Intelligent alarm system for internet of things lock based on adaptive learning

Abstract

The invention relates to the technical field of intelligent locks, in particular to an intelligent alarm system for an internet of things lock based on adaptive learning, which comprises an acquisition module and an identification module, the determining module is used for determining whether an unlocking behavior is an abnormal behavior or not according to the type of a time period where a timestamp of the unlocking behavior of a user is located and the unlocking frequency in the time period of the unlocking behavior; the evaluation module is used for evaluating the real-time risk level of the abnormal behavior; the response module is used for determining a response strategy according to the real-time risk level of the abnormal behavior; the adjusting module is used for determining whether to adjust the risk judgment parameter or not according to the effective rate of the response strategy in the preset period; according to the invention, the accuracy of analyzing the abnormal behaviors of the user is improved, so that the accuracy of evaluating the real-time risk of the abnormal behaviors is improved, and the use safety and the user experience of the intelligent Internet of Things lock are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart locks, and in particular to an intelligent alarm system for an Internet of Things lock based on adaptive learning. Background Art

[0002] Traditional IoT locks usually only have simple unlocking and locking functions and rely on fixed keys or preset passwords. They cannot adapt to the user's actual usage habits and environmental changes, and it is difficult to meet the modern user's needs for convenience and personalization. For example, in a family scenario, family members have different schedules and usage habits, but traditional locks cannot automatically adjust security strategies based on these differences, which may cause inconvenience in use or potential safety hazards in some cases. Traditional IoT locks cannot perceive changes in the surrounding environment and potential security threats in real time. For situations such as violent prying, abnormal vibrations, and drastic changes in ambient temperature and humidity, they cannot respond in time and send alarms to users, resulting in a poor user experience.

[0003] On the basis of the above technical problems, Chinese Patent Publication No.: CN114202831A discloses a family care alarm method based on a smart lock, which includes receiving a care setting instruction, constructing a care setting table according to the care setting instruction, receiving unlocking record information, and associating it with the corresponding care member, setting and executing a timed task; when the timed task starts, querying the care end time point in the care setting table that has not yet arrived and is closest to the current system time, and obtaining the minimum alarm time point; the timed task enters the thread blocking stage; when the current time reaches the minimum alarm time point, the thread blocking stage ends, and enters the care query mode to query whether the corresponding care member has unlocking records within the effective time; if not, an alarm message is issued. The family care alarm method based on a smart lock of the invention can timely receive alarm information of other family members who have not returned home within the preset care time, so as to facilitate timely safety protection processing.

[0004] This shows that the analysis of the user's abnormal behavior is not accurate enough, which leads to low accuracy of real-time risk assessment of abnormal behavior and poor user experience. Summary of the invention

[0005] To this end, the present invention provides an intelligent alarm system for IoT locks based on adaptive learning, so as to overcome the problem in the prior art that the analysis of the user's abnormal behavior is not accurate enough, resulting in low accuracy of real-time risk assessment of abnormal behavior and poor user experience.

[0006] To achieve the above object, the present invention provides an intelligent alarm system for IoT locks based on adaptive learning, comprising:

[0007] A collection module, which is used to collect the user's unlocking data;

[0008] An identification module, which is connected to the acquisition module, is used to pre-determine whether to turn on the abnormal behavior identification function according to the smart IoT lock wake-up mode, and to identify abnormal unlocking behavior according to the time period type and unlocking frequency of the user's unlocking behavior;

[0009] An evaluation module, which is connected to the identification module and is used to evaluate the real-time risk level of the abnormal behavior according to the frequency of occurrence of the abnormal behavior and the rate of change of the environmental parameters of the environment in which the intelligent IoT interlock is located;

[0010] A response module, which is connected to the evaluation module and is used to determine a response strategy according to the real-time risk level, wherein the response strategy is to optimize the judgment criteria of abnormal behavior, temporarily lock some unlocking modes, or lock the smart IoT lock and issue an alarm;

[0011] An adjustment module, which is connected to the response module and is used to determine whether to adjust the risk determination parameters according to the effectiveness of the response strategy within a preset period;

[0012] Among them, the environmental parameters include temperature parameters, humidity parameters and vibration parameters, and the risk judgment parameters include the preset occurrence frequency of abnormal behavior for defining the real-time risk level, and, in the abnormal behavior identification process, the optimization coefficient of the preset unlocking frequency for comparison with the unlocking frequency.

[0013] Furthermore, the recognition module pre-determines whether to enable the abnormal behavior recognition function, including:

[0014] Under the condition that the smart IoT lock wake-up mode is a valid wake-up mode, determine to identify abnormal behavior of unlocking behavior;

[0015] Among them, the effective awakening method includes that the smart IoT lock is turned on when it is automatically awakened or the smart IoT lock is continuously in the awakened state.

[0016] Furthermore, the abnormal behavior recognition function of the recognition module includes:

[0017] If the time period type of the timestamp of the user's unlocking behavior is an abnormal time period or the unlocking frequency in the time period of the unlocking behavior is greater than or equal to the preset unlocking frequency, the unlocking behavior is determined to be an abnormal behavior.

[0018] Furthermore, the abnormal time period is a time period in which the user has not performed any unlocking behavior, and the preset unlocking frequency is determined according to the user's historical unlocking frequency in the same time period.

[0019] Further, in response to the occurrence of abnormal behavior, the evaluation module evaluates the real-time risk level of the abnormal behavior according to the frequency of occurrence of the abnormal behavior within a first preset time period and the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within a second preset time period;

[0020] If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is greater than the preset rate of change, the evaluation module evaluates the real-time risk level of the abnormal behavior as level one;

[0021] If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is less than or equal to the preset rate of change and the frequency of occurrence of the abnormal behavior within the first preset time period is greater than the preset frequency of occurrence, the evaluation module evaluates the real-time risk level of the abnormal behavior as level 2;

[0022] If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is less than or equal to the preset change rate and the frequency of occurrence of abnormal behavior within the first preset time period is less than or equal to the preset occurrence frequency, the evaluation module evaluates the real-time risk level of the abnormal behavior as level three.

[0023] Furthermore, the preset change rate is determined based on a historical average value of the change rate of the environmental parameters of the environment within a second preset time period when the intelligent IoT lock does not exhibit abnormal behavior.

[0024] Furthermore, the response module determines a response strategy according to the real-time risk level of the abnormal behavior, including:

[0025] If the real-time risk level of the abnormal behavior is level one, the response module determines to lock the smart IoT lock and issue an alarm; if the real-time risk level of the abnormal behavior is level two, the response module determines to temporarily lock some unlocking methods; if the real-time risk level of the abnormal behavior is level three, the response module determines to optimize the abnormal behavior judgment parameters.

[0026] Furthermore, the response module determines to optimize the preset unlocking frequency with the optimization coefficient under the condition of determining to optimize the abnormal behavior determination parameter.

[0027] Furthermore, the adjustment module determines whether to adjust the risk determination parameter, including:

[0028] If the effectiveness of the response strategy within the preset period is less than the preset effectiveness, the adjustment module determines to adjust the risk determination parameters.

[0029] Furthermore, the adjustment amount of the preset occurrence frequency is positively correlated with the effectiveness of the response strategy within the preset period, and the adjustment amount of the optimization coefficient is negatively correlated with the effectiveness of the response strategy within the preset period.

[0030] Compared with the prior art, the beneficial effect of the present invention is that the present invention can distinguish between intentional awakening of the smart IoT lock and unintentional awakening of the smart IoT lock by judging whether the awakening method is effective. For example, when the smart IoT lock has subsequent opening behavior or a continuous awakening state when it is automatically awakened, it means that the user intentionally awakens the smart IoT lock. At this time, abnormal behavior recognition can timely discover and prevent illegal opening attempts. If the opening behavior is under an invalid awakening method, it is likely that the smart IoT lock is awakened by an erroneous operation. The system does not perform abnormal behavior recognition to avoid false alarms, and at the same time, it will not waste system resources on unnecessary detection. This design can reduce unnecessary abnormal behavior recognition caused by erroneous operation or unexpected situations.

[0031] For example, when a user is just passing by a smart IoT lock without any intention to use it, the system will not identify abnormal behavior, thus avoiding the inconvenience caused to the user by frequent false alarms and improving the user experience of using the smart IoT lock. The above method improves the accuracy of the analysis of user abnormal behavior and thus improves the safety of using the smart IoT lock.

[0032] Furthermore, the present invention can more accurately identify possible abnormal unlocking behaviors by dividing abnormal time periods and analyzing unlocking frequencies. For unlocking that occurs in a time period when the user has never unlocked the door before, and the unlocking frequency exceeds the normal range, the system can promptly determine it as abnormal behavior, greatly improving the ability to identify potential security risks. Clear abnormal time periods and reasonable unlocking frequency judgment standards can effectively reduce the occurrence of false alarms. The system will not trigger false alarms due to some accidental behaviors within the normal range, thereby saving system resources and improving overall operating efficiency. The above method improves the accuracy of user abnormal behavior analysis and thereby improves the safety of use of smart IoT locks and enhances user experience.

[0033] Furthermore, the present invention evaluates the real-time risk level of abnormal behavior by combining the frequency of abnormal behavior within the first preset time and the rate of change of environmental parameters within the second preset time, thereby achieving a comprehensive consideration of risks from two important dimensions of time and environment. In this way, the severity of abnormal behavior can be judged more comprehensively and accurately, avoiding the one-sidedness caused by a single factor evaluation. The real-time risk level of abnormal behavior is divided into level one, level two and level three according to different conditions, and the risk can be quickly graded in a short time, which enables the system to take corresponding security response measures according to different risk levels, thereby improving the response speed and processing efficiency to risks. Considering the rate of change of temperature, humidity and vibration of the environment in which the smart IoT lock is located within the second preset time, the system can perceive changes in the surrounding environment in a timely manner, which is very important for preventing abnormal behavior caused by environmental factors. For example, a sudden increase in temperature may mean that someone is trying to unlock the lock with a hot tool, and abnormal vibration may mean that someone is violently destroying the smart IoT lock, etc. The above method improves the accuracy of the analysis of user abnormal behavior, thereby improving the safety of using the smart IoT lock and enhancing the user experience.

[0034] Furthermore, the present invention determines different response strategies according to the real-time risk level of abnormal behavior, and can accurately respond to risks of different degrees. For the first level with high risk, the smart IoT lock is locked and an alarm is issued, which can quickly prevent potential security threats and promptly notify relevant personnel to deal with them. For the second level with medium risk, some unlocking methods are temporarily locked, which can not only limit abnormal behavior to a certain extent, but also retain some emergency unlocking methods. For the third level with low risk, the abnormal behavior judgment parameters are optimized, which can make the system more intelligently adapt to the actual situation and improve the accuracy of risk judgment. By taking different response measures for different risk levels, the system has higher flexibility and adaptability, and can be adjusted and optimized according to specific circumstances to ensure that the safety of the smart IoT lock can be effectively guaranteed in various complex environments. The design of hierarchical response and optimized parameters can reduce the occurrence of false alarms. For low-risk abnormal behavior, unnecessary alarms and interference can be avoided by optimizing parameters, thereby improving the user experience. At the same time, for high-risk and medium-risk abnormal behaviors, decisive measures can be taken to protect the property safety of users and enhance the user's trust in smart IoT locks.

[0035] Furthermore, the present invention determines whether to adjust the preset occurrence frequency and the optimization coefficient by comparing the effectiveness of the response strategy within a preset period with the preset efficiency, so that the system can continuously perform self-optimization according to the actual operating conditions. As time goes by and data accumulates, the system can more accurately adapt to different usage scenarios and risk conditions, and improve the overall performance. The adjustment amount of the preset occurrence frequency is positively correlated with the effectiveness of the response strategy, which means that when the response strategy is effective, the preset occurrence frequency can be appropriately relaxed to reduce misjudgments; the adjustment amount of the optimization coefficient is negatively correlated with the effectiveness of the response strategy, which means that when the response strategy is not effective, the optimization coefficient can be increased to more actively optimize the abnormal behavior judgment parameters. By adaptively adjusting the parameters, the system can continuously optimize the abnormal behavior judgment criteria and reduce the occurrence of false alarms and missed alarms, which can avoid unnecessary interference and trouble to users. At the same time, it can also ensure that alarms are issued in time when real security risks occur, thereby improving users' trust and satisfaction with smart IoT locks. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a schematic diagram of the structure of an intelligent alarm system for IoT locks based on adaptive learning according to an embodiment of the present invention;

[0037] Figure 2 A flowchart of a method for determining whether to perform abnormal behavior recognition on a user's unlocking behavior based on adaptive learning in an IoT lock intelligent alarm system according to an embodiment of the present invention;

[0038] Figure 3 A flowchart of a recognition module in an adaptive learning-based IoT lock intelligent alarm system according to an embodiment of the present invention for determining whether an unlocking behavior is an abnormal behavior;

[0039] Figure 4 The present invention is a flowchart of the adjustment module in the IoT lock intelligent alarm system based on adaptive learning according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0041] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0042] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0043] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0044] See also Figure 1-Figure 4 As shown, Figure 1 This is a schematic diagram of the structure of an intelligent alarm system for IoT locks based on adaptive learning according to an embodiment of the present invention; Figure 2 A flowchart of a method for determining whether to perform abnormal behavior recognition on a user's unlocking behavior based on adaptive learning in an IoT lock intelligent alarm system according to an embodiment of the present invention; Figure 3 A flowchart of a recognition module in an adaptive learning-based IoT lock intelligent alarm system according to an embodiment of the present invention for determining whether an unlocking behavior is an abnormal behavior; Figure 4 The present invention is a flowchart of the adjustment module in the IoT lock intelligent alarm system based on adaptive learning according to an embodiment of the present invention.

[0045] The present invention provides an intelligent alarm system for IoT locks based on adaptive learning, comprising:

[0046] A collection module, which is used to collect the user's unlocking data;

[0047] An identification module, which is connected to the acquisition module, is used to pre-determine whether to turn on the abnormal behavior identification function according to the smart IoT lock wake-up mode, and to identify abnormal unlocking behavior according to the time period type and unlocking frequency of the user's unlocking behavior;

[0048] An evaluation module, which is connected to the identification module and is used to evaluate the real-time risk level of the abnormal behavior according to the frequency of occurrence of the abnormal behavior and the rate of change of the environmental parameters of the environment in which the intelligent IoT interlock is located;

[0049] A response module, which is connected to the evaluation module and is used to determine a response strategy according to the real-time risk level, wherein the response strategy is to optimize the judgment criteria of abnormal behavior, temporarily lock some unlocking modes, or lock the smart IoT lock and issue an alarm;

[0050] An adjustment module, which is connected to the response module and is used to determine whether to adjust the risk determination parameters according to the effectiveness of the response strategy within a preset period;

[0051] Among them, the environmental parameters include temperature parameters, humidity parameters and vibration parameters, and the risk judgment parameters include the preset occurrence frequency of abnormal behavior for defining the real-time risk level, and, in the abnormal behavior identification process, the optimization coefficient of the preset unlocking frequency for comparison with the unlocking frequency.

[0052] The unlocking data of the user in the embodiment of the present invention includes but is not limited to "unlocking method, unlocking time and unlocking times", and the division of time periods includes an active period (6:00-23:00) divided into time periods with one hour as the dividing line, for example, 7:00-8:00 is divided into one time period, and a quiet period (23:00-6:00) is divided into time periods with 30 minutes as the dividing line, for example, 1:00-1:30 is divided into one time period.

[0053] Specifically, the recognition module pre-determines whether to enable the abnormal behavior recognition function, including determining whether to perform abnormal behavior recognition on the unlocking behavior according to the smart IoT lock wake-up mode;

[0054] If the smart IoT lock wake-up mode is a valid wake-up mode, the identification module determines to perform abnormal behavior identification on the unlocking behavior;

[0055] If the smart IoT lock wake-up mode is an invalid wake-up mode, the identification module determines not to perform abnormal behavior identification on the unlocking behavior.

[0056] The smart IoT lock described in the embodiment of the present invention has an automatic wake-up function when it detects a person approaching. The judgment of whether the wake-up method of the smart IoT lock is valid includes determining that the wake-up method of the smart IoT lock is valid when the smart IoT lock is opened during automatic wake-up or the smart IoT lock is continuously in a wake-up state, and determining that the wake-up method of the smart IoT lock is invalid when there is no subsequent opening behavior during automatic wake-up and the smart IoT lock is not continuously in a wake-up state. For example, when a user approaches the smart IoT lock, an attempt is made to open the smart IoT lock to determine that the wake-up method of the smart IoT lock is valid, or when the user approaches the smart IoT lock, there is no opening behavior of the smart IoT lock but the user does not leave, so that the smart IoT lock is continuously in a wake-up state, determining that the wake-up method of the smart IoT lock is valid.

[0057] The present invention can distinguish between intentional awakening of the smart IoT lock and unintentional awakening of the smart IoT lock by judging whether the awakening method is effective. For example, when the smart IoT lock has subsequent opening behavior or a continuous awakening state when it is automatically awakened, it means that the user intends to awaken the smart IoT lock. At this time, abnormal behavior recognition can timely discover and prevent illegal opening attempts. If the opening behavior is under an invalid awakening method, it is likely that the smart IoT lock is awakened by an incorrect operation. The system does not perform abnormal behavior recognition to avoid false alarms and will not waste system resources on unnecessary detection. This design can reduce unnecessary abnormal behavior recognition caused by incorrect operations or unexpected situations. For example, when the user just passes by the smart IoT lock without intending to use it, the system will not perform abnormal behavior recognition, thereby avoiding the trouble caused by frequent false alarms to the user and improving the user experience of using the smart IoT lock. The above method improves the accuracy of user abnormal behavior analysis and thus improves the safety of use of the smart IoT lock.

[0058] Specifically, the identification module determines whether the unlocking behavior is abnormal according to the time period type of the timestamp of the unlocking behavior of the user and the unlocking frequency within the time period of the unlocking behavior, under the condition of determining to identify the abnormal behavior of the unlocking behavior of the user;

[0059] If the time period type of the timestamp of the user's unlocking behavior is an abnormal time period or the unlocking frequency in the time period of the unlocking behavior is greater than or equal to the preset unlocking frequency, the identification module determines that the unlocking behavior is an abnormal behavior;

[0060] If the time period type of the timestamp of the user's unlocking behavior is a normal time period and the unlocking frequency in the time period of the unlocking behavior is less than the preset unlocking frequency, the identification module determines that the unlocking behavior is a normal behavior.

[0061] The abnormal time period in the embodiment of the present invention is a time period in which the user has not performed any unlocking behavior. For example, the user has never performed any unlocking behavior between 4 and 5 in the morning, and 4 to 5 o'clock is the abnormal time period. If the timestamp of the unlocking behavior is 4:17, the time period in which the unlocking behavior occurs is 4 to 4:30, and the time period is within the time period of 4 to 5 o'clock. At this time, the time period in which the unlocking behavior occurs is determined to be the abnormal time period. The preset unlocking frequency is the average value of the historical unlocking frequency of the user in the same time period. The unlocking frequency is the ratio of the number of unlocking times to the length of the unlocking time period. For example, the number of unlocking times by the user in the time period of 8 to 9 o'clock is 6, and the length of the unlocking time period is 60 minutes, then the unlocking frequency is 0.1, but the above values ​​are not limited to this, and those skilled in the art can also adjust the value according to actual needs.

[0062] By dividing the abnormal time periods and analyzing the unlocking frequency, the present invention can more accurately identify possible abnormal unlocking behaviors. For unlocking that occurs in a time period when the user has never unlocked the door before, and the situation where the unlocking frequency exceeds the normal range, the system can promptly determine it as abnormal behavior, which greatly improves the ability to identify potential security risks. Clear abnormal time periods and reasonable unlocking frequency judgment standards can effectively reduce the occurrence of false alarms. The system will not trigger false alarms due to some accidental behaviors within the normal range, thereby saving system resources and improving overall operating efficiency. The above method improves the accuracy of the analysis of user abnormal behavior, thereby improving the safety of use of smart IoT locks and enhancing user experience.

[0063] Specifically, the evaluation module evaluates the real-time risk level of the abnormal behavior based on the frequency of occurrence of the abnormal behavior within a first preset time period and the rate of change of the environmental parameters of the environment in which the intelligent IoT interlocking system is located within a second preset time period under the condition that the unlocking behavior is determined to be an abnormal behavior;

[0064] If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is greater than the preset rate of change, the real-time risk level of the abnormal behavior is evaluated as level one;

[0065] If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is less than or equal to the preset change rate and the frequency of occurrence of abnormal behavior within the first preset time period is greater than the preset frequency, the real-time risk level of the abnormal behavior is evaluated as level 2;

[0066] If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is less than or equal to the preset change rate and the frequency of occurrence of abnormal behavior within the first preset time period is less than or equal to the preset frequency, the real-time risk level of the abnormal behavior is assessed to be level three.

[0067] In the embodiment of the present invention, the first preset time length is set to 10 days, the second preset time length is set to 60 seconds, the preset change rate is the historical average of the rate of change of the environmental parameters of the environment in the second preset time length when the intelligent IoT lock does not have abnormal behavior, the frequency of abnormal behavior in the first preset time length is the ratio of the number of abnormal behavior occurrences to the first preset time length. For example, abnormal behavior occurs 3 times in the first preset time length, and the first preset time length is 10 days, then the frequency of abnormal behavior in the first preset time length is 0.3, and the preset occurrence frequency is the average of the frequencies of abnormal behavior in several first preset time lengths. For example, the average of the frequencies of abnormal behavior in 3 first preset time lengths is selected as the preset occurrence frequency, but the above value is not limited to this, and those skilled in the art may also adjust the value according to actual needs.

[0068] The environmental parameters of the environment in which the smart IoT lock is located in the embodiment of the present invention can be obtained by detecting the temperature sensor, humidity sensor and vibration sensor arranged inside the smart IoT lock. The change rate of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period includes but is not limited to "temperature parameter change rate, humidity parameter change rate and vibration parameter change rate".

[0069] The present invention evaluates the real-time risk level of abnormal behavior by combining the frequency of abnormal behavior within a first preset time and the rate of change of environmental parameters within a second preset time, thereby realizing a comprehensive consideration of risks from two important dimensions of time and environment. In this way, the severity of abnormal behavior can be judged more comprehensively and accurately, avoiding the one-sidedness caused by a single factor evaluation. The real-time risk level of abnormal behavior is divided into level one, level two and level three according to different conditions, and the risk can be quickly graded in a short time, which enables the system to take corresponding security response measures according to different risk levels, thereby improving the response speed and processing efficiency to risks. Considering the rate of change of environmental parameters such as temperature, humidity, vibration, etc. in the environment where the smart IoT lock is located within the second preset time, the system can perceive changes in the surrounding environment in a timely manner, which is very important for preventing abnormal behavior caused by environmental factors. For example, a sudden increase in temperature may mean that someone is trying to unlock the lock with a hot tool, and abnormal vibration may mean that someone is violently destroying the smart IoT lock, etc. The above method improves the accuracy of the analysis of user abnormal behavior, thereby improving the safety of using the smart IoT lock and enhancing the user experience.

[0070] Specifically, the response module determines the response strategy according to the real-time risk level of the abnormal behavior under the condition of determining the response strategy;

[0071] If the real-time risk level of the abnormal behavior is level one, the response module determines to lock the smart IoT lock and issue an alarm;

[0072] If the real-time risk level of the abnormal behavior is level 2, the response module determines to temporarily lock part of the unlocking modes;

[0073] If the real-time risk level of the abnormal behavior is level three, the response module determines to optimize abnormal behavior determination parameters.

[0074] The unlocking method described in the embodiment of the present invention includes but is not limited to "entering a password, Bluetooth and fingerprint recognition".

[0075] Specifically, the response module determines to optimize the preset unlocking frequency with the optimization coefficient under the condition of optimizing the abnormal behavior determination parameter.

[0076] In the embodiment of the present invention, the value range of the optimization coefficient is set to 0.84-0.97, and the value of the optimization coefficient is preferably 0.88, but the above value is not limited thereto, and those skilled in the art may also adjust the value according to actual needs.

[0077] The present invention determines different response strategies according to the real-time risk level of abnormal behavior, and can accurately respond to risks of different degrees. For the first level of high risk, the intelligent IoT lock is locked and an alarm is issued, which can quickly prevent potential security threats and promptly notify relevant personnel to deal with them. For the second level of medium risk, some unlocking methods are temporarily locked, which can not only limit abnormal behaviors to a certain extent, but also retain some emergency unlocking methods. For the third level of low risk, the abnormal behavior judgment parameters are optimized, so that the system can adapt to the actual situation more intelligently and improve the accuracy of risk judgment. By taking different response measures for different risk levels, the system has higher flexibility and adaptability, and can be adjusted and optimized according to specific circumstances to ensure that the safety of the intelligent IoT lock can be effectively guaranteed in various complex environments. The design of hierarchical response and optimized parameters can reduce the occurrence of false alarms. For low-risk abnormal behaviors, unnecessary alarms and interference can be avoided by optimizing parameters, thereby improving the user experience. At the same time, for high-risk and medium-risk abnormal behaviors, decisive measures can be taken to protect the property safety of users and enhance the user's trust in the intelligent IoT lock.

[0078] Specifically, the adjustment module determines whether to adjust the risk determination parameter, including determining whether to adjust the risk determination parameter according to a comparison result between the effectiveness of the response strategy within a preset period and a preset effectiveness rate;

[0079] If the effectiveness of the response strategy within the preset period is less than the preset effectiveness, the adjustment module determines that the risk determination parameter needs to be adjusted;

[0080] If the effectiveness of the response strategy within the preset period is greater than or equal to the preset effectiveness, the adjustment module determines that there is no need to adjust the risk determination parameters.

[0081] In the embodiment of the present invention, the preset period is set to 20 days, and the effectiveness of the response strategy within the preset period is the ratio of the effective number of times the response strategy occurs to the total number of times the response strategy occurs. The judgment that the response strategy occurs effectively includes but is not limited to "abnormal behavior is blocked after an alarm is issued, abnormal behavior is blocked after the partial unlocking method is locked, and the number of abnormal behaviors increases within the first preset time period after the abnormal behavior judgment parameters are optimized". The preset effectiveness is the average value of the effectiveness of the response strategies within several preset periods. For example, the average value of the effectiveness of the response strategies within 3 preset periods is selected as the preset effectiveness. However, the above value is not limited to this, and those skilled in the art may also adjust the value according to actual needs.

[0082] Specifically, under the condition that the adjustment module determines that the risk determination parameter needs to be adjusted, the adjustment module determines to adjust the preset occurrence frequency with a first adjustment coefficient and to adjust the optimization coefficient with a second adjustment coefficient.

[0083] In the embodiment of the present invention, the value range of the first adjustment coefficient is set to 0.82-0.96, and the value of the first adjustment coefficient is preferably 0.9. The value range of the second adjustment coefficient is set to 1.04-1.19, and the value of the second adjustment coefficient is preferably 1.12. The adjustment amount of the preset occurrence frequency is positively correlated with the effectiveness of the response strategy within the preset period, and the adjustment amount of the optimization coefficient is negatively correlated with the effectiveness of the response strategy within the preset period. However, the above values ​​are not limited to this, and those skilled in the art may also adjust the values ​​according to actual needs.

[0084] The present invention determines whether to adjust the preset occurrence frequency and the optimization coefficient by comparing the effectiveness of the response strategy within a preset period with the preset efficiency, so that the system can continuously perform self-optimization according to the actual operating conditions. As time goes by and data accumulates, the system can more accurately adapt to different usage scenarios and risk conditions, and improve the overall performance. The adjustment amount of the preset occurrence frequency is positively correlated with the effectiveness of the response strategy, which means that when the response strategy is effective, the preset occurrence frequency can be appropriately relaxed to reduce misjudgment; the adjustment amount of the optimization coefficient is negatively correlated with the effectiveness of the response strategy, which means that when the response strategy is not effective, the optimization coefficient can be increased to more actively optimize the abnormal behavior judgment parameters. By adaptively adjusting the parameters, the system can continuously optimize the abnormal behavior judgment criteria and reduce the occurrence of false alarms and missed alarms, which can avoid unnecessary interference and trouble to users, while also ensuring that alarms are issued in time when real safety risks occur, thereby improving users' trust and satisfaction with smart IoT locks.

[0085] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An intelligent alarm system for IoT locks based on adaptive learning, characterized in that: include: A collection module, which is used to collect the user's unlocking data; An identification module, which is connected to the acquisition module, is used to pre-determine whether to turn on the abnormal behavior identification function according to the smart IoT lock wake-up mode, and to identify abnormal unlocking behavior according to the time period type and unlocking frequency of the user's unlocking behavior; An evaluation module, which is connected to the identification module and is used to evaluate the real-time risk level of the abnormal behavior according to the frequency of occurrence of the abnormal behavior and the rate of change of the environmental parameters of the environment in which the intelligent IoT interlock is located; A response module, which is connected to the evaluation module and is used to determine a response strategy according to the real-time risk level, wherein the response strategy is to optimize the judgment criteria of abnormal behavior, temporarily lock some unlocking modes, or lock the smart IoT lock and issue an alarm; An adjustment module, which is connected to the response module and is used to determine whether to adjust the risk determination parameters according to the effectiveness of the response strategy within a preset period; Among them, the environmental parameters include temperature parameters, humidity parameters and vibration parameters, and the risk judgment parameters include the preset occurrence frequency of abnormal behavior for defining the real-time risk level, and, in the abnormal behavior identification process, the optimization coefficient of the preset unlocking frequency for comparison with the unlocking frequency.

2. The adaptive learning-based IoT interlock intelligent alarm system according to claim 1 is characterized in that: The recognition module pre-determines whether to enable the abnormal behavior recognition function, including: Under the condition that the smart IoT lock wake-up mode is a valid wake-up mode, determine to identify abnormal behavior of unlocking behavior; Among them, the effective awakening method includes that the smart IoT lock is turned on when it is automatically awakened or the smart IoT lock is continuously in the awakened state.

3. The adaptive learning-based IoT interlock intelligent alarm system according to claim 2 is characterized in that: The abnormal behavior recognition function of the recognition module includes: If the time period type of the timestamp of the user's unlocking behavior is an abnormal time period or the unlocking frequency in the time period of the unlocking behavior is greater than or equal to the preset unlocking frequency, the unlocking behavior is determined to be an abnormal behavior.

4. The adaptive learning-based IoT interlock intelligent alarm system according to claim 3 is characterized in that: The abnormal time period is a time period during which the user has not performed any unlocking behavior, and the preset unlocking frequency is determined according to the historical unlocking frequency of the user in the same time period.

5. The adaptive learning-based IoT interlock intelligent alarm system according to claim 4 is characterized in that: In response to the occurrence of abnormal behavior, the evaluation module evaluates the real-time risk level of the abnormal behavior according to the frequency of occurrence of the abnormal behavior within a first preset time period and the rate of change of environmental parameters of the environment in which the intelligent IoT interlock is located within a second preset time period; If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is greater than the preset rate of change, the evaluation module evaluates the real-time risk level of the abnormal behavior as level one; If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is less than or equal to the preset rate of change and the frequency of occurrence of the abnormal behavior within the first preset time period is greater than the preset frequency of occurrence, the evaluation module evaluates the real-time risk level of the abnormal behavior as level 2; If the rate of change of the environmental parameters of the environment in which the smart IoT lock is located within the second preset time period is less than or equal to the preset change rate and the frequency of occurrence of abnormal behavior within the first preset time period is less than or equal to the preset occurrence frequency, the evaluation module evaluates the real-time risk level of the abnormal behavior as level three.

6. The adaptive learning-based IoT interlock intelligent alarm system according to claim 5 is characterized in that: The preset change rate is determined based on the historical average value of the change rate of the environmental parameters of the environment within the second preset time period when the intelligent IoT lock does not have abnormal behavior.

7. The adaptive learning-based IoT interlock intelligent alarm system according to claim 6 is characterized in that: The response module determines a response strategy according to the real-time risk level of the abnormal behavior, including: If the real-time risk level of the abnormal behavior is level one, the response module determines to lock the smart IoT lock and issue an alarm; if the real-time risk level of the abnormal behavior is level two, the response module determines to temporarily lock some unlocking methods; if the real-time risk level of the abnormal behavior is level three, the response module determines to optimize the abnormal behavior judgment parameters.

8. The adaptive learning-based IoT interlock intelligent alarm system according to claim 7 is characterized in that: The response module determines to optimize the preset unlocking frequency with the optimization coefficient under the condition of determining to optimize the abnormal behavior determination parameter.

9. The adaptive learning-based IoT interlock intelligent alarm system according to claim 8 is characterized in that: The adjustment module determines whether to adjust the risk determination parameter, including: If the effectiveness of the response strategy within the preset period is less than the preset effectiveness, the adjustment module determines to adjust the risk determination parameters.

10. The adaptive learning-based IoT interlock intelligent alarm system according to claim 9 is characterized in that: The adjustment amount of the preset occurrence frequency is positively correlated with the effectiveness of the response strategy within the preset period, and the adjustment amount of the optimization coefficient is negatively correlated with the effectiveness of the response strategy within the preset period.

Images