An internet of things device monitoring system and method based on big data

Abstract

The application relates to the technical field of cloud computing, and discloses an Internet of Things equipment monitoring system and method based on big data, which comprises an Internet of Things equipment monitoring environment identification module, a cloud processing strategy switching module, a cloud processing risk identification module, a cloud early warning information prompting module and an equipment monitoring maintenance tracking module, performs Internet of Things communication signal evaluation by acquiring Internet of Things equipment communication signal data, identifies abnormal targets based on the evaluation results, switches cloud processing strategies to perform cloud processing, monitors the cloud state in real time, acquires a risk index of a non-identification target unit and a risk index of an identification target tracking unit, performs risk evaluation according to the risk identification results, the cloud sends early warning information prompts according to the risk evaluation results, and the non-identification target unit and the identification abnormal target unit are maintained and tracked respectively, so that comprehensive monitoring and maintenance of the Internet of Things equipment are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud computing, more particularly to an Internet of Things device monitoring system and method based on big data. BACKGROUND

[0002] With the rapid development of Internet of Things technology, the number and application scenarios of Internet of Things devices are increasing, especially in the field of monitoring, through Internet of Things devices, regional monitoring is realized according to big data technology, but the dispersion and complexity of Internet of Things devices bring great challenges to the monitoring and maintenance of devices, traditional device monitoring methods often rely on manual inspection and regular maintenance, which is not only inefficient, but also difficult to find and handle device failures in time, in the prior art, there are some Internet of Things device monitoring systems, but these systems mostly have problems such as limited monitoring range and insufficient data processing capacity, especially when facing large-scale Internet of Things devices, these systems often have difficulty in realizing comprehensive monitoring and real-time warning of the devices.

[0003] On the other hand, although the monitoring devices based on Internet of Things can realize real-time monitoring of the monitoring area, they often only focus on monitoring itself and ignore in-depth analysis of Internet of Things communication signals, this single monitoring method may still perform real-time monitoring on the monitoring area when the quality of the Internet of Things communication signals is poor, on this basis, the system attempts to process and analyze the monitoring data using big data technology, however, due to the lack of consideration of the quality of the communication signals, this approach may cause the monitoring system to produce false alarms in some cases, that is, to falsely issue alarms, or to miss reports, that is, to fail to timely discover and report real abnormal situations, thereby affecting the efficiency of the entire monitoring system and the trust of users. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides an Internet of Things device monitoring system based on big data to solve the problems existing in the background art.

[0005] The present application provides the following technical solution: an Internet of Things device monitoring system based on big data, comprising: an Internet of Things device monitoring environment identification module, a cloud processing strategy switching module, a cloud processing risk identification module, a cloud early warning information prompting module and a device monitoring and maintenance tracking module;

[0006] The Internet of Things device monitoring environment identification module comprises an Internet of Things communication signal data acquisition unit and an Internet of Things communication signal evaluation unit, the Internet of Things communication signal data acquisition unit acquires Internet of Things device communication signal data, and the Internet of Things communication signal evaluation unit evaluates the Internet of Things communication signals according to the data;

[0007] The cloud processing strategy switching module includes a non-identified target unit and an identified target tracking unit, identifies an abnormal target based on an Internet of Things communication signal evaluation result of an Internet of Things device monitoring environment identification module, switches a cloud processing strategy for cloud processing;

[0008] The cloud processing risk identification module, when performing cloud processing, monitors a cloud state in real time, obtains a non-identified target unit risk index and an identified target tracking unit risk index, and performs risk evaluation according to a risk identification result;

[0009] The cloud early warning information prompting module, the cloud sends an early warning information prompt according to a risk evaluation result, including a non-identified target unit risk early warning information prompt and an identified abnormal target unit risk early warning information prompt;

[0010] The device monitoring and maintenance tracking module performs maintenance tracking on the non-identified target unit and the identified abnormal target unit according to the early warning information prompt, locates a maintenance tracking result, and transmits the maintenance tracking result to a man-machine interaction.

[0011] Preferably, the Internet of Things device monitoring environment identification module, an Internet of Things communication signal data acquisition unit obtains a receiving power of a communication signal at an environment sensor interface through an environment sensor, and the environment sensor performs communication denoising processing on the Internet of Things communication signal when performing Internet of Things communication signal data acquisition;

[0012] The Internet of Things communication signal evaluation unit evaluates the Internet of Things communication signal based on the receiving power of the communication signal at the environment sensor interface, and the calculation formula is: , wherein represents an Internet of Things communication signal evaluation value, represents a receiving power of a communication signal at an environment sensor interface, represents a preset reference receiving power of a communication signal, represents a preset receiving strength of a communication signal;

[0013] When the Internet of Things communication signal evaluation value is greater than or equal to a preset evaluation threshold value, the Internet of Things communication signal evaluation result is a normal communication signal, and a cloud hierarchical processing strategy switching instruction is sent to the cloud hierarchical processing strategy switching module; when the Internet of Things communication signal evaluation value is less than the preset evaluation threshold value, the Internet of Things communication signal evaluation result is an abnormal communication signal, and a cloud early warning information prompt instruction is sent to the cloud early warning information prompting module.

[0014] Preferably, in the cloud processing strategy switching module: the non-identified target unit is used for cloud security processing on the Internet of Things device; and the identified target tracking unit is used for cloud tracking processing on the identified target.

[0015] When the evaluation result of the Internet of Things communication signal is that the communication signal is normal, target recognition is performed on the monitoring area, when no mobile target appears in the monitoring area, the Internet of Things device is processed by the cloud end security processing unit without target recognition; when a mobile target appears in the monitoring area, a double-unit processing strategy is switched to utilize the no-target recognition unit and the target recognition and tracking unit for double-unit cloud end processing.

[0016] Preferably, the specific content of the no-target recognition unit for cloud end security processing of the Internet of Things device is as follows:

[0017] The data packets on the link from the Internet of Things device network to the cloud end network are divided into n sub-packets with a length of T, i = 1, 2, 3,..., n, wherein i represents the number of sub-packets;

[0018] The network situation of each sub-packet is analyzed according to the time window, and the network situation awareness index of each sub-packet is obtained, and the calculation formula is: , wherein represents the network situation awareness index of each sub-packet in the time window with a length of T, represents the number of attacks on each sub-packet in the time window with a length of T, represents the number of unpatched vulnerabilities of each sub-packet in the time window with a length of T, represents the total number of vulnerabilities generated by each sub-packet in the time window with a length of T, and represents the weight coefficient of the network situation awareness index.

[0019] Preferably, the specific content of the target recognition and tracking unit for cloud end tracking processing of the identified target is as follows:

[0020] The video stream of the identified target is obtained by the camera, and the frame processing of the video stream is performed to obtain M frame frame images of the identified target, m = 1, 2, 3,..., M, wherein m represents the number of frame images;

[0021] The pixel point coordinates in the target contour region in the frame image are calculated according to the formula: , wherein k represents the number of pixel points in the target contour region in each frame image, k = 1, 2, 3,..., K, K represents the total number of pixel points in the target contour region in each frame image, and the centroid point coordinates of each frame image are calculated according to the formula:

[0022] , wherein represents the abscissa of the centroid point of each frame image;

[0023] , wherein​ a longitudinal coordinate of a centroid point of each sub-frame image;

[0024] connecting the centroid point coordinates of each sub-frame image in sequence to obtain a cloud tracking path curve of the identified target.

[0025] Preferably, in the cloud processing risk identification module, the specific content of obtaining the no-identified target unit risk index and the identified target tracking unit risk index and performing risk assessment according to the risk identification result is as follows:

[0026] According to the network situation awareness index of each sub-data packet obtained by the no-identified target unit, the number of risk data packets is judged: when the network situation awareness index is greater than a preset risk judgment threshold, the data packet is judged as a risk data packet;

[0027] The no-identified target unit risk index is calculated, and the calculation formula is: , wherein the no-identified target unit risk index, the number of risk data packets judged by the no-identified target unit, the total number of data packets segmented by the no-identified target unit;

[0028] The identified target tracking unit risk index is calculated according to the cloud tracking path curve obtained by the identified target tracking unit, and the calculation formula is: , wherein the identified target tracking unit risk index, the interval time of the identified target from the mth centroid point to the m+1th centroid point.

[0029] Preferably, the specific content of performing risk assessment according to the risk identification result is as follows:

[0030] When the no-identified target unit risk index is greater than a preset cloud first threshold, the first risk assessment result is high risk, otherwise, the first risk assessment result is low risk;

[0031] When the identified target tracking unit risk index is greater than a preset cloud second threshold, the second risk assessment result is high risk, otherwise, the second risk assessment result is low risk.

[0032] Preferably, the cloud early warning information prompting module, according to the first risk assessment result, issues an Internet of Things device network early warning information prompt; according to the second risk assessment result, an identified target behavior early warning information prompt is issued.

[0033] Preferably, the device monitoring and maintenance tracking module, according to the early warning information prompt, respectively maintains and tracks the no-identified target unit and the identified abnormal target unit, locates the maintenance and tracking result and transmits it to the man-machine interaction end.

[0034] A big data-based Internet of Things device monitoring method, comprising the following steps:

[0035] Step S01: Obtain Internet of Things device communication signal data through an Internet of Things communication signal data acquisition unit, and perform Internet of Things communication signal evaluation according to the data;

[0036] Step S02: Identify an abnormal target based on the Internet of Things communication signal evaluation result, and switch a cloud processing strategy for cloud processing;

[0037] Step S03: Real-time monitor a cloud state, obtain a risk index of a non-identified target unit and a risk index of an identified target tracking unit, and perform risk evaluation according to a risk identification result;

[0038] Step S04: The cloud sends a warning information prompt according to the risk evaluation result;

[0039] Step S05: Maintain and track the non-identified target unit and the identified abnormal target unit according to the warning information prompt.

[0040] Technical effects and advantages of the present application:

[0041] The present application obtains the communication signal data of the Internet of Things device through the Internet of Things communication signal data acquisition unit, evaluates the data, discovers the communication signal abnormality in time, switches the cloud processing strategy, performs targeted processing on the abnormal target, simultaneously, real-time monitors the cloud state, obtains the risk index of the non-identified target unit and the risk index of the identified target tracking unit, and performs risk evaluation according to the risk identification result, thereby further improving the safety and reliability of the system;

[0042] When the risk evaluation result shows high risk, the cloud warning information prompt module sends a warning information prompt in time, so that the relevant personnel can quickly take measures for processing, finally, the device monitoring and maintenance tracking module maintains and tracks the non-identified target unit and the identified abnormal target unit according to the warning information prompt, thereby ensuring the normal operation and safety of the device. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 It is a structural schematic view of a big data-based Internet of Things device monitoring system.

[0044] Figure 2 It is a flow schematic view of a big data-based Internet of Things device monitoring method. DETAILED DESCRIPTION

[0045] The technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application. In addition, the forms of the structures described in the following embodiments are only examples, and the big data-based Internet of Things device monitoring system and method involved in the present application are not limited to the structures described in the following embodiments. All other embodiments obtained by those of ordinary skill in the art without creative labor belong to the scope of protection of the present application.

[0046] As shown in Figure 1 The present application provides a big data-based Internet of Things device monitoring system, which comprises an Internet of Things device monitoring environment identification module, a cloud processing strategy switching module, a cloud processing risk identification module, a cloud early warning information prompting module, and a device monitoring maintenance tracking module.

[0047] The Internet of Things device monitoring environment identification module comprises an Internet of Things communication signal data acquisition unit and an Internet of Things communication signal evaluation unit. The Internet of Things communication signal data acquisition unit acquires Internet of Things device communication signal data, and the Internet of Things communication signal evaluation unit evaluates the Internet of Things communication signal based on the data.

[0048] The cloud processing strategy switching module comprises a non-identification target unit and an identification target tracking unit. Based on the Internet of Things communication signal evaluation result of the Internet of Things device monitoring environment identification module, the abnormal target is identified, and the cloud processing strategy is switched for cloud processing.

[0049] The cloud processing risk identification module monitors the cloud state in real time when performing cloud processing, acquires the risk index of the non-identification target unit and the risk index of the identification target tracking unit, and performs risk evaluation based on the risk identification result.

[0050] The cloud early warning information prompting module sends early warning information prompts based on the risk evaluation result, including non-identification target unit risk early warning information prompts and identification abnormal target unit risk early warning information prompts.

[0051] The device monitoring maintenance tracking module performs maintenance tracking on the non-identification target unit and the identification abnormal target unit based on the early warning information prompts, locates the maintenance tracking results, and transmits them to the man-machine interaction end.

[0052] In this embodiment, it needs to be specifically pointed out that the Internet of Things device monitors the environment identification module, the Internet of Things communication signal data collection unit obtains the received power of the communication signal at the environment sensor interface through the environment sensor, and the environment sensor performs communication denoising processing on the Internet of Things communication signal when collecting the Internet of Things communication signal data. Noise is an unavoidable problem in the signal collection process, which will affect the authenticity and accuracy of the data. Through communication denoising processing, the influence of noise on the signal can be effectively reduced, thereby improving the accuracy of the collected data, which is crucial for the Internet of Things system because accurate data is the basis for subsequent analysis and decision-making.

[0053] The Internet of Things communication signal evaluation unit evaluates the Internet of Things communication signal based on the received power of the communication signal at the environment sensor interface, and the calculation formula is: , wherein represents the Internet of Things communication signal evaluation value, represents the received power of the communication signal at the environment sensor interface, represents the preset reference received power of the communication signal, represents the preset received intensity of the communication signal, wherein represents the received intensity of the communication signal at the environment sensor interface, and the ratio of the preset received intensity of the communication signal reflects the quality of the Internet of Things communication signal.

[0054] When the received power of the communication signal at the environment sensor interface is 0.2 mW, and the preset reference received power of the communication signal is 0.1 mW, the received intensity of the communication signal at the environment sensor interface is:

[0055] When the Internet of Things communication signal evaluation value is greater than or equal to the preset evaluation threshold, the Internet of Things communication signal evaluation result is normal communication signal, and the cloud hierarchical processing strategy switching instruction is sent to the cloud hierarchical processing strategy switching module. When the Internet of Things communication signal evaluation value is less than the preset evaluation threshold, the Internet of Things communication signal evaluation result is abnormal communication signal, and the cloud early warning information prompt instruction is sent to the cloud early warning information prompt module.

[0056] In this embodiment, it needs to be specifically pointed out that in the cloud processing strategy switching module: the no identification target unit is used for cloud security processing of the Internet of Things device; and the identification target tracking unit is used for cloud tracking processing of the identification target.

[0057] ​When the evaluation result of the Internet of Things communication signal is that the communication signal is normal, target recognition is performed on the monitoring area, when no mobile target appears in the monitoring area, the Internet of Things device is processed by the cloud through the non-identified target unit; when a mobile target appears in the monitoring area, a double-unit processing strategy is switched to use the non-identified target unit and the identified target tracking unit for double-unit cloud processing.

[0058] In this embodiment, it needs to be specifically pointed out that the specific content of the non-identified target unit for cloud security processing of the Internet of Things device is as follows:

[0059] The data packets on the link from the Internet of Things device network to the cloud network are divided into n sub-packets according to a time window with a length of T, i=1, 2, 3,..., n, wherein i represents the number of sub-packets;

[0060] According to the time window, the network situation of each sub-packet is analyzed, and the network situation awareness index of each sub-packet is obtained, and the calculation formula is: , wherein represents the network situation awareness index of each sub-packet in the time window with a length of T, and the larger the network situation awareness index is, the higher the risk is, represents the number of attacks on each sub-packet in the time window with a length of T, represents the number of unpatched vulnerabilities of each sub-packet in the time window with a length of T, represents the total number of vulnerabilities generated by each sub-packet in the time window with a length of T, and represents the weight coefficient of the network situation awareness index, and ;

[0061] The parameter setting time window T=1h, the weight coefficient of the network situation awareness index , , the number of attacks on each sub-packet in the time window with a length of T , the number of unpatched vulnerabilities of each sub-packet in the time window with a length of T , and the total number of vulnerabilities generated by each sub-packet in the time window with a length of T . .

[0062] In this embodiment, it needs to be specifically pointed out that the specific content of the identified target tracking unit for cloud tracking processing of the identified target is as follows:

[0063] The video stream of the identified target is obtained through the camera, the video stream is frame-processed to obtain M frames of frame images of the identified target, m=1, 2, 3,..., M, wherein m represents the number of frame images;

[0064] According to the pixel point coordinates in the target contour region in the frame image , wherein k represents the pixel point number in the target contour region in each frame image, k = 1, 2, 3, …, K, K represents the total number of pixel points in the target contour region in each frame image, and the centroid point coordinates of each frame image are calculated :

[0065] , wherein represents the horizontal coordinate of the centroid point of each frame image;

[0066] , wherein represents the vertical coordinate of the centroid point of each frame image;

[0067] The centroid point coordinates of each frame image are sequentially connected to obtain a cloud tracking path curve of the identified target.

[0068] In this embodiment, it is specifically pointed out that the specific content of the cloud processing risk identification module for obtaining the no-identified target unit risk index and the identified target tracking unit risk index and performing risk assessment according to the risk identification result is as follows:

[0069] According to the network situation awareness index of each sub-data packet obtained by the no-identified target unit, the number of risk data packets is judged: when the network situation awareness index is greater than a preset risk judgment threshold, the data packet is judged as a risk data packet;

[0070] The no-identified target unit risk index is calculated, and the calculation formula is: , wherein represents the no-identified target unit risk index, represents the number of risk data packets judged by the no-identified target unit, represents the total number of data packets cut by the no-identified target unit;

[0071] The identified target tracking unit risk index is calculated according to the cloud tracking path curve obtained by the identified target tracking unit, and the calculation formula is: , wherein represents the identified target tracking unit risk index, represents the interval time of the identified target from the mth centroid point to the m+1th centroid point, wherein represents the interval time of the identified target from the 1st centroid point to the 2nd centroid point, represents the interval time of the identified target from the 2nd centroid point to the 3rd centroid point, represents the interval time of the identified target from the 3rd centroid point to the 4th centroid point, The interval time from the M-1th centroid point to the Mth centroid point is represented as the identification target.

[0072] In this embodiment, it needs to be specifically pointed out that the specific content of risk assessment according to the risk identification result is as follows:

[0073] When the risk index of the non-identification target unit is greater than the preset cloud first threshold value, the first risk assessment result is high risk, otherwise, the first risk assessment result is low risk;

[0074] When the risk index of the identification target tracking unit is greater than the preset cloud second threshold value, the second risk assessment result is high risk, otherwise, the second risk assessment result is low risk.

[0075] In this embodiment, it needs to be specifically pointed out that the cloud early warning information prompting module issues Internet of Things device network early warning information prompts according to the first risk assessment result; and issues identification target behavior early warning information prompts according to the second risk assessment result.

[0076] In this embodiment, it needs to be specifically pointed out that the device monitoring and maintenance tracking module performs maintenance tracking on the non-identification target unit and the identification abnormal target unit according to the early warning information prompts respectively, locates the maintenance tracking results and transmits them to the man-machine interaction end.

[0077] As shown in Figure 2 In this embodiment, it needs to be specifically pointed out that a big data-based Internet of Things device monitoring method includes the following steps:

[0078] Step S01: Obtain Internet of Things device communication signal data through an Internet of Things communication signal data acquisition unit, and perform Internet of Things communication signal evaluation according to the data;

[0079] Step S02: Identify abnormal targets based on the Internet of Things communication signal evaluation results, switch cloud processing strategies for cloud processing;

[0080] Step S03: Real-time monitor the cloud state, obtain the risk index of the non-identification target unit and the risk index of the identification target tracking unit, and perform risk assessment according to the risk identification results;

[0081] Step S04: The cloud issues early warning information prompts according to the risk assessment results;

[0082] Step S05: Perform maintenance tracking on the non-identification target unit and the identification abnormal target unit according to the early warning information prompts respectively.

[0083] In this embodiment, it needs to be specifically pointed out that the difference between the present embodiment and the prior art is mainly that the communication signal data acquisition unit of the Internet of Things acquires the communication signal data of the Internet of Things device, and evaluates the data to discover communication signal abnormalities in time, so as to switch the cloud processing strategy, and process the abnormal target in a targeted manner. At the same time, the cloud state is monitored in real time, the risk index of the non-identification target unit and the risk index of the identification target tracking unit are obtained, and the risk is evaluated according to the risk identification result, thereby further improving the safety and reliability of the system.

[0084] When the risk assessment result shows high risk, the cloud early warning information prompt module will timely issue early warning information prompt, so that the relevant personnel can quickly take measures to process, finally, the equipment monitoring and maintenance tracking module carries out maintenance tracking to the non-identification target unit and the identification abnormal target unit according to the early warning information prompt, and ensures the normal operation and safety of the equipment.

[0085] Finally: the above only for the preferred embodiments of the present application, and not for limiting the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application, should be included in the protection scope of the present application.

[0086] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A big data based internet of things device monitoring system, characterized in that: The application relates to an IoT device monitoring environment identification module, a cloud processing strategy switching module, a cloud processing risk identification module, a cloud early warning information prompting module and a device monitoring maintenance tracking module. The IoT device monitoring environment identification module comprises an IoT communication signal data acquisition unit and an IoT communication signal evaluation unit, the IoT communication signal data acquisition unit acquires IoT device communication signal data, and the IoT communication signal evaluation unit evaluates the data. The cloud processing strategy switching module comprises a non-identification target unit and an identification target tracking unit, which identify abnormal targets based on the IoT communication signal evaluation result of the IoT device monitoring environment identification module and switch the cloud processing strategy for cloud processing. In the cloud processing strategy switching module, the non-identification target unit is used for cloud security processing of the IoT device, and the identification target tracking unit is used for cloud tracking processing of the identification target. When the IoT communication signal evaluation result is a normal communication signal, target identification is performed on the monitoring area, when no mobile target appears in the monitoring area, the non-identification target unit is used for cloud security processing of the IoT device, and when a mobile target appears in the monitoring area, a double-unit processing strategy is switched to perform double-unit cloud processing by using the non-identification target unit and the identification target tracking unit. The specific content of the non-identification target unit for cloud security processing of the IoT device is as follows: the data packets on the cloud network link from the IoT device network are divided into n sub-packets with a length of T, i=1, 2, 3,..., n, wherein i represents the number of sub-packets. The specific content of the identification target tracking unit for cloud tracking processing of the identification target is as follows: a video stream of the identification target is acquired by using a camera, frame processing is performed on the video stream to acquire M frames of frame images of the identification target, m=1, 2, 3,..., M, wherein m represents the number of frame images, the centroid point coordinates of each frame image are sequentially connected to acquire a cloud tracking path curve of the identification target. The cloud processing risk identification module monitors the cloud state in real time when performing cloud processing, acquires the risk index of the non-identification target unit and the risk index of the identification target tracking unit, and performs risk evaluation according to the risk identification result. The cloud early warning information prompting module sends early warning information prompts according to the risk evaluation result, including non-identification target unit risk early warning information prompts and identification abnormal target unit risk early warning information prompts. According to the time window, network situation analysis is performed on each sub data packet to obtain a network situation awareness index of each sub data packet, and a calculation formula is as follows: Wherein represents the network situation awareness index of each sub data packet in a time window with a length of T, represents the number of attacks on each sub data packet in a time window with a length of T, represents the number of unpatched vulnerabilities of each sub data packet in a time window with a length of T, represents the total number of vulnerabilities generated by each sub data packet in a time window with a length of T, and represents a weight coefficient of the network situation awareness index. The device monitoring maintenance tracking module performs maintenance tracking on the non-identification target unit and the identification abnormal target unit according to the early warning information prompts, locates the maintenance tracking result and transmits the maintenance tracking result to a man-machine interaction terminal. The IoT device monitoring environment identification module, the IoT communication signal data acquisition unit acquires the receiving power of the communication signal at the environment sensor interface through an environment sensor, and the environment sensor performs communication denoising processing on the IoT communication signal when performing IoT communication signal data acquisition. According to the pixel point coordinates in the target contour region in the frame image , wherein k represents the pixel point number in the target contour region in each frame image, k = 1, 2, 3, …, K, K represents the total number of pixel points in the target contour region in each frame image, and the centroid point coordinates of each frame image are calculated : wherein denotes the horizontal coordinate of the center of mass point of the respective subframe image; wherein denotes the longitudinal coordinate of the center of mass point of the respective subframe image; ​ ​ ​ ​ 2. The big data based IoT device monitoring system according to claim 1, wherein: ​ The Internet of Things communication signal evaluation unit evaluates the Internet of Things communication signal based on the received power of the communication signal at the environment sensor interface, and the calculation formula is: wherein represents the Internet of Things communication signal evaluation value, represents the received power of the communication signal at the environment sensor interface, represents the preset reference received power of the communication signal, represents the preset received intensity of the communication signal; When the Internet of Things communication signal evaluation value is greater than or equal to a preset evaluation threshold, the Internet of Things communication signal evaluation result is a normal communication signal, and a cloud hierarchical processing strategy switching instruction is sent to a cloud hierarchical processing strategy switching module; when the Internet of Things communication signal evaluation value is less than the preset evaluation threshold, the Internet of Things communication signal evaluation result is an abnormal communication signal, and a cloud early warning information prompt instruction is sent to a cloud early warning information prompt module.

3. The big data based IoT device monitoring system according to claim 1, wherein: In the cloud processing risk identification module, the specific content of obtaining the non-identification target unit risk index and the identification target tracking unit risk index and performing risk assessment according to the risk identification result is as follows: According to the network situation awareness index of each sub-data packet obtained by the non-identification target unit, the number of risk data packets is determined: when the network situation awareness index is greater than a preset risk judgment threshold, the data packet is determined as a risk data packet; The risk index of the non-identified target unit is calculated, and the calculation formula is: wherein represents the risk index of the non-identified target unit, represents the number of risk data packets determined by the non-identified target unit, represents the total number of data packets segmented by the non-identified target unit; The risk index of the identification target tracking unit is calculated according to a cloud tracking path curve acquired by the identification target tracking unit, and the calculation formula is: wherein represents the risk index of the identification target tracking unit, represents the interval time of the identification target from the mth centroid point to the m+1th centroid point.

4. The big data based IoT device monitoring system of claim 3, wherein: The specific content of performing risk assessment according to the risk identification result is as follows: When the non-identification target unit risk index is greater than a preset cloud first threshold, the first risk assessment result is high risk, otherwise, the first risk assessment result is low risk; When the identification target tracking unit risk index is greater than a preset cloud second threshold, the second risk assessment result is high risk, otherwise, the second risk assessment result is low risk.

5. The big data based IoT device monitoring system according to claim 1, wherein: The cloud early warning information prompt module sends an Internet of Things device network early warning information prompt according to the first risk assessment result, and sends an identification target behavior early warning information prompt according to the second risk assessment result.

6. The big data based IoT device monitoring system according to claim 1, wherein: The device monitoring and maintenance tracking module performs maintenance tracking on the non-identification target unit and the identification abnormal target unit according to the early warning information prompt, respectively, locates the maintenance tracking result, and transmits the maintenance tracking result to the man-machine interaction end. 7.A big data based IoT device monitoring method, used for a big data based IoT device monitoring system according to any one of claims 1-6. The method comprises the following steps: Step S01: obtaining Internet of Things device communication signal data through an Internet of Things communication signal data acquisition unit, and performing Internet of Things communication signal evaluation according to the data; Step S02: identifying abnormal targets based on the Internet of Things communication signal evaluation result, switching a cloud processing strategy, and performing cloud processing; Step S03: real-time monitoring of a cloud state, obtaining a non-identification target unit risk index and an identification target tracking unit risk index, and performing risk assessment according to a risk identification result; Step S04: sending an early warning information prompt according to a risk assessment result; Step S05: performing maintenance tracking on the non-identification target unit and the identification abnormal target unit according to the early warning information prompt, respectively.

Images