Internet of Things system based on big data and AI

Abstract

The invention relates to the technical field of Internet of Things, and specifically discloses an Internet of Things system based on big data and AI, comprising: a preliminary determination module for determining a preliminary intrusion determination result of each sensor of the Internet of Things system; the processing module is used for obtaining all normal sensors and all abnormal sensors based on the initial intrusion judgment results of all the sensors; the calculation module is used for calculating an intrusion risk assessment value of each normal sensor and an intrusion risk reference value of each abnormal sensor; the final judgment module is used for obtaining final intrusion judgment results of all the normal sensors based on the intrusion risk assessment values of all the normal sensors and the intrusion risk reference values of all the abnormal sensors; and the detection module is used for obtaining an intrusion detection result of the Internet of Things system based on the final intrusion judgment results of all the normal sensors. According to the invention, the intrusion detection capability of the sensor in the Internet of Things system is improved, and the safe operation of the Internet of Things is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to an Internet of Things system based on big data and AI. Background Art

[0002] At present, with the rapid development of Internet of Things technology, more and more devices, systems and applications are connected to the Internet, forming a huge and complex network system. Although this system has greatly improved production efficiency, life convenience and data interaction capabilities, it also brings severe security challenges. Sensors in the Internet of Things system are key nodes for data collection and transmission. Their security and reliability are directly related to the stable operation and data security of the entire system. However, in practical applications, sensors often face various potential intrusion threats, such as malicious attacks, data tampering, illegal access, etc. These threats may not only destroy the normal functions of sensors, but also cause sensitive data leakage, which will have a serious impact on the entire Internet of Things system. Therefore, how to effectively detect the intrusion behavior of Internet of Things sensors has become an important issue to ensure the safe operation of the Internet of Things system.

[0003] However, the existing IoT systems based on big data and AI simply link the underlying IoT data from different systems to form new data samples for machine learning and AI decision analysis, opening up data channels for multi-system integration of the IoT, increasing data correlation between independent systems, and providing data support for upper-level business systems. However, they do not consider how to accurately judge whether each sensor in the IoT system has been invaded. For example, the patent publication number is "CN111787066B" and the patent name is "A IoT data platform based on big data and AI". Its method includes the following steps: data acquisition node server cluster (Godzilla-Agent), transit node server cluster (Godzilla-Broker), central coordination node server cluster (Godzilla-Coordinator), data analysis node server cluster (Godzilla-Processor), API open platform server cluster (Godzilla-API), alarm notification server cluster (Godzilla-Notifier), AI server cluster (Godzilla-AI) and platform management server cluster (Godzilla-Admin), etc. This system is compatible with IoT terminal devices of various communication protocols, can quickly access point data, and combine data fusion processing of different dimensions, associate the underlying IoT data from different systems together, and form new data samples for machine learning and AI decision analysis, which opens up data channels for multi-system integration of the IoT, increases data correlation between independent systems, provides data support for upper-level business systems, and forms an integrated solution for IoT data acquisition, storage, processing and AI decision-making. However, this patent only links the underlying IoT data from different systems to form new data samples for machine learning and AI decision analysis, opens up data channels for multi-system integration of the IoT, increases data correlation between independent systems, and provides data support for upper-level business systems. However, it does not consider how to accurately determine whether each sensor in the IoT system has been invaded.

[0004] Therefore, the present invention proposes an Internet of Things system based on big data and AI. Summary of the invention

[0005] The present invention provides an Internet of Things system based on big data and AI, which is used to accurately obtain the preliminary intrusion judgment result of each sensor of the Internet of Things system according to all types of transmission parameters of all sensors of the Internet of Things system at each moment in a preset time period before the current moment, so as to facilitate the subsequent classification of all sensors of the Internet of Things system, obtain all normal sensors and all abnormal sensors according to the preliminary intrusion judgment result of all sensors of the Internet of Things system, and then obtain all supplementary parameters of each type of transmission parameters of all abnormal sensors according to all types of transmission parameters of all abnormal sensors at all moments in the preset time period before the current moment and AI technology, so as to supplement each type of transmission parameters of all abnormal sensors, so as to facilitate the accurate calculation of the intrusion risk reference value of the subsequent abnormal sensors, and All types of transmission parameters at all times within a preset time period are used to obtain the intrusion risk assessment value of each normal sensor, thereby quantifying the possibility of a single normal sensor being potentially invaded. According to all supplementary parameters of each type of transmission parameters of each abnormal sensor, the intrusion risk reference value of each abnormal sensor is obtained, thereby quantifying the possibility of a single abnormal sensor being potentially invaded. Finally, according to the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors, the final intrusion judgment results of all normal sensors are accurately obtained. According to the preliminary intrusion judgment results of each sensor of the Internet of Things system and the final intrusion judgment results of all normal sensors, the intrusion detection results of the Internet of Things system are obtained, thereby improving the intrusion detection capability of sensors in the Internet of Things system and ensuring the safe operation of the Internet of Things.

[0006] The present invention provides an Internet of Things system based on big data and AI, comprising:

[0007] A preliminary determination module, used to obtain a preliminary intrusion determination result of each sensor of the Internet of Things system based on all class transmission parameters of all sensors of the Internet of Things system at each moment within a preset time period before the current moment;

[0008] A processing module, used for obtaining all normal sensors and all abnormal sensors based on preliminary intrusion determination results of all sensors of the IoT system, and obtaining all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters and AI technology at all times within a preset time period before the current time.

[0009] A calculation module, used to obtain an intrusion risk assessment value of each normal sensor based on all types of transmission parameters of each normal sensor at all times within a preset time period before the current time, and to obtain an intrusion risk reference value of each abnormal sensor based on all supplementary parameters of each type of transmission parameters of each abnormal sensor;

[0010] A final determination module, used to obtain final intrusion determination results of all normal sensors based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors;

[0011] The detection module is used to obtain the intrusion detection result of the Internet of Things system based on the preliminary intrusion determination result of each sensor of the Internet of Things system and the final intrusion determination result of all normal sensors.

[0012] Preferably, the preliminary determination module of the Internet of Things system based on big data and AI includes:

[0013] The acquisition submodule is used to obtain all class transmission parameters of all sensors in the IoT system at each moment in a preset time period before the current moment, wherein all class transmission parameters include transmission time, packet loss rate and sending frequency;

[0014] The preliminary determination submodule is used to obtain a preliminary intrusion determination result of each sensor of the Internet of Things system based on all class transmission parameters of all sensors of the Internet of Things system at each moment within a preset time period before the current moment.

[0015] Preferably, the IoT system based on big data and AI has a preliminary determination submodule, including:

[0016] A judging unit, used to judge whether the difference between each type of transmission parameter of each sensor of the Internet of Things system at each moment within a preset time period before the current moment and the corresponding type of transmission parameters at all adjacent moments is less than a preset difference of the corresponding type of transmission parameter, and if so, regard the corresponding moment as an abnormal moment of the corresponding type of transmission parameter of the corresponding sensor, otherwise, regard the corresponding moment as a non-abnormal moment of the corresponding type of transmission parameter of the corresponding sensor;

[0017] The preliminary judgment unit is used to regard the need for further detection as the preliminary intrusion judgment result of each sensor of the Internet of Things system when the number of abnormal moments of all types of transmission parameters of each sensor of the Internet of Things system is 0, and to regard the existence of intrusion as the preliminary intrusion judgment result of each sensor of the Internet of Things system when the number of abnormal moments of all types of transmission parameters of each sensor of the Internet of Things system is not 0.

[0018] Preferably, the Internet of Things system based on big data and AI, the processing module includes:

[0019] The discrimination submodule is used to treat all sensors in the IoT system that are initially judged to need further detection as normal sensors, and treat all sensors in the IoT system that are initially judged to have been invaded as abnormal sensors;

[0020] The processing submodule is used to obtain all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters of all abnormal sensors at all times within a preset time period before the current moment and AI technology.

[0021] Preferably, in an Internet of Things system based on big data and AI, the processing submodule obtains all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters of all abnormal sensors at all times within a preset time period before the current time and AI technology, including:

[0022] Determine whether the number of abnormal moments of each type of transmission parameter of each abnormal sensor is greater than a preset number threshold. If so, take the average value of the corresponding type of transmission parameter of the corresponding abnormal sensor at all times within a preset time period before the current time as a supplementary parameter of the corresponding type of transmission parameter of the corresponding abnormal sensor;

[0023] Otherwise, all the moments of the corresponding abnormal sensor in the preset time period before the current moment are regarded as the horizontal coordinate value, and the corresponding class transmission parameter of the corresponding abnormal sensor at the corresponding moment in the preset time period before the current moment is regarded as the vertical coordinate value, to obtain all the sample points of the corresponding class transmission parameter of the corresponding abnormal sensor;

[0024] All sample points with all non-abnormal moments of the corresponding class of transmission parameters as horizontal coordinate values ​​in all sample points of each type of transmission parameters of each abnormal sensor are regarded as all non-abnormal sample points of the corresponding class of transmission parameters of the corresponding abnormal sensor, and all sample points with all abnormal moments of the corresponding class of transmission parameters as horizontal coordinate values ​​in all sample points of each type of transmission parameters of each abnormal sensor are regarded as all abnormal sample points of the corresponding class of transmission parameters of the corresponding abnormal sensor;

[0025] The mean value of the distance between each abnormal sample point of each type of transmission parameter of each abnormal sensor and all non-abnormal sample points of the corresponding type of transmission parameter of the corresponding abnormal sensor is regarded as the supplementary sample distance of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor, and the mean value of the supplementary sample distance of all abnormal sample points of each type of transmission parameter of each abnormal sensor is regarded as the supplementary distance of each type of transmission parameter of each abnormal sensor;

[0026] Taking each abnormal sample point of each type of transmission parameter of each abnormal sensor as the origin and the supplementary distance of the corresponding type of transmission parameter of the corresponding abnormal sensor as the radius, a supplementary circle of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor is obtained. Based on the supplementary circles of all abnormal sample points of each type of transmission parameters of all abnormal sensors and AI technology, all supplementary parameters of each type of transmission parameters of all abnormal sensors are obtained.

[0027] Preferably, the Internet of Things system based on big data and AI obtains all supplementary parameters of each type of transmission parameters of all abnormal sensors based on the supplementary circle of all abnormal sample points of each type of transmission parameters of all abnormal sensors and AI technology, including:

[0028] The maximum value of the distance between each abnormal sample point of each type of transmission parameter of each abnormal sensor and all non-abnormal sample points of the corresponding type of transmission parameter of the corresponding abnormal sensor is taken as the supplementary reference distance of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor, and the mean value of the supplementary reference distances of all abnormal sample points of each type of transmission parameter of each abnormal sensor is taken as the reference distance of each type of transmission parameter of each abnormal sensor;

[0029] Taking each abnormal sample point of each type of transmission parameter of each abnormal sensor as the origin and the reference distance of the corresponding type of transmission parameter of the corresponding abnormal sensor as the radius, a reference circle of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor is obtained, and the non-overlapping area of ​​the reference circle and the supplementary circle of each abnormal sample point of each type of transmission parameter of each abnormal sensor is regarded as the reference area of ​​each abnormal sample point of each type of transmission parameter of each abnormal sensor;

[0030] Based on the reference area, the supplementary circle and the AI ​​technology of each abnormal sample point of each type of transmission parameter of each abnormal sensor, all supplementary parameters of each type of transmission parameter of each abnormal sensor are obtained.

[0031] Preferably, the computing module of the Internet of Things system based on big data and AI includes:

[0032] The first calculation submodule is used to obtain the intrusion risk assessment value of each normal sensor based on all class transmission parameters of each normal sensor at all times within a preset time period before the current time, that is:

[0033]

[0034] Wherein, β is the intrusion risk assessment value of the normal sensor currently calculated, maxA is the maximum value of the transmission time of the normal sensor currently calculated at all times in the preset time period before the current moment, minA is the minimum value of the transmission time of the normal sensor currently calculated at all times in the preset time period before the current moment, maxB is the maximum value of the packet loss rate of the normal sensor currently calculated at all times in the preset time period before the current moment, minB is the minimum value of the packet loss rate of the normal sensor currently calculated at all times in the preset time period before the current moment, maxC is the maximum value of the sending frequency of the normal sensor currently calculated at all times in the preset time period before the current moment, minC is the minimum value of the sending frequency of the normal sensor currently calculated at all times in the preset time period before the current moment, ln is the natural logarithm, and the value of the natural constant e is 2.718;

[0035] The second calculation submodule is used to obtain the intrusion risk reference value of each abnormal sensor based on all supplementary parameters of each type of transmission parameters of each abnormal sensor, that is:

[0036]

[0037] Among them, γ is the intrusion risk reference value of the currently calculated abnormal sensor, maxF is the maximum value of all supplementary parameters of the currently calculated transmission time of the abnormal sensor, minF is the minimum value of all supplementary parameters of the currently calculated transmission time of the abnormal sensor, maxG is the maximum value of all supplementary parameters of the currently calculated packet loss rate of the abnormal sensor, minG is the minimum value of all supplementary parameters of the currently calculated packet loss rate of the abnormal sensor, maxH is the maximum value of all supplementary parameters of the currently calculated transmission frequency of the abnormal sensor, and minH is the minimum value of all supplementary parameters of the currently calculated transmission frequency of the abnormal sensor.

[0038] Preferably, the final determination module of the Internet of Things system based on big data and AI includes:

[0039] A third calculation submodule is used to obtain a risk determination value of each normal sensor based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors;

[0040] The final judgment submodule is used to determine whether the risk judgment value of each normal sensor is greater than the preset risk judgment threshold. If so, the existence of intrusion is regarded as the final intrusion judgment result of the corresponding normal sensor. Otherwise, further detection is required as the final intrusion judgment result of the corresponding normal sensor.

[0041] Preferably, in the Internet of Things system based on big data and AI, the method in which the third calculation submodule obtains the risk determination value of each normal sensor based on the intrusion risk assessment value of all normal sensors and the intrusion risk reference value of all abnormal sensors includes:

[0042]

[0043] Among them, ρ is the risk judgment value of the normal sensor currently calculated, β1 is the intrusion risk assessment value of the normal sensor currently calculated, maxβ is the maximum value of the intrusion risk assessment values ​​of all normal sensors, minβ is the minimum value of the intrusion risk assessment values ​​of all normal sensors, γ0 is the mean of the intrusion risk reference values ​​of all abnormal sensors, ln is the natural logarithm, and the value of the natural constant e is 2.718.

[0044] Preferably, the IoT system based on big data and AI, the detection module includes:

[0045] When the preliminary intrusion determination result of each sensor of the IoT system is that there is an intrusion, the control alarm will be used as the intrusion detection result of the IoT system;

[0046] When the preliminary intrusion determination result of each sensor of the IoT system is that further detection is required, and the final intrusion determination result of the corresponding sensor is that intrusion exists, the alarm of the control alarm is regarded as the intrusion detection result of the IoT system;

[0047] When the preliminary intrusion judgment result of each sensor of the Internet of Things system is that further detection is required, and the final intrusion judgment result of the corresponding sensor is that further detection is required, the number of the corresponding sensor will be sent to the administrator backend as the intrusion detection result of the Internet of Things system.

[0048] The beneficial effects of the present invention compared with the prior art are as follows: according to all types of transmission parameters of all sensors of the Internet of Things system at each moment in a preset time period before the current moment, the preliminary intrusion judgment result of each sensor of the Internet of Things system is accurately obtained, which is convenient for the subsequent classification of all sensors of the Internet of Things system; according to the preliminary intrusion judgment result of all sensors of the Internet of Things system, all normal sensors and all abnormal sensors are obtained; and then according to all types of transmission parameters of all abnormal sensors at all moments in a preset time period before the current moment and AI technology, all supplementary parameters of each type of transmission parameters of all abnormal sensors are obtained, so as to supplement each type of transmission parameters of all abnormal sensors, which is convenient for the subsequent accurate calculation of the intrusion risk reference value of the abnormal sensor; according to the preset time period of each normal sensor before the current moment, the intrusion risk reference value of each abnormal sensor is calculated; All types of transmission parameters at all times in the time period are used to obtain the intrusion risk assessment value of each normal sensor, so as to quantify the possibility of potential intrusion of a single normal sensor. According to all supplementary parameters of each type of transmission parameters of each abnormal sensor, the intrusion risk reference value of each abnormal sensor is obtained, so as to quantify the possibility of potential intrusion of a single abnormal sensor. Finally, according to the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors, the final intrusion judgment results of all normal sensors are accurately obtained. According to the preliminary intrusion judgment results of each sensor of the Internet of Things system and the final intrusion judgment results of all normal sensors, the intrusion detection results of the Internet of Things system are obtained, so as to improve the intrusion detection capability of sensors in the Internet of Things system and ensure the safe operation of the Internet of Things.

[0049] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the written application document.

[0050] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0052] Figure 1 Schematic diagram of an Internet of Things system based on big data and AI in an embodiment of the present invention;

[0053] Figure 2 Detailed schematic diagram of the preliminary determination module in the embodiment of the present invention. DETAILED DESCRIPTION

[0054] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0055] Embodiment 1:

[0056] The present invention provides an Internet of Things system based on big data and AI, referring to Figure 1 ,include:

[0057] A preliminary determination module, used to obtain a preliminary intrusion determination result of each sensor of the Internet of Things system based on all class transmission parameters of all sensors of the Internet of Things system at each moment within a preset time period before the current moment;

[0058] A processing module, used for obtaining all normal sensors and all abnormal sensors based on preliminary intrusion determination results of all sensors of the IoT system, and obtaining all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters and AI technology at all times within a preset time period before the current time.

[0059] A calculation module, used to obtain an intrusion risk assessment value of each normal sensor based on all types of transmission parameters of each normal sensor at all times within a preset time period before the current time, and to obtain an intrusion risk reference value of each abnormal sensor based on all supplementary parameters of each type of transmission parameters of each abnormal sensor;

[0060] A final determination module, used to obtain final intrusion determination results of all normal sensors based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors;

[0061] The detection module is used to obtain the intrusion detection result of the Internet of Things system based on the preliminary intrusion determination result of each sensor of the Internet of Things system and the final intrusion determination result of all normal sensors.

[0062] In this embodiment, the Internet of Things system is a network system that connects various physical devices, sensors, software, and other technologies through the Internet, enabling them to communicate with each other and exchange data.

[0063] In this embodiment, the sensor is a core component of the Internet of Things system and is responsible for sensing and measuring various physical quantities, such as temperature, humidity, pressure, light, etc., and converting this information into digital signals for processing and analysis by computers or other devices.

[0064] In this embodiment, the preset time period is a preset time period for obtaining all types of transmission parameters of all sensors, for example, 10 minutes.

[0065] In this embodiment, each moment is a moment selected from a preset time period before the current moment, and the time length between each adjacent moment is the same (for example, 1 s).

[0066] In this embodiment, the preliminary intrusion determination result is a determination result obtained based on all types of transmission parameters of all sensors of the Internet of Things system at each moment in a preset time period before the current moment, which can preliminarily reflect whether each sensor of the Internet of Things system has been invaded, wherein the preliminary intrusion determination result includes the existence of intrusion and the need for further detection.

[0067] In this embodiment, normal sensors are all sensors in the Internet of Things system whose preliminary intrusion determination results indicate that further detection is required.

[0068] In this embodiment, the abnormal sensors are all sensors in the Internet of Things system whose preliminary intrusion determination results are that there is intrusion.

[0069] In this embodiment, the AI ​​technology is an existing machine learning model technology, such as generative adversarial networks (GANs) or reinforcement learning models.

[0070] In this embodiment, the supplementary parameters are parameters that can be used to supplement each type of transmission parameters of all abnormal sensors.

[0071] In this embodiment, the intrusion risk assessment value of the normal sensor is a value evaluated based on all types of transmission parameters of the single normal sensor at all times within a preset time period before the current time, and can represent the possibility of the single normal sensor being potentially invaded.

[0072] In this embodiment, the intrusion risk reference value of the abnormal sensor is a reference value obtained based on all supplementary parameters of each type of transmission parameters of the single abnormal sensor, and can characterize the possibility of the single abnormal sensor being potentially invaded.

[0073] In this embodiment, the final intrusion determination result of the normal sensor is a determination result obtained based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors, which can ultimately reflect whether all normal sensors have been invaded, wherein the final intrusion determination result includes the existence of intrusion and the need for further detection.

[0074] The beneficial effects of the above technology are: according to all types of transmission parameters of all sensors of the Internet of Things system at each moment in a preset time period before the current moment, the preliminary intrusion judgment result of each sensor of the Internet of Things system is accurately obtained, which is convenient for the subsequent classification of all sensors of the Internet of Things system; according to the preliminary intrusion judgment result of all sensors of the Internet of Things system, all normal sensors and all abnormal sensors are obtained; and then according to all types of transmission parameters of all abnormal sensors at all moments in the preset time period before the current moment and AI technology, all supplementary parameters of each type of transmission parameters of all abnormal sensors are obtained, so as to supplement each type of transmission parameters of all abnormal sensors, which is convenient for the subsequent accurate calculation of the intrusion risk reference value of the abnormal sensor ... All types of transmission parameters at all times are used to obtain the intrusion risk assessment value of each normal sensor, so as to quantify the possibility of potential intrusion of a single normal sensor. According to all supplementary parameters of each type of transmission parameters of each abnormal sensor, the intrusion risk reference value of each abnormal sensor is obtained, so as to quantify the possibility of potential intrusion of a single abnormal sensor. Finally, according to the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors, the final intrusion judgment results of all normal sensors are accurately obtained. According to the preliminary intrusion judgment results of each sensor of the Internet of Things system and the final intrusion judgment results of all normal sensors, the intrusion detection results of the Internet of Things system are obtained, so as to improve the intrusion detection capability of sensors in the Internet of Things system and ensure the safe operation of the Internet of Things.

[0075] Embodiment 2:

[0076] Based on Example 1, the IoT system based on big data and AI, the preliminary determination module, reference Figure 2 ,include:

[0077] The acquisition submodule is used to obtain all class transmission parameters of all sensors in the IoT system at each moment in a preset time period before the current moment, wherein all class transmission parameters include transmission time, packet loss rate and sending frequency;

[0078] The preliminary determination submodule is used to obtain a preliminary intrusion determination result of each sensor of the Internet of Things system based on all class transmission parameters of all sensors of the Internet of Things system at each moment within a preset time period before the current moment.

[0079] In this embodiment, the transmission time at each moment is the average value of the time taken by each sensor of the Internet of Things system to send or receive all data packets between each moment and the adjacent previous moment within a preset time period before the current moment.

[0080] In this embodiment, the packet loss rate at each moment is the ratio of the number of data packets lost by each sensor in the sending and receiving process of the IoT system between each moment and the adjacent previous moment within a preset time period before the current moment to the total number of data packets sent and received.

[0081] In this embodiment, the sending frequency is the quotient of the number of data packets sent by each sensor of the Internet of Things system between each moment and the adjacent previous moment within a preset time period before the current moment and the time interval between the corresponding moment and the adjacent previous moment.

[0082] The beneficial effects of the above technology are: clarifying the specific parameter items of all types of transmission parameters of all sensors in the Internet of Things system at each moment in the preset time period before the current moment, and accurately obtaining the preliminary intrusion judgment results of each sensor in the Internet of Things system based on all types of transmission parameters of all sensors in the Internet of Things system at each moment in the preset time period before the current moment, so as to facilitate the subsequent classification of all sensors in the Internet of Things system.

[0083] Embodiment 3:

[0084] On the basis of Example 2, the Internet of Things system based on big data and AI preliminarily determines the sub-modules, including:

[0085] A judging unit, used to judge whether the difference between each type of transmission parameter of each sensor of the Internet of Things system at each moment within a preset time period before the current moment and the corresponding type of transmission parameters at all adjacent moments is less than a preset difference of the corresponding type of transmission parameter, and if so, regard the corresponding moment as an abnormal moment of the corresponding type of transmission parameter of the corresponding sensor, otherwise, regard the corresponding moment as a non-abnormal moment of the corresponding type of transmission parameter of the corresponding sensor;

[0086] The preliminary judgment unit is used to regard the need for further detection as the preliminary intrusion judgment result of each sensor of the Internet of Things system when the number of abnormal moments of all types of transmission parameters of each sensor of the Internet of Things system is 0, and to regard the existence of intrusion as the preliminary intrusion judgment result of each sensor of the Internet of Things system when the number of abnormal moments of all types of transmission parameters of each sensor of the Internet of Things system is not 0.

[0087] In this embodiment, the preset difference is a preset difference value used to obtain the abnormal time and the non-abnormal time of each type of transmission parameter of each sensor, and each type of transmission parameter corresponds to a preset difference value separately.

[0088] The beneficial effects of the above technology are: based on all types of transmission parameters of all sensors in the Internet of Things system at each moment in a preset time period before the current moment and the preset difference of each type of transmission parameters, the abnormal moment and non-abnormal moment of each type of transmission parameter of each sensor are accurately obtained, and then based on the abnormal moment and non-abnormal moment of each type of transmission parameter of each sensor, the preliminary intrusion judgment result of each sensor in the Internet of Things system is accurately obtained, which is convenient for the subsequent classification of all sensors in the Internet of Things system. This embodiment gives in detail a specific method for accurately obtaining the preliminary intrusion judgment result of each sensor in the Internet of Things system based on all types of transmission parameters of all sensors in the Internet of Things system at each moment in a preset time period before the current moment.

[0089] Embodiment 4:

[0090] On the basis of Example 3, the Internet of Things system based on big data and AI, the processing module includes:

[0091] The discrimination submodule is used to treat all sensors in the IoT system that are initially judged to need further detection as normal sensors, and treat all sensors in the IoT system that are initially judged to have been invaded as abnormal sensors;

[0092] The processing submodule is used to obtain all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters of all abnormal sensors at all times within a preset time period before the current moment and AI technology.

[0093] The beneficial effects of the above technology are: based on all preliminary intrusion judgment results in the Internet of Things system, all normal sensors and all abnormal sensors are accurately obtained, and then based on all types of transmission parameters of all abnormal sensors at all times in a preset time period before the current moment and AI technology, all supplementary parameters of each type of transmission parameters of all abnormal sensors are obtained, and each type of transmission parameters of all abnormal sensors are supplemented, which facilitates the accurate calculation of the intrusion risk reference value of the subsequent abnormal sensors.

[0094] Embodiment 5:

[0095] On the basis of Example 4, in the Internet of Things system based on big data and AI, the processing submodule obtains all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters of all abnormal sensors at all times within a preset time period before the current moment and AI technology, including:

[0096] Determine whether the number of abnormal moments of each type of transmission parameter of each abnormal sensor is greater than a preset number threshold. If so, take the average value of the corresponding type of transmission parameter of the corresponding abnormal sensor at all times within a preset time period before the current time as a supplementary parameter of the corresponding type of transmission parameter of the corresponding abnormal sensor;

[0097] Otherwise, all the moments of the corresponding abnormal sensor in the preset time period before the current moment are regarded as the horizontal coordinate value, and the corresponding class transmission parameter of the corresponding abnormal sensor at the corresponding moment in the preset time period before the current moment is regarded as the vertical coordinate value, to obtain all the sample points of the corresponding class transmission parameter of the corresponding abnormal sensor;

[0098] All sample points with all non-abnormal moments of the corresponding class of transmission parameters as horizontal coordinate values ​​in all sample points of each type of transmission parameters of each abnormal sensor are regarded as all non-abnormal sample points of the corresponding class of transmission parameters of the corresponding abnormal sensor, and all sample points with all abnormal moments of the corresponding class of transmission parameters as horizontal coordinate values ​​in all sample points of each type of transmission parameters of each abnormal sensor are regarded as all abnormal sample points of the corresponding class of transmission parameters of the corresponding abnormal sensor;

[0099] The mean value of the distance between each abnormal sample point of each type of transmission parameter of each abnormal sensor and all non-abnormal sample points of the corresponding type of transmission parameter of the corresponding abnormal sensor is regarded as the supplementary sample distance of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor, and the mean value of the supplementary sample distance of all abnormal sample points of each type of transmission parameter of each abnormal sensor is regarded as the supplementary distance of each type of transmission parameter of each abnormal sensor;

[0100] Taking each abnormal sample point of each type of transmission parameter of each abnormal sensor as the origin and the supplementary distance of the corresponding type of transmission parameter of the corresponding abnormal sensor as the radius, a supplementary circle of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor is obtained. Based on the supplementary circles of all abnormal sample points of each type of transmission parameters of all abnormal sensors and AI technology, all supplementary parameters of each type of transmission parameters of all abnormal sensors are obtained.

[0101] In this embodiment, the preset number threshold is a pre-set number threshold, for example, 10.

[0102] In this embodiment, the supplementary sample distance of the abnormal sample point is the average value of the distance between each abnormal sample point of each type of transmission parameter of each abnormal sensor and all non-abnormal sample points of the corresponding type of transmission parameter of the corresponding abnormal sensor.

[0103] In this embodiment, the supplementary distance of each type of transmission parameter of the abnormal sensor is the average value of the supplementary sample distances of all abnormal sample points of each type of transmission parameter of each abnormal sensor.

[0104] The beneficial effects of the above technology are as follows: according to all classes of transmission parameters of all abnormal sensors at all times within a preset time period before the current moment, all sample points of all classes of transmission parameters of all abnormal sensors are obtained, and then according to all sample points of all classes of transmission parameters of all abnormal sensors, all abnormal sample points and all non-abnormal sample points of all classes of transmission parameters of all abnormal sensors are obtained; according to all abnormal sample points and all non-abnormal sample points of all classes of transmission parameters of all abnormal sensors, the supplementary circle of all abnormal sample points of each class of transmission parameters of all abnormal sensors is obtained; finally, according to the supplementary circle of all abnormal sample points of each class of transmission parameters of all abnormal sensors, all supplementary parameters of each class of transmission parameters of all abnormal sensors are obtained, so as to supplement each class of transmission parameters of all abnormal sensors. This embodiment gives in detail a specific method for obtaining all supplementary parameters of each class of transmission parameters of all abnormal sensors according to all classes of transmission parameters of all abnormal sensors at all times within a preset time period before the current moment.

[0105] Embodiment 6:

[0106] On the basis of Example 5, the Internet of Things system based on big data and AI, based on the supplementary circles of all abnormal sample points of each type of transmission parameters of all abnormal sensors and AI technology, obtains all supplementary parameters of each type of transmission parameters of all abnormal sensors, including:

[0107] The maximum value of the distance between each abnormal sample point of each type of transmission parameter of each abnormal sensor and all non-abnormal sample points of the corresponding type of transmission parameter of the corresponding abnormal sensor is taken as the supplementary reference distance of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor, and the mean value of the supplementary reference distances of all abnormal sample points of each type of transmission parameter of each abnormal sensor is taken as the reference distance of each type of transmission parameter of each abnormal sensor;

[0108] Taking each abnormal sample point of each type of transmission parameter of each abnormal sensor as the origin and the reference distance of the corresponding type of transmission parameter of the corresponding abnormal sensor as the radius, a reference circle of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor is obtained, and the non-overlapping area of ​​the reference circle and the supplementary circle of each abnormal sample point of each type of transmission parameter of each abnormal sensor is regarded as the reference area of ​​each abnormal sample point of each type of transmission parameter of each abnormal sensor;

[0109] Based on the reference area, the supplementary circle and the AI ​​technology of each abnormal sample point of each type of transmission parameter of each abnormal sensor, all supplementary parameters of each type of transmission parameter of each abnormal sensor are obtained.

[0110] In this embodiment, the supplementary reference distance of the abnormal sample point is the maximum value of the distance between each abnormal sample point of each type of transmission parameter of each abnormal sensor and all non-abnormal sample points of the corresponding type of transmission parameter of the corresponding abnormal sensor.

[0111] In this embodiment, the reference distance of each type of transmission parameter of the abnormal sensor is the average value of the supplementary reference distances of all abnormal sample points of each type of transmission parameter of each abnormal sensor.

[0112] In this embodiment, based on the reference area, supplementary circle and AI technology of each abnormal sample point of each type of transmission parameter of each abnormal sensor, the specific steps of obtaining all supplementary parameters of each type of transmission parameter of each abnormal sensor are as follows:

[0113] The standard deviation of the density distribution of all non-abnormal sample points in the reference area of ​​each abnormal sample point of each type of transmission parameter of each abnormal sensor is analyzed, and then a preset number of new points (for example, 10, etc.) are placed in the supplementary circle of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor according to AI technology (generative adversarial networks (GANs) or reinforcement learning models, etc.), and the standard deviation of the density distribution of all new points in the supplementary circle of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor after the new points are placed is consistent with the standard deviation of the density distribution of all non-abnormal sample points in the reference area of ​​the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor, and the ordinate values ​​of all new points in the supplementary circle of each abnormal sample point of each type of transmission parameter of each abnormal sensor are used as all supplementary parameters of the corresponding type of transmission parameter of the corresponding abnormal sensor.

[0114] The beneficial effect of the above technology is: according to the supplementary circle of all abnormal sample points of each type of transmission parameters of all abnormal sensors and the AI ​​technology, all supplementary parameters of each type of transmission parameters of all abnormal sensors are obtained, and each type of transmission parameters of all abnormal sensors are supplemented. This embodiment gives in detail a specific method for obtaining all supplementary parameters of each type of transmission parameters of all abnormal sensors according to the supplementary circle of all abnormal sample points of each type of transmission parameters of all abnormal sensors and the AI ​​technology.

[0115] Embodiment 7:

[0116] On the basis of Example 1, the Internet of Things system based on big data and AI, the computing module includes:

[0117] The first calculation submodule is used to obtain the intrusion risk assessment value of each normal sensor based on all class transmission parameters of each normal sensor at all times within a preset time period before the current time, that is:

[0118]

[0119] Wherein, β is the intrusion risk assessment value of the normal sensor currently calculated, maxA is the maximum value of the transmission time of the normal sensor currently calculated at all times in the preset time period before the current moment, minA is the minimum value of the transmission time of the normal sensor currently calculated at all times in the preset time period before the current moment, maxB is the maximum value of the packet loss rate of the normal sensor currently calculated at all times in the preset time period before the current moment, minB is the minimum value of the packet loss rate of the normal sensor currently calculated at all times in the preset time period before the current moment, maxC is the maximum value of the sending frequency of the normal sensor currently calculated at all times in the preset time period before the current moment, minC is the minimum value of the sending frequency of the normal sensor currently calculated at all times in the preset time period before the current moment, ln is the natural logarithm, and the value of the natural constant e is 2.718;

[0120] The second calculation submodule is used to obtain the intrusion risk reference value of each abnormal sensor based on all supplementary parameters of each type of transmission parameters of each abnormal sensor, that is:

[0121]

[0122] Among them, γ is the intrusion risk reference value of the currently calculated abnormal sensor, maxF is the maximum value of all supplementary parameters of the currently calculated transmission time of the abnormal sensor, minF is the minimum value of all supplementary parameters of the currently calculated transmission time of the abnormal sensor, maxG is the maximum value of all supplementary parameters of the currently calculated packet loss rate of the abnormal sensor, minG is the minimum value of all supplementary parameters of the currently calculated packet loss rate of the abnormal sensor, maxH is the maximum value of all supplementary parameters of the currently calculated transmission frequency of the abnormal sensor, and minH is the minimum value of all supplementary parameters of the currently calculated transmission frequency of the abnormal sensor.

[0123] The beneficial effects of the above technology are: based on all types of transmission parameters of each normal sensor at all times within a preset time period before the current moment, the intrusion risk assessment value of each normal sensor is obtained, thereby quantifying the possibility of each normal sensor being potentially invaded, and based on all supplementary parameters of each type of transmission parameters of each abnormal sensor, the intrusion risk reference value of each abnormal sensor is obtained, thereby quantifying the possibility of each abnormal sensor being potentially invaded. This embodiment gives in detail a specific method for quantifying the possibility of a normal sensor or an abnormal sensor being potentially invaded.

[0124] Embodiment 8:

[0125] On the basis of Example 1, the final determination module of the Internet of Things system based on big data and AI includes:

[0126] A third calculation submodule is used to obtain a risk determination value of each normal sensor based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors;

[0127] The final judgment submodule is used to determine whether the risk judgment value of each normal sensor is greater than the preset risk judgment threshold. If so, the existence of intrusion is regarded as the final intrusion judgment result of the corresponding normal sensor. Otherwise, further detection is required as the final intrusion judgment result of the corresponding normal sensor.

[0128] In this embodiment, the risk determination value of a normal sensor is a numerical value determined based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors, which can characterize the risk level (possibility) of each normal sensor being potentially invaded in the entire Internet of Things system. The higher the risk determination value, the greater the possibility that the corresponding normal sensor is potentially invaded in the entire Internet of Things system.

[0129] In this embodiment, the preset risk judgment threshold is a threshold of a risk judgment value preset to obtain a final intrusion judgment result of a normal sensor.

[0130] The beneficial effects of the above technology are: based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors, the risk judgment value of each normal sensor is accurately obtained, and then based on the risk judgment value of each normal sensor, the final intrusion judgment result of each normal sensor is accurately obtained, which is convenient for subsequent acquisition of intrusion detection results of the Internet of Things system.

[0131] Embodiment 9:

[0132] On the basis of Example 8, in the Internet of Things system based on big data and AI, the third calculation submodule obtains the risk determination value of each normal sensor based on the intrusion risk assessment value of all normal sensors and the intrusion risk reference value of all abnormal sensors, including:

[0133]

[0134] Among them, ρ is the risk judgment value of the normal sensor currently calculated, β1 is the intrusion risk assessment value of the normal sensor currently calculated, maxβ is the maximum value of the intrusion risk assessment values ​​of all normal sensors, minβ is the minimum value of the intrusion risk assessment values ​​of all normal sensors, γ0 is the mean of the intrusion risk reference values ​​of all abnormal sensors, ln is the natural logarithm, and the value of the natural constant e is 2.718.

[0135] The beneficial effect of the above technology is: based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors, the risk determination value of each normal sensor is accurately obtained. This embodiment gives in detail a specific method for quantifying the risk level of potential intrusion of each normal sensor in the entire Internet of Things system.

[0136] Embodiment 10:

[0137] On the basis of Example 1, the Internet of Things system based on big data and AI, the detection module includes:

[0138] When the preliminary intrusion determination result of each sensor of the IoT system is that there is an intrusion, the control alarm will be used as the intrusion detection result of the IoT system;

[0139] When the preliminary intrusion determination result of each sensor of the IoT system is that further detection is required, and the final intrusion determination result of the corresponding sensor is that intrusion exists, the alarm of the control alarm is regarded as the intrusion detection result of the IoT system;

[0140] When the preliminary intrusion judgment result of each sensor of the Internet of Things system is that further detection is required, and the final intrusion judgment result of the corresponding sensor is that further detection is required, the number of the corresponding sensor will be sent to the administrator backend as the intrusion detection result of the Internet of Things system.

[0141] The beneficial effects of the above technology are: based on the preliminary intrusion judgment results of each sensor of the Internet of Things system and the final intrusion judgment results of all normal sensors, the intrusion detection results of the Internet of Things system are accurately obtained, the intrusion detection capabilities of sensors in the Internet of Things system are improved, and the safe operation of the Internet of Things is ensured.

[0142] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention, and the present invention is also intended to include these changes and modifications.

Claims

1. An Internet of Things system based on big data and AI, characterized in that: include: A preliminary determination module, used to obtain a preliminary intrusion determination result of each sensor of the Internet of Things system based on all class transmission parameters of all sensors of the Internet of Things system at each moment within a preset time period before the current moment; A processing module, used for obtaining all normal sensors and all abnormal sensors based on preliminary intrusion determination results of all sensors of the IoT system, and obtaining all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters and AI technology at all times within a preset time period before the current time. A calculation module, used to obtain an intrusion risk assessment value of each normal sensor based on all types of transmission parameters of each normal sensor at all times within a preset time period before the current time, and to obtain an intrusion risk reference value of each abnormal sensor based on all supplementary parameters of each type of transmission parameters of each abnormal sensor; A final determination module, used to obtain final intrusion determination results of all normal sensors based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors; The detection module is used to obtain the intrusion detection result of the Internet of Things system based on the preliminary intrusion determination result of each sensor of the Internet of Things system and the final intrusion determination result of all normal sensors.

2. The Internet of Things system based on big data and AI according to claim 1, characterized in that: The preliminary determination module includes: The acquisition submodule is used to obtain all class transmission parameters of all sensors in the IoT system at each moment in a preset time period before the current moment, wherein all class transmission parameters include transmission time, packet loss rate and sending frequency; The preliminary determination submodule is used to obtain a preliminary intrusion determination result of each sensor of the Internet of Things system based on all class transmission parameters of all sensors of the Internet of Things system at each moment within a preset time period before the current moment.

3. The Internet of Things system based on big data and AI according to claim 2, characterized in that: The preliminary determination submodule includes: A judging unit, used to judge whether the difference between each type of transmission parameter of each sensor of the Internet of Things system at each moment within a preset time period before the current moment and the corresponding type of transmission parameters at all adjacent moments is less than a preset difference of the corresponding type of transmission parameter, and if so, regard the corresponding moment as an abnormal moment of the corresponding type of transmission parameter of the corresponding sensor, otherwise, regard the corresponding moment as a non-abnormal moment of the corresponding type of transmission parameter of the corresponding sensor; The preliminary judgment unit is used to regard the need for further detection as the preliminary intrusion judgment result of each sensor of the Internet of Things system when the number of abnormal moments of all types of transmission parameters of each sensor of the Internet of Things system is 0, and to regard the existence of intrusion as the preliminary intrusion judgment result of each sensor of the Internet of Things system when the number of abnormal moments of all types of transmission parameters of each sensor of the Internet of Things system is not 0.

4. The Internet of Things system based on big data and AI according to claim 3, characterized in that: Processing module, including: The discrimination submodule is used to treat all sensors in the IoT system that are initially judged to need further detection as normal sensors, and treat all sensors in the IoT system that are initially judged to have been invaded as abnormal sensors; The processing submodule is used to obtain all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters of all abnormal sensors at all times within a preset time period before the current moment and AI technology.

5. The Internet of Things system based on big data and AI according to claim 4, characterized in that: The processing submodule obtains all supplementary parameters of each type of transmission parameters of all abnormal sensors based on all types of transmission parameters of all abnormal sensors at all times within a preset time period before the current time and AI technology, including: Determine whether the number of abnormal moments of each type of transmission parameter of each abnormal sensor is greater than a preset number threshold. If so, take the average value of the corresponding type of transmission parameter of the corresponding abnormal sensor at all times within a preset time period before the current time as a supplementary parameter of the corresponding type of transmission parameter of the corresponding abnormal sensor; Otherwise, all the moments of the corresponding abnormal sensor in the preset time period before the current moment are regarded as the horizontal coordinate value, and the corresponding class transmission parameter of the corresponding abnormal sensor at the corresponding moment in the preset time period before the current moment is regarded as the vertical coordinate value, to obtain all the sample points of the corresponding class transmission parameter of the corresponding abnormal sensor; All sample points with all non-abnormal moments of the corresponding class of transmission parameters as horizontal coordinate values ​​in all sample points of each type of transmission parameters of each abnormal sensor are regarded as all non-abnormal sample points of the corresponding class of transmission parameters of the corresponding abnormal sensor, and all sample points with all abnormal moments of the corresponding class of transmission parameters as horizontal coordinate values ​​in all sample points of each type of transmission parameters of each abnormal sensor are regarded as all abnormal sample points of the corresponding class of transmission parameters of the corresponding abnormal sensor; The mean value of the distance between each abnormal sample point of each type of transmission parameter of each abnormal sensor and all non-abnormal sample points of the corresponding type of transmission parameter of the corresponding abnormal sensor is regarded as the supplementary sample distance of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor, and the mean value of the supplementary sample distance of all abnormal sample points of each type of transmission parameter of each abnormal sensor is regarded as the supplementary distance of each type of transmission parameter of each abnormal sensor; Taking each abnormal sample point of each type of transmission parameter of each abnormal sensor as the origin and the supplementary distance of the corresponding type of transmission parameter of the corresponding abnormal sensor as the radius, a supplementary circle of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor is obtained. Based on the supplementary circles of all abnormal sample points of each type of transmission parameters of all abnormal sensors and AI technology, all supplementary parameters of each type of transmission parameters of all abnormal sensors are obtained.

6. The Internet of Things system based on big data and AI according to claim 5, characterized in that: Based on the supplementary circles of all abnormal sample points of each type of transmission parameters of all abnormal sensors and AI technology, all supplementary parameters of each type of transmission parameters of all abnormal sensors are obtained, including: The maximum value of the distance between each abnormal sample point of each type of transmission parameter of each abnormal sensor and all non-abnormal sample points of the corresponding type of transmission parameter of the corresponding abnormal sensor is taken as the supplementary reference distance of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor, and the mean value of the supplementary reference distances of all abnormal sample points of each type of transmission parameter of each abnormal sensor is taken as the reference distance of each type of transmission parameter of each abnormal sensor; Taking each abnormal sample point of each type of transmission parameter of each abnormal sensor as the origin and the reference distance of the corresponding type of transmission parameter of the corresponding abnormal sensor as the radius, a reference circle of the corresponding abnormal sample point of the corresponding type of transmission parameter of the corresponding abnormal sensor is obtained, and the non-overlapping area of ​​the reference circle and the supplementary circle of each abnormal sample point of each type of transmission parameter of each abnormal sensor is regarded as the reference area of ​​each abnormal sample point of each type of transmission parameter of each abnormal sensor; Based on the reference area, the supplementary circle and the AI ​​technology of each abnormal sample point of each type of transmission parameter of each abnormal sensor, all supplementary parameters of each type of transmission parameter of each abnormal sensor are obtained.

7. The Internet of Things system based on big data and AI according to claim 1, characterized in that: Computing module, including: The first calculation submodule is used to obtain the intrusion risk assessment value of each normal sensor based on all class transmission parameters of each normal sensor at all times within a preset time period before the current time, that is: Wherein, β is the intrusion risk assessment value of the normal sensor currently calculated, maxA is the maximum value of the transmission time of the normal sensor currently calculated at all times in the preset time period before the current moment, minA is the minimum value of the transmission time of the normal sensor currently calculated at all times in the preset time period before the current moment, maxB is the maximum value of the packet loss rate of the normal sensor currently calculated at all times in the preset time period before the current moment, minB is the minimum value of the packet loss rate of the normal sensor currently calculated at all times in the preset time period before the current moment, maxC is the maximum value of the sending frequency of the normal sensor currently calculated at all times in the preset time period before the current moment, minC is the minimum value of the sending frequency of the normal sensor currently calculated at all times in the preset time period before the current moment, ln is the natural logarithm, and the value of the natural constant e is 2.718; The second calculation submodule is used to obtain the intrusion risk reference value of each abnormal sensor based on all supplementary parameters of each type of transmission parameters of each abnormal sensor, that is: Among them, γ is the intrusion risk reference value of the currently calculated abnormal sensor, maxF is the maximum value of all supplementary parameters of the currently calculated transmission time of the abnormal sensor, minF is the minimum value of all supplementary parameters of the currently calculated transmission time of the abnormal sensor, maxG is the maximum value of all supplementary parameters of the currently calculated packet loss rate of the abnormal sensor, minG is the minimum value of all supplementary parameters of the currently calculated packet loss rate of the abnormal sensor, maxH is the maximum value of all supplementary parameters of the currently calculated transmission frequency of the abnormal sensor, and minH is the minimum value of all supplementary parameters of the currently calculated transmission frequency of the abnormal sensor.

8. The Internet of Things system based on big data and AI according to claim 1, characterized in that: The final judgment module includes: A third calculation submodule is used to obtain a risk determination value of each normal sensor based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors; The final judgment submodule is used to determine whether the risk judgment value of each normal sensor is greater than the preset risk judgment threshold. If so, the existence of intrusion is regarded as the final intrusion judgment result of the corresponding normal sensor. Otherwise, further detection is required as the final intrusion judgment result of the corresponding normal sensor.

9. The Internet of Things system based on big data and AI according to claim 8, characterized in that: The method of obtaining the risk determination value of each normal sensor by the third calculation submodule based on the intrusion risk assessment values ​​of all normal sensors and the intrusion risk reference values ​​of all abnormal sensors includes: Among them, ρ is the risk judgment value of the normal sensor currently calculated, β1 is the intrusion risk assessment value of the normal sensor currently calculated, maxβ is the maximum value of the intrusion risk assessment values ​​of all normal sensors, minβ is the minimum value of the intrusion risk assessment values ​​of all normal sensors, γ0 is the mean of the intrusion risk reference values ​​of all abnormal sensors, ln is the natural logarithm, and the value of the natural constant e is 2.

718.

10. The Internet of Things system based on big data and AI according to claim 1, characterized in that: Detection module, including: When the preliminary intrusion determination result of each sensor of the IoT system is that there is an intrusion, the control alarm will be used as the intrusion detection result of the IoT system; When the preliminary intrusion determination result of each sensor of the IoT system is that further detection is required, and the final intrusion determination result of the corresponding sensor is that intrusion exists, the alarm of the control alarm is regarded as the intrusion detection result of the IoT system; When the preliminary intrusion judgment result of each sensor of the Internet of Things system is that further detection is required, and the final intrusion judgment result of the corresponding sensor is that further detection is required, the number of the corresponding sensor will be sent to the administrator backend as the intrusion detection result of the Internet of Things system.

Images