Sharing intelligent device state real-time monitoring method and system based on Internet of Things

Abstract

The invention discloses a shared intelligent device state real-time monitoring method and system based on the Internet of Things, and relates to the technical field of device intelligent monitoring, and the method comprises the steps: enabling a shared intelligent device to be connected with a gateway, obtaining a device message, calculating a message arrival rate and an uploading equivalent constant, correcting the message time through double-layer time anchoring, and calculating a real-time AoI, a device AoDT is further calculated; constructing a device feature, predicting a device target feature by using a q-TCN model, calculating a device anomaly score in combination with freshness gating of the device feature, and grading the device anomaly through a local decision response of the device; and performing real-time monitoring display and equipment maintenance according to the equipment exception classification, and forming equipment maintenance record storage. According to the invention, by realizing a closed loop from data to action, the reliability and real-time performance of monitoring and maintenance of the shared equipment are improved, and the operation and maintenance safety of the equipment is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of intelligent equipment monitoring technology, and in particular to a method and system for real-time monitoring of the status of shared intelligent equipment based on the Internet of Things. Background Technology

[0002] In recent years, with the high-density deployment of shared smart devices (such as shared charging, shared mobility terminals, and shared security nodes) in urban spaces, the remote operation and maintenance of these devices has evolved from "offline inspection + periodic reporting" to an "Internet of Things (IoT) + edge computing (MEC)" architecture. The typical technical path is: the edge side samples at fixed intervals, and the gateway or cloud side completes decoding, aggregation, and alarm threshold determination. Network indicators such as latency, packet loss rate, and availability are used to roughly characterize the monitoring quality. However, industry applications often still rely on the simplistic assumption of "freshness upon arrival," rarely including "upload link, queuing / computing service, and twin refresh" in the same metric. Furthermore, there is a lack of engineering solutions that integrate "freshness - anomaly detection - local handling - maintenance closed loop." Existing technologies lack closed-loop responses oriented towards on-site constraints, mainly relying on unified alarms on the cloud side. They lack hierarchical local real-time monitoring and processing strategies, making it difficult to achieve millisecond-level loss mitigation at the edge side. Moreover, they often directly apply time-series models to raw or coarsely aligned data, leading to systemic false alarms / missed alarms during link congestion or batch retransmission. Summary of the Invention

[0003] In view of the aforementioned existing problems, the present invention is proposed.

[0004] Therefore, this invention provides a method and system for real-time monitoring of the status of shared smart devices based on the Internet of Things, which solves the problems of existing technologies lacking closed-loop response to field constraints, relying mainly on unified alarms on the cloud side, lacking hierarchical local real-time monitoring and processing strategies, making it difficult to stop losses at the edge in milliseconds, and often applying time-series models directly to raw or coarsely aligned data, resulting in systemic false alarms / missed alarms during link congestion or batch retransmission.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for real-time monitoring of the status of shared smart devices based on the Internet of Things, comprising: The shared smart device is connected to the gateway to obtain device messages and calculate message arrival rate and upload equivalent constant. The message time is corrected through two-layer time anchoring and the real-time AoI is calculated. The device AoDT is further calculated. Construct equipment features and use the q-TCN model to predict equipment target features. Combine the freshness gating of equipment features to calculate equipment anomaly scores, and classify equipment anomalies by local decision response of equipment. Based on the classification of equipment anomalies, real-time monitoring and display of equipment and equipment maintenance are carried out, and equipment maintenance records are generated and stored.

[0006] As a preferred embodiment of the IoT-based real-time monitoring method for the status of shared smart devices described in this invention, the step of connecting the shared smart device to the gateway to obtain device packets and calculating the packet arrival rate and upload equivalent constant refers to establishing a TLS 1.3 bidirectional authentication connection between the shared smart device and the gateway, collecting device packets through the gateway, and the edge gateway counting the number of device packet arrivals for each device using a fixed sliding window W. And calculate the instantaneous arrival rate. ; Robust arrival rate is calculated by suppressing jitter through exponential smoothing. ; For each received device message, the edge gateway determines the timing based on the reception time. and the time of transmission within the message Calculate the uplink delay of the message and take the median within a fixed sliding window W as the equivalent uplink constant. .

[0007] As a preferred embodiment of the IoT-based real-time monitoring method for the status of shared intelligent devices described in this invention, wherein: the message time is corrected by two-layer time anchoring and the real-time AoI is calculated, and the device AoDT is further calculated based on the message uplink delay and the upload equivalent constant to calculate the instantaneous deviation of the device relative to the edge gateway. Update the bias estimate using exponential smoothing. With drift estimation ; The message time is corrected by combining bias estimation and drift estimation; If two consecutive device messages for the same device i satisfy both the time proximity and the state proximity, then delete the latter device message. For each valid device message, calculate the real-time AoI and simultaneously calculate the message arrival interval. System dwell time ; PAoI is approximated by the rolling mean of the peak AoI calculated based on the real-time AoI. The AoDT of the device is estimated by combining robust arrival rate and upload equivalent constant.

[0008] As a preferred embodiment of the IoT-based real-time monitoring method for the status of shared intelligent devices described in this invention, the following steps are described: constructing device features and using a q-TCN model to predict device target features, and combining the freshness gating of device features to calculate device anomaly scores, which in turn define the single-frame features of the device. ; A freshness-gating mechanism is used to set gating hysteresis for single-frame features of the device. ; A single-branch causal one-dimensional convolution is constructed using the q-TCN model; The q-TCN model is trained using pinball loss, and the predicted target features are output based on the trained q-TCN model. And calculate the anomaly score for each feature dimension based on the predicted target features. ; Anomaly scores across all dimensions Summation is performed to obtain the original anomaly score. The freshness penalty is calculated based on the freshness gating hysteresis and added to the original anomaly score to calculate the device anomaly score.

[0009] As a preferred embodiment of the IoT-based real-time monitoring method for the status of shared intelligent devices according to the present invention, the step of classifying and setting anomaly activation thresholds through local device decision-making responses to device anomalies is described. and the closing threshold Define the number of consecutive triggers required. Build an exception counter ; Equipment malfunctions are classified into different levels based on their malfunction scores.

[0010] As a preferred embodiment of the IoT-based real-time monitoring method for the status of shared smart devices described in this invention, the step of real-time monitoring and displaying and maintaining devices according to the device anomaly classification refers to visually displaying device anomalies in real time according to the device anomaly classification and notifying relevant personnel to monitor and maintain the corresponding shared devices.

[0011] As a preferred embodiment of the IoT-based real-time monitoring method for the status of shared smart devices described in this invention, the step of forming and storing device maintenance records refers to forming a device maintenance log after device maintenance and storing the data before and after device maintenance in a shared database.

[0012] Secondly, the present invention provides a real-time monitoring system for the status of shared smart devices based on the Internet of Things, comprising: The data collection module is used to connect the shared smart device to the gateway to obtain device messages and calculate message arrival rate and upload equivalence constant. It corrects message time through two-layer time anchoring and calculates real-time AoI, and further calculates device AoDT. The anomaly detection module is used to construct equipment features and predict equipment target features using the q-TCN model. It calculates equipment anomaly scores by combining the freshness gating of equipment features and classifies equipment anomalies through local decision-making of equipment. The maintenance record module is used to monitor and display equipment in real time according to the level of equipment anomalies and to perform equipment maintenance, and to generate and store equipment maintenance records.

[0013] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the method for real-time monitoring of the status of shared smart devices based on the Internet of Things as described in the first aspect of the present invention.

[0014] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the method for real-time monitoring of the status of shared smart devices based on the Internet of Things as described in the first aspect of the present invention.

[0015] The beneficial effects of this invention are as follows: This invention achieves a closed loop from data to action by statistically analyzing device data packets at the gateway end and combining it with dual-layer time anchoring to correct timestamps, and by constructing a device feature model with real-time AoI for quantile prediction and anomaly classification processing. This improves the reliability and real-time performance of shared device monitoring and maintenance, and ensures the operational safety of the devices. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of the method for real-time monitoring of the status of shared smart devices based on the Internet of Things in Example 1.

[0018] Figure 2 This is a structural diagram of the IoT-based shared smart device status real-time monitoring system in Example 1. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] Example 1, referring to Figure 1 and Figure 2 This is the first embodiment of the present invention, which provides a method for real-time monitoring of the status of shared smart devices based on the Internet of Things, including the following steps: S1. Connect the shared smart device to the gateway to obtain device messages and calculate message arrival rate and upload equivalent constant. Correct message time through two-layer time anchoring and calculate real-time AoI. Further calculate device AoDT. Specifically, connecting the shared smart device to the gateway to obtain device messages and calculating the message arrival rate and upload equivalent constant involves establishing a TLS 1.3 bidirectional authentication connection between the shared smart device and the gateway, collecting device messages through the gateway. These device messages include device identifiers, time, device data (sensor data types), spatial permissions, security certificates, etc. The edge gateway counts the number of device message arrivals for each device using a fixed sliding window W (60 seconds). And calculate the instantaneous arrival rate. :

[0023] Robust arrival rate is calculated by suppressing jitter through exponential smoothing. :

[0024] in For smoothing coefficients, The statistical step size is fixed at 1 second. The smoothed, robust arrival rate; For each received device message, the edge gateway determines the timing based on the reception time. and the time of transmission within the message Calculate the uplink delay of the message and take the median within a fixed sliding window W as the equivalent uplink constant. :

[0025]

[0026] in This is to delay the uplink time of the message.

[0027] By collecting and statistically analyzing device messages, and calculating message arrival rate and upload equivalence constant, the system can monitor the status and operation of devices in real time. Through exponential smoothing to calculate robust arrival rate, it effectively suppresses data fluctuations caused by short-term network jitter or instantaneous traffic changes, enabling the system to maintain stable performance even in highly volatile environments. The exponentially smoothed robust arrival rate provides stable message traffic prediction, avoiding the impact of instantaneous changes on system load prediction and improving the responsiveness of the device monitoring system. By calculating upload latency and applying median smoothing, it can effectively filter noise under poor network conditions, maintaining the stability of device monitoring.

[0028] Furthermore, by correcting the message time through dual-layer time anchoring and calculating the real-time AoI, the device AoDT index is further calculated based on the message uplink latency and the upload equivalent constant to calculate the instantaneous deviation of the device relative to the edge gateway:

[0029] in This refers to the instantaneous deviation between the device clock and the edge gateway; Update the bias estimate using exponential smoothing. With drift estimation :

[0030]

[0031] in The deviation coefficient, The drift coefficient was set experimentally. and The initial values ​​for the deviation and drift estimates from the previous time step are set using the device and gateway latency. The time of the previous message reception; The message time is corrected by combining bias estimation and drift estimation:

[0032] in The generation time within the device message. To correct message time; If two consecutive device messages from the same device i simultaneously satisfy both time proximity and state proximity, the latter device message is deleted. Time proximity means the difference in the corrected message times of two consecutive device messages is less than a set time proximity threshold. State proximity means the difference in the content of two consecutive device messages is less than a set state proximity threshold. For example:

[0033] in and The content of two consecutive device messages, To set a state proximity threshold; For each valid device message, calculate the real-time AoI (Age of Information) and simultaneously calculate the interval between adjacent arrivals of the messages. System dwell time :

[0034]

[0035] in For the real-time AoI of the nth device message, The time of receiving the nth device message. This is the completion time of the nth device message, i.e., the system completion time for processing the device message. The approximate PAoI is calculated by using the rolling mean of the peak AoI based on the real-time AoI:

[0036] in and The adjacent arrival interval and system dwell time of the kth valid message are obtained by scrolling within a fixed sliding window W; Combining robust arrival rate and upload equivalent constant estimation of device AoDT (digital twin delay):

[0037] in For device AoDT, This represents the robust arrival rate average for all devices.

[0038] By using dual-layer time anchoring to correct message deviations and drifts, the device clock can be precisely aligned with the gateway clock, effectively solving the clock synchronization problem between the device and the gateway. In multi-device, complex environments, synchronization differences between the device clock and the edge gateway can lead to timestamp errors, affecting data accuracy and subsequent analysis. By correcting the timestamps, subsequent calculations (such as real-time AoI) can be ensured. The accuracy of AoI (AoDT) provides a reliable data foundation for equipment management and maintenance. Real-time AoI calculation, by comparing the reception time and generation time of equipment messages, can reflect the "freshness" of equipment data in real time. Peak AoI, by rolling averaging the arrival interval and dwell time of adjacent equipment messages, provides a more stable and long-term assessment of equipment information quality, accurately capturing the timeliness of equipment information and avoiding misjudgments caused by outdated data. Especially in scenarios with high-frequency equipment updates, it ensures the timeliness of equipment status and improves the system's response speed and accuracy. Based on real-time AoI and robust arrival rate, this invention introduces the calculation of equipment anomaly scores, which are filtered through freshness gating and the equipment is locally graded according to the anomaly score. Local graded response can react quickly when equipment anomalies occur, improving the intelligence and real-time performance of equipment management, avoiding the delay and burden in traditional remote monitoring methods. Through real-time response, the losses caused by equipment failures can be minimized, improving equipment stability and operation and maintenance efficiency.

[0039] S2. Construct equipment features and use the q-TCN model to predict equipment target features. Combine the freshness gating of equipment features to calculate equipment anomaly scores, and classify equipment anomalies by responding to equipment anomalies through local equipment decision-making. Specifically, device features are constructed and the q-TCN model is used to predict device target features. The device anomaly score is calculated by combining the freshness gating of these features with the definition of single-frame device features. :

[0040] in For device status data, including location and battery level, The age is real-time, obtained through real-time AoI; A freshness-gating mechanism is used to set gating hysteresis for single-frame features of the device. :

[0041] in For the gating hysteresis at time t-1, , , as well as These are the thresholds for enabling and disabling, respectively. A single-branch causal one-dimensional convolution is constructed using the q-TCN model, with a depth of 5 and a kernel length of K. The dilation coefficient sequence is set. and number of channels ,in The output is the predicted target value at the next time step, i.e., the predicted feature at the next time step. The output of the causal dilated convolution at time t of the d-th layer is:

[0042] in For the output of layer d, Let m be the weight of the m-th convolutional kernel in the d-th layer. For bias, For ReLU function, This is the output of the previous layer, with the initial input being the device's single-frame features. ; The q-TCN model is trained using pinball loss, and the predicted target features are output based on the trained q-TCN model. and output quantile predictions. (Taking features) The next-time prediction of a subset of the target features is given, and anomaly scores for each feature dimension are calculated based on the predicted target features. :

[0043] in , as well as quantiles The j-th dimension parameter of the quantile prediction output, To predict the j-th dimension parameter of the target feature, To prevent the number from being divided by zero; Anomaly scores across all dimensions Summation is performed to obtain the original anomaly score. The freshness penalty is calculated based on the freshness gating hysteresis and then added to the original anomaly score to calculate the device anomaly score:

[0044]

[0045] in As a penalty for lack of freshness, and For AoDT and PAoI at time t+1, through Obtain The penalty coefficient is... This represents the equipment malfunction score.

[0046] The construction of device features includes device status information (such as location and battery level), instantaneous age, AoDT, and PAoI. These features are used to predict the next time step using a q-TCN model. The q-TCN model captures the dynamic changes in device status by combining causal convolution and dilated convolution with quantile prediction. By accurately predicting device features, this invention can effectively capture the changing trends of device status, especially providing higher accuracy in predicting the timeliness and abnormal changes of device data. This helps to identify potential faults in advance and handle the device in a timely manner, improving the stability and reliability of device operation. The freshness gating mechanism dynamically determines whether to allow device message features to enter the q-TCN model for processing based on the device's AoDT and PAoI values. The gating mechanism will only allow the data when it is "fresh". According to calculations, the freshness gating mechanism ensures that the system is not affected by outdated or invalid data when processing real-time data, thereby improving the accuracy of anomaly detection. This mechanism can effectively improve the real-time response capability of the system, especially in large-scale equipment environments, and can reduce unnecessary computational burden. By calculating the anomaly score of the equipment and correcting the anomaly score in combination with the freshness penalty mechanism, the system can not only detect equipment anomalies, but also adjust the impact of anomalies according to the freshness of the data. By comprehensively considering the freshness of equipment data and the anomaly score, the system can respond to equipment anomalies more intelligently, the impact of outdated data is effectively reduced, and the anomaly detection results are more reliable. In addition, the introduction of freshness penalty allows the system to maintain higher accuracy and efficiency when facing high-frequency data streams, avoiding misjudgments caused by insufficient data timeliness.

[0047] Furthermore, by implementing a tiered approach to device anomalies through local decision-making, anomaly activation thresholds can be set. and the closing threshold Define the number of consecutive triggers required. Build an exception counter :

[0048] Equipment anomalies are classified into different levels based on their anomaly scores:

[0049] in Equipment anomalies are classified into three levels: 0, 1, and 2, which respectively represent freshness fault reset (immediately refresh and reset the model to obtain prediction output), minor anomaly (local buzzer warning, increased maintenance priority), and severe anomaly (power outage isolation, reporting the fault location, and waiting for maintenance).

[0050] By setting anomaly activation and deactivation thresholds, intelligent adjustment of device anomaly responses is achieved. High-intensity anomaly handling is only triggered when a device anomaly persists for a predetermined number of times, avoiding false alarms and over-response, and improving the stability and efficiency of the device management system. Through device anomaly classification, this invention can classify device responses according to the severity of the anomaly. For minor anomalies, only warnings and priority increases are issued; for severe anomalies, device isolation is immediately implemented to ensure that the system can prevent the escalation of serious problems in real time. This classified response makes device management more efficient and intelligent, avoids resource waste, and improves device maintenance efficiency. Combined with a freshness gating mechanism and anomaly score correction, it ensures that only fresh and valid device data is used to calculate the anomaly score. The freshness penalty mechanism minimizes the impact of outdated data by adjusting the anomaly score. The freshness gating mechanism ensures that the system's response to device anomalies is more accurate, avoiding false alarms or missed alarms caused by outdated device data. Through the freshness penalty mechanism, the system can appropriately adjust the anomaly score for "fresh" device data, further improving the accuracy of anomaly detection and optimizing maintenance response.

[0051] S3. Real-time monitoring and display of equipment anomalies based on their severity levels, and maintenance of equipment, and storage of equipment maintenance records; Specifically, real-time monitoring and maintenance based on equipment anomaly classification refers to visually displaying equipment anomalies in real time according to their classification and notifying relevant personnel to perform corresponding shared equipment monitoring and maintenance.

[0052] Furthermore, the formation of equipment maintenance record storage refers to the creation of equipment maintenance logs after equipment maintenance, recording the data before and after equipment maintenance and storing it in a shared database.

[0053] This embodiment also provides a real-time monitoring system for the status of shared smart devices based on the Internet of Things, including: The data collection module is used to connect the shared smart device to the gateway to obtain device messages and calculate message arrival rate and upload equivalence constant. It corrects message time through two-layer time anchoring and calculates real-time AoI, and further calculates device AoDT. The anomaly detection module is used to construct equipment features and predict equipment target features using the q-TCN model. It calculates equipment anomaly scores by combining the freshness gating of equipment features and classifies equipment anomalies through local decision-making of equipment. The maintenance record module is used to monitor and display equipment in real time according to the level of equipment anomalies and to perform equipment maintenance, and to generate and store equipment maintenance records.

[0054] This embodiment also provides a computer device applicable to the real-time monitoring method for the status of shared smart devices based on the Internet of Things (IoT), comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the real-time monitoring method for the status of shared smart devices based on the IoT proposed in the above embodiment.

[0055] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0056] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the method for real-time monitoring of the status of shared smart devices based on the Internet of Things as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0057] In summary, this invention achieves a closed loop from data to action by statistically analyzing device data packets at the gateway end, combining dual-layer time anchoring to correct timestamps, and constructing a device feature-based tiered prediction and anomaly classification processing system using real-time AoI. This improves the reliability and real-time performance of shared device monitoring and maintenance, and ensures the operational safety of the devices.

[0058] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for real-time monitoring of the status of shared smart devices based on the Internet of Things, characterized in that: include, The shared smart device is connected to the gateway to obtain device messages and calculate message arrival rate and upload equivalent constant. The message time is corrected through two-layer time anchoring and the real-time AoI is calculated. The device AoDT is further calculated. Construct equipment features and use the q-TCN model to predict equipment target features. Combine the freshness gating of equipment features to calculate equipment anomaly scores, and classify equipment anomalies by local decision response of equipment. Based on the classification of equipment anomalies, real-time monitoring and display of equipment and equipment maintenance are carried out, and equipment maintenance records are generated and stored.

2. The method for real-time monitoring of the status of shared intelligent devices based on the Internet of Things as described in claim 1, characterized in that: The process of connecting the shared smart device to the gateway to obtain device packets and calculating packet arrival rate and upload equivalence constant refers to establishing a TLS 1.3 bidirectional authentication connection between the shared smart device and the gateway, collecting device packets through the gateway, and the edge gateway counting the number of device packets arriving for each device using a fixed sliding window W. And calculate the instantaneous arrival rate. ; Robust arrival rate is calculated by suppressing jitter through exponential smoothing. ; For each received device message, the edge gateway determines the timing based on the reception time. and the time of transmission within the message Calculate the uplink delay of the message and take the median within a fixed sliding window W as the equivalent uplink constant. .

3. The method for real-time monitoring of the status of shared smart devices based on the Internet of Things as described in claim 2, characterized in that: The process involves correcting message time through dual-layer time anchoring and calculating real-time AoI. Further calculation of device AoDT refers to the instantaneous deviation of the device relative to the edge gateway based on the message uplink delay and upload equivalent constant. Update the bias estimate using exponential smoothing. With drift estimation ; The message time is corrected by combining bias estimation and drift estimation; If two consecutive device messages for the same device i satisfy both the time proximity and the state proximity, then delete the latter device message. For each valid device message, calculate the real-time AoI and simultaneously calculate the message arrival interval. System dwell time ; PAoI is approximated by the rolling mean of the peak AoI calculated based on the real-time AoI. The AoDT of the device is estimated by combining robust arrival rate and upload equivalent constant.

4. The method for real-time monitoring of the status of shared smart devices based on the Internet of Things as described in claim 3, characterized in that: The process involves constructing device features and using the q-TCN model to predict device target features. The freshness gating of these features is then used to calculate the device anomaly score, defining the single-frame features of the device. ; A freshness-gating mechanism is used to set gating hysteresis for single-frame features of the device. ; A single-branch causal one-dimensional convolution is constructed using the q-TCN model; The q-TCN model is trained using pinball loss, and the predicted target features are output based on the trained q-TCN model. And calculate the anomaly score for each feature dimension based on the predicted target features. ; Anomaly scores across all dimensions Summation is performed to obtain the original anomaly score. The freshness penalty is calculated based on the freshness gating hysteresis and added to the original anomaly score to calculate the device anomaly score.

5. The method for real-time monitoring of the status of shared intelligent devices based on the Internet of Things as described in claim 4, characterized in that: The method of classifying and setting anomaly activation thresholds through device local decision-making and response to device anomalies is described. and the closing threshold Define the number of consecutive triggers required. Build an exception counter ; Equipment malfunctions are classified into different levels based on their malfunction scores.

6. The method for real-time monitoring of the status of shared intelligent devices based on the Internet of Things as described in claim 5, characterized in that: The aforementioned real-time monitoring, display, and maintenance based on equipment anomaly classification refers to visually displaying equipment anomalies in real time according to their classification and notifying relevant personnel to monitor and maintain the corresponding shared equipment.

7. The method for real-time monitoring of the status of shared intelligent devices based on the Internet of Things as described in claim 6, characterized in that: The aforementioned equipment maintenance record storage refers to the creation of an equipment maintenance log after equipment maintenance, and the recording of data before and after equipment maintenance to a shared database.

8. A real-time monitoring system for the status of shared smart devices based on the Internet of Things (IoT), based on the real-time monitoring method for the status of shared smart devices based on the IoT as described in any one of claims 1 to 7, characterized in that: include, The data collection module is used to connect the shared smart device to the gateway to obtain device messages and calculate message arrival rate and upload equivalence constant. It corrects message time through two-layer time anchoring and calculates real-time AoI, and further calculates device AoDT. The anomaly detection module is used to construct equipment features and predict equipment target features using the q-TCN model. It calculates equipment anomaly scores by combining the freshness gating of equipment features and classifies equipment anomalies through local decision-making of equipment. The maintenance record module is used to monitor and display equipment in real time according to the level of equipment anomalies and to perform equipment maintenance, and to generate and store equipment maintenance records.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the method for real-time monitoring of the status of shared smart devices based on the Internet of Things as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the real-time monitoring method for the status of shared smart devices based on the Internet of Things as described in any one of claims 1 to 7.

Images