A method and system for WiFi non-sensing detection of a cellular internet of things DRX cycle adaptation

Abstract

The application relates to the technical field of Internet of Things, in particular to a WiFi non-inductive detection method and system for cellular Internet of Things DRX cycle adaptation; the method comprises the following steps: constructing a continuous time modeling cellular monitoring function and a radio frequency priority model by analyzing DRX cycle parameters of an Internet of Things terminal, identifying a safe time window for radio frequency switching, and generating a scheduling decision in combination with hardware switching overhead; a dynamic safe window prediction mechanism is designed, a WiFi scanning window sequence is generated through weighted history correction and multi-cycle sliding prediction; a complete WiFi scanning task is split into multiple micro scanning fragments, executable fragments are dynamically matched according to the length of each safe window, and the scanning task is incrementally completed across multiple DRX cycles; and through an asynchronous cache and fragment fusion mechanism, scanning results are reliably integrated into a complete scanning view. The application effectively guarantees the cooperative scheduling of terminal cellular radio frequency resources and the reliable execution of a WiFi scanning task.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, specifically to a WiFi-free detection method and system for DRX periodic adaptation in cellular IoT. Background Technology

[0002] In narrowband IoT applications, terminal devices typically need to have both cellular communication capabilities and WiFi environment awareness to meet diverse business needs such as location services, intelligent operation and maintenance, and seamless connectivity. To achieve long battery life, narrowband IoT terminals widely adopt cellular discontinuous reception mechanisms, which reduce power consumption by periodically listening to the paging channel, allowing the device to remain in sleep mode most of the time. WiFi network detection, as an important means of obtaining information about the surrounding environment, requires the terminal to periodically scan multiple channels to discover available access points and collect signal characteristics.

[0003] Chinese invention patent application CN120434750A discloses a low-power communication method and system based on a 5G RedCap portable WiFi network, including: adaptive control of DRX cycle based on WiFi access traffic prediction, setting wake-up control for the RedCap communication terminal module, setting adaptive switching for bandwidth, and setting switching operation for cellular communication standard; by introducing a DRX cycle adaptive control mechanism based on WiFi access traffic prediction and an uplink-downlink linkage intelligent wake-up strategy into the 5G RedCap communication terminal module, the DRX sleep and wake-up cycle can be dynamically adjusted according to the actual service load status to complete energy-saving management on the cellular side.

[0004] However, in order to save material costs and PCB space, many dual-mode terminals adopt a single-RF front-end single-antenna architecture (i.e., cellular communication and WiFi detection share a single RF transceiver link and antenna) in order to save material costs and PCB space under the design trend of low cost and low power consumption of IoT terminals. Under this architecture, RF resources can only serve one mode of cellular or WiFi at any given time, and the two have a strict time-division mutual exclusion relationship at the physical layer. If the switch to WiFi band is forcibly made during cellular listening, it will directly lead to the loss of paging messages, wireless link failure, or even network disconnection. Conversely, if WiFi scanning is not performed for a long time due to waiting for cellular idle, the environmental sensing function will fail.

[0005] Furthermore, with the increasing complexity of IoT business scenarios, terminals have placed higher demands on the reliability of cellular communication and the real-time performance of WiFi sensing. Efficient coordination between cellular and WiFi functions has become a key direction for multi-mode terminal design. How to achieve deep integration of WiFi scanning tasks and cellular listening behavior under the inherent non-continuous reception timing constraints of narrowband IoT terminals, so as to ensure the stable online status of cellular links and achieve continuous detection of WiFi channels, is an important issue that needs to be addressed in the technological evolution of this field. Summary of the Invention

[0006] The purpose of this invention is to address the problems existing in the background technology by proposing a WiFi non-contact detection method and system for DRX period adaptation in cellular IoT.

[0007] The technical solution of this invention: a WiFi non-intrusive detection method for DRX periodic adaptation in cellular IoT, comprising the following specific implementation steps: S1. By analyzing the cellular discontinuous reception cycle parameters of IoT terminals, a cellular listening function and radio frequency priority model on a continuous time axis are constructed, a continuous safe time window for radio frequency switching is identified, and radio frequency collaborative scheduling decisions are generated in combination with radio frequency hardware switching overhead. S2. Based on the security time window and cellular monitoring status, a dynamic security window prediction mechanism is designed to generate a security window sequence that can perform WiFi scanning through weighted historical correction and multi-cycle sliding prediction. S3. The complete WiFi scanning task is divided into several micro scanning segments. The executable scanning segments are dynamically matched according to the actual length of each security window, and the scanning task is completed incrementally across several non-continuous reception cycles. S4. Through asynchronous caching and fragment fusion mechanism, the scan fragment results completed in several cycles are deduplicated, weighted and merged and timestamped to generate a complete WiFi scan view.

[0008] Preferably, in step S1, the specific process of constructing the cellular listening function and radio frequency priority model on a continuous time axis is as follows: The system analyzes the length of the discontinuous reception cycle, the paging monitoring start offset, and the active monitoring duration of the terminal's current cellular communication. It then maps the discrete protocol configuration parameters to a cellular monitoring state function on a continuous time axis. During each discontinuous reception cycle, the function takes a value of 1 between the monitoring start time and the monitoring end time to indicate that the cellular monitoring state must be maintained, and takes a value of 0 at other times to indicate that it is in the cellular monitoring idle segment. A radio frequency resource occupancy priority function is established based on the weight coefficients of cellular communication tasks and WiFi scanning tasks, as well as the task status factor. By comparing the numerical values ​​of cellular communication priority and WiFi scanning priority, it is determined whether radio frequency resources should be allocated to cellular communication or WiFi scanning at any given time, thereby achieving radio frequency resource competition and coordination under the condition of prioritizing cellular communication.

[0009] Preferably, the identification process for continuous safety time windows is as follows: By combining the cellular listening function and the radio frequency priority function, it is determined whether WiFi scanning is allowed at the current time. When cellular communication does not require listening but WiFi scanning is required, a continuously available radio frequency switching time segment is identified, and the start and end times of the segment and the duration between them are calculated. The time overhead required for the radio frequency hardware to switch between the cellular band and the WiFi band is obtained. This time overhead includes the time to switch from the cellular band to the WiFi band and the time to recover from the WiFi band to the cellular band. The actual executable scan time is obtained by subtracting the above-mentioned switching overhead from the available radio frequency window length. The actual executable scan time is compared with the minimum scan execution time threshold. When the actual executable scan time is greater than or equal to the minimum scan execution time threshold, an RF switching scheduling command is generated, which enables the RF front end to switch to the WiFi band to perform the detection task within the safe window and resume cellular listening before the window ends.

[0010] Preferably, in step S2, the specific implementation of the dynamic safety window prediction mechanism is as follows: The time interval within each discontinuous reception cycle that meets the following three conditions is defined as the initial security window: the cellular link is in a non-monitoring state, the radio frequency priority allows the WiFi scanning task to be executed, and the window length is greater than the minimum execution time of a single scan. The length of this window is then calculated. A weighted historical correction mechanism is adopted to combine the current window length with the average actual available time in the same historical period. By adjusting the weights to balance the importance of the current measurement and historical experience, the corrected safe window length is obtained. The safe window for several future discontinuous reception periods is recursively predicted based on the corrected window length, and a confidence index is calculated for each predicted window to characterize the prediction reliability. This confidence index is determined based on the degree of deviation between the historical window length and the predicted window length. The window intervals and confidence levels of several future periods are integrated to form a safe window sequence. The legality of the predicted window is verified by combining the radio frequency switching overhead and the minimum scan time threshold, and scheduling interface information including the window start time, window end time and actual executable scan time is generated.

[0011] Preferably, in step S3, the specific method for dividing the complete WiFi scanning task into several micro-scanning segments is as follows: The set of WiFi channels currently supported by the terminal is structurally modeled. This set of channels is generated by the terminal's WiFi protocol stack based on the regional wireless spectrum specifications and the chip's support capabilities. The complete scanning task is abstracted into a set of tasks consisting of several channel scanning subtasks. Each channel scanning subtask needs to complete two operations: switching the radio frequency to the target channel and listening to beacon frames or probing response frames on the target channel. The complete scanning task is divided into several independently executable micro-scanning slices. Each scanning slice contains one or more consecutive channel scanning subtasks. Standardized description parameters are established for each slice, including the sequence of scanning subtasks within the slice and the total execution time of the slice, so that the scanning task can be scheduled across cycles in the form of micro-scanning units.

[0012] Preferably, in step S3, the specific process of dynamically matching executable scan slices is as follows: Based on the predicted actual executable scan time of the safety window, an executable scan segment evaluation mechanism is established. The total execution time of each scan segment is matched and judged with the current safety window length. Only scan segments whose total execution time is strictly less than the actual executable scan time of the current safety window are allowed to enter the scheduling candidate queue. Based on the terminal's historical scanning records, known WiFi channel distribution, and regional spectrum usage statistics, each channel is assigned an environmental probability weight, which reflects the historical number of times an access point has been detected on the corresponding channel. By combining channel environment weights to optimize the channel order in scanning slices, channels with higher channel environment probability weights are scanned first, thereby improving WiFi detection efficiency under limited security window resources.

[0013] Preferably, in step S3, the specific implementation method for completing the scanning task incrementally across cycles is as follows: The session state table is scanned to record the completed scanned fragments and fragments to be executed, and a suitable fragment is selected from the queue to be executed when a new security window arrives. For each slice to be executed, a window matching priority is calculated. This priority is determined based on the degree of matching between the slice execution time and the window length, as well as the environmental probability weight of the channel contained in the slice. The slices are selected for execution within the current safe window based on the matching priority from high to low, and the scan progress pointer is updated after the slice execution is completed. When all channel scans are completed or the cumulative execution time of a scan session exceeds the maximum allowed scan session time, the current scan task is terminated and a new scan task is restarted to ensure the real-time nature of the scan results.

[0014] Preferably, in step S4, the specific implementation of the asynchronous caching and sharding fusion mechanism is as follows: After each scan segment is completed, the observed access point information is written to the segment-level buffer. This information includes the access point's basic service set identifier, service set identifier, received signal strength, scan timestamp, and the confidence level of the security window in which the current scan segment is located. The hierarchical storage structure includes temporary shard cache and persistent session cache. Temporary shard cache is used to store the current shard scan results to support fast fusion and asynchronous updates. Persistent session cache is organized by scan session and indexes each shard result to support cross-cycle fusion and historical data comparison. The scanning results within each segment are deduplicated. When the same access point appears multiple times in the same segment, the observation record with the strongest signal strength or the latest timestamp is retained. The signal strength of the repeated access points is initially fused by window confidence weighted averaging. At the same time, the first and last occurrence times of the access point are recorded.

[0015] Preferably, in step S4, the specific process of generating a complete WiFi scan view is as follows: After all scan slices have been initially fused, all slice results are aggregated to form a complete scan session result set; For access points that repeatedly appear across segments, a weighted fusion calculation is performed again to determine the signal quality score. This score is based on the average signal strength, frequency of occurrence, and cumulative observation time of the access points. The integrated information for each access point includes the basic service set identifier, service set identifier, weighted fused signal strength value, signal quality score, first occurrence time, last occurrence time, and cumulative observation time. The merged sharding results are passed to the upper-layer application through an asynchronous interface, and a final scan report containing a complete list of access points and the above comprehensive information is generated at the end of the scan session. When a scan segment is not completed due to window preemption or interruption, the incomplete segment number is automatically recorded, and the segment is executed first in the next available safe window to achieve incremental compensation scanning and ensure the integrity of the final scan session.

[0016] The technical solution of this invention: A WiFi non-intrusive detection system with DRX periodic adaptation for cellular IoT, comprising: Memory; processor; The memory stores computer programs; When the processor executes the computer program, it implements the WiFi non-contact detection method for DRX periodic adaptation in cellular IoT as described in any one of claims 1 to 9.

[0017] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects: This invention designs a WiFi seamless detection method and system for cellular IoT DRX cycle adaptation. Under the premise that cellular and WiFi share a single radio frequency, it creates an innovative collaborative scheduling and control system method, realizing the WiFi seamless detection capability of narrowband IoT terminals while maintaining cellular connectivity. By accurately analyzing the cellular DRX timing sequence to establish a radio frequency priority scheduling mechanism, it ensures that WiFi scanning tasks are executed only within a safe window when cellular communication is idle, eliminating cellular link interruptions or paging loss caused by radio frequency switching, and guaranteeing the continuity and reliability of critical services of narrowband IoT terminals. Furthermore, by employing a dynamic safe window prediction and segmented scanning execution strategy, the complete WiFi channel detection task is intelligently decomposed into multiple micro-scanning units. It adaptively matches the fragmented available time window within each DRX cycle, enabling the scanning process to be completed progressively across cycles, significantly improving the terminal's utilization efficiency of scarce idle time slots. Through asynchronous caching and fragmented fusion mechanisms, it deduplicates, weights, and integrates timestamps of the scattered observation access point information, presenting it to upper-layer applications as a continuous and complete scanning session view, achieving complete transparency of the complex scheduling at the lower layer to upper-layer applications. This invention effectively controls overall power consumption while achieving continuous environmental awareness, extending the battery life of narrowband IoT devices. This invention has high engineering adaptability and is compatible with existing cellular and WiFi protocol stacks, providing reliable technical support for the deep integration of narrowband IoT in fields such as smart cities and asset tracking. Attached Figure Description

[0018] Figure 1 This is a flowchart of a WiFi non-contact detection method for DRX periodic adaptation in cellular IoT proposed in this invention. Detailed Implementation

[0019] Example 1, as Figure 1 As shown, the present invention proposes a WiFi non-intrusive detection method for DRX periodic adaptation in cellular IoT, which includes the following specific implementation steps: S1. Construct a basic mechanism for radio frequency (RF) collaborative scheduling under the condition of maintaining cellular connectivity. By performing continuous-time modeling of the cellular DRX (Discontinuous Reception) listening timing, establish the RF priority relationship between cellular communication and WiFi scanning. Based on this, identify continuous time windows available for RF handover. Simultaneously, combine the RF hardware handover overhead to generate the final scheduling decision, enabling the terminal to temporarily switch to the WiFi band to perform probing tasks within the time segment allowed by the cellular protocol, while maintaining the cellular connectivity state at the logical layer without interruption. This provides a stable RF resource scheduling foundation for subsequent security window prediction and scan segmentation scheduling. The specific implementation process is as follows: S11. By parsing the protocol configurations such as the DRX cycle parameters, paging monitoring start offset, and active monitoring duration of the terminal's current cellular communication, the monitoring timing parameters that originally existed discretely in the protocol stack are mapped to cellular monitoring state functions on a continuous time axis. This enables the system to determine at any time whether the terminal is in a time segment where it must maintain cellular monitoring, and to establish a complete cellular monitoring time model accordingly, providing an accurate time basis for subsequent radio frequency resource scheduling and window identification. Specifically: After the terminal completes cellular network attachment and enters the RRC connected state or idle state, the protocol stack first parses the current cellular communication listening behavior; Specifically, the MAC layer reads the DRX parameters of the network configuration from the RRC layer, including information such as the DRX period length, active listening duration, and paging listening start offset; These discrete parameters are converted into a listening function on a continuous time axis. A continuous time axis is maintained locally on the terminal, and the cellular listening function is constructed with the DRX period as the basic unit. ; in, This represents the cellular monitoring state function of the terminal at time t. When the value is 1, it means that the terminal must maintain the cellular monitoring state and is not allowed to leave the cellular frequency band. When the value is 0, it means that the terminal is in the cellular monitoring idle segment. This indicates the start time of paging monitoring within the current DRX cycle. This parameter originates from the paging timing configuration issued by the cellular network during the RRC configuration phase. This indicates the DRX cycle length, which is configured by the base station according to the terminal's power-saving strategy. For example, in NB-IoT (Narrowband Internet of Things) or LTE-M systems, it is typically from hundreds of milliseconds to several seconds. This parameter represents the duration for which the terminal must maintain active listening in each DRX cycle. It corresponds to OnDurationTimer and is configured by the network side. k represents the DRX cycle number, used to characterize the kth DRX cycle, and its value is a non-negative integer. It should be noted that RRC is the Radio Resource Control layer in the cellular communication protocol stack, responsible for connection management, radio resource configuration, and mobility control between the terminal and the base station. In narrowband IoT systems, the RRC layer mainly maintains two states: in the RRC connected state, the terminal maintains a dedicated connection with the base station, enabling bidirectional data transmission; in the RRC idle state, the terminal only periodically listens to the paging channel to receive downlink data notifications. The RRC layer configures DRX parameters to the MAC layer, including key information such as DRX period length, active listening duration OnDurationTimer, and paging timing offset. These parameters together determine when the terminal must maintain cellular listening and when it can enter sleep mode. It should be noted that the MAC layer, or Media Access Control layer, is located in the lower half of the data link layer of the protocol stack. In cellular communication, it is responsible for resource scheduling, timing control, and physical layer access management. S12. Based on the cellular listening time model, establish a radio frequency resource occupancy priority function. By setting the weight coefficients of cellular communication tasks and WiFi scanning tasks, as well as task status factors, different wireless tasks are uniformly mapped to the same priority calculation model. Priority comparison is used to determine whether radio frequency resources should be preferentially allocated to cellular communication or WiFi scanning at any given time, thereby realizing a radio frequency resource competition and coordination mechanism under the condition of prioritizing cellular communication. Specifically: After obtaining the cellular listening function, a radio frequency resource occupancy priority model is further established to solve the resource competition problem between cellular communication and WiFi detection, and different wireless tasks are uniformly mapped to the priority function for comparison. Define RF task priorities: ; The state factors of the WiFi scanning task are determined by the listening function: ; The value is 1 when the scan task is in a waiting state, and 0 otherwise. At any given time, the allocation of radio frequency resources is determined by comparing priority functions: ; If this condition is met, radio frequency must be allocated to cellular communication; WiFi scanning can only obtain radio frequency resources when the priority of cellular communication is 0. S13. The cellular listening function and the radio frequency priority function are jointly used to determine whether WiFi scanning is permitted at the current time. If cellular communication does not require listening and WiFi scanning is required, continuously available radio frequency switching time segments are identified. The duration of these time segments is calculated to form a safe time window that can support temporary radio frequency switching, providing clear time constraints for subsequent WiFi scanning task scheduling. Specifically: After the priority function is established, the time segment in which WiFi scanning can be performed is further identified; Radio frequency (RF) is only allowed to switch to the WiFi band when cellular communication does not require eavesdropping and WiFi scanning is needed. Therefore, an RF availability determination function is defined: ; Calculate the length of the continuous available segment, assuming the start time of the current available window is... End time Then the window length is: ; in, Indicates whether radio frequency switching is allowed at time t. The time indicates that the current time is within the available range, and a WiFi scan can be performed. This indicates that the radio frequency must maintain cellular communication. Indicates the length of the continuously available radio frequency window; It should be noted that, through the above calculations, the maximum time segment during which radio frequency can leave the cellular band within each DRX cycle can be obtained. The available window can be accurately calculated through protocol timing prediction, so that WiFi scanning is strictly constrained by the rhythm of cellular communication. S14. After identifying the available radio frequency window, further consider the time overhead required for the radio frequency hardware to switch between the cellular band and the WiFi band. By calculating the executable scan time and comparing it with the minimum scan execution time threshold, a radio frequency switching scheduling command is generated when the execution conditions are met. This allows the radio frequency front-end to switch to the WiFi band to perform the detection task within the safe window and resume cellular listening before the window ends, thus forming a complete radio frequency collaborative scheduling decision mechanism. Specifically: After obtaining an available RF window, further consideration is given to the RF hardware switching overhead to determine whether to perform a WiFi scan; Since the RF front-end requires a certain amount of time to switch between cellular and WiFi bands, an RF handover compensation model is introduced: Calculate the actual executable scan time: ; Set the minimum scan execution time threshold: ; When the conditions are met, the radio frequency scheduler sends a switching command to the radio frequency control layer, causing the radio frequency to switch to the WiFi band to perform scanning at the beginning of the window; This indicates the time required to perform a minimum scan operation, such as the time required to scan a single WiFi channel. This parameter is obtained from WiFi chip scanning performance statistics. When the scan is complete or the window is about to close, the radio frequency immediately returns to the cellular band to continue listening; in, This indicates the effective execution time available for WiFi scanning within the current window; This indicates the radio frequency band switching time, including the switching overhead from cellular to WiFi and WiFi back to cellular. This parameter is derived from the hardware characteristics of the terminal's radio frequency chip.

[0020] S2. Based on the available radio frequency window and cellular listening status output in step S1, a dynamic security window prediction method is designed. Through continuous time modeling, historical correction, multi-cycle sliding prediction, and legality verification, an executable WiFi scanning time period is generated to achieve seamless, predictable, and reliable scheduling of WiFi scanning under the DRX cycle, while ensuring cellular connection continuity. The specific implementation process is as follows: S21. Using the continuously available radio frequency window and cellular listening function output in step S1, define the time period during which WiFi scanning can be performed within each DRX cycle as the initial security window, and calculate the window length to provide a basis for subsequent dynamic adjustments, specifically: Define a safe window (SW) for a time period that meets the following conditions: Condition 1: The cellular link is in a non-monitoring state (i.e.) ); Condition 2: Radio frequency priority allows WiFi scanning tasks to execute (i.e.) ); Condition 3: The window length is greater than the minimum execution time for a single scan. ; Specifically, it is expressed as follows: ; in, This represents the initial safety window for the k-th DRX cycle; This indicates the window length, which is the maximum duration during which WiFi scanning is allowed. Indicates the start time of the window, determined by the starting point of the available radio frequency window in step S1. supply; This indicates the end time of the window and the corresponding start time of the next cellular monitoring session; S22. Combining the initial window length of the current period with the statistical average of similar historical windows, the window length is dynamically adjusted through weighted correction to adapt to network scheduling fluctuations and changes in terminal status, ensuring that WiFi scanning is performed within the safe window, i.e.: A weighted historical correction mechanism is adopted, which combines the current window length with historical window statistics: ; ; in, Indicates the corrected safety window length; This represents the average actual available time for the same historical period; M represents the number of historical periods. This indicates the adjustment weights, used to balance the importance of current measurements with historical experience. ; This represents the actual available safety window length in the i-th historical DRX cycle, which is obtained by recording the actual executable scan time during system operation. It should be noted that, considering the impact of unforeseen events, such as sudden paging or emergency cellular dispatch, if it is found that the current window may be occupied beyond the threshold, the window length will be shortened immediately to ensure that scanning will never affect cellular monitoring. S23. Based on the corrected window length, recursively predict the safe windows for several future DRX cycles, and calculate the confidence level for each window to form a safe window sequence that can be scheduled across cycles, providing a planning basis for WiFi scanning slice execution, specifically: The sliding prediction mechanism predicts safe windows for several future periods and generates a sequence of consecutive windows for the scheduler to reference. The process is as follows: Forecast window: ; Confidence assessment is performed, and window confidence is introduced. Characterize the reliability of predictions: ; Generate a window sequence, integrating the windows and confidence levels for the next N periods to form a safe window sequence: ; Where N represents the number of DRX cycles predicted in the future; This represents the prediction safety window interval for the (k+i)th DRX period; This represents a prediction safety window object, which contains three pieces of information: window start time, window end time, and window prediction confidence. This represents the confidence index for the (k+i)th prediction safety window; S24. Combining RF switching overhead and minimum scan time threshold, the validity of the prediction window is verified, and a WiFi scan scheduling interface is generated. The executable window and execution time are passed to the scheduling module to ensure that the scan task is safe and can be executed. Specifically: Determine execution time, taking into account radio frequency switching overhead. Calculate the actual executable time: ; Perform a validity check to determine if the minimum scan time requirement is met. : Valid window conditions: ; Generate a scheduling interface to generate an internal scheduling interface for the protocol stack for valid windows: ; in, This represents the actual execution time available for WiFi scanning in the (k+i)th security window; This indicates the radio frequency band switching time, which includes the time to switch from the cellular band to the WiFi band and the time to return from the WiFi band to the cellular band. This parameter is determined by the hardware characteristics of the terminal's radio frequency chip.

[0021] S3. Construct a WiFi scanning task fragmentation and cross-cycle incremental execution mechanism for cellular DRX security windows. The complete WiFi scanning process is decomposed into multiple independently executable micro-scanning fragments, and the executable fragments are dynamically selected based on the actual length of each security window, allowing the scanning process to be completed incrementally cycle by cycle. Simultaneously, by combining execution time prediction, channel priority evaluation, and historical scan status records, the continuity and adaptability of scan scheduling are achieved. The specific implementation process is as follows: S31. Construct a WiFi scanning task fragmentation model, dividing the traditional full-channel scanning task into multiple independent scanning fragments. Each fragment contains only a single or a small number of WiFi channels, and establishes standardized execution time description parameters for each fragment. This allows the scanning task to be scheduled in the form of micro-scanning units, thereby enabling the execution time of a single scan to accurately match the length of the cellular DRX security window. Specifically: When the terminal needs to detect WiFi networks, the WiFi scanning task is structured and modeled, that is, the complete scanning task is abstracted into a task set consisting of multiple channel scanning sub-tasks; Define the set of WiFi channels that the terminal currently supports scanning as follows: ; For each channel, the system needs to perform the following two operations: switch the radio frequency to the target channel and listen for Beacon or Probe Response frames on that channel; Therefore, the scan time for each channel can be expressed as: ; The complete scan task execution time is: ; Due to the execution time of a full scan task Much larger than the length of a single DRX security window, therefore the scan task was broken down into multiple independently executable subtasks: ; To further improve scanning efficiency, environmental priority analysis is performed on the channels. Based on the terminal's historical scanning records, known WiFi channel distribution, and regional spectrum usage statistics, an environmental probability weight is assigned to each channel: ; Where C represents the WiFi channel set, which describes all the WiFi channels that the terminal needs to detect during a complete scan. This set is generated by the terminal's WiFi protocol stack according to the wireless spectrum specifications of the current region and the chip's support capabilities. For example, in the 2.4GHz band, it usually contains channels 1 to 13, while in the 5GHz band, multiple scannable channels are generated according to the regulatory domain configuration. This represents the j-th specific channel in the WiFi channel set, which is the wireless channel that the terminal needs to switch to and listen to one by one during the scanning process; This represents the total number of WiFi channels that need to be scanned, which is the number of elements in set C; Indicates that the terminal is in the channel The time required to perform one scan operation; Indicates that the radio frequency switches from the current channel to the channel. Time required; This indicates the minimum time required to listen for Beacon frames on this channel; This represents the total time required to perform a complete WiFi scan, which is the sum of the scan times for all channels. Indicates the channel during historical scans The number of times the access point was detected; S32. Based on the security window length information predicted in step S2, a scan segment executability evaluation mechanism is established. The expected execution time of each scan segment is matched with the current security window length. Only scan segments whose execution time is strictly less than the security window length are allowed to enter the scheduling candidate queue, ensuring that no scanning behavior will encroach on cellular listening time. Specifically: After obtaining the set of scanning subtasks, the execution time of the safe window is predicted based on step S2. Dynamically generate scan fragments, ensuring that each fragment can be executed within a single safe window. Define scan slices: ; The fragment execution time is: ; To ensure that the scan segmentation can be completed within the safe window, the following must be met: ; And generate shards according to the following rules: ; Simultaneously, the channel order in the slice is optimized by incorporating channel environment weights: ; in, This represents the m-th scan segment; Indicates the composition of scan slices The sequence of scan subtasks, where a is the index of the first scan task in the slice and b is the index of the last scan task; Indicates scan slices The total execution time is the sum of the execution times of all scanning subtasks in this slice; Indicates scan slices The maximum number of channels included; Indicates the weighting coefficient; n represents the number of consecutive scan subtasks that can be contained in a scan slice, i.e., within the safety window. The number of channel scans that can be performed continuously within the specified range; S33. Design a scanning progress maintenance mechanism across DRX cycles. Record completed scanning segments and pending segments in the scanning session state table. When a new security window arrives, select a suitable segment from the pending queue for execution. This allows a complete WiFi scan to be gradually accumulated and completed over multiple DRX cycles, thereby achieving continuity in the scanning process and stability in system scheduling. Specifically: Define the prediction window: ; Further, calculate the window matching priority for each slice: ; in, This represents the (k+i)th DRX cycle safety window predicted in step S2. S34. Establish a segmented scheduling optimization mechanism based on channel priority and environmental change awareness. By combining historical scan results, channel congestion status, and terminal mobility trends, the execution order of scan segments is dynamically adjusted to prioritize the scanning of channels with a higher probability of available WiFi access points. This improves WiFi detection efficiency and scan result quality within a limited security window. Specifically: Maintain the scan progress pointer: ; Each time a scan slice completes: ; When the following conditions are met: When this time is reached, it indicates that all channel scans are complete; To avoid the results becoming invalid due to excessively long scan times, a maximum scan session duration is defined: ; If the time limit is exceeded, the scan task will be reinitialized. Simultaneously record the execution time of each slice: And update the prediction model for the next cycle window to improve the accuracy of subsequent scan scheduling; in, This indicates the cumulative execution time of the current scanning session, used to control the duration of the entire scanning process and prevent the scanning from spanning too many DRX cycles, which could lead to a decrease in data timeliness. This indicates the maximum allowed scan session time. When the scan session duration exceeds this threshold, the current scan is terminated and a new scan task is restarted to ensure the real-time nature of the scan results.

[0022] S4. Implement WiFi scanning tasks in a segmented, cross-DRX cycle manner. Through asynchronous caching, segment fusion, and session generation mechanisms, present the scan results reliably, continuously, and transparently to upper-layer applications, while ensuring the monitoring security and scan integrity of cellular links. The specific implementation process is as follows: S41. After each scan slice is completed, the observed AP information (including BSSID, RSSI, timestamp, and window confidence) is written to the slice cache, and a hierarchical storage structure is established simultaneously to support subsequent fusion and cross-cycle management, ensuring that the scan results are available immediately and traceable. Specifically: Each scan segment After completing the scan in its corresponding security window, the observed WiFi access point information is immediately written to the fragment-level cache. : ; Meanwhile, to support future incremental fusion and historical comparison, a two-tier storage strategy is introduced: Temporary shard cache: Stores the current shard scan results, supporting fast merging and asynchronous updates; Persistent session cache: Organized by scan session, indexing each shard result, supporting cross-period fusion and historical data comparison; in, This represents the set of scan results for the m-th scan segment, used to temporarily store information about all WiFi access points scanned in this segment; This represents the j-th WiFi access point entity that was scanned. This represents the BSSID address of the j-th access point. The BSSID is usually the MAC address of the access point's wireless interface and is an important parameter that uniquely identifies the AP in a WiFi network. In this embodiment, it is mainly used for: AP deduplication, multi-scan fusion, and cross-segment result matching. The service set identifier (WiFi network name) of the j-th access point is used to represent the WiFi network name visible to the user. This represents the received signal strength of the j-th access point in the current scan. This represents the timestamp when the j-th access point was scanned; This indicates the confidence level of the security window where the current scanned fragment is located; S42. The scan results within each segment are deduplicated, retaining the strongest signal or the most recent observation record. A preliminary fusion of the RSSI of repeated APs is performed using a window confidence-weighted average, while recording the first and last occurrence times to provide a reliable basis for subsequent cross-segment fusion. Specifically: Since the same AP may appear repeatedly in different scan slices or different windows, it is necessary to deduplicate the results within each slice and calculate the initial fusion value, i.e.: Deduplication within the same partition: ; Weighted signal strength fusion: If the same AP appears repeatedly in multiple window segments, a window confidence-weighted average is used. ; Temporal information fusion: recording the first occurrence And the most recent occurrence ; in, This represents the set of fragmented scan results after deduplication; This represents the signal strength of the j-th AP observed in the i-th scan; This represents the final signal strength value of the j-th AP after weighted fusion; This indicates the number of times a specific AP was observed during the entire scan process; This represents the confidence weight of the window in which the i-th scan observation is located; This indicates the first appearance time of the j-th AP in the scanning session, used to determine the network appearance time and scanning coverage area; This represents the time when the j-th AP was last scanned in the scanning session, used to determine the network duration and stability. S43. Summarize all fragmented results to form a complete WiFi scanning session. For duplicate APs, re-weight and fuse them to calculate signal quality indicators. Simultaneously, integrate the timestamps to generate a unified session-level scanning view, achieving completeness and reliability of cross-DRX cycle scanning. Specifically: After all shards have completed initial fusion, the sharding results are aggregated into a complete scan session. : ; The duplicate APs are weighted and fused again to form session-level RSSI and signal quality score (QS): ; In addition, comprehensive information for each AP is generated: ; in, This represents the complete set of WiFi scan session results, i.e., the total set of results after merging all scan segments; K represents the total number of scan segments after the complete scan task is split. This represents the signal quality score of the j-th access point, used to evaluate the stability and reliability of the WiFi network; This represents the cumulative time that the j-th AP is observed in the scanning session, calculated as the duration of the AP from the first scan to the last scan; This indicates the total duration of the entire WiFi scanning session, from the start of the first scan segment to the completion of the last scan segment; S44. The merged sharding results are passed to the upper-layer application via an asynchronous interface. Simultaneously, a final scan report is generated at the end of the session. Combined with fault tolerance and incremental compensation mechanisms, incomplete sharding is ensured to be executed in subsequent windows, achieving underlying sharding, upper-layer transparency, and scan integrity assurance. Specifically: An asynchronous result update mechanism is set up: after each shard is completed, the fusion result is immediately passed to the upper layer application through the scheduling interface without waiting for the session to complete; the result obtained by the upper layer is reliable data after weighted fusion, which is presented as a complete scan effect; Generate final session: After all fragments are completed, the fragment results are integrated into a final session and returned to the upper layer as a scan report; the report includes a complete AP list, signal strength, quality score, occurrence timestamp, and other information; Configure an exception handling mechanism: If a segment is not completed or the window is preempted and interrupted, the number of the incomplete segment is automatically recorded; the segment is executed first in the next available safe window to achieve incremental compensation scanning; ensure the integrity of the final session and do not affect the cellular link.

[0023] Example 2: The present invention proposes a cellular IoT DRX period adaptation WiFi non-intrusive detection system, which is used to execute the cellular IoT DRX period adaptation WiFi non-intrusive detection method proposed in Example 1, comprising: Memory; processor; A computer program stored in the memory and capable of running on the processor; The processor executes a computer program to implement a WiFi non-contact detection method for DRX periodic adaptation in the above embodiment 1.

[0024] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A WiFi non-intrusive detection method for DRX periodic adaptation in cellular IoT, characterized in that, The specific implementation steps include the following: S1. By analyzing the cellular discontinuous reception cycle parameters of IoT terminals, a cellular listening function and radio frequency priority model on a continuous time axis are constructed, a continuous safe time window for radio frequency switching is identified, and radio frequency collaborative scheduling decisions are generated in combination with radio frequency hardware switching overhead. S2. Based on the security time window and cellular monitoring status, a dynamic security window prediction mechanism is designed to generate a security window sequence that can perform WiFi scanning through weighted historical correction and multi-cycle sliding prediction. S3. The complete WiFi scanning task is divided into several micro scanning segments. The executable scanning segments are dynamically matched according to the actual length of each security window, and the scanning task is completed incrementally across several non-continuous reception cycles. S4. Through asynchronous caching and fragment fusion mechanism, the scan fragment results completed in several cycles are deduplicated, weighted and merged and timestamped to generate a complete WiFi scan view.

2. The WiFi non-contact detection method for DRX periodic adaptation in cellular IoT according to claim 1, characterized in that, In step S1, the specific process of constructing the cellular listening function and radio frequency priority model on a continuous time axis is as follows: The system analyzes the length of the discontinuous reception cycle, the paging monitoring start offset, and the active monitoring duration of the terminal's current cellular communication. It then maps the discrete protocol configuration parameters to a cellular monitoring state function on a continuous time axis. During each discontinuous reception cycle, the function takes a value of 1 between the monitoring start time and the monitoring end time to indicate that the cellular monitoring state must be maintained, and takes a value of 0 at other times to indicate that it is in the cellular monitoring idle segment. A radio frequency resource occupancy priority function is established based on the weight coefficients of cellular communication tasks and WiFi scanning tasks, as well as the task status factor. By comparing the numerical values ​​of cellular communication priority and WiFi scanning priority, it is determined whether radio frequency resources should be allocated to cellular communication or WiFi scanning at any given time, thereby achieving radio frequency resource competition and coordination under the condition of prioritizing cellular communication.

3. The WiFi non-contact detection method for DRX periodic adaptation in cellular IoT according to claim 2, characterized in that, The process for identifying continuous safety time windows is as follows: By combining the cellular listening function and the radio frequency priority function, it is determined whether WiFi scanning is allowed at the current time. When cellular communication does not require listening but WiFi scanning is required, a continuously available radio frequency switching time segment is identified, and the start and end times of the segment and the duration between them are calculated. The time overhead required for the radio frequency hardware to switch between the cellular band and the WiFi band is obtained. This time overhead includes the time to switch from the cellular band to the WiFi band and the time to recover from the WiFi band to the cellular band. The actual executable scan time is obtained by subtracting the above-mentioned switching overhead from the available radio frequency window length. The actual executable scan time is compared with the minimum scan execution time threshold. When the actual executable scan time is greater than or equal to the minimum scan execution time threshold, an RF switching scheduling command is generated, which enables the RF front end to switch to the WiFi band to perform the detection task within the safe window and resume cellular listening before the window ends.

4. The WiFi non-contact detection method for DRX periodic adaptation in cellular IoT according to claim 3, characterized in that, In step S2, the specific implementation of the dynamic safety window prediction mechanism is as follows: The time interval within each discontinuous reception cycle that meets the following three conditions is defined as the initial security window: the cellular link is in a non-monitoring state, the radio frequency priority allows the WiFi scanning task to be executed, and the window length is greater than the minimum execution time of a single scan. The length of this window is then calculated. A weighted historical correction mechanism is adopted to combine the current window length with the average actual available time in the same historical period. By adjusting the weights to balance the importance of the current measurement and historical experience, the corrected safe window length is obtained. The safe window for several future discontinuous reception periods is recursively predicted based on the corrected window length, and a confidence index is calculated for each predicted window to characterize the prediction reliability. This confidence index is determined based on the degree of deviation between the historical window length and the predicted window length. The window intervals and confidence levels of several future periods are integrated to form a safe window sequence. The legality of the predicted window is verified by combining the radio frequency switching overhead and the minimum scan time threshold, and scheduling interface information including the window start time, window end time and actual executable scan time is generated.

5. The WiFi non-contact detection method for DRX periodic adaptation in cellular IoT according to claim 4, characterized in that, In step S3, the specific method for dividing the complete WiFi scanning task into several micro-scanning segments is as follows: The set of WiFi channels currently supported by the terminal is structurally modeled. This set of channels is generated by the terminal's WiFi protocol stack based on the regional wireless spectrum specifications and the chip's support capabilities. The complete scanning task is abstracted into a set of tasks consisting of several channel scanning subtasks. Each channel scanning subtask needs to complete two operations: switching the radio frequency to the target channel and listening to beacon frames or probing response frames on the target channel. The complete scanning task is divided into several independently executable micro-scanning slices. Each scanning slice contains one or more consecutive channel scanning subtasks. Standardized description parameters are established for each slice, including the sequence of scanning subtasks within the slice and the total execution time of the slice, so that the scanning task can be scheduled across cycles in the form of micro-scanning units.

6. The WiFi non-contact detection method for DRX periodic adaptation in cellular IoT according to claim 5, characterized in that, In step S3, the specific process of dynamically matching executable scan fragments is as follows: Based on the predicted actual executable scan time of the safety window, an executable scan segment evaluation mechanism is established. The total execution time of each scan segment is matched and judged with the current safety window length. Only scan segments whose total execution time is strictly less than the actual executable scan time of the current safety window are allowed to enter the scheduling candidate queue. Based on the terminal's historical scanning records, known WiFi channel distribution, and regional spectrum usage statistics, each channel is assigned an environmental probability weight, which reflects the historical number of times an access point has been detected on the corresponding channel. By combining channel environment weights to optimize the channel order in scanning slices, channels with higher channel environment probability weights are scanned first, thereby improving WiFi detection efficiency under limited security window resources.

7. A WiFi non-contact detection method for DRX periodic adaptation in cellular IoT according to claim 6, characterized in that, In step S3, the specific implementation method for completing the scanning task incrementally across cycles is as follows: The session state table is scanned to record the completed scanned fragments and fragments to be executed, and a suitable fragment is selected from the queue to be executed when a new security window arrives. For each slice to be executed, a window matching priority is calculated. This priority is determined based on the degree of matching between the slice execution time and the window length, as well as the environmental probability weight of the channel contained in the slice. The slices are selected for execution within the current safe window based on the matching priority from high to low, and the scan progress pointer is updated after the slice execution is completed. When all channel scans are completed or the cumulative execution time of a scan session exceeds the maximum allowed scan session time, the current scan task is terminated and a new scan task is restarted to ensure the real-time nature of the scan results.

8. A WiFi non-contact detection method for DRX periodic adaptation in cellular IoT according to claim 7, characterized in that, In step S4, the specific implementation method of the asynchronous caching and sharding fusion mechanism is as follows: After each scan segment is completed, the observed access point information is written to the segment-level buffer. This information includes the access point's basic service set identifier, service set identifier, received signal strength, scan timestamp, and the confidence level of the security window in which the current scan segment is located. The hierarchical storage structure includes temporary shard cache and persistent session cache. Temporary shard cache is used to store the current shard scan results to support fast fusion and asynchronous updates. Persistent session cache is organized by scan session and indexes each shard result to support cross-cycle fusion and historical data comparison. The scanning results within each segment are deduplicated. When the same access point appears multiple times in the same segment, the observation record with the strongest signal strength or the latest timestamp is retained. The signal strength of the repeated access points is initially fused by window confidence weighted averaging. At the same time, the first and last occurrence times of the access point are recorded.

9. A WiFi non-contact detection method for DRX periodic adaptation in cellular IoT according to claim 8, characterized in that, In step S4, the specific process of generating a complete WiFi scan view is as follows: After all scan slices have been initially fused, all slice results are aggregated to form a complete scan session result set; For access points that repeatedly appear across segments, a weighted fusion calculation is performed again to determine the signal quality score. This score is based on the average signal strength, frequency of occurrence, and cumulative observation time of the access points. The integrated information for each access point includes the basic service set identifier, service set identifier, weighted fused signal strength value, signal quality score, first occurrence time, last occurrence time, and cumulative observation time. The merged sharding results are passed to the upper-layer application through an asynchronous interface, and a final scan report containing a complete list of access points and the above comprehensive information is generated at the end of the scan session. When a scan segment is not completed due to window preemption or interruption, the incomplete segment number is automatically recorded, and the segment is executed first in the next available safe window to achieve incremental compensation scanning and ensure the integrity of the final scan session.

10. A WiFi-free detection system for cellular IoT DRX cycle adaptation, comprising: Memory; processor; The memory stores computer programs; When the processor executes the computer program, it implements the WiFi non-contact detection method for DRX periodic adaptation in cellular IoT as described in any one of claims 1 to 9.

Images