Article loss early warning method and device, electronic equipment, storage medium and program product

Abstract

The invention provides an article loss early warning method and device, electronic equipment, a storage medium and a program product. The method comprises the steps of determining the probability that a label is in an abnormal state based on a pre-constructed label abnormal state probability calculation model in combination with historical state information and real-time state information of the label of a to-be-managed article; judging whether the state of the to-be-managed article is an abnormal state or not based on the probability that the label is in the abnormal state; and when the state of the to-be-managed article is an abnormal state, sending abnormal prompt information.

Description

Technical Field

[0001] This application relates to the field of Internet of Things (IoT) technology, and more particularly to a method, device, electronic device, storage medium, and program product for early warning of lost items. Background Technology

[0002] In related technologies, personal item management is mostly done manually or by using GPS locators, Bluetooth, or Ultra Wide Band (UWB) trackers. However, the item retrieval solutions in these technologies can usually only be found after the item is discovered to be lost, which increases the feasibility of item retrieval. Summary of the Invention

[0003] This application provides a method, device, electronic device, storage medium, and program product for early warning of lost items, which can reduce the difficulty of finding items.

[0004] The technical solution of this application embodiment is implemented as follows: This application provides a method for issuing a lost item warning, the method comprising: Based on a pre-built probability calculation model for abnormal tag states, the probability of a tag being in an abnormal state is determined by combining the historical and real-time status information of the tags of the items to be managed. Based on the probability that the tag is in an abnormal state, determine whether the state of the item to be managed is abnormal. If the status of the item to be managed is abnormal, an abnormal prompt message will be issued.

[0005] This application embodiment also provides a lost item warning device, the device comprising: The first processing module is used to determine the probability that the tag is in an abnormal state based on a pre-built tag abnormal state probability calculation model, combined with the historical and real-time state information of the tags of the items to be managed. The second processing module is used to determine whether the status of the item to be managed is abnormal based on the probability that the tag is in an abnormal state; and to issue an abnormal prompt message if the status of the item to be managed is abnormal.

[0006] This application also provides an electronic device, which includes a processor and a memory for storing a computer program that can run on the processor; wherein the processor is used to run the computer program to execute any of the above-described item loss warning methods.

[0007] This application also provides a computer storage medium storing a computer program, which, when executed by a processor, implements any of the above-described methods for early warning of lost items.

[0008] This application also provides a computer program product, including a computer program that, when executed by a processor, implements any of the above-described methods for early warning of lost items.

[0009] The embodiments of this application have the following beneficial effects: Based on the historical and real-time status information of the tags of the items to be managed, the embodiments of this application can determine the probability that the tags are in an abnormal state, thereby more accurately judging whether the status of the items to be managed is abnormal. When the status of the items to be managed is abnormal, an abnormal prompt message can be issued in a timely manner. In this way, it is not necessary to search for the items after they are found to be lost, but the items can be found in a timely manner based on the abnormal prompt message, realizing timely early warning of lost items and reducing the feasibility of finding items. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating a lost item warning method according to an embodiment of this application; Figure 2 This is an interactive flowchart illustrating the process of implementing a lost item alert in this embodiment of the application. Figure 3 This is another flowchart illustrating the item loss warning method according to an embodiment of this application; Figure 4 This is a schematic diagram of the structure of the item loss warning device according to an embodiment of this application; Figure 5 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] In related technologies, personal item management is mostly achieved through manual management or by using GPS locators, Bluetooth, or UWB trackers. Taking Bluetooth trackers as an example, the solution involves placing the Bluetooth tracker next to the item to be tracked and pairing it with an application on a mobile device. Once paired, the tracking function is activated. The Bluetooth tracker maintains a Bluetooth connection with the phone and sends its status information and signal strength data to the phone in real time or periodically. When the item needs to be located, the user opens the mobile application, which calculates the relative distance and direction between the tracker and the phone based on the data sent by the Bluetooth tracker. The user can then use this information and follow the instructions provided by the application to gradually approach the item's location until it is found.

[0012] Future cellular passive IoT will leverage the wide-area coverage of base stations to achieve item management and location tracking. A cellular passive IoT system consists of cellular passive tags, user mobile phones, reader / writer devices (including base stations, relay nodes, and auxiliary nodes), a core network, a platform, and a user application (App). After purchasing tags, users affix them to the items to be managed and register the tags via the mobile app, binding the tag's identification information to their mobile phone number. After registration, the cellular passive core network or platform stores the mapping between the tag's identification information and the mobile phone number. During daily use, reader / writer devices (including cellular base stations, relay devices, and auxiliary devices) periodically or as needed inventory all tags within their communication coverage area, recording the tag inventory information and its mapping to the reader / writer device, and reporting the inventory information to the core network and platform. The core network or platform summarizes and saves the information for subsequent tag retrieval.

[0013] The related technology-based solutions for item management and lost and found have at least the following problems: 1) Usually, the search can only be carried out after the item is discovered to be missing. There is no early warning function when the user has not discovered the item is missing, which increases the difficulty of finding the item.

[0014] 2) Although some item finding systems (such as AirTag) have a separation reminder function to prevent items from being lost, the function is relatively simple. It will generate a lot of false alarms in normal situations such as brief separation, which will reduce the user experience. In addition, the corresponding functions need to be manually set by the user. If there are many items marked, it will cause great inconvenience to the user. It is also not easy to adjust in time when the user's environment changes.

[0015] 3) With the widespread deployment of cellular passive IoT systems, wide-area tag retrieval becomes more difficult because cellular passive base stations cannot maintain a continuous connection with the tags. When a tag's location changes, it struggles to promptly synchronize its status with the base station. Therefore, by the time a user discovers a lost item, a considerable amount of time has often passed, increasing the difficulty of finding the item.

[0016] In view of the technical problems existing in related technologies, this application proposes technical solutions based on its embodiments.

[0017] This application provides a method, device, electronic device, storage medium, and program product for early warning of lost items. In this application, tags can be deployed on items to be managed, and the location of the items identified by the tags can be located using ubiquitous cellular base stations or relay nodes. For example, tags can be attached to the items to be managed; the tags deployed on the items can be cellular passive tags. In this application, based on a cellular passive IoT system, by collecting and analyzing tag information, user mobile phone information, location information, historical movement trajectories, etc., a scenario-oriented weighted probability calculation method can be used to obtain the probability that the tag is in an abnormal state, thereby determining the state of the items to be managed. When the state of the items to be managed is determined to be abnormal, an alarm is promptly issued to the user to help the user quickly retrieve the items.

[0018] The embodiments of this application will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the embodiments provided herein are merely illustrative of the embodiments of this application and are not intended to limit the embodiments of this application. Furthermore, the embodiments provided below are some embodiments for implementing this application, and not all embodiments for implementing this application. Unless otherwise specified, the technical solutions described in the embodiments of this application can be implemented in any combination.

[0019] It should be noted that, in the embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a method or apparatus that includes a list of elements includes not only the elements expressly described, but also other elements not expressly listed, or elements inherent to implementing the method or apparatus. Without further limitations, an element defined by the phrase "comprising a..." does not exclude the presence of other related elements (e.g., steps in the method or units in the apparatus, such as portions of circuitry, processors, programs, or software, etc.) in the method or apparatus that includes that element.

[0020] The item loss warning method provided in this application includes a series of steps, but the item loss warning method provided in this application is not limited to the steps described. Similarly, the item loss warning device provided in this application includes a series of modules, but the device provided in this application is not limited to the modules explicitly described, and may also include modules that need to be set up for obtaining relevant information or processing based on the information.

[0021] This application proposes a method for early warning of lost items, which can be applied to electronic devices such as terminals and servers. Figure 1 This is a flowchart illustrating a lost item warning method according to an embodiment of this application, such as... Figure 1 As shown, the process includes: Step 101: Based on the pre-built tag abnormal state probability calculation model, combine the historical and real-time status information of the tags of the items to be managed to determine the probability that the tag is in an abnormal state.

[0022] In this embodiment of the application, the tag abnormal state probability calculation model is used to calculate the probability that the tag is in an abnormal state. In practical applications, the historical and real-time status information of the tags of the items to be managed can be input into the tag abnormal state probability calculation model, and the probability that the tag is in an abnormal state can be calculated through the tag abnormal state probability calculation model.

[0023] Step 102: Based on the probability that the tag is in an abnormal state, determine whether the status of the item to be managed is abnormal.

[0024] Understandably, since the tags are deployed on the items to be managed, the probability that a tag is in an abnormal state can reflect the state of the items to be managed. In some embodiments, this step is implemented as follows: when the probability that a tag is in an abnormal state is greater than or equal to an abnormal probability threshold, the state of the item to be managed is determined to be abnormal.

[0025] Here, the anomaly probability threshold can be set according to actual needs. For example, the initial anomaly probability threshold can be set to 50%. After the initial setting, the anomaly probability threshold can also be dynamically adjusted. Understandably, when the probability of a tag being in an abnormal state is greater than or equal to the anomaly probability threshold, the tag can be considered to be in a relatively high probability of being in an abnormal state, thus accurately determining the state of the item to be managed as abnormal.

[0026] If the probability of a tag being in an abnormal state is less than the abnormal probability threshold, the status of the item to be managed can be determined to be normal.

[0027] Step 103: If the status of the item to be managed is abnormal, issue an abnormal prompt message.

[0028] In this embodiment, the anomaly alert message is used to notify the user that the item to be managed may be lost; that is, the anomaly alert message is a loss warning message. When the item to be managed is in an abnormal state, the anomaly alert message can be issued through text, sound, or other means.

[0029] In practical applications, steps 101 to 103 can be implemented based on a processor, which can be at least one of the following: Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), Controller, Microcontroller, and Microprocessor.

[0030] As can be seen, the embodiments of this application can determine the probability that the tag is in an abnormal state based on the historical and real-time status information of the tag of the item to be managed, thereby accurately judging whether the status of the item to be managed is abnormal. If the status of the item to be managed is abnormal, an abnormal prompt message can be issued in a timely manner. In this way, it is not necessary to search for the item after it is found to be lost, but the item can be found in a timely manner based on the abnormal prompt message, realizing timely early warning of lost items and reducing the feasibility and difficulty of finding items.

[0031] Regarding the implementation of step 101, in some embodiments, one or more of the first probability, second probability, third probability, and fourth probability can be calculated based on the tag abnormal state probability calculation model, combined with the historical and real-time state information of the tag of the item to be managed; then, the probability that the tag is in an abnormal state is determined based on one or more of the first probability, second probability, third probability, and fourth probability.

[0032] Among them, the first probability represents the probability that the tag and the associated device of the item to be managed are in a separate state, the second probability represents the probability that the location trajectory of the item to be managed is an unexpected trajectory, the third probability represents the probability that the moving speed of the item to be managed is abnormal, and the fourth probability represents the probability that the tag is separated from the tag's associated tag set.

[0033] In this embodiment, the associated device of the item to be managed can be predetermined. Under normal circumstances, the item to be managed and the associated device should be in an inseparable state. For example, the associated device of the item to be managed can be a mobile terminal such as a mobile phone. If the tag and the associated device of the item to be managed are in a separate state, it can be considered that there is a certain possibility that the item to be managed may be lost.

[0034] Unexpected trajectories can be pre-determined trajectories. For example, the historical location trajectory of the item to be managed can be determined based on the historical status information of the item's tags, and then the unexpected trajectory can be determined based on the historical location trajectory of the item to be managed. The associated tag set can include one or more tags. The associated tag set can be determined based on the distance between different tags. For example, for the tags of the item to be managed, the associated tag set can be formed by selecting tags that are relatively close to the tags of the item to be managed based on the distance between different tags.

[0035] In practical applications, after periodically or on-demand tag inventory, the reading and writing device can report all the tag information counted to the core network. The total tag information counted can be denoted as S, where S = {RID, {TagID}. i ,T j}}.

[0036] In an integrated architecture, RID represents the identifier of the read / write device. In a separate architecture, RID represents the identifier of the information receiving device, which can be a base station, a deployed fixed UE, a user's mobile phone, etc. TagID i This represents the identifier of the i-th tag detected, where i ranges from 1 to q, and q represents the number of tags detected by the reader / writer. T i Represents TagID i The timestamp. For the i-th tag counted, the first probability, second probability, third probability, and fourth probability can be denoted as P, respectively. i1 P i2 P i3 P i4 .

[0037] As can be seen, the embodiments of this application can determine the probability of a tag being in an abnormal state relatively accurately and reasonably based on the probability that the tag and the associated device of the item to be managed are in a separated state, the probability that the location trajectory of the item to be managed is an unexpected trajectory, the probability that the moving speed of the item to be managed is abnormal, or the probability that the tag is separated from the tag's associated tag set.

[0038] In some embodiments, the first probability can be calculated based on a first distance between the tag and the associated device of the item to be managed. Understandably, the first distance between the tag and the associated device of the item to be managed can be used to accurately determine whether the tag and the associated device of the item to be managed are in a separated state. Therefore, based on the first distance between the tag and the associated device of the item to be managed, the probability that the tag and the associated device of the item to be managed are in a separated state can be calculated relatively accurately, i.e., the first probability can be calculated relatively accurately.

[0039] Regarding the implementation of determining the first distance, in some embodiments, when the associated device of the item to be managed is a mobile terminal, the relative distance between the reading / writing device corresponding to the tag and the base station where the associated device is located can be determined as the first distance.

[0040] Here, when the associated device of the item to be managed is a mobile terminal, the reading and writing device corresponding to the tag can be a base station, relay node, or auxiliary node, etc.

[0041] Understandably, the location of a mobile terminal may change constantly. In this case, it is not necessary to determine the location of the mobile terminal in real time. Instead, the fixed location of the base station where the mobile terminal is located can be easily determined, and the relative distance between the reading and writing device corresponding to the tag and the base station where the mobile terminal is located can be obtained. That is, the first distance can be determined based on the fixed location of the base station where the mobile terminal is located, which is easy to implement.

[0042] In some embodiments, the first probability can be determined as 0 when the first distance is equal to 0; the first probability can be determined as a positive number less than 1 when the first distance is less than a preset first distance threshold; and the first probability can be determined as 1 when the first distance is greater than or equal to the first distance threshold.

[0043] Here, the first distance threshold can be set according to actual needs.

[0044] Understandably, when the first distance is equal to 0, it can be assumed that the tag and the associated device of the item to be managed are in the same position, and thus it can be accurately determined that the tag and the associated device of the item to be managed are not in a separated state; when the first distance is greater than or equal to the first distance threshold, it can be assumed that the tag and the associated device of the item to be managed are far apart, and thus it can be accurately determined that the tag and the associated device of the item to be managed are in a separated state; when the first distance is greater than 0 and less than the preset first distance threshold, it can be assumed that the tag and the associated device of the item to be managed may be in a separated state or may not be in a separated state. In this case, the probability that the tag and the associated device of the item to be managed are in a separated state can be reasonably determined as a positive number less than 1.

[0045] When the first distance is greater than 0 and less than a preset first distance threshold, one way to determine the first probability is to determine the first probability based on the ratio of the first distance to the first distance threshold. For example, the first probability is positively correlated with the ratio of the first distance to the first distance threshold.

[0046] As can be seen, since the first distance is the distance between the tag and the associated device of the item to be managed, the ratio of the first distance to the first distance threshold can accurately reflect the possibility that the tag and the associated device of the item to be managed are in a separate state to a certain extent. Therefore, based on the ratio of the first distance to the first distance threshold, the first probability can be determined relatively accurately.

[0047] In an exemplary application scenario, the associated device for the item to be managed is a mobile phone, and the first distance is denoted as [missing information]. The first distance threshold can be denoted as For the i-th tag identified in the inventory, the core network can extract the RID (Recognition ID) of the reading / writing device at the time of the most recent tag inventory (time t). t According to RID t Find the BSID of the base station where the tag is located. t , base station BSID t Location Loc t As the position of the label at time t, Loc t = (x t y t If a network topology of direct base station connection or uplink auxiliary connection is adopted, then Loc t For the location of the base station, if a network topology of relay direct connection and downlink auxiliary connection is adopted, then Loc t For relay nodes or auxiliary nodes RID t The location. When cellular passive IoT systems are deployed densely, tag information may be received by multiple read / write devices at the same time, resulting in multiple Locs at time t. t Simultaneously, the core network searches for the mobile phone number of the managed device in the registration information and finds the BSID of the base station where that mobile phone is located at time t. i , base station BSID i Location as user's phone location (LocPhone) t LocPhone t =(xp t yp t The core network can use the base station BSID. t Location Loc t User's mobile phone location (LocPhone) t Once the information is sent to the platform, the platform can calculate the relative distance between the i-th tag found in the inventory and its corresponding mobile phone at time t. And calculate the first probability P i1 .

[0048] If BSID t =BSID iThis indicates that the tag and the mobile phone are within the same base station coverage area at time t, and can be considered... That is, the tag and the phone are in the same location and not separated. In this case, the first probability P i1 =0. If BSID t With BSID i If they are not equal, it means that at time t, the tag and the mobile phone are within the coverage area of ​​different base stations. In this case, the locations of all read / write devices up to the tag at time t can be inventoried. t Clustering is performed to obtain the center location of all read / write devices connected to the tag. This center location is used as the tag's location. Then, the relative distance between the tag and the base station where the mobile phone is located is calculated. , .when At this point, the tag and the phone can be considered to be in a separate state, and the first probability P is... i1 =1. If If the tag is considered to be close to the phone, then the first probability P is given. i1 equal .

[0049] In some embodiments, the second probability can be calculated based on the current position of the tag and the position of the tag within the first historical time period.

[0050] For example, the first historical time period can be determined based on the time period in which the current moment occurs. For instance, if the current moment is the b-th time period of today, the first historical time period could be the b-th time period of the previous day or the b-th time period of a day a week ago. When the current moment is time t, the time period in which the current moment occurs could be... The length of the time period in which the current moment occurs is .

[0051] Understandably, the location of the tag of the item to be managed within the first historical time period can be used as the basis for determining the expected trajectory of the item to be managed. Therefore, based on the location of the tag at the current moment and the location of the tag within the first historical time period, the probability that the location trajectory of the item to be managed is an unexpected trajectory can be determined relatively accurately, that is, the second probability can be determined relatively accurately.

[0052] In some embodiments, the implementation of calculating the second probability based on the position of the tag at the current moment and the position of the tag within the first historical time period can be carried out by clustering the positions of the tags within the first historical time period to obtain clusters; and the second probability can be calculated based on the second distance between the position of the tag at the current moment and the center position of the cluster.

[0053] For example, at time t, the location record set of the tag within the first historical time period can be obtained. Density clustering is then performed on the locations in this set. The cluster radius can be set based on the density of read / write devices at the corresponding locations within the first historical time period. If the read / write devices are densely deployed, the cluster radius is smaller; if the read / write device density coefficient (such as the base station density coefficient) is high, the cluster radius is larger. The minimum number of points included in the clustering algorithm can be set based on the total number of location records extracted from the historical locations. Multiple clusters are obtained through calculation. The cluster containing the most sample points is selected as the valid cluster. The coordinates of the center location of the cluster are LocH. j LocH j =(x j ,y j ).

[0054] Understandably, after clustering the positions of tags within the first historical time period to obtain clusters, the center position of the cluster can be used as the basis for determining the expected trajectory of the item to be managed. Therefore, based on the second distance between the current position of the tag and the center position of the cluster, the probability that the location trajectory of the item to be managed is an unexpected trajectory can be determined more accurately, that is, the second probability can be determined more accurately.

[0055] In some embodiments, the second probability can be calculated based on the second distance to determine the second probability. This can be achieved by calculating a third distance based on the second distance between the current label's position and the center of the cluster, and then calculating the second probability based on this third distance. The third distance is positively correlated with the second distance and the time decay coefficient, while the time decay coefficient is negatively correlated with the first time interval. The first time interval represents the time interval between the current moment and a historical moment, which is determined based on the historical time records corresponding to the cluster.

[0056] In this embodiment of the application, the time decay coefficient can be denoted as: Historical moments can be denoted as t. h For example, historical moment t h It could be the median of the time recorded at the location corresponding to a cluster. In one example, it could be the time decay coefficient at time t. .

[0057] It can be seen that, considering the timeliness of historical locations, the closer a historical location is to the current time, the greater its influence on the cluster center. Based on this, in the embodiments of this application, the third distance can be determined more reasonably by comprehensively considering the time interval between the current time and the historical time, as well as the second distance, and then the second probability can be determined more reasonably based on the third distance.

[0058] In some embodiments, when the third distance is less than or equal to a preset lower distance limit, the second probability can be determined as 0; when the third distance is greater than the preset lower distance limit and less than the preset upper distance limit, the second probability can be determined as a positive number less than 1; and when the third distance is greater than or equal to the preset upper distance limit, the second probability can be determined as 1.

[0059] Here, the preset lower distance limit and preset upper distance limit can be set according to actual needs.

[0060] Understandably, when the third distance is less than or equal to the preset lower distance limit, it can be considered that the current position of the tag is close to the center of the cluster, or the recording time of the cluster center is close to the current time. Therefore, the location trajectory of the item to be managed can be determined as the expected trajectory; that is, the probability that the location trajectory of the item to be managed is an unexpected trajectory can be accurately determined as 0. When the third distance is greater than or equal to the preset upper distance limit, it can be considered that the current position of the tag is far from the center of the cluster, or the recording time of the cluster center is far from the current time. Therefore, the location trajectory of the item to be managed can be determined as an unexpected trajectory; that is, the probability that the location trajectory of the item to be managed is an unexpected trajectory can be accurately determined as 1. When the third distance is greater than the preset lower distance limit and less than the preset upper distance limit, it can be considered that the location trajectory of the item to be managed may be either the expected trajectory or an unexpected trajectory. In this case, the probability that the location trajectory of the item to be managed is an unexpected trajectory can be reasonably determined as a positive number less than 1.

[0061] When the third distance is greater than a preset lower distance limit and less than a preset upper distance limit, one way to determine the second probability is to determine the second probability based on the ratio of the third distance to the preset upper distance limit. For example, the second probability is positively correlated with the ratio of the third distance to the preset upper distance limit.

[0062] It can be seen that the third distance is determined by comprehensively considering the time interval between the current moment and the historical moment, as well as the second distance. The second distance represents the distance between the current position of the label and the center position of the cluster reflecting the historical position set. Therefore, the ratio of the third distance to the preset distance upper limit can accurately reflect the possibility that the position trajectory of the item to be managed is an unexpected trajectory to a certain extent. Thus, based on the ratio of the third distance to the preset distance upper limit, the second probability can be determined relatively accurately.

[0063] In an exemplary application scenario, the current time is time t, and the third distance is denoted as . The preset lower limit of distance is denoted as The preset distance limit is denoted as . .when If the tag is not within the historical movement path range, a deviation may have occurred, and the location trajectory of the item to be managed is an unexpected trajectory. In this case, the second probability P... i2 =1. If If the label deviates from its historical trajectory, then the second probability P is considered to be... i2 equal .if If the label is within the historical movement path range and no abnormal deviation has occurred, then the second probability P is considered to be within that range. i2 =0. To improve the accuracy of calculating the second probability, multiple historical time points can be selected, and P can be calculated separately for each point. i2 And take the average of multiple second probabilities.

[0064] In some embodiments, the speed of the tag at the current moment and the speed confidence interval of the tag can be determined. When the speed of the tag at the current moment is within the speed confidence interval, the third probability is determined to be 0. When the speed of the tag at the current moment is not within the speed confidence interval, the third probability is determined to be 1.

[0065] Understandably, when the speed of the tag at the current moment is within the speed confidence interval, the reliability of the speed of the tag at the current moment can be considered high, and the movement speed of the item under management can be considered normal. Therefore, the probability that the movement speed of the item under management is abnormal can be determined to be 0 with relatively high accuracy, that is, the third probability can be determined to be 0 with relatively high accuracy. When the speed of the tag at the current moment is not within the speed confidence interval, the reliability of the speed of the tag at the current moment can be considered low, and the movement speed of the item under management can be considered abnormal. Therefore, the probability that the movement speed of the item under management is abnormal can be determined to be 1 with relatively high accuracy, that is, the third probability can be determined to be 1 with relatively high accuracy.

[0066] In some embodiments, the velocity confidence interval of a tag can be determined based on the tag's position sequence within a second historical time period; and the velocity confidence interval can be determined based on the mean, standard deviation, and skewness of the velocity sequence.

[0067] For example, the second historical time period can be determined based on the time when the tag was last inventoried. For instance, the second historical time period can be the time period during which the tag was inventoried n+1 times consecutively before the last inventory.

[0068] As can be seen, the embodiments of this application can determine the velocity sequence of the tag within the second historical time period, and reasonably determine the velocity confidence interval of the tag based on a comprehensive consideration of the mean, standard deviation and skewness of the tag velocity sequence.

[0069] For example, the platform can obtain the times of the tag's two most recent inventory checks, calculate the time difference between the two times, and calculate the tag's speed *v* at the current moment based on the relative distance and time difference between the reading and writing devices at the time of the two most recent inventory checks. The platform can also obtain the times of the tag's *n+1* consecutive inventory checks prior to the most recent inventory check. It can also obtain the position coordinate sequence of the tag during the n+1 consecutive inventory counts before the most recent inventory count. Then, based on the time of the tag being inventoried n+1 times consecutively, and position coordinate sequence Calculate the average speed between the i-th and (i+1)-th inventory checks during the process of the tag being checked n+1 times consecutively. Average speed It can be calculated using formula (1).

[0070] (1) in, This represents the time of the (i+1)th inventory check during a series of (n+1) consecutive inventory checks of the tag. This represents the time of the i-th inventory count during the process of the label being counted n+1 times consecutively. This represents the coordinates of the (i+1)th time the tag is counted during a series of (n+1) consecutive counts. This represents the coordinates of the i-th item being inventoried during the process of the label being inventoried n+1 times consecutively.

[0071] Construct a speed sequence based on the average speed of two consecutive inventory checks during the process of the tag being checked n+1 times consecutively. Calculate the mean of the velocity sequence. Standard deviation Skewing S, exemplarily, is the mean of a velocity sequence. Standard deviation The skewness S can be calculated using formulas (2) to (4). (2) (3) (4) The platform calculates the speed confidence interval of the tags based on relevant data [v] min ,v max ].

[0072] In the first example, when the value of n is large (e.g., n is greater than or equal to 30) and the absolute value of skewness S is small (e.g., ... When the value is less than or equal to 0.5, the overall speed of the tags approximately follows a normal distribution. , ,in, The quantiles of the standard normal distribution are calculated using a confidence level of at least 95% (confidence levels can be set as needed). This allows us to determine the velocity confidence interval.

[0073] In the second example, when the value of n is small (e.g. And the absolute value of the skewness S is slightly large (e.g. When the label's velocity follows a t-distribution, in this case, , .in, For the quantiles of the t-distribution with n-1 degrees of freedom, calculate the speed confidence interval at a confidence level of not less than 95%.

[0074] In the third example, when the absolute value of the skewness S is significantly larger (e.g., When the velocity sequence follows a skewed distribution, the velocity sequence is sorted from smallest to largest to form a new ordered velocity sequence. Calculate the median vm of the ordered velocity sequence, vm = v0 (n+1) / 2 Or, vm=(vo n / 2 +vo n / 2+1 ) / 2. At this point, a confidence level of at least 95% can be used to calculate the velocity confidence interval, that is, the median vm of the ordered velocity sequence can be placed at v0. k1 and vo k2 The probability between them is denoted as .exist In the case of vo, calculate k1 and vo k2 ,make , , where k1 and k2 are the corresponding position numbers in the ordered velocity sequence.

[0075] After calculating the velocity confidence interval, the velocity confidence interval is used as the decision threshold. When At this point, it can be assumed that the label speed has not changed abnormally, and the third probability P i3 =0, otherwise, the third probability P i3 =1.

[0076] In some embodiments, the fourth probability can be calculated by determining the target base station to which the tag belongs at each inventory based on the tag's multiple inventory information; determining the associated tag set based on each tag within the coverage area of ​​the target base station to which the tag belongs at each inventory; determining the first tag set based on each tag within the coverage area of ​​the base station to which the tag belongs at the most recent inventory; and calculating the fourth probability based on the first tag set and the associated tag set.

[0077] In practical applications, when tags are inventoried multiple times, the target base station to which the tag belongs at each inventory can be determined. This can be understood as the target base station being the same or different base stations across multiple inventories. At each inventory, a base station coverage tag list is constructed based on the tags within the coverage area of ​​the target base station. These base station coverage tag lists from multiple inventory times are combined to obtain a combined tag list, which can then be used to generate an associated tag set. Duplicate tag identifiers are allowed in the combined tag list. In one example, the combined tag list can be directly used as the associated tag set; in another example, a subset of tags can be selected from the combined tag list to form the associated tag set.

[0078] It can be seen that the tags within the coverage area of ​​the base station to which the tags belong during the most recent tag inventory belong to the same base station and are relatively close to each other. Therefore, by analyzing the tags within the coverage area of ​​the base station to which the tags belong during the most recent tag inventory, as well as the associated tag set, the probability of a tag leaving the associated tag set can be reasonably determined.

[0079] In some embodiments, the tag set for determining the associated tag set can be determined from the tags within the coverage area of ​​the target base station to which the tag belongs during each tag inventory. The tag set whose inventory count is greater than the set count is then determined as the associated tag set.

[0080] For example, when each tag is inventoried, the tags of the items to be managed can be filtered out from the tags within the coverage area of ​​the target base station to which the tag belongs. Then, from the filtered tags, the set of tags whose inventory counts exceed a set number is determined as the associated tag set. That is, the managed tag set does not include the tags of the items to be managed.

[0081] For example, in the combined tag list, the tags of the items to be managed can be filtered first, and then the number of times each tag's identifier appears in the filtered tag list can be determined. Tags that appear more than a set number of times can be filtered out, and the filtered tags can be combined into an associated tag set.

[0082] It can be seen that the tags in the associated tag set are those that have been inventoried more than the set number of times. Therefore, it can be considered that the tags in the associated tag set and the tags of the items to be managed are frequently inventoried. In this way, it can be determined to a certain extent that the tags in the associated tag set and the tags of the items to be managed are more related. Thus, the embodiments of this application can more accurately determine the associated tag set.

[0083] In some embodiments, the intersection of the first tag set and the associated tag set can be determined; the fourth probability is calculated based on the first number of tags in the intersection, and the fourth probability is negatively correlated with the first number.

[0084] Understandably, when the number of tags in the intersection of the first tag set and the associated tag set is small, it can be assumed that the tags of the items to be managed are more likely to be removed from the associated tag set. Therefore, based on the first number of tags in the intersection of the first tag set and the associated tag set, the probability of a tag being removed from the associated tag set can be determined more accurately, that is, the fourth probability can be determined more accurately.

[0085] In an exemplary application scenario, the platform determines the fourth probability by performing correlation analysis on the location and status data between different tags and identifying anomalies by combining historical behavior patterns. First, the platform can determine the fourth probability based on the TagID of the i-th tag being inventoried. i Establish a list of relationships, for those identified by TagID i The tag can be used to determine the base station to which the tag belongs during the first to m-th inventory checks. Based on the identifiers of all tags within the coverage area of ​​the base station that checks the tag each time, a base station coverage tag list is constructed. For tags in the base station coverage tag list, the TagID can be removed. i The updated base station coverage label list is obtained, and then an associated label set is generated based on the updated base station coverage label list. The TagID of the labels in the base station coverage label list can also be retained. i .

[0086] When i ranges from 1 to m, the base station coverage tag list constructed from the identifiers of all tags within the coverage area of ​​the base station for the i-th inventory can be denoted as M. i M i ={BSID i ,{TagID x}},{TagID x} represents the set of identifiers for all tags within the coverage area of ​​the base station where the tag is inventoried for the i-th time. Let M1 to M... m Combining tags yields a combined tag list M, where duplicate tags are allowed. Calculate the frequency C of each tag in M ​​(i.e., the number of times the corresponding tag has been counted). x Tags that appear more than a set number C0 times are grouped into an associated tag set G, where G = {TagID} x |C x >=C0}, TagIDx ∈M.

[0087] For those identified by TagID i The tag can be used to determine the base station that performs the inventory of the tag in the (m+1)th inventory. Based on the identifiers of all tags within the coverage area of ​​the base station that performs the inventory of the tag in the (m+1)th inventory, a first tag set G1 is constructed, and the number of tags in the first tag set G1 can be denoted as C1. For the identifiers of the tags in the first tag set G1, the TagID can be removed. i This yields the updated first tag set G1. For the tags in the first tag set G1, the TagID can also be retained. i For example, the identifier of a tag in the base station coverage tag list does not include TagID. i At that time, the first tag set G1 also does not include TagID. i The identifier of a tag in the base station coverage tag list includes TagID. i At that time, the first tag set G1 also includes TagID. i .

[0088] After determining the associated label set G and the first label set G1, the fourth probability P can be calculated according to formula (5). i4 .

[0089] (5) in, This indicates the number of labels in G. This represents the number of labels in the intersection of G and G1.

[0090] In some embodiments, the probability of a tag being in an abnormal state can be obtained by weighted summation of the first, second, third, and fourth probabilities based on one or more of the first, second, third, and fourth probabilities.

[0091] This application proposes a method for early warning of lost items in cellular passive IoT. The method sets anomaly probability calculation weight based on the user's behavioral characteristics in different application scenarios, and calculates the weighted comprehensive probability of an item's abnormal condition by combining user behavioral characteristics in different scenarios and tag information collected by reading and writing devices, thereby determining in advance whether the item is likely to be lost and promptly alerting the user.

[0092] For example, corresponding weights can be set for the abnormal behaviors shown in Table 1. It should be noted that the abnormal behaviors are not limited to the behaviors recorded in Table 1, and can be expanded as needed in practical applications.

[0093] Table 1

[0094] For example, the weight of the first probability is W1, the weight of the second probability is W2, the weight of the third probability is W3, and the weight of the fourth probability is W4, according to the formula... The probability that a tag is in an abnormal state can be calculated. .

[0095] As can be seen, the embodiments of this application can determine the probability of a tag being in an abnormal state relatively accurately and reasonably by comprehensively considering the probability that the tag and the associated device of the item to be managed are in a separate state, the probability that the location trajectory of the item to be managed is an unexpected trajectory, the probability that the moving speed of the item to be managed is abnormal, and the probability that the tag is separated from the associated tag set.

[0096] In some embodiments, the weight of each of the first, second, third, and fourth probabilities can be determined based on the user's behavioral characteristics in the target application scenario.

[0097] For example, based on dimensions such as activity area and movement speed, target application scenarios can be divided into the following three categories: fixed-stay scenarios, public activity scenarios, and transportation scenarios. Fixed-stay scenarios include office buildings, hotels, etc.; public activity scenarios include shopping malls, supermarkets, hospitals, etc.; and transportation scenarios include buses, subways, etc.

[0098] For example, a weight list as shown in Table 2 can be established based on the characteristics of fixed-stay scenarios, public activity scenarios, and transportation scenarios.

[0099] Table 2

[0100] In fixed-station scenarios, users' mobile phones are less likely to be separated from the items to be managed, and the changes in movement speed are relatively small. The movement trajectory will show a certain regularity, and the set of associated tags is relatively stable. Therefore, the same weight can be set for the first probability, the second probability, the third probability, and the fourth probability.

[0101] In public activity scenarios, users' phones are rarely separated from the items to be managed, and the speed of the managed items is relatively stable. Therefore, W1 and W3 are set to larger values. However, in this scenario, the movement of items exhibits strong randomness, and the group variation of the associated tag set is significant. Therefore, W2 and W4 need to be set to smaller values.

[0102] In transportation scenarios, users' mobile phones are rarely separated from the items to be managed, and users' movements show strong regularity. Therefore, W1 and W2 should be set to larger values. However, the speed of the items to be managed and the group of associated tag sets may vary greatly, so W3 and W4 should be set to smaller values.

[0103] It can be seen that the behavioral characteristics of users in the target application scenario can reflect the importance of various abnormal behaviors to a certain extent. Therefore, the embodiments of this application can determine the probability weights of various abnormal behaviors more reasonably based on the behavioral characteristics of users in the target application scenario. That is, it can more reasonably determine the probability weight of the tag being separated from the associated device of the item to be managed, the probability weight of the location trajectory of the item to be managed being an unexpected trajectory, the probability weight of the movement speed of the item to be managed being abnormal, and the probability weight of the tag of the item to be managed being separated from the associated tag set.

[0104] In this embodiment of the application, the weight of the probability of each abnormal behavior can also be updated. In some embodiments, historical state data of the items to be managed can be obtained, and the historical state data is used to record the state of the items to be managed within a third historical time period; then, the weight of each probability among the first probability, the second probability, the third probability, and the fourth probability is updated according to the number of times the normal state of the items to be managed occurs in the historical state data, and the total number of times the normal state and the abnormal state of the items to be managed occur in the historical state data.

[0105] It can be seen that the number of times the normal state of the items to be managed appears in the historical state data, as well as the total number of times the normal and abnormal states of the items to be managed appear in the historical state data, can reflect the possibility of the items to be managed being abnormal in the application scenario to a certain extent. Therefore, based on the number of times the normal state of the items to be managed appears in the historical state data, as well as the total number of times the normal and abnormal states of the items to be managed appear in the historical state data, it is helpful to accurately update the weights of various probabilities of the tags of the items to be managed being in abnormal states.

[0106] In this embodiment, the weighting of abnormal behavior needs to be determined by considering the risk level, frequency of occurrence, and scope of impact of the event. A "basic weight + dynamic correction" model is adopted, based on the application scenario and historical behavior, to ensure that the weight allocation matches the actual application scenario. For example, in... Under the constraints, a weight list as shown in Table 2 can be established for application scenarios and historical behaviors.

[0107] For example, after establishing the weight list shown in Table 2, the platform can adjust the weight of each probability by combining the historical status data of the items to be managed, where i takes values ​​from 1 to 4 in sequence. The modified weight can be denoted as The formula for updating the weight of each probability is Equation (6).

[0108] (6) in, This indicates the number of times the normal state of the item to be managed occurs in the historical status data. This represents the total number of times the normal and abnormal states of the items to be managed occurred in the historical status data.

[0109] In this embodiment of the application, the anomaly probability threshold can be denoted as P0. For example, if the label is in an abnormal state, the probability... If the value is greater than or equal to P0, an abnormal notification message can be generated, and a warning message can be sent to the user's APP (or other mobile phone number reserved by the user) to which the item to be managed is bound, so that the user can take appropriate measures in a timely manner.

[0110] In this embodiment, the initial anomaly probability threshold can be set to 50%. When determining whether the status of the managed item is abnormal based on the anomaly probability threshold set at the initial time, there may be a high false positive rate and a low false negative rate. In this case, the anomaly probability threshold can be updated. The platform can update the anomaly probability threshold based on historical data statistical analysis and combined with the actual user usage scenario. In some embodiments, the status of the managed item can be recorded after each determination of the status of the managed item; then, the anomaly probability threshold can be updated based on the total number of states of the managed item in the fourth historical time period and the first time the state of the managed item was incorrect in the fourth historical time period.

[0111] As can be seen, the embodiments of this application can reasonably adjust the anomaly probability threshold based on the total number of states of the items to be managed in the fourth historical time period and the first number of times the state of the items to be managed in the fourth historical time period is incorrect.

[0112] In some embodiments, when the total number of managed items in the first and fourth historical time periods exceeds a proportional threshold, the current anomaly probability threshold is incremented to obtain the updated anomaly probability threshold.

[0113] For example, for fixed-stay scenarios, public activity scenarios, and transportation scenarios, the number of false alarms N for whether an item is lost within a certain period (such as one day) can be counted by combining actual user feedback and subsequent data analysis. f And the total number of times N determines whether an item is lost. t N fN represents the first instance of an error occurring in the status of managed items within the historical timeframe of the fourth region. t The total number of items to be managed. Calculate the false positive rate F, F = N. f / N t When the false positive rate F is greater than or equal to 10%, the anomaly probability threshold P0 is increased by 1 / N. t For the anomaly probability threshold, the update process can be repeated continuously until the false positive rate F is less than 10%.

[0114] It can be seen that when the total number of managed items in the first and fourth historical time periods exceeds the proportion threshold, it indicates that the number of times the managed items' status is incorrect in the fourth historical time period. At this time, by incrementing the current anomaly probability threshold, the anomaly probability threshold can be increased. When judging the status of managed items based on the increased anomaly probability threshold, it helps to reduce the number of times the status of managed items is incorrect in the future.

[0115] In this embodiment of the application, considering that the number of tags deployed in actual application scenarios is large, only the tags that meet the specific needs, specific types, specific objects, etc. that users care about can be analyzed. For example, data analysis can be performed only on the tags of objects that a user needs to pay attention to (such as ID cards, keys, etc.), instead of analyzing all the tags of the user, thereby reducing network communication and computing overhead.

[0116] Figure 2 This is an interactive flowchart illustrating the implementation of a lost item alert in this application embodiment, such as... Figure 2 As shown, the process includes: Step 21: The APP sends a tag registration request to the platform.

[0117] Here, after obtaining a tag, users complete tag registration through the APP, bind the tag to their mobile phone number, and activate the tag.

[0118] Step 22: The reading and writing device periodically performs tag inventory.

[0119] The reader / writer periodically inventories the tags within its coverage area, and the tags can report their own information to the reader / writer. For example, the reader / writer can be a base station or a user equipment (UE) such as a mobile phone.

[0120] Step 23: The reader / writer receives the tag information.

[0121] Step 24: The reading and writing device reports the tag information to the core network.

[0122] In this step, the reading and writing device summarizes and reports all tag information within its coverage area to the core network.

[0123] Step 25: The core network synchronizes information with the platform.

[0124] The implementation method for this step is as follows: The core network aggregates the information reported by each read / write device, searches for the corresponding mobile phone number in the registration information based on the tag ID, and retrieves the base station information and mobile phone location at the corresponding time. Simultaneously, it retrieves the base station information where the tag is located at the corresponding time, as well as the location of the read / write device, and synchronously sends the retrieved information to the platform.

[0125] Step 26: The platform determines whether the status of the item to be managed is abnormal.

[0126] The implementation method of this step is as follows: the platform calculates the information synchronized by the core network to obtain the probability that the tag is in an abnormal state, and then determines whether the status of the item to be managed is abnormal based on the probability that the tag is in an abnormal state.

[0127] Step 27: Send an abnormal warning message to the APP.

[0128] The implementation method of this step is as follows: When the platform determines that the status of the item to be managed is abnormal, it sends an abnormal warning message to the user's registered mobile APP to remind the user that the item to be managed may have been lost, so that the user can retrieve it in time.

[0129] Step 28: The platform regularly updates the weight of each probability and the threshold for anomalies.

[0130] The implementation method for this step has been explained in the aforementioned content.

[0131] Figure 3 This is another flowchart illustrating the lost item warning method according to an embodiment of this application, such as... Figure 3As shown, the core network can synchronize information to the platform's information receiving module (finding and sending related information to the information receiving module). The platform's information receiving module receives the synchronized information and forwards it to the first probability calculation module, the second probability calculation module, the third probability calculation module, and the fourth probability calculation module (which can be expanded to more similar calculation modules). These modules calculate various probabilities in parallel and extract feature information from different dimensions. The first, second, third, and fourth probabilities are calculated using these modules respectively. Based on these probabilities, the probability of a tag being in an abnormal state can be calculated. The platform determines whether the managed item is in an abnormal state based on the probability of the tag being in an abnormal state and the abnormal probability threshold. If the managed item is in a normal state, the process returns to the platform's information receiving module to receive and forward the synchronized information. If the managed item is in an abnormal state, an APP alert is triggered, reporting the abnormal situation to the relevant user or management terminal and initiating a reminder to the user. The platform regularly adjusts the weight of each probability and the threshold for anomalies based on historical data and user feedback.

[0132] In summary, this application proposes a method for early warning of lost items in cellular passive IoT. Based on the abnormal behaviors of tags in different scenarios (including but not limited to the four types of abnormal behaviors shown in Table 1) set by the platform, and based on reported tag inventory information, the platform calculates the probability of each abnormal behavior by the tag, including information such as the phone's location, tag location, relative distance between the tag and the phone, the tag's historical movement status, tag movement speed, and associated tag set. The calculated probabilities are then aggregated and weighted to obtain the probability that the tag is in an abnormal state, and timely alerts can be sent to the user. In this application, based on the characteristics of different application scenarios and combined with the risk level, frequency of occurrence, and scope of impact of events, a "basic weight + dynamic correction" model matching the weight values ​​is established based on the application scenario and historical behavior. This dynamically adjusts the weight of each probability and the abnormal probability threshold in different application scenarios to improve the accuracy of the judgment. This application also proposes an interaction process and platform data processing process for the method of early warning of lost items based on cellular passive IoT. By adding a platform information processing step to the existing cellular passive IoT communication process, the platform performs data processing and modeling to achieve anomaly judgment and user alerts.

[0133] Compared with the solutions of related technologies, the embodiments of this application have at least the following advantages: 1) Data analysis can be performed using historical tag information to build a probability calculation model for tag abnormal states. By combining historical and real-time tag state information, the probability that the tag is in an abnormal state at the current moment can be calculated, and users can be promptly alerted to potential anomalies. This can prevent the loss of items due to users not noticing them for a long time, thereby improving the value of cellular passive IoT network services.

[0134] 2) This application proposes various probability weights and abnormal probability threshold calculation methods, which can combine the characteristics of the application scenario and the characteristics of the user's historical behavior to form a "basic weight + dynamic correction" model, realize the distinction and judgment of abnormal behavior in different application scenarios, and dynamically adjust the parameters to match the environment and the user's real behavior.

[0135] 3) No modifications are required on the tag side or the reader / writer side, and it is compatible with cellular passive IoT air interface protocols.

[0136] 4) The core network only needs to search and match the tag information and mobile phone information. The model building and calculation will be carried out on the platform side, without excessively increasing the burden on the core network.

[0137] 5) The platform only needs to collect, process and analyze data related to tags and mobile phones through software updates. No hardware modification is required. It only needs to expand on the basis of existing software functions.

[0138] Passive Internet of Things (IoT) technology has garnered widespread attention in the industry, particularly the integration of passive IoT with cellular networks, which represents a crucial future development area. This integration can provide end-to-end, network-wide item management capabilities for personal user applications seeking lost persons or items. Current solutions such as GPS and Bluetooth suffer from limitations in application environments, short communication distances, short terminal standby times, and a lack of early warning mechanisms. Therefore, adopting cellular passive IoT technology for managing massive amounts of personal belongings is highly feasible. This application proposes an item loss warning method based on cellular passive IoT, widely applicable to various indoor and outdoor environments. It provides users with cellular network-based item management warning services, helping them promptly detect abnormal states of managed items and facilitating timely retrieval, demonstrating broad market application prospects.

[0139] Those skilled in the art will understand that, in the above-described method of the specific implementation, the order in which each step is written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0140] Based on the item loss warning method proposed in the foregoing embodiments, this application also proposes an item loss warning device. Figure 4 This is a schematic diagram of the structure of the lost item warning device according to an embodiment of this application, as shown below. Figure 4 As shown, the device includes: The first processing module 401 is used to determine the probability that the tag is in an abnormal state based on a pre-built tag abnormal state probability calculation model, combined with the historical and real-time state information of the tags of the items to be managed. The second processing module 402 is used to determine whether the status of the item to be managed is abnormal based on the probability that the tag is in an abnormal state; and to issue an abnormal prompt message if the status of the item to be managed is abnormal.

[0141] In some embodiments, the first processing module 401 determines the probability that the tag is in an abnormal state based on a pre-built tag abnormal state probability calculation model and by combining the historical and real-time state information of the tags of the item to be managed, including: Based on the aforementioned tag abnormal state probability calculation model, and combining the historical and real-time state information of the tags of the items to be managed, one or more of the following probabilities are calculated: a first probability, a second probability, a third probability, and a fourth probability. The first probability represents the probability that the tag is separated from the associated device of the item to be managed; the second probability represents the probability that the location trajectory of the item to be managed is an unexpected trajectory; the third probability represents the probability that the moving speed of the item to be managed is abnormal; and the fourth probability represents the probability that the tag is separated from the set of associated tags of the tag. The probability that the tag is in an abnormal state is determined based on one or more of the first probability, the second probability, the third probability, and the fourth probability.

[0142] In some embodiments, the first processing module 401 is specifically used to calculate the first probability based on a first distance between the tag and the associated device of the item to be managed.

[0143] In some embodiments, the first processing module 401 calculates the first probability based on a first distance between the tag and the associated device of the item to be managed, including: When the first distance is equal to 0, the first probability is determined to be 0; when the first distance is greater than 0 and less than a preset first distance threshold, the first probability is determined to be a positive number less than 1; when the first distance is greater than or equal to the first distance threshold, the first probability is determined to be equal to 1.

[0144] In some embodiments, the first processing module 401 is specifically used to determine the first probability based on the ratio of the first distance to the first distance threshold.

[0145] In some embodiments, the associated device is a mobile terminal; the first processing module 401 is further configured to determine the relative distance between the read / write device corresponding to the tag and the base station where the associated device is located as the first distance.

[0146] In some embodiments, the first processing module 401 is specifically used to calculate the second probability based on the position of the tag at the current time and the position of the tag within a first historical time period.

[0147] In some embodiments, the first processing module 401 calculates the second probability based on the current position of the tag and the position of the tag within the first historical time period, including: Cluster the locations of the tags within the first historical time period to obtain clusters; The second probability is calculated based on the second distance between the current position of the label and the center position of the cluster.

[0148] In some embodiments, the first processing module 401 calculates the second probability based on the second distance between the current position of the label and the center position of the cluster, including: Based on the second distance between the current position of the label and the center position of the cluster, a third distance is calculated. The third distance is positively correlated with the second distance and the time decay coefficient. The time decay coefficient is negatively correlated with the first time interval. The first time interval represents the time interval between the current moment and the historical moment. The historical moment is determined based on the historical time corresponding to the cluster. The second probability is calculated based on the third distance.

[0149] In some embodiments, the first processing module 401 calculates the second probability based on the third distance, including: When the third distance is less than or equal to the preset lower distance limit, the second probability is determined to be 0; when the third distance is greater than the preset lower distance limit and less than the preset upper distance limit, the second probability is determined to be a positive number less than 1; when the third distance is greater than or equal to the preset upper distance limit, the second probability is determined to be equal to 1.

[0150] In some embodiments, the first processing module 401 is specifically used to determine the second probability based on the ratio of the third distance to the preset distance upper limit.

[0151] In some embodiments, the first processing module 401 is specifically used to determine the speed of the tag at the current moment and the speed confidence interval of the tag; when the speed of the tag at the current moment is within the speed confidence interval, the third probability is determined to be 0; when the speed of the tag at the current moment is not within the speed confidence interval, the third probability is determined to be 1.

[0152] In some embodiments, the first processing module 401 is specifically used to determine the velocity sequence of the tag based on the position sequence of the tag within a second historical time period; and to determine the velocity confidence interval of the tag based on the mean, standard deviation, and skewness of the velocity sequence.

[0153] In some embodiments, the first processing module 401 is specifically configured to: determine the target base station to which the tag belongs at each tag inventory based on the tag's multiple inventory information; determine the associated tag set based on each tag within the coverage area of ​​the target base station to which the tag belongs at each tag inventory; determine the first tag set based on each tag within the coverage area of ​​the base station to which the tag belongs at the most recent tag inventory; and calculate the fourth probability based on the first tag set and the associated tag set.

[0154] In some embodiments, the first processing module 401 determines the associated tag set based on each tag within the coverage area of ​​the target base station to which the tag belongs during each tag inventory, including: determining, among each tag within the coverage area of ​​the target base station to which the tag belongs during each tag inventory, tags that have been inventoried more than a set number of times; and determining the set of tags that have been inventoried more than a set number of times as the associated tag set.

[0155] In some embodiments, the first processing module 401 calculates the fourth probability based on the first tag set and the associated tag set, including: determining the intersection of the first tag set and the associated tag set; and calculating the fourth probability based on a first number of tags in the intersection, wherein the fourth probability is negatively correlated with the first number.

[0156] In some embodiments, the first processing module 401 determines the probability that the tag is in an abnormal state based on one or more of the first probability, the second probability, the third probability, and the fourth probability, including: performing a weighted summation of the first probability, the second probability, the third probability, and the fourth probability to obtain the probability that the tag is in an abnormal state.

[0157] In some embodiments, the first processing module 401 is further configured to determine the weight of each of the first probability, the second probability, the third probability, and the fourth probability based on the user's behavioral characteristics in the target application scenario.

[0158] In some embodiments, the first processing module 401 is further configured to acquire historical status data of the item to be managed, wherein the historical status data is used to record the status of the item to be managed within a third historical time period. The first processing module 401 is further configured to update the weight of each of the first probability, the second probability, the third probability, and the fourth probability based on the number of times the normal state of the item to be managed appears in the historical state data and the total number of times the normal state and the abnormal state of the item to be managed appears in the historical state data.

[0159] In some embodiments, the second processing module 402 determines whether the state of the item to be managed is abnormal based on the probability that the tag is in an abnormal state, including: when the probability that the tag is in an abnormal state is greater than or equal to an abnormal probability threshold, determining that the state of the item to be managed is abnormal.

[0160] In some embodiments, the second processing module 402 is further configured to record the status of the item to be managed after each determination of the status of the item to be managed; and update the anomaly probability threshold based on the total number of statuses of the item to be managed within the fourth historical time period and the first number of times the status of the item to be managed was incorrect within the fourth historical time period.

[0161] In some embodiments, the second processing module 402 is specifically used to increment the current anomaly probability threshold to obtain the updated anomaly probability threshold when the total number of states of the items to be managed within the first number of times and the fourth historical time period is greater than the proportion threshold.

[0162] In practical applications, the first processing module 401 and the second processing module 402 can be implemented based on a processor and a communication device.

[0163] It should be noted that the description of the above device embodiments is similar to the description of the above method embodiments, and has similar beneficial effects. For technical details not disclosed in the device embodiments of this application, please refer to the description of the method embodiments of this application for understanding.

[0164] It should be noted that, in the embodiments of this application, if the above-described methods are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a terminal, server, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0165] Correspondingly, this application embodiment further provides a computer program product, the computer program product including computer executable instructions, which are used to implement any of the item loss warning methods provided in this application embodiment.

[0166] Accordingly, this application embodiment further provides a computer storage medium storing computer-executable instructions, which are used to implement any of the item loss warning methods provided in the above embodiments.

[0167] This application also provides an electronic device. Figure 5 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of this application, as shown below. Figure 5 As shown, the electronic device 50 may include: Memory 501 is used to store executable instructions; The processor 502 is used to implement any of the above-mentioned congestion control methods when executing executable instructions stored in the memory 501.

[0168] The processor 502 mentioned above can be at least one of ASIC, DSP, DSPD, PLD, FPGA, CPU, controller, microcontroller, and microprocessor.

[0169] The aforementioned computer-readable storage medium and memory 501 may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM), etc.; or it may be various terminals that include one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.

[0170] In some embodiments, the functions or modules of the apparatus provided in this application can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0171] The description of the various embodiments above tends to emphasize the differences between the various embodiments. The similarities or similarities between them can be referred to, and for the sake of brevity, they will not be repeated here.

[0172] The methods disclosed in the various method embodiments provided in this application can be arbitrarily combined to obtain new method embodiments without conflict.

[0173] The features disclosed in the various product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0174] The features disclosed in the various method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0175] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0176] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims. All of these forms are within the protection scope of this application.

Claims

1. A method for issuing an early warning for lost items, characterized in that, The method includes: Based on a pre-built probability calculation model for abnormal tag states, the probability of a tag being in an abnormal state is determined by combining the historical and real-time status information of the tags of the items to be managed. Based on the probability that the tag is in an abnormal state, determine whether the state of the item to be managed is abnormal. If the status of the item to be managed is abnormal, an abnormal prompt message will be issued.

2. The method according to claim 1, characterized in that, The pre-built tag abnormal state probability calculation model, which combines historical and real-time status information of the tags of the items to be managed, determines the probability that the tag is in an abnormal state, including: Based on the aforementioned tag abnormal state probability calculation model, and combining the historical and real-time state information of the tags of the items to be managed, one or more of the following probabilities are calculated: a first probability, a second probability, a third probability, and a fourth probability. The first probability represents the probability that the tag is separated from the associated device of the item to be managed; the second probability represents the probability that the location trajectory of the item to be managed is an unexpected trajectory; the third probability represents the probability that the moving speed of the item to be managed is abnormal; and the fourth probability represents the probability that the tag is separated from the set of associated tags of the tag. The probability that the tag is in an abnormal state is determined based on one or more of the first probability, the second probability, the third probability, and the fourth probability.

3. The method according to claim 2, characterized in that, Calculating the first probability includes: The first probability is calculated based on the first distance between the tag and the associated device of the item to be managed.

4. The method according to claim 3, characterized in that, The calculation of the first probability based on the first distance between the tag and the associated device of the item to be managed includes: When the first distance is equal to 0, the first probability is determined to be 0; when the first distance is greater than 0 and less than a preset first distance threshold, the first probability is determined to be a positive number less than 1; when the first distance is greater than or equal to the first distance threshold, the first probability is determined to be equal to 1.

5. The method according to claim 4, characterized in that, The step of determining that the first probability is a positive number less than 1 when the first distance is less than a preset first distance threshold includes: The first probability is determined based on the ratio of the first distance to the first distance threshold.

6. The method according to claim 3, characterized in that, The associated device is a mobile terminal; the method further includes: determining the relative distance between the reader / writer device corresponding to the tag and the base station where the associated device is located as the first distance.

7. The method according to claim 2, characterized in that, Calculating the second probability includes: The second probability is calculated based on the current position of the tag and the position of the tag within the first historical time period.

8. The method according to claim 7, characterized in that, The calculation of the second probability based on the current position of the tag and the position of the tag within the first historical time period includes: Cluster the locations of the tags within the first historical time period to obtain clusters; The second probability is calculated based on the second distance between the current position of the label and the center position of the cluster.

9. The method according to claim 8, characterized in that, The calculation of the second probability based on the second distance between the current position of the label and the center position of the cluster includes: Based on the second distance between the current position of the label and the center position of the cluster, a third distance is calculated. The third distance is positively correlated with the second distance and the time decay coefficient. The time decay coefficient is negatively correlated with the first time interval. The first time interval represents the time interval between the current moment and the historical moment. The historical moment is determined based on the historical time corresponding to the cluster. The second probability is calculated based on the third distance.

10. The method according to claim 9, characterized in that, The calculation of the second probability based on the third distance includes: When the third distance is less than or equal to the preset lower distance limit, the second probability is determined to be 0; when the third distance is greater than the preset lower distance limit and less than the preset upper distance limit, the second probability is determined to be a positive number less than 1; when the third distance is greater than or equal to the preset upper distance limit, the second probability is determined to be equal to 1.

11. The method according to claim 10, characterized in that, When the third distance is greater than a preset lower distance limit and less than a preset upper distance limit, determining that the second probability is a positive number less than 1 includes: The second probability is determined based on the ratio of the third distance to the preset upper limit of distance.

12. The method according to claim 2, characterized in that, Calculating the third probability includes: Determine the current velocity of the tag and the velocity confidence interval of the tag; When the speed of the tag at the current moment is within the speed confidence interval, the third probability is determined to be 0; when the speed of the tag at the current moment is not within the speed confidence interval, the third probability is determined to be 1.

13. The method according to claim 12, characterized in that, Determining the velocity confidence interval of the label includes: The velocity sequence of the tags is determined based on the position sequence of the tags within the second historical time period; Based on the mean, standard deviation, and skewness of the velocity sequence, the velocity confidence interval of the label is determined.

14. The method according to claim 2, characterized in that, Calculating the fourth probability includes: Based on the multiple inventory information of the tags, the target base station to which the tags belong at each inventory time is determined; The associated tag set is determined based on each tag within the coverage area of ​​the target base station to which the tag belongs at each tag inventory; the first tag set is determined based on each tag within the coverage area of ​​the base station to which the tag belongs at the most recent tag inventory. The fourth probability is calculated based on the first tag set and the associated tag set.

15. The method according to claim 14, characterized in that, The process of determining the associated tag set based on each tag within the coverage area of ​​the target base station to which the tag belongs during each tag inventory includes: During each tag inventory, among the tags within the coverage area of ​​the target base station to which the tag belongs, identify the tags that have been inventoried more than a set number of times; The set of tags that have been inventoried more than a set number of times is determined as the associated tag set.

16. The method according to claim 15, characterized in that, The calculation of the fourth probability based on the first tag set and the associated tag set includes: Determine the intersection of the first tag set and the associated tag set; Based on the first number of labels in the intersection, the fourth probability is calculated, and the fourth probability is negatively correlated with the first number.

17. The method according to claim 2, characterized in that, Determining the probability that the tag is in an abnormal state based on one or more of the first probability, second probability, third probability, and fourth probability includes: The probability that the label is in an abnormal state is obtained by weighted summation of the first probability, the second probability, the third probability and the fourth probability.

18. The method according to claim 17, characterized in that, The method further includes: Based on the user's behavioral characteristics in the target application scenario, determine the weight of each of the first, second, third, and fourth probabilities.

19. The method according to claim 18, characterized in that, The method further includes: Obtain historical status data of the item to be managed, wherein the historical status data is used to record the status of the item to be managed within a third historical time period; Based on the number of times the normal state of the item to be managed appears in the historical state data, and the total number of times the normal and abnormal states of the item to be managed appear in the historical state data, the weight of each probability among the first probability, the second probability, the third probability, and the fourth probability is updated.

20. The method according to claim 1, characterized in that, The step of determining whether the status of the item to be managed is abnormal based on the probability that the tag is in an abnormal state includes: When the probability that the label is in an abnormal state is greater than or equal to the abnormal probability threshold, the state of the item to be managed is determined to be abnormal.

21. The method according to claim 20, characterized in that, The method further includes: After each determination of the status of the item to be managed, the status of the item to be managed is recorded; The anomaly probability threshold is updated based on the total number of states of the items to be managed within the fourth historical time period and the first time an error occurred in the state of the items to be managed within the fourth historical time period.

22. The method according to claim 21, characterized in that, The step of updating the anomaly probability threshold based on the total number of managed items in the fourth historical time period and the first occurrence of an error in the managed item's status in the fourth historical time period includes: When the total number of states of the items to be managed within the first number of times and the fourth historical time period is greater than the proportion threshold, the current anomaly probability threshold is incremented to obtain the updated anomaly probability threshold.

23. A lost item warning device, characterized in that, The device includes: The first processing module is used to determine the probability that the tag is in an abnormal state based on a pre-built tag abnormal state probability calculation model, combined with the historical and real-time state information of the tags of the items to be managed. The second processing module is used to determine whether the status of the item to be managed is abnormal based on the probability that the tag is in an abnormal state; and to issue an abnormal prompt message if the status of the item to be managed is abnormal.

24. An electronic device, characterized in that, The electronic device includes a processor and a memory for storing computer programs capable of running on the processor; wherein, The processor is used to run the computer program to perform the method according to any one of claims 1 to 22.

25. A computer storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 22.

26. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 22.

Images