A network state monitoring management method and system based on the Internet of Things

Abstract

This invention provides a network status monitoring and management method and system based on the Internet of Things (IoT), relating to the field of IoT technology. The method includes: collecting historical status information of networks accessible to IoT terminals within a historical time range, as well as real-time status information of the networks currently accessed by the IoT terminals; constructing a network monitoring model based on the collected historical status information of accessible networks; using the network monitoring model to identify real-time network status information and obtain a first identification result; analyzing the first identification result and taking corresponding control measures. The network monitoring model is obtained by training a neural network using the collected historical network status information; the network monitoring model is used to identify real-time network status information of accessible networks, completing network anomaly detection and analysis, and taking corresponding control measures to ensure the normal operation of the network.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to a network status monitoring and management method and system based on IoT. Background Technology

[0002] In recent years, the rapid development of online platforms has made the internet an indispensable part of people's daily learning, life, and work. The Internet of Things (IoT), an important component of the new generation of information technology, is an extension and expansion of the internet that can connect any object to the internet for information exchange and communication.

[0003] However, the widespread adoption of IoT technology has also brought significant risks to IoT network security. Existing IoT technologies generally focus on functional implementation, which inevitably leads to many network anomalies that prevent the network from functioning properly. Therefore, ensuring network availability and versatility is of paramount importance.

[0004] Therefore, the present invention provides a network status monitoring and management method and system based on the Internet of Things. Summary of the Invention

[0005] This invention provides a network status monitoring and management method and system based on the Internet of Things (IoT). The method uses historical network status information to train a neural network to obtain a network monitoring model. The network monitoring model is used to identify real-time network status information of accessible networks, complete the detection and analysis of network anomalies, and take corresponding control measures to ensure the normal operation of the network.

[0006] This invention provides a network status monitoring and management method based on the Internet of Things, comprising:

[0007] Step 1: Collect historical status information of the network accessibility of IoT terminals within a historical time range, as well as real-time status information of the current network access of the IoT terminals;

[0008] Step 2: Based on the collected historical status information of accessible networks, construct a network monitoring model;

[0009] Step 3: Using the network monitoring model, identify the real-time network status information to obtain the first identification result;

[0010] Step 4: Analyze the first identification result and take corresponding control measures.

[0011] Preferably, a network monitoring model is constructed based on the collected historical status information of accessible networks, including:

[0012] Step 11: Determine the target network traffic characteristics of IoT terminals accessing the network within the historical time range;

[0013] Step 12: Obtain the queue length, queue delay, and packet loss rate of IoT terminals accessing the network within a historical time range;

[0014] Step 13: Use the acquired target network traffic characteristics as the first parameter, and the queue length, queue delay, and packet loss rate of the inbound and outbound queues as the first state information;

[0015] Step 14: Parse the first state information to obtain the first state data;

[0016] Step 15: Train the neural network using the first parameter and the first state data to obtain the network monitoring model.

[0017] Preferably, the target network traffic characteristics include: network traffic generated by the IoT terminal accessing the network, the total amount of all TCP and UDP traffic, the sum of the message lengths of the uplink and downlink traffic accessing the network, the number of target IP addresses interacting with the IoT terminal, and the duration of the network traffic.

[0018] Preferably, the network monitoring model is used to identify real-time network status information to obtain a first identification result, including:

[0019] The system obtains real-time network status information of the networks that the IoT terminal can currently access, and analyzes the real-time parameters and real-time status information.

[0020] The real-time status information is parsed to obtain real-time status data;

[0021] The real-time parameters and real-time status data are input into the network monitoring model for identification. If at least one of the various parameters in the real-time parameters exceeds the corresponding preset threshold, it is determined that there is a network anomaly in the current network. At this time, the current network is detected for anomaly based on the detection method that matches the parameter that exceeds the corresponding preset threshold, and the detection result is output as the first identification result.

[0022] If none of the parameters in the real-time parameters exceed the preset threshold, then the real-time status data is identified.

[0023] If any data in the real-time status data exceeds the corresponding warning threshold, the current network is judged to be abnormal; otherwise, the current network status is judged to be normal. In this case, the judgment result of the current network status is output as the first identification result.

[0024] Preferably, anomaly detection is performed on the current network based on a detection method that matches parameters exceeding a corresponding preset threshold, and the detection result is output as the first identification result, including:

[0025] Step 21: Extract the first parameter within the same time zone as the current time within the historical time range, and divide it into several groups of comparison parameters according to the daily segmentation dimension;

[0026] Step 22: Standardize all comparison parameters and real-time parameters, and calculate the anomaly scores between the standardized real-time parameter data and the standardized comparison parameter data for all groups:

[0027]

[0028] Where S represents the anomaly score between the standardized real-time parameter data and the standardized comparison parameter data of the i-th group; (d i -f) represents the standardized comparison parameter data of the i-th group. i The edit distance difference between the normalized real-time parameter data f and the edit distance d. i Let f represent the standardized comparison parameter data of the i-th group, i∈[1,N], where N represents the number of comparison parameter groups after segmentation; f represents the standardized real-time parameter data. It represents the error coefficient in the calculation process of the distance difference between the standardized comparison parameter data and the real-time parameter data of the i-th group;

[0029] Step 23: Compare the calculated anomaly score with the anomaly threshold;

[0030] If the abnormal score is not greater than the abnormal threshold, the current network is judged to be normal;

[0031] Otherwise, determine that there is an anomaly in the current network.

[0032] Preferably, the first identification result is analyzed, and corresponding control measures are taken, including:

[0033] Extract the first identification result. If there is a detection result indicating abnormal current network traffic, set a routing strategy to implement traffic scheduling.

[0034] Otherwise, if the current network status is determined to be normal based on the first identification result, the current communication transmission rules shall be maintained.

[0035] When the current network status is determined to be abnormal and congested based on the first identification result, the queuing delay of the data packets being queued is checked, and the priority of the data packets and the data packet capacity are adjusted and controlled based on the warning threshold and the preset congestion threshold to limit the transmission rate.

[0036] If the current network is in a state of continuous congestion, discard data packets whose queuing delay is greater than the preset congestion threshold;

[0037] If the current network status is in a congestion-relieved state, update the warning threshold and the preset congestion threshold.

[0038] Preferably, if there is a detection result indicating abnormal current network traffic, the process of setting a routing policy to implement traffic scheduling also includes:

[0039] Acquire basic information about currently accessible networks and collect real-time network traffic status;

[0040] Estimate the bandwidth requirements corresponding to the real-time network traffic of the accessible network;

[0041] If the bandwidth requirement is not less than the preset bandwidth threshold, the scheduler searches for all possible paths for the current traffic, calculates the congestion degree of each path, and selects the path with the lowest congestion degree as the optimal forwarding path.

[0042]

[0043] Among them, Y j V is represented as the congestion degree of the j-th path; k1 The load is represented by the real-time network traffic at time point k1; t represents the traffic transmission duration; D j μ1 represents the bandwidth capacity of the j-th path; μ2 represents the probability factor of heavy network load; μ3 represents the probability factor of tight bandwidth resources for the j-th path.

[0044] The gain of the transmission path of the current network is compared with the traffic scheduling cost. If the loss of gain is greater than the loss of traffic scheduling cost, the optimal forwarding path is adopted for traffic transmission as a routing strategy to achieve traffic scheduling.

[0045] Otherwise, maintain the current network traffic transmission path for accessible networks.

[0046] This invention provides a network status monitoring and management system based on the Internet of Things, comprising:

[0047] Information acquisition module: used to collect historical status information of the network that the IoT terminal can access within a historical time range, as well as real-time status information of the network currently accessed by the IoT terminal;

[0048] Model building module: used to build a network monitoring model based on the collected historical status information of the access network;

[0049] Model recognition module: used to identify real-time network status information through the network monitoring model and obtain a first recognition result;

[0050] Control and management module: used to analyze the first identification results and make corresponding control measures.

[0051] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0052] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0053] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0054] Figure 1 This invention provides a network status monitoring and management method based on the Internet of Things (IoT).

[0055] Figure 2 This invention relates to a network status monitoring and management system based on the Internet of Things (IoT). Detailed Implementation

[0056] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0057] This invention provides a network status monitoring and management method based on the Internet of Things, such as... Figure 1 As shown, it includes:

[0058] Step 1: Collect historical status information of the network accessibility of IoT terminals within a historical time range, as well as real-time status information of the current network access of the IoT terminals;

[0059] Step 2: Based on the collected historical status information of accessible networks, construct a network monitoring model;

[0060] Step 3: Using the network monitoring model, identify the real-time network status information to obtain the first identification result;

[0061] Step 4: Analyze the first identification result and take corresponding control measures.

[0062] In this embodiment, the Internet of Things (IoT) terminal refers to a device in the Internet of Things that connects the sensor network layer and the transmission network layer, and realizes data collection and data transmission to the network layer.

[0063] In this embodiment, the status information includes network traffic characteristics, the queue lengths of the network's inbound and outbound queues, queuing delay, and packet loss rate.

[0064] In this embodiment, the network monitoring model is obtained by training a neural network using historical state information.

[0065] In this embodiment, the first identification result is obtained by inputting real-time network status information of the accessible network into the network monitoring model for identification, and targeted control measures are taken based on the first identification result to ensure the normal operation of the network, such as controlling data packet capacity, limiting transmission rate, and traffic scheduling.

[0066] The beneficial effects of the above technical solution are: a network monitoring model is obtained by training a neural network through the collected historical network status information; the network monitoring model is used to identify the real-time network status information of the accessible network, complete the detection and analysis of network anomalies, and take corresponding control measures to ensure the normal operation of the network.

[0067] This invention provides a network status monitoring and management method based on the Internet of Things (IoT), which constructs a network monitoring model based on collected historical status information of accessible networks, including:

[0068] Step 11: Determine the target network traffic characteristics of IoT terminals accessing the network within the historical time range;

[0069] Step 12: Obtain the queue length, queue delay, and packet loss rate of IoT terminals accessing the network within a historical time range;

[0070] Step 13: Use the acquired target network traffic characteristics as the first parameter, and the queue length, queue delay, and packet loss rate of the inbound and outbound queues as the first state information;

[0071] Step 14: Parse the first state information to obtain the first state data;

[0072] Step 15: Train the neural network using the first parameter and the first state data to obtain the network monitoring model.

[0073] In this embodiment, the target network traffic characteristics mainly include network traffic generated by IoT terminals accessing the network, the total amount of all TCP and UDP traffic, the sum of the message lengths of uplink and downlink traffic accessing the network, the number of target IP addresses interacting with IoT terminals, and the duration of network traffic, which are used as the first parameter to train the neural network.

[0074] In this embodiment, the first state information refers to the queue length, queue delay, and packet loss rate of IoT terminals accessing the network within a historical time range; the first state data is obtained by translating the first state information.

[0075] The beneficial effects of the above technical solution are: by using the target network traffic characteristics as the first parameter, the network monitoring model is obtained by training a neural network with the first state data obtained by parsing the queue length, queue delay and packet loss rate of the inbound and outbound queues. This enables effective monitoring of network anomalies.

[0076] This invention provides a network status monitoring and management method based on the Internet of Things (IoT). The method utilizes a network monitoring model to identify real-time network status information and obtain a first identification result, including:

[0077] The system obtains real-time network status information of the networks that the IoT terminal can currently access, and analyzes the real-time parameters and real-time status information.

[0078] The real-time status information is parsed to obtain real-time status data;

[0079] The real-time parameters and real-time status data are input into the network monitoring model for identification. If at least one of the various parameters in the real-time parameters exceeds the corresponding preset threshold, it is determined that there is a network anomaly in the current network. At this time, the current network is detected for anomaly based on the detection method that matches the parameter that exceeds the corresponding preset threshold, and the detection result is output as the first identification result.

[0080] If none of the parameters in the real-time parameters exceed the preset threshold, then the real-time status data is identified.

[0081] If any data in the real-time status data exceeds the corresponding warning threshold, the current network is judged to be abnormal; otherwise, the current network status is judged to be normal. In this case, the judgment result of the current network status is output as the first identification result.

[0082] In this embodiment, the real-time network status information mainly includes the current network traffic characteristics, the queuing length of the network's inbound and outbound queues, queuing delay, and packet loss rate.

[0083] In this embodiment, real-time parameters refer to the network traffic generated by the currently accessible network, the total amount of all TCP and UDP traffic, the sum of the packet lengths of the uplink and downlink traffic of the access network, the number of target IP addresses interacting with the IoT terminal, and the duration of network traffic; real-time status information refers to the current queue lengths of the inbound and outbound queues of the accessible network, the queue delay, and the packet loss rate.

[0084] In this embodiment, real-time status data refers to the data obtained by parsing, or translating, real-time status information.

[0085] In this embodiment, the preset thresholds corresponding to various parameters in the real-time parameters are pre-set.

[0086] In this embodiment, for example, if there is an accessible network 1, and the sum of the message lengths of the currently generated uplink traffic and downlink traffic is greater than the corresponding preset threshold, it is determined that there is a network anomaly in the current network. At this time, the anomaly detection of the currently accessible network 1 is performed by a detection method that matches the sum of the message lengths of the uplink traffic and downlink traffic.

[0087] In this embodiment, for example, if there is an accessible network 2, and the real-time parameters input into the network monitoring model do not exceed the corresponding preset threshold, the network monitoring model is used to identify the real-time status data of the accessible network 2.

[0088] At this time, the packet loss rate in the real-time status data of the accessible network 2 exceeds the corresponding warning threshold, so it is determined that there is an anomaly in the accessible network.

[0089] In this embodiment, the warning thresholds corresponding to various types of data in the real-time status data are pre-set. The beneficial effects of the above technical solution are: by inputting the real-time parameters and real-time status information of the accessible network into the network monitoring model, and combining the preset thresholds and warning thresholds for identification; the network status of the current network is determined based on the obtained identification results, thereby realizing effective monitoring of network anomalies and laying the foundation for subsequent determination of control measures.

[0090] This invention provides a network status monitoring and management method based on the Internet of Things (IoT). The method performs anomaly detection on the current network based on a detection method that matches parameters exceeding a corresponding preset threshold, and outputs the detection result as a first identification result. The method includes:

[0091] Step 21: Extract the first parameter within the same time zone as the current time within the historical time range, and divide it into several groups of comparison parameters according to the daily segmentation dimension;

[0092] Step 22: Standardize all comparison parameters and real-time parameters, and calculate the anomaly scores between the standardized real-time parameter data and the standardized comparison parameter data for all groups:

[0093]

[0094] Where S represents the anomaly score between the standardized real-time parameter data and the standardized comparison parameter data of the i-th group; (d i -f) represents the standardized comparison parameter data of the i-th group. i The edit distance difference between the normalized real-time parameter data f and the edit distance d. i Let f represent the standardized comparison parameter data of the i-th group, i∈[1,N], where N represents the number of comparison parameter groups after segmentation; f represents the standardized real-time parameter data. It represents the error coefficient in the calculation process of the distance difference between the standardized comparison parameter data and the real-time parameter data of the i-th group;

[0095] Step 23: Compare the calculated anomaly score with the anomaly threshold;

[0096] If the abnormal score is not greater than the abnormal threshold, the current network is judged to be normal;

[0097] Otherwise, determine that there is an anomaly in the current network.

[0098] In this embodiment, the comparison parameter is the parameter obtained by dividing the first parameter within the same time range as the current time within the historical time range according to the daily segmentation dimension. The daily segmentation dimension refers to dividing the first parameter within the historical time range according to the day.

[0099] In this embodiment, the standardization of the comparison parameters and real-time parameters is to unify the units of measurement, which is beneficial for subsequent calculation operations.

[0100] In this embodiment, the abnormal threshold is preset, typically set to 0.6;

[0101] In this embodiment, the abnormal score S ranges from (0,1);

[0102] In this embodiment, for example, if the abnormal score of the second group of standardized comparison parameters and real-time parameters is 0.2, which is not greater than the abnormal threshold, then the current network is judged to be normal.

[0103] The beneficial effects of the above technical solution are: by standardizing several comparison parameters obtained by dividing the first parameter into dimensions on a daily basis and then combining them with real-time parameters, anomaly scores are obtained by combining them with formulas; by using the obtained anomaly scores to determine whether the current network is normal, effective monitoring of the network status is achieved.

[0104] This invention provides a network status monitoring and management method based on the Internet of Things (IoT), which analyzes the first identification result and takes corresponding control measures, including:

[0105] Extract the first identification result. If there is a detection result indicating abnormal current network traffic, set a routing strategy to implement traffic scheduling.

[0106] Otherwise, if the current network status is determined to be normal based on the first identification result, the current communication transmission rules shall be maintained.

[0107] When the current network status is determined to be abnormal and congested based on the first identification result, the queuing delay of the data packets being queued is checked, and the priority of the data packets and the data packet capacity are adjusted and controlled based on the warning threshold and the preset congestion threshold to limit the transmission rate.

[0108] If the current network is in a state of continuous congestion, discard data packets whose queuing delay is greater than the preset congestion threshold;

[0109] If the current network status is in a congestion-relieved state, update the warning threshold and the preset congestion threshold.

[0110] In this embodiment, traffic scheduling refers to the process of adjusting the traffic of the current network according to the routing strategy to achieve traffic control. The main content of the routing strategy is to change the transmission path of the current network.

[0111] The beneficial effects of the above technical solution are: by analyzing the obtained identification results, it is possible to determine the areas of the current network that are abnormal, and to take corresponding control measures to effectively mitigate network anomalies, thereby ensuring the normal operation of the network.

[0112] This invention provides a network status monitoring and management method based on the Internet of Things. If an anomaly in current network traffic is detected, the method further includes setting a routing policy to implement traffic scheduling, and includes:

[0113] Acquire basic information about currently accessible networks and collect real-time network traffic status;

[0114] Estimate the bandwidth requirements corresponding to the real-time network traffic of the accessible network;

[0115] If the bandwidth requirement is not less than the preset bandwidth threshold, the scheduler searches for all possible paths for the current traffic, calculates the congestion degree of each path, and selects the path with the lowest congestion degree as the optimal forwarding path.

[0116]

[0117] Among them, Y j V is represented as the congestion degree of the j-th path; k1 The load is represented by the real-time network traffic at time point k1; t represents the traffic transmission duration; D j μ1 represents the bandwidth capacity of the j-th path; μ2 represents the probability factor of heavy network load; μ3 represents the probability factor of tight bandwidth resources for the j-th path.

[0118] The gain of the transmission path of the current network is compared with the traffic scheduling cost. If the loss of gain is greater than the loss of traffic scheduling cost, the optimal forwarding path is adopted for traffic transmission as a routing strategy to achieve traffic scheduling.

[0119] Otherwise, maintain the current network traffic transmission path for accessible networks.

[0120] In this embodiment, the congestion level Yj The range of values ​​for is (0,1);

[0121] In this embodiment, the basic data of the network that can be accessed mainly refers to the network topology, number of hosts, link bandwidth, amount of traffic that may be generated, bandwidth requirement range of traffic, and selectable forwarding paths of traffic in the network; the real-time network traffic status mainly refers to the current amount of data transmitted in the network and network traffic demand information. Since the network traffic status is dynamic, the real-time network traffic status is collected periodically.

[0122] In this embodiment, the preset bandwidth threshold is pre-set based on the current basic network information.

[0123] In this embodiment, for example, if there are possible paths 1, 2, and 3 for the current traffic, the congestion degree of paths 1, 2, and 3 is calculated using a formula. The calculated results are compared pairwise to find that path 1 has the smallest congestion degree. At this time, path 1 is taken as the optimal forwarding path for the current traffic.

[0124] In this embodiment, for example, there is an accessible network 3, whose current traffic transmission path A, after adopting a routing strategy, results in the optimal forwarding path B. Since the gain loss of the current accessible network 3 transmission path is greater than the loss of traffic scheduling cost, the optimal forwarding path B is selected for traffic transmission.

[0125] The beneficial effects of the above technical solution are as follows: the bandwidth requirement is estimated by using basic information data of the accessible network and real-time network traffic status; based on the condition that the bandwidth requirement is met, all possible paths of the current traffic and the congestion degree of each path are obtained, and the path with the lowest congestion degree is selected as the optimal path; based on the comparison between the gain loss of the transmission path of the accessible network and the traffic scheduling cost loss, the transmission path of the accessible network is determined, which serves as a control measure to effectively respond to the detection results of abnormal current network traffic.

[0126] This invention provides a network status monitoring and management system based on the Internet of Things, such as... Figure 2 As shown, it includes:

[0127] Information acquisition module: used to collect historical status information of the network that the IoT terminal can access within a historical time range, as well as real-time status information of the network currently accessed by the IoT terminal;

[0128] Model building module: used to build a network monitoring model based on the collected historical status information of the access network;

[0129] Model recognition module: used to identify real-time network status information through the network monitoring model and obtain a first recognition result;

[0130] Control and management module: used to analyze the first identification results and make corresponding control measures.

[0131] The beneficial effects of the above technical solution are: a network monitoring model is obtained by training a neural network with collected historical network status information; the network monitoring model is used to identify the real-time network status information of accessible networks, complete the detection and analysis of network anomalies, and take corresponding control measures to ensure the normal operation of the network.

[0132] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A network status monitoring and management method based on the Internet of Things, characterized in that, include: Step 1: Collect historical status information of the network accessibility of IoT terminals within a historical time range, as well as real-time status information of the current network access of the IoT terminals; Step 2: Based on the collected historical status information of accessible networks, construct a network monitoring model; Step 3: Using the network monitoring model, identify the real-time network status information to obtain the first identification result; Step 4: Analyze the first identification result and take corresponding control measures; Using the network monitoring model, real-time network status information is identified to obtain a first identification result, including: The system obtains real-time network status information of the networks that the IoT terminal can currently access, and analyzes the real-time parameters and real-time status information. The real-time status information is parsed to obtain real-time status data; The real-time parameters and real-time status data are input into the network monitoring model for identification. If at least one of the various parameters in the real-time parameters exceeds the corresponding preset threshold, it is determined that there is a network anomaly in the current network. At this time, the current network is detected for anomaly based on the detection method that matches the parameter that exceeds the corresponding preset threshold, and the detection result is output as the first identification result. If none of the parameters in the real-time parameters exceed the preset threshold, then the real-time status data is identified. If any data in the real-time status data exceeds the corresponding warning threshold, the current network is judged to be abnormal; otherwise, the current network status is judged to be normal. In this case, the judgment result of the current network status is output as the first identification result. Real-time parameters refer to the network traffic generated by the currently accessible network, the total amount of all TCP and UDP traffic, the sum of the packet lengths of the uplink and downlink traffic to the access network, the number of target IP addresses interacting with IoT terminals, and the duration of network traffic; real-time status information refers to the current queue lengths of the inbound and outbound queues of the accessible network, the queue delay, and the packet loss rate.

2. The network status monitoring and management method based on the Internet of Things as described in claim 1, characterized in that, Based on the collected historical status information of accessible networks, a network monitoring model is constructed, including: Step 11: Determine the target network traffic characteristics of IoT terminals accessing the network within the historical time range; Step 12: Obtain the queue length, queue delay, and packet loss rate of IoT terminals accessing the network within a historical time range; Step 13: Use the acquired target network traffic characteristics as the first parameter, and the queue length, queue delay, and packet loss rate of the inbound and outbound queues as the first state information; Step 14: Parse the first state information to obtain the first state data; Step 15: Train the neural network using the first parameter and the first state data to obtain the network monitoring model.

3. The network status monitoring and management method based on the Internet of Things as described in claim 2, characterized in that, The target network traffic characteristics include: network traffic generated by IoT terminals accessing the network, the total amount of all TCP and UDP traffic, the sum of the message lengths of uplink and downlink traffic accessing the network, the number of target IP addresses interacting with IoT terminals, and the duration of network traffic.

4. The network status monitoring and management method based on the Internet of Things as described in claim 1, characterized in that, Anomaly detection is performed on the current network based on a detection method that matches parameters exceeding a corresponding preset threshold, and the detection result is output as the first identification result, including: Step 21: Extract the first parameter within the same time zone as the current time within the historical time range, and divide it into several groups of comparison parameters according to the daily segmentation dimension; Step 22: Standardize all comparison parameters and real-time parameters, and calculate the anomaly scores of the standardized real-time parameter data and all standardized comparison parameter data: Where S represents the anomaly score between the standardized real-time parameter data and the i-th standardized comparison parameter data; Represented as the standardized comparison parameter data of the i-th group. With standardized real-time parameter data Edit distance difference; This represents the standardized comparison parameter data for the i-th group. Where N represents the number of comparison parameter groups after segmentation; f represents the standardized real-time parameter data; It represents the error coefficient in the calculation process of the distance difference between the standardized comparison parameter data and the real-time parameter data of the i-th group; Step 23: Compare the calculated anomaly score with the anomaly threshold; If the abnormal score is not greater than the abnormal threshold, the current network is judged to be normal; Otherwise, determine that there is an anomaly in the current network.

5. The network status monitoring and management method based on the Internet of Things as described in claim 1, characterized in that, Analyze the first identification result and take corresponding control measures, including: Extract the first identification result. If there is a detection result indicating abnormal current network traffic, set a routing strategy to implement traffic scheduling. Otherwise, if the current network status is determined to be normal based on the first identification result, the current communication transmission rules shall be maintained. When the current network status is determined to be abnormal and congested based on the first identification result, the queuing delay of the data packets being queued is checked, and the priority of the data packets and the data packet capacity are adjusted and controlled based on the warning threshold and the preset congestion threshold to limit the transmission rate. If the current network is in a state of continuous congestion, discard data packets whose queuing delay is greater than the preset congestion threshold; If the current network status is in a congestion-relieved state, update the warning threshold and the preset congestion threshold.

6. A network status monitoring and management system based on the Internet of Things (IoT), applied in the network status monitoring and management method based on the IoT as described in claim 1, characterized in that, include: Information acquisition module: used to collect historical status information of the network that the IoT terminal can access within a historical time range, as well as real-time status information of the network currently accessed by the IoT terminal; Model building module: used to build a network monitoring model based on the collected historical status information of the access network; Model recognition module: used to identify real-time network status information through the network monitoring model and obtain a first recognition result; Control and management module: used to analyze the first identification result and take corresponding control measures.

Images