subways and high-speed trains, mobile terminals need to continuously synchronize data with servers. Existing technologies use heartbeat mechanisms or timed polling schemes, which do not distinguish between wireless signal quality and link stability. Under any network access conditions, as long as the terminal completes network access, it directly initiates a synchronization request. When the terminal travels through areas with weak signal coverage or when frequent cell switching occurs due to high-speed movement, the terminal continues to retry establishing TCP connections and sending small data packets, generating a large amount of invalid communication behavior. This leads to increased terminal RF power consumption, reduced battery life, and core network signaling congestion. Summary of the Invention
[0003] To address the aforementioned technical problems, this disclosure provides a method, apparatus, device, medium, and product for cross-network resource scheduling, achieving cost reduction and efficiency improvement in cross-network resource scheduling.
[0004] A first aspect of this disclosure provides a cross-network resource scheduling method applied to a terminal, the method comprising: Based on the terminal's dwell behavior pattern when entering the current cell, the predicted stable dwell time of the terminal when entering the current cell is obtained; Based on the predicted stable dwell time, a stable connection window is identified, and resource-intensive data synchronization operations are performed centrally during the stable connection window period.
[0005] A second aspect of this disclosure provides a cross-network resource scheduling apparatus applied to a terminal, the apparatus comprising: The acquisition module is configured to obtain the predicted stable dwell time of the terminal in the current cell based on the dwell behavior pattern of the terminal entering the current cell; The window module is configured to identify a stable connection window based on the predicted stable dwell time, and to perform resource-intensive data synchronization operations centrally during the stable connection window period.
[0006] A third aspect of this disclosure provides an electronic device, including: At least one processor; Memory for storing the at least one processor-executable instruction; The at least one processor is used to execute the instructions to implement the above-described method.
[0007] A fourth aspect of this disclosure provides a computer-readable storage medium that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the methods described above.
[0008] A fifth aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described above.
[0009] The above-mentioned at least one technical solution adopted in the embodiments of this disclosure can achieve the following beneficial effects: by using the terminal's own historical movement patterns to predict stable connection windows, replacing the traditional real-time perception and blind retry mechanism, the number of invalid network connection attempts can be significantly reduced without sacrificing synchronization timeliness, thereby reducing the workload of the terminal's radio frequency module and overall power consumption. Attached Figure Description
[0010] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.
[0011] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0012] Figure 1 A flowchart illustrating a cross-network resource scheduling method provided in this embodiment of the disclosure; Figure 2 A flowchart illustrating yet another cross-network resource scheduling method provided in this embodiment of the disclosure; Figure 3 A schematic diagram of the structure of a cross-network resource scheduling device provided in an embodiment of this disclosure; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure; Figure 5 This is a schematic diagram of the structure of an exemplary computer system provided in an embodiment of the present disclosure. Detailed Implementation
[0013] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.
[0014] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.
[0015] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.
[0016] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.
[0017] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".
[0018] The following is combined Figures 1-5 This disclosure describes the cross-network resource scheduling method, apparatus, device, medium, and product provided in the embodiments of this disclosure.
[0019] Figure 1 This is a flowchart illustrating a cross-network resource scheduling method provided in an embodiment of this disclosure, as shown below. Figure 1 As shown, the methods applied to the terminal include: S101. Based on the dwell behavior pattern of the terminal entering the current cell, obtain the predicted stable dwell time of the terminal entering the current cell; First, the patterns of terminal dwell behavior when entering the current cell are obtained by collecting network dwell feature data and terminal mobility feature data, which include: terminal cellular network status, reference signal receiving power, geographical location, and cell dwell time sequence information.
[0020] Furthermore, based on the collected network dwell time feature data and terminal mobility feature data, the dwell time behavior patterns are learned offline, and a local base station behavior database is constructed and dynamically updated iteratively.
[0021] Based on the unique identifier of the current cell, query the local base station behavior database to see if there is a corresponding historical dwell time record. If there is a corresponding historical dwell time record, retrieve the historical average dwell time stored in the local base station behavior database as the predicted stable dwell time. If there is no corresponding historical dwell time record, retrieve the historical average dwell time of neighboring cells stored in the local base station behavior database based on the geographical location of the current cell as the predicted stable dwell time.
[0022] S102. Based on the predicted stable dwell time, a stable connection window is identified, and resource-intensive data synchronization operations are performed centrally during the stable connection window period.
[0023] In this step, a stable connection window is identified based on the predicted stable dwell time, and resource-intensive data synchronization operations are performed centrally during the stable connection window period; if no stable connection window is identified based on the predicted stable dwell time, no network request is initiated.
[0024] When the predicted stable dwell time exceeds the preset stable dwell threshold, the current dwell time of the terminal constitutes a stable connection window. During the stable connection window period, resource-intensive data synchronization operations are performed centrally. The start time of the stable connection window period is... = , The current moment is the termination time of the stable connection window. For the current moment The result is obtained by summing the predicted stable dwell time.
[0025] The process involves performing resource-intensive data synchronization operations within a stable connection window, including: starting the synchronization service process at a pre-set time based on the start time of the stable connection window; reading the local data change logs to be synchronized when the stable connection window is detected, and sorting all synchronization tasks according to priority; dynamically adapting the data packet size based on the duration of the stable connection window; uploading local cached data in batches according to the sorted priority queue and the adaptively adjusted data packet size within the effective time interval of the stable connection window; and closing state synchronization before the stable connection window ends, sending a synchronization state summary packet based on the reserved time, closing the network connection, and entering sleep mode.
[0026] In this embodiment of the disclosure, by using the terminal's own historical movement patterns to predict stable connection windows, instead of the traditional real-time perception and blind retry mechanism, the number of invalid network connection attempts can be significantly reduced without sacrificing synchronization timeliness, thereby reducing the workload of the terminal's radio frequency module and overall power consumption.
[0027] Figure 2This is a flowchart illustrating another cross-network resource scheduling method provided in an embodiment of the present disclosure, as shown below. Figure 2 As shown, the method includes: S201. The terminal continuously collects and locally stores runtime context information; During normal operation, the terminal adopts a fusion acquisition method that combines fixed-period sampling with event-driven sampling to obtain real-time information on the terminal's cellular network status, reference signal reception power, geographical location, and cell dwell time. All data is encrypted and stored in the terminal's local non-volatile storage space to ensure that it is not lost when power is off and that the data is secure and reliable.
[0028] This step collects real and complete spatiotemporal network observation data from the terminal, overcoming the limitations of existing technologies that rely solely on instantaneous network conditions and cannot predict future link stability. This provides fundamental data source support for subsequent tasks such as mining user mobility and dwell patterns, building a local base station behavior database, and predicting communication stability windows. Specific data collection content is as follows: (1) Cell Identity (Cell ID): The terminal extracts the Cell Identity from the system messages (SIB1, SIB3, etc.) broadcast by the cellular network. This Cell ID is used to uniquely identify cells covered by different base stations and accurately distinguish the coverage areas of different base stations. This data collection operation can achieve grouping and aggregation of behavioral data at the cell level, providing accurate dimensional basis for subsequent targeted statistics on the camping characteristics of each cell and distinguishing the communication stability of different coverage areas.
[0029] (2) Reference Signal Received Power (RSRP): The terminal reads the reference signal received power (RSRP) value (unit: dBm) measured by the physical layer in real time. The higher the value, the stronger the signal. It is used to characterize the cellular link quality of the current cell and intuitively reflects the strength and stability of the network signal during the stay. The continuously collected RSRP time-series data can form the original sample set of signal quality for each cell, providing core network indicator support for subsequent statistical analysis of the steady-state characteristics of cell signals and judgment of the stability of communication links.
[0030] (3) Terminal geographic location coordinates: The terminal preferentially uses the Global Navigation Satellite System (GNSS) module to obtain latitude and longitude coordinates. If GPS is unavailable and positioning fails, auxiliary positioning methods are activated in stages: Wi-Fi fingerprint KNN (K-Nearest Neighbors) matching positioning, Bluetooth Beacon triangulation positioning, and multi-neighbor base station location fingerprint lookup positioning, to ensure that location information can be effectively obtained in complex scenarios such as indoors, tunnels, and underground.
[0031] Among them, Wi-Fi fingerprint KNN matching positioning: KNN classification is performed based on a known list of (AP, wireless access point) and their RSSI (Received Signal Strength Indicator) vectors. Specifically, when the terminal fails to locate via GNSS, it scans the surrounding set of Wi-Fi access points, extracts the signal strength feature vectors of multiple APs, and performs similarity matching with the local historical Wi-Fi fingerprint database through the K-nearest neighbor algorithm, thereby realizing the terminal location estimation in the absence of satellite signal.
[0032] Bluetooth Beacon triangulation: This method estimates distance using iBeacon broadcast signal strength. iBeacon is a short-range broadcast positioning protocol based on Bluetooth Low Energy technology, which can be fixedly deployed indoors as positioning anchor points. The specific positioning logic is as follows: The terminal scans all nearby Wi-Fi accessible devices (APs, wireless access points), collects the RSSI values of each AP to form a multi-dimensional signal feature vector, and performs KNN similarity classification matching between the real-time feature vector and a locally pre-stored geographic location fingerprint database. Simultaneously, Bluetooth Beacon triangulation assists in positioning: the fixed iBeacon anchor points continuously broadcast Bluetooth signals. The terminal receives the signals and, based on the attenuation pattern of the Bluetooth RSSI signal strength, estimates the straight-line distance between itself and each iBeacon anchor point. Combining the distance data from at least three anchor points, the triangulation algorithm calculates the terminal's precise geographic location, compensating for the insufficient accuracy of Wi-Fi fingerprint positioning.
[0033] Base station coarse positioning: This involves estimating the location based on a lookup table of location fingerprints formed by multiple neighboring Cell IDs. Finally, adaptation is achieved through base station coarse positioning: In extremely weak coverage scenarios where Wi-Fi and Bluetooth signals are absent, the terminal scans and collects the unique Cell IDs of the current serving cell and multiple neighboring cells, forming a cellular network fingerprint feature of the current location. This fingerprint is then matched against a pre-stored (cell combination-geographical location) fingerprint mapping table on the terminal to retrieve the geographic location that best matches the current neighboring cell combination features. This is used to estimate the terminal's current location, achieving positioning in scenarios without satellite or short-range wireless signals.
[0034] The above multi-mode, full-scenario positioning data collection can realize the spatial correlation between terminal movement trajectory and community coverage, providing location data support for subsequent cold start dwell prediction of new communities, neighboring area behavior matching, and accurate mining of spatial dwell patterns.
[0035] (4) Cell stay start and end timestamps and stay duration: The terminal is driven by cell handover and RRC (Radio Resource Control) connection reconstruction events to accurately record the complete life cycle of a single cell stay: whenever the terminal completes RRC connection reconstruction or hands over to a new Cell ID, the entry time is recorded. When cell leaving or handover to the current cell is detected, the exit time is recorded. The duration of a single stay is defined as:
[0036] in d This indicates the actual dwell time of the terminal in the cell, in seconds, and serves as the basic input for subsequent statistical analysis. Accurately collecting dwell time information and calculating dwell time can precisely characterize the terminal's dwell status in each cell. Dwell time, as a core basic metric for evaluating the stability of cell dwelling, is the core basis for subsequent prediction of communication stability windows.
[0037] The terminal configures a standard UTC (Coordinated Universal Time) timestamp for each data collection record, and organizes it in a structured manner using Cell ID as an index to form a standardized time-series log. This log includes fields such as event time, unique cell identifier, RSRP, latitude and longitude, and event type, comprehensively recording the spatiotemporal and network status changes of the terminal during each cell stay. This unified, structured data organization method enables the standardized processing of massive amounts of fragmented collected data, ensuring the accuracy and efficiency of subsequent data statistics and pattern mining. To adapt to the limited local storage resources of the terminal, this solution adopts a finite-length circular buffer storage mechanism, setting the maximum number of log entries to N (default N=5000, which can be adaptively configured according to the terminal's storage limit and system parameters). When the log entries reach the limit, the oldest historical records are automatically discarded, retaining only the latest valid data. This storage mechanism avoids local storage overflow and dynamically removes outdated and invalid historical data, ensuring that the retained data matches the user's latest mobile usage habits and guaranteeing the real-time accuracy of subsequent pattern statistics and window prediction.
[0038] To adapt to the limited local storage resources of the terminal, this embodiment adopts a finite-length circular buffer storage mechanism, setting the maximum number of log entries to N (default N=5000, which can be adaptively configured according to the terminal storage limit and system parameters). When the maximum number of log entries is reached, the oldest historical records are automatically discarded, and only the latest valid resident data is retained. This storage mechanism avoids local storage overflow on the terminal and can dynamically remove outdated and invalid historical behavior data, ensuring that the retained data matches the user's latest mobile resident habits and guaranteeing the real-time accuracy of subsequent pattern statistics and window prediction.
[0039] The structured data storage format is shown in Table 1: Table 1 - Signal Measurement Data Table
[0040] This step utilizes a silent, event-driven, and periodic fusion data collection method adapted to all scenarios to accurately collect and persistently store data across four core dimensions: unique cell identifier, signal quality, geographic location, and dwell time sequence. Simultaneously, a circular buffer storage mechanism dynamically optimizes the stored data volume, eliminating invalid and outdated data, ultimately forming a raw dataset of terminal dwell behavior that is time-ordered, comprehensive in dimensions, and closely aligned with users' real-time behavioral habits.
[0041] This step lays the data foundation for subsequent autonomous terminal statistics of cell behavior characteristics, iterative updates of the local base station behavior database, mining of user mobility and dwell patterns, and accurate identification of stable connection windows (SCW), thus enabling differentiated intelligent synchronous scheduling.
[0042] S202. Establish a local base station behavior database based on historical data; Based on the terminal cell dwell time, reference signal received power, and spatiotemporal timing data collected and persisted in step S201, a lightweight, self-iterably updating local base station behavior database is constructed on the terminal. This step involves feature extraction and pattern refinement from discrete, raw single-time dwell logs, transforming raw observation data into structured statistical features that characterize cell communication characteristics and user travel habits, thus addressing the issue that raw logs cannot be directly used for communication stability prediction. The established local base station behavior database can continuously iterate and converge with long-term terminal use, adaptively fitting typical user travel patterns and gradually converging to these patterns over time. It possesses unsupervised self-learning capabilities on the terminal side, providing standardized feature basis for subsequent stable connection window prediction and differentiated network resource scheduling.
[0043] For example, after the terminal device is powered on, it first initializes a local SQLite database instance as a dedicated local base station behavior database. This lightweight local database architecture requires no cloud computing power and enables the terminal to independently build and update the database offline. The database uses a unique Cell ID as the primary key to create independent records, with each record corresponding to a unique Cell ID. This enables the collection and management of behavioral features at the cell level. The database storage fields and corresponding statistical dimensions are shown in Table 2.
[0044] Table 2 - Community Residency Data Fields Table
[0045] The above-mentioned multi-dimensional field design breaks through the limitations of evaluation based on single dwell time. It constructs a cell profile from multiple dimensions such as dwell time, signal quality, access frequency, and timeliness, comprehensively depicting the communication stability and user behavior habits of different base station coverage areas, thereby ensuring the comprehensiveness and accuracy of subsequent window prediction.
[0046] The local base station behavior database adopts an event-driven, first-in-first-out sliding window update mechanism. It is triggered by the completion of each complete cell dwell behavior of the terminal to achieve dynamic updating of feature data and elimination of old data. The specific execution process is as follows: (1) Calculate the dwell time each time the terminal completes a complete dwell cycle (from the start of the event type entry until the end of the exit). d ; The terminal monitors the end of a single complete cell dwell cycle, i.e., from the start of the cell entry event to the end of the cell exit event, and calculates the precise dwell time based on the preceding and following UTC timestamps. d This provides the latest sample data for database updates. By using real dwell events as the update trigger, it ensures that each feature iteration closely matches the user's actual movement behavior, avoiding invalid data iterations.
[0047] (2) Check if a record with the corresponding Cell ID exists in the database; Using the current cell ID as the primary key, the system iterates through the local base station behavior database to check if there are any historical records of the corresponding cell, achieving accurate matching and updates, avoiding data confusion across cells, and ensuring the independence of single-cell features.
[0048] (3) If it exists, read duration_samples and deserialize it into an array. ; If the local base station behavior database already contains historical records corresponding to the Cell ID, the `duration_samples` field in the local base station behavior database is read and deserialized to restore the historical dwell time sample array D. By parsing historical samples, the latest samples can be superimposed on the user's past behavior data to achieve continuous convergence and optimization of features.
[0049] (4) If Then add directly. ; Set a maximum sample size M (example value M=10) to dynamically manage the sample array. When the number of historical samples k is less than the maximum size M, directly append the current dwell time d to the end of the sample array to gradually accumulate cell dwell behavior samples and improve the initial behavioral characteristics.
[0050] (5) If Remove Shift the remaining elements to the left and insert at the end. ; When the number of historical samples k is greater than or equal to the maximum capacity M, the oldest historical sample is automatically removed, and the remaining samples are shifted to the left and the latest residence time is inserted at the end of the array. d The purpose of this sliding window mechanism is to eliminate outdated and old behavioral data, retain recent and valid behavioral samples, ensure that the database features always match users' recent travel habits, and improve the model's adaptive capabilities.
[0051] (6) Update historical average dwell time ; Based on the updated residency sample array, the average residency time per cell was recalculated. The calculation formula is:
[0052] in, This indicates that the total number of duration_samples is retrieved, which is the total number of currently valid resident samples. co is a counting function. This represents the i-th dwell time sample in duration_samples. The mean value is used to update the data, enabling iterative quantification of the cell's regular dwell time capability, which serves as the core quantitative indicator for subsequently judging the stability of the communication window.
[0053] (7) Update the frequency counter; The frequency of visits to a community is cumulatively updated daily, continuously recording the number of times a user visits the community each day. This quantifies user trajectory preferences and the probability of frequent visits, providing data support for the stable weighting of different communities.
[0054] (8) Update last_seen to the current time; Update the last_seen timestamp to the end time of this cell stay, marking the latest active status of the cell. This is used to determine the timeliness of data and filter valid historical samples, thereby avoiding outdated data from interfering with the prediction results.
[0055] (9) Reserialize the updated community residency data field table into duration_samples and write it back to the database.
[0056] The updated array of resident samples is reserialized and written to the duration_samples field, and all statistical fields are updated synchronously to complete a single round of database iteration, thereby ensuring that the local base station behavior database data is real-time, effective, and accurate.
[0057] This step is based on the original spatiotemporal network historical data collected in step S201. A multi-dimensional local base station behavior database is constructed offline on the terminal, and the samples and statistical features are continuously updated through an event-driven sliding window iteration mechanism.
[0058] This step transforms fragmented, single-visit raw data into a structured, quantifiable, and iterative cell behavior feature model, achieving end-side self-learning convergence of user travel patterns and cell communication characteristics without cloud-based training or manual configuration. It solves the problem of traditional network scheduling relying solely on instantaneous signals and failing to predict future link stability. Furthermore, the sample sliding update mechanism ensures the database remains adapted to dynamically changing user travel habits, providing core feature support and quantitative basis for accurately predicting stable communication connection windows and achieving differentiated intelligent resource synchronization scheduling.
[0059] S203. When a terminal accesses a new cell, query its historical average dwell time. This step, based on the local base station behavior database built in step S202, enables rapid querying of dwell time and cold start prediction after a terminal accesses a new cell.
[0060] For unfamiliar cells without records in the database, the terminal cannot directly obtain historical dwell time statistics, resulting in the inability to identify the stability window and the failure of the synchronization scheduling strategy. This step provides a reasonable initial value for dwell time prediction in scenarios without direct historical records by using a hierarchical adaptation of direct table lookup of existing cells and geographic mapping prediction of new cells. This compensates for the lack of features in new cells and further improves the completeness and robustness of communication stability window prediction.
[0061] Whenever a terminal completes an RRC connection reconstruction or successfully switches to a new serving cell, it immediately parses the cellular network broadcast message to obtain the current cell ID, and uses the cell ID as the retrieval key to query the local base station behavior database in real time to see if there is a corresponding historical camping record, thus achieving rapid feature matching after accessing a new cell.
[0062] If a valid record for the current Cell ID exists in the database, then the record stored in the database will be read directly. The historical average dwell time field serves as a direct basis for judging the current dwell stability of the cell, enabling rapid identification and scheduling decisions of stable windows based on existing historical data.
[0063] If no corresponding record for the current Cell ID is found in the database, the cell is determined to be an unfamiliar cell that is being accessed for the first time. The terminal automatically creates a new entry for the Cell ID in the database and initializes all statistical fields to null or zero values. At the same time, the cell is marked as being in the first observation state and no synchronization action is triggered. During the initial observation phase of the cell, the terminal does not trigger any data synchronization scheduling action to avoid making incorrect scheduling decisions based on blank and invalid features and to ensure network scheduling stability.
[0064] To address the issue of prediction failure caused by the lack of historical data in unfamiliar cells, this embodiment of the present disclosure adopts a geographic proximity mapping prediction mechanism. Based on spatial location similarity and consistency of motion state, it reuses mature statistical features of neighboring high-frequency cells to generate estimated dwell time, thus filling the blind spot of cold start prediction.
[0065] The specific triggering condition is: the geographical location of the cell being accessed for the first time. The spatial distance between the cell and any cell in the set of high-frequency resident cells already stored in the database is less than a preset geographical proximity threshold. And the directions of motion are consistent (the angle between the velocity vectors is calculated based on continuous position difference). If the geographic mapping prediction mechanism is activated, a preliminary estimate will be made using historical statistical data of similar locations.
[0066] in, This is a set of neighboring candidate cells that simultaneously satisfy both spatial and directional constraints. Let be the matching weight corresponding to the j-th neighboring candidate cell, and let be the weight. The weighting decreases exponentially with increasing distance, ensuring that the closer the neighborhood, the higher the reference weight of the neighborhood characteristics, which aligns with actual travel and stay patterns. The weighting calculation formula is as follows:
[0067] in, This represents the Euclidean distance between the current location coordinates of the new residential area and the historical location coordinates of its neighboring residential areas. The coordinates of the current location of the cell. Let j be the coordinates of the location of the j-th historical cell. This parameter represents the attenuation rate control parameter, with a default value of 100. It is used to adjust the attenuation rate of spatial distance on the weight, ensuring that the statistical characteristics of nearby cells play a dominant role, while the characteristics of distant cells decay rapidly and become ineffective, thus avoiding interference from irrelevant cell data with the prediction results.
[0068] In this embodiment, a dual adaptation of hierarchical query and geographic mapping prediction is performed for newly accessed cells. For existing cells, historical statistical features can be directly reused to quickly complete stability prediction. For unfamiliar cells accessed for the first time, mature dwell characteristics of neighboring high-frequency cells are reasonably reused through spatial proximity matching, motion direction verification, and distance-weighted mapping algorithms to generate reliable initial values for dwell time prediction. Hierarchical query and geographic mapping prediction effectively solve the problem of cold start prediction blind spots caused by the lack of historical samples in new cells, make up for the deficiency that pure historical statistical models can only adapt to existing cells, achieve full coverage of various terminal dwell scenarios, greatly improve the completeness and adaptability of communication stability window prediction, and provide reliable algorithmic support for differentiated network synchronization scheduling in all scenarios.
[0069] S204. Determine whether a stable connection window is formed; This step, based on the local base station behavior database constructed in step S202, calculates the historical average dwell time characteristics of cells based on the historical dwell time sample set stored in the database, and completes the identification and time boundary definition of the stable connection window (SCW). By filtering stable time periods with continuous communication capabilities through the long-term dwell behavior patterns of users, synchronization tasks are performed only within the reliable window, avoiding problems such as synchronization failure, retransmission redundancy, and power waste caused by short-term cell handover, fast fading coverage, and instantaneous weak network, thus achieving accurate, efficient, and low-power network synchronization scheduling.
[0070] Only when historical statistical data from the local base station behavior database indicates that the terminal has long-term and stable dwell characteristics in the current cell is it determined that reliable communication conditions are available for that period of time, allowing communication tasks such as data synchronization to be performed. This determination method is based on the user's actual travel and dwell patterns, which is closer to the actual communication link stability change trend than instantaneous signal detection, and has higher predictive accuracy.
[0071] The specific determination method is as follows: read the historical average dwell time corresponding to the current serving cell. The value is then compared with a preset stable duration threshold T. Threshold T supports scene-adaptive configuration and remote iterative optimization, with a default baseline value set to 30 seconds. It also supports local configuration file customization and OTA remote dynamic updates, adapting to the dwell characteristics of different mobile scenarios. An example scenario adaptation strategy is as follows: In high-speed rail scenarios, where terminal cell switching is frequent and single-cell dwell time is short, threshold T can be lowered to 20 seconds; in urban bus scenarios, where movement speed is moderate and single-cell coverage dwell time is more stable, threshold T can be raised to 40 seconds, achieving scenario-differentiated judgment.
[0072] If the conditions are met:
[0073] If the current cell is determined to be a high-frequency, frequently visited area or a stable passageway for the user, then the current dwell time of the terminal constitutes a stable connection window (SCW). This determination indicates that the current cell link has continuous and stable communication duration guarantees, can securely carry communication tasks such as data synchronization and signaling interaction, and has no risk of sudden disconnection or rapid handover.
[0074] After determining the generation of a stable connection window, the SCW is precisely time-bounded to determine the effective time interval for stable communication. This provides a timing basis for subsequent precise window scheduling.
[0075] Among them, the start time of the window This value represents the current time. ,Right now: That is, the current effective time of the judgment is used as the start time of the stable connection window, so as to ensure real-time synchronization between window judgment and scheduling execution, without any lag or deviation.
[0076] Window termination time This value represents the historical average dwell time. With the current moment Adding them together, we get: This means using the current time plus the historical average dwell time of the cell as the window cutoff time, and predicting the stable end point of the current dwell time based on the user's historical behavior patterns to achieve forward-looking window interval estimation.
[0077] This disclosure establishes a quantifiable, configurable, and scenario-adaptive SCW stability window determination system based on locally accumulated cell dwell statistics. It filters dwell periods with reliable communication capabilities through an average dwell time threshold and precisely defines the temporal boundaries of the stable connection window. This solves the technical drawbacks of traditional wireless network scheduling, which lacks prediction, filtering, and blindly executes tasks. It achieves centralized scheduling for stable windows and differentiated scheduling for silent hibernation in unstable windows. This significantly improves the success rate and stability of wireless communication synchronization tasks while effectively reducing terminal power consumption losses caused by invalid retransmissions and frequent signaling interactions. It is suitable for various mobile scenarios such as high-speed rail, buses, and daily commutes, greatly improving the overall energy efficiency ratio of terminal cellular network communication.
[0078] S205. Intelligent scheduling of synchronous tasks based on stable connection windows; This step, based on the terminal's successful determination of a stable connection window (SCW), executes the data synchronization task according to the process of pre-processing, hierarchical uploading, and adaptive packet control. The specific steps are as follows: S2051. Perform stable connection window pre-wake-up and transmission pre-preparation; Based on the predicted start time of SCW, the terminal starts the synchronization pre-preparation process WT seconds in advance (default 10 seconds, configurable), actively wakes up the synchronization service process in the background that is in a dormant state, including the Android JobScheduler scheduled task and the iOS Background Fetch background fetch task, activates the cellular data link and initiates a server session creation request.
[0079] Three preparatory tasks were completed simultaneously: First, the local TLS session key was loaded and the encrypted session context was reused to prepare for encrypted data transmission; second, the OAuth2 Token was refreshed in advance to avoid the risk of credential expiration; and third, the server's DNS domain name was pre-resolved and the IP address was cached to eliminate network handshake delays during the transmission phase and ensure that data transmission could begin immediately after the SCW window was officially opened.
[0080] S2052, Take over scheduling authority and organize the hierarchical synchronization task queue; When the detection occurs during the SCW period, the terminal takes over the system scheduling control, disabling the traditional random synchronization mechanism. The terminal reads the backlog of data change logs awaiting synchronization in batches and sorts all synchronization tasks according to high, medium, and low priorities.
[0081] High-priority tasks: These involve critical data that must be delivered, such as account security, payment vouchers, emergency alerts, and unread notifications. Medium-priority tasks: regular message push, log reporting, location trajectory retransmission, etc. Low-priority tasks: personalized recommendation caching, non-real-time backup, offline reading content download, etc.
[0082] Build a priority task queue: } in, Indicates the first The priority level for each task can be selected as High (H), Medium (M), or Low (L). For its data body, the scheduler according to Retrieve tasks in descending order to ensure timely uploading of critical data.
[0083] Strictly follow the priority descending scheduling rules, prioritize the execution of high-priority core tasks such as account security, payment vouchers, and emergency alarms, and then process medium-priority tasks such as ordinary logs and trajectory retransmission, as well as low-priority tasks such as cache updates and offline resource synchronization, to ensure that core data is delivered first.
[0084] S2053, Dynamically adapting data packet size based on stable connection window duration; Based on the current SCW prediction duration PD, the optimal transmission data packet size is dynamically calculated using an adaptive packet size algorithm. The calculation formula is as follows:
[0085] Where MTU is the maximum transmission unit of the current link, PD is the predicted duration of this SCW, and BP represents the empirical throughput baseline, which is a configurable system parameter with a default value of 500 to comprehensively consider the impact of air interface rate, queuing delay and bit error rate.
[0086] If the SCW duration is short, the data packet size is automatically compressed to reduce data fragmentation and disassembly operations, avoiding transmission timeouts and retransmission losses. If the SCW duration is sufficient, the single packet transmission size is appropriately increased to improve overall data throughput efficiency and adapt to real-time link transmission conditions.
[0087] S2054. Perform batch hierarchical data synchronization within a stable connection window.
[0088] Within the valid SCW time interval, local cached data is uploaded in batches according to the sorted priority queue and the adaptively adjusted packet size. Synchronization operations are performed entirely within a stable communication window, avoiding link fluctuations and cell handover interference during unstable periods, and completing differentiated scheduling and transmission of all synchronization tasks.
[0089] This disclosure embodiment achieves pre-warming, hierarchical scheduling, and adaptive transmission of synchronization tasks through standardized and time-sequential step-by-step operations. It makes full use of stable SCW communication resources, eliminates problems such as invalid synchronization, bandwidth contention, and transmission timeout, and effectively improves the success rate of data synchronization and terminal operating efficiency in mobile scenarios.
[0090] S206. Before the stable connection window ends, close the state synchronization reserved time CT seconds, send the current synchronization state summary packet, and then close the network connection and enter sleep mode. This step is the final closed-loop stage of the Stable Connection Window (SCW) synchronization process. It executes status feedback and orderly network shutdown / sleep operations as the SCW window is about to end. Its purpose is to complete the integrity verification of a single synchronization task, retain the synchronization status data, and proactively release network resources, shut down radio frequency and background tasks before the window expires. This avoids transmission anomalies caused by link fluctuations at the end of the window, while simultaneously achieving extremely low-power standby for the terminal. The specific operation process is as follows: S2601, Stable connection window end-of-term reserved trigger synchronization status receipt; The terminal predicts the termination time based on the current SCW and reserves CT seconds before the window ends to trigger the cleanup process. CT is a configurable system parameter with a default value of 5 seconds. By pre-judging the window failure node with a fixed reserved time, it avoids data reporting failures caused by upcoming cell handover or link instability, ensuring that the cleanup action is reliably completed within a stable link.
[0091] S2602. Construct and report a lightweight synchronization state summary packet; Within the reserved window period, the terminal assembles a lightweight status summary packet in standard JSON format. This packet is used to send complete operational information for this SCW synchronization back to the server, providing data support for the server to verify synchronization integrity and iterate cloud scheduling strategies. The core content of the summary packet includes the start time, end time, total amount of successfully uploaded data, and the terminal's network status and unique identifier of the cell during the synchronization period. Specifically, it covers: the start and end timestamps of this synchronization (UTC), the number of successfully uploaded tasks, the current cell ID, and the RSRP signal quality range during the synchronization period, enabling traceability, verifiability, and optimization of a single synchronization action.
[0092] S2603. Orderly release of network and system resources; After the terminal completes the upload of the status summary packet, it immediately performs standardized resource release operations, sequentially shutting down all active network and background scheduling resources: closing all established Socket communication connections and clearing the transmission session context; terminating the timer task and background scheduling process corresponding to this synchronization; actively requesting the terminal to enter deep sleep mode through the system API; and cutting off the power supply to the cellular radio frequency module to completely eliminate redundant background operation and radio frequency idle consumption.
[0093] S2604, enters low-power standby mode and waits for the next trigger.
[0094] After completing all resource shutdown operations, the terminal exits the current SCW synchronization process, enters an ultra-low power standby state, stops all active network behavior, and silently waits for the wake-up trigger signal of the next stable connection window, realizing the energy-saving scheduling logic of wake-up on demand and shutdown after use.
[0095] This embodiment of the disclosure ensures that a complete operational log is retained for each data synchronization through a window-end reservation receipt mechanism, facilitating cloud-based verification of synchronization integrity and continuous optimization of scheduling strategies. Simultaneously, through a standardized and orderly resource release and RF power-down sleep mechanism, it precisely controls the terminal's operating status after synchronization, completely resolving the issues of background persistence, RF idle running, and unnecessary power consumption after synchronization in traditional solutions. This step, in conjunction with the aforementioned pre-wake-up, hierarchical transmission, and adaptive packet control mechanisms, constructs a complete end-side time-selective synchronization process, further improving the solution's reliability, traceability, and low-power performance of the terminal.
[0096] S207. If a stable connection window is not entered, synchronization will be temporarily suspended, and change records will only be accumulated in the background without initiating network requests.
[0097] This step, through a silent strategy, eliminates invalid attempts on low-quality links, significantly reducing air interface resource consumption and terminal power consumption.
[0098] For cell access events that do not meet the SCW conditions (i.e.) The terminal strictly implements the "silent policy," does not perform any data interaction operations, and only appends all data to be uploaded to the local incremental log file, retaining metadata (timestamp, type, priority) for subsequent sorting processing.
[0099] When a terminal determines that the average historical dwell time of a cell has not reached a preset threshold and a stable connection window cannot be formed, it immediately executes a silent strategy, ceasing all network connection and data synchronization requests. This avoids invalid transmissions and retransmissions in scenarios with weak networks, short dwell times, and frequent handovers, effectively saving air interface resources and reducing terminal power consumption. Simultaneously, the terminal locally appends records all data changes generated by the service, storing them losslessly as incremental logs, and retaining metadata such as timestamps, data types, and task priorities to ensure data integrity and order. After a stable SCW window is subsequently connected, hierarchical batch synchronization can be quickly completed based on the retained logs. This achieves differentiated scheduling, enabling opportunistic synchronization during stable windows and silent data accumulation during unstable windows. It solves the problems of poor reliability and high energy consumption associated with traditional indiscriminate synchronization, balancing data synchronization integrity with low-power terminal performance.
[0100] The above description is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements, optimizations and modifications can be made without departing from the principle of the present invention, and these should also be considered within the scope of protection of the present invention.
[0101] Figure 3 This is a schematic diagram of the structure of a cross-network resource scheduling device provided in an embodiment of the present disclosure, as shown below. Figure 3 As shown, the device 300 includes: The acquisition module 301 is configured to obtain the predicted stable dwell time of the terminal in the current cell based on the dwell behavior pattern of the terminal entering the current cell. Window module 302 is configured to identify a stable connection window based on the predicted stable dwell time, and to centrally perform resource-intensive data synchronization operations during the stable connection window period.
[0102] In some embodiments, the device 300 further includes: The module is configured to learn the patterns of stationary behavior offline based on the collected network stationary feature data and terminal mobility feature data, and to build and dynamically update the local base station behavior database.
[0103] In some embodiments, the acquisition module 301 includes: The query module is configured to query the local base station behavior database to see if there is a corresponding historical camping record based on the obtained unique identifier of the current cell. The first prediction module is configured to retrieve data from the local base station behavior database if a corresponding historical record exists. Historical average length of stay is used as a predictor of stable length of stay; The second prediction module is configured to, if no corresponding historical dwell record exists, obtain neighboring cells stored in the local base station behavior database based on the current cell's geographical location. Historical average dwell time is used as a predictor of stable dwell time.
[0104] In some embodiments, the window module 302 is further configured such that when the predicted stable dwell time exceeds a preset stable dwell threshold, the current dwell time of the terminal constitutes a stable connection window, and resource-intensive data synchronization operations are centrally executed within the stable connection window period. The start time of the stable connection window period is... = , The current moment is the termination time of the stable connection window. For the current moment The result is obtained by summing the predicted stable dwell time.
[0105] In some embodiments, the window module 302 includes: The wake-up module is configured to start the wake-up synchronization service process in advance based on the start time of the stable connection window. The sorting module is configured to read the local data change log to be synchronized when it is detected that the data is within a stable connection window, and sort all synchronization tasks according to priority. The data packet module is configured to dynamically adapt the data packet size based on the duration of a stable connection window. The upload module is configured to upload locally cached data in batches within the valid time interval of a stable connection window, according to a sorted priority queue and with adaptively adjusted data packet size.
[0106] In some embodiments, the window module 302 further includes: The shutdown module is configured to disable state synchronization before the stable connection window ends, send a synchronization state summary packet based on the reserved time, close the network connection and enter hibernation.
[0107] In some embodiments, the device 300 further includes: The hibernation module is configured to not initiate a network request if a stable connection window cannot be identified based on the predicted stable dwell time.
[0108] The specific implementation process of the functions and roles of each module in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.
[0109] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure, such as... Figure 4As shown, this disclosure also provides an electronic device 400, which includes at least one processor 401 and a memory 402 coupled to the processor 401. The memory 402 is used to store at least one processor 401 executable instructions, wherein the at least one processor 401 is used to execute the instructions to implement the steps of the method described above in this disclosure.
[0110] The processor 401 described above can also be referred to as a Central Processing Unit (CPU), which can be an integrated circuit chip with signal processing capabilities. Each step in the method described in this embodiment can be implemented by the integrated logic circuitry in the hardware of the processor 401 or by instructions in software form. The processor 401 described above can be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method in conjunction with this embodiment can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can be located in the memory 402, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor 401 reads information from the memory 402 and, in conjunction with its hardware, completes the steps of the method described above.
[0111] Figure 5 This is a schematic diagram of an exemplary computer system provided by an embodiment of the present disclosure. Various operations / processes according to embodiments of the present disclosure, implemented via software and / or firmware, can be transmitted from a storage medium or network to a computer system with a dedicated hardware architecture, for example... Figure 5 The computer system 500 shown is equipped with the programs that constitute the software. When various programs are installed, the computer system is able to perform various functions, including those described above.
[0112] Computer system 500 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of this disclosure described and / or claimed herein.
[0113] like Figure 5 As shown, the computer system 500 includes a computing unit 501, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the computer system 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.
[0114] Multiple components in the computer system 500 are connected to the I / O interface 505, including: an input unit 506, an output unit 507, a storage unit 508, and a communication unit 509. The input unit 506 can be any type of device capable of inputting information into the computer system 500. The input unit 506 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 507 can be any type of device capable of presenting information and may include, but is not limited to, a monitor, speaker, video / audio output terminal, vibrator, and / or printer. The storage unit 508 may include, but is not limited to, a hard disk and an optical disk. The communication unit 509 allows the computer system 500 to exchange information / data with other devices via a network such as the Internet, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth™ devices, Wi-Fi devices, WiMax devices, cellular communication devices, and / or the like.
[0115] The computing unit 501 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running analytical algorithm models, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above. For example, in some embodiments, the methods described above in the embodiments of this disclosure can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed on an electronic device via ROM 502 and / or communication unit 509. In some embodiments, the computing unit 501 can be configured to perform the methods described above in the embodiments of this disclosure by any other suitable means (e.g., by means of firmware).
[0116] This disclosure provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the methods described in this disclosure.
[0117] Computer-readable storage media can be volatile memory, such as random-access memory (RAM); or non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); or devices that include one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.
[0118] It should be noted that the computer-readable storage medium described in this disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), or any suitable combination thereof.
[0119] Embodiments of this disclosure provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for constructing and updating a knowledge graph.
[0120] In embodiments of this disclosure, computer program code for performing the operations of this disclosure can be written in one or more programming languages or a combination thereof. These programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, and C++, as well as conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on a computer, partially on a computer, as a standalone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0121] The modules, components, or units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules, components, or units do not necessarily constitute a limitation on the module, component, or unit itself.
[0122] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary hardware logic components that can be used include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), etc.
[0123] It should be noted that, in this document, terms such as "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 a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0124] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.
Claims
1. A cross-network resource scheduling method, characterized in that, Applied to a terminal, the method includes: Based on the terminal's dwell behavior pattern when entering the current cell, the predicted stable dwell time of the terminal when entering the current cell is obtained; Based on the predicted stable dwell time, a stable connection window is identified, and resource-intensive data synchronization operations are performed centrally during the stable connection window period.
2. The method according to claim 1, characterized in that, Before obtaining the predicted stable dwell time of the terminal in the current cell based on the dwell behavior pattern of the terminal entering the current cell, the method further includes: Based on the collected network dwell time feature data and terminal mobility feature data, the dwell time behavior patterns are learned offline, and a local base station behavior database is constructed and dynamically updated iteratively.
3. The method according to claim 2, characterized in that, The step of obtaining the predicted stable dwell time of the terminal in the current cell based on the dwell behavior pattern of the terminal entering the current cell includes: Based on the unique identifier of the current cell, query the local base station behavior database to see if there is a corresponding historical camping record; If a corresponding historical record exists, then retrieve the data stored in the local base station behavior database. Historical average length of stay is used as a predictor of stable length of stay; If no corresponding historical residency record exists, then the neighboring cells stored in the local base station behavior database are retrieved based on the current cell's geographical location. Historical average dwell time is used as a predictor of stable dwell time.
4. The method according to any one of claims 1-3, characterized in that, The step of identifying a stable connection window based on the predicted stable dwell time, and then centrally performing resource-intensive data synchronization operations within the stable connection window period, includes: When the predicted stable dwell time exceeds a preset stable dwell threshold, the current dwell time of the terminal constitutes a stable connection window. During the stable connection window period, resource-intensive data synchronization operations are performed centrally. The start time of the stable connection window period is... = , The current moment is the termination time of the stable connection window. For the current moment The result is obtained by summing the predicted stable dwell time.
5. The method according to claim 4, characterized in that, The centralized execution of resource-intensive data synchronization operations during a stable connection window includes: Based on the start time of the stable connection window, the wake-up synchronization service process is started in advance at a preset time. When a stable connection window is detected, the local data change log to be synchronized is read, and all synchronization tasks are sorted by priority. Dynamically adapt data packet size based on the duration of a stable connection window; Within the valid time interval of a stable connection window, local cached data is uploaded in batches according to the sorted priority queue and the adaptively adjusted data packet size.
6. The method according to claim 5, characterized in that, The method of centrally performing resource-intensive data synchronization operations within a stable connection window also includes: Before the stable connection window ends, state synchronization is turned off, a state summary packet for this synchronization is sent based on the reserved time, and the network connection is closed and the system enters sleep mode.
7. The method according to claim 1, characterized in that, The method further includes: If a stable connection window cannot be identified based on the predicted stable dwell time, no network request will be initiated.
8. A cross-network resource scheduling device, applied to a terminal, characterized in that, The device includes: The acquisition module is configured to obtain the predicted stable dwell time of the terminal in the current cell based on the dwell behavior pattern of the terminal entering the current cell; The window module is configured to identify a stable connection window based on the predicted stable dwell time, and to centrally perform resource-intensive data synchronization operations during the stable connection window period.
9. An electronic device, characterized in that, include: At least one processor; Memory for storing the at least one processor-executable instruction; The at least one processor is configured to execute the instructions to implement the method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the method as described in any one of claims 1-7.
11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-7.
