Low-power-consumption long-distance communication method based on Internet of Things

Abstract

The invention relates to the technical field of Internet of Things communication, in particular to an Internet of Things-based low-power-consumption remote communication method, which comprises the following steps of: acquiring data information of each Internet of Things node to form a communication behavior corresponding to each Internet of Things node; the method comprises the following steps: acquiring communication times and associated parameters of each communication behavior, and deriving a state label of each Internet of Things node according to a relative coverage range and a relative distance of each Internet of Things node; performing abnormal behavior analysis on the Internet of Things nodes in the different state labels based on the dormancy instruction received by each Internet of Things node, and evaluating the index deviation degree of each Internet of Things node under the abnormal communication behavior; generating a parameter processing sequence based on the index deviation degree of each Internet of Things node; and performing anomaly detection on the parameter processing sequence according to the number of abnormal nodes on each Internet of Things node during transmission, and adjusting the control strategy of each Internet of Things node according to the distribution difference of each abnormal node. And the reliability and efficiency of low-power-consumption remote communication are realized.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of Internet of Things communication, in particular to a low-power long-distance communication method based on the Internet of Things. BACKGROUND

[0002] With the popularity of Internet of Things devices, more and more Internet of Things devices are applied in various fields. Internet of Things devices are mainly used for terminal data collection, and are mostly powered by batteries. Most of the time, local data collection is realized, and data is transmitted to the Internet of Things management platform through wired or wireless means at regular intervals or in abnormal scenarios. However, when the Internet of Things node communicates at a long distance, the high transmission power easily leads to a difficult balance between energy consumption and communication quality.

[0003] For example, CN116669018A discloses a data processing method and device based on Internet of Things communication. When privacy data is detected in a mobile terminal, the mobile terminal no longer stores the privacy data in the mobile terminal itself, but transfers the privacy data to a region where the mobile terminal frequently stays, and a high-stability Internet of Things device. Further, to avoid the problem that a malicious user obtains the interaction record with the Internet of Things device through the mobile terminal and then tracks the privacy data, the application selects an Internet of Things device that has not interacted with the mobile terminal before, and selects another high-stability Internet of Things device as a transmission intermediary between the mobile terminal and the storage Internet of Things device, so that the mobile terminal and the storage Internet of Things device no longer directly interact, but indirectly obtain the privacy data through the transmission Internet of Things device.

[0004] For example, CN117793793A discloses an Internet of Things communication method, device and electronic device. The method includes receiving broadcast information of a transmission channel of the Internet of Things; determining a congestion level of the transmission channel and a delay level of transmission information of the transmission channel based on the broadcast information; determining a target device level corresponding to a device allowed to send information on the transmission channel based on the congestion level; in response to the target device level including the device level of the to-be-sent device, determining a target delay time length based on the delay level; and controlling the to-be-sent device to transmit information to the Internet of Things through the transmission channel according to the target delay time length.

[0005] The prior art respectively describes adjusting the retrieval request of the Internet of Things device by the transmission distance, and determining the control of the Internet of Things transmission by the demonstration time length during the communication congestion. However, the prior art ignores the working state of the Internet of Things device, which leads to the problem that the Internet of Things device cannot adapt to abnormal cooperative processing of multiple nodes after simple data collection in the sleep state, resulting in increased energy consumption and low communication anomaly recognition efficiency. SUMMARY

[0006] In order to solve the above technical problems, the technical scheme adopted by the present application is: a low-power long-distance communication method based on the Internet of Things, comprising: S1, collecting data information of each Internet of Things node based on initial configuration parameters and network topology of the Internet of Things node, and forming corresponding communication behaviors of each Internet of Things node.

[0007] S2, by acquiring the communication times and associated parameters of each communication behavior, analyzing the communication quality of each Internet of Things node during execution of the sleep instruction, and deriving the state label of each Internet of Things node based on the relative coverage range and relative distance of each Internet of Things node.

[0008] S3, based on the sleep instruction received by each Internet of Things node, performing abnormal behavior analysis on the Internet of Things nodes in different state labels, and evaluating the index deviation of each Internet of Things node under abnormal communication behavior.

[0009] S4, based on the index deviation of each Internet of Things node, identifying the path loss and signal-to-noise ratio of the current communication behavior, and generating a parameter processing sequence based on the differences of the communication behavior at multiple time points.

[0010] S5, according to the number of abnormal nodes existing on each Internet of Things node at the time of transmission, performing abnormal detection on the parameter processing sequence, and adjusting the control strategy of each Internet of Things node based on the distribution difference of each abnormal node.

[0011] The beneficial effects of the present application are: first, the present application generates a dynamic state label by analyzing the relative coverage range, distance and sleep instruction execution time window of the node, to identify the state of each Internet of Things node under the sleep instruction, and the classification processing of multiple devices under the Internet of Things node, and combines the time window translation sliding mechanism to adapt to the form of the Internet of Things node changing with time, and improve the accuracy of the state label.

[0012] Second, the present application constructs an element relationship network, combines communication links, signal strength and other elements, captures interference elements and locally traces, quantifies the index deviation, and preliminarily identifies the abnormal nodes existing on the current Internet of Things node. Then, according to the difference value of the state label, the path loss and the signal-to-noise ratio, the residual damage processing sequence is selected, and the currently preliminarily identified abnormal nodes are screened by merging, non-merging and preset deviation, to determine the relative relationship of the Internet of Things nodes in multiple environmental scenarios, to ensure the processing form and identification content of different data.

[0013] Thirdly, the application adjusts the control strategy of different Internet of Things nodes by respectively processing the continuity problem and local concentration problem of abnormal nodes, etc. by the number of abnormal nodes and the abnormal type of abnormal nodes, so as to shorten the abnormal response time, reduce the energy consumption of the whole Internet of Things node, and improve the abnormal processing efficiency of the Internet of Things node after executing the sleep instruction. BRIEF DESCRIPTION OF DRAWINGS

[0014] The application will be further described below in combination with the drawings and embodiments.

[0015] Figure 1 It is a flowchart of a low-power long-distance communication method based on the Internet of Things.

[0016] Figure 2 It is a flowchart of step S1 of a low-power long-distance communication method based on the Internet of Things.

[0017] Figure 3 It is a flowchart of step S2 of a low-power long-distance communication method based on the Internet of Things.

[0018] Figure 4 It is a flowchart of step S3 of a low-power long-distance communication method based on the Internet of Things.

[0019] Figure 5 It is a flowchart of step S4 of a low-power long-distance communication method based on the Internet of Things.

[0020] Figure 6 It is a flowchart of step S5 of a low-power long-distance communication method based on the Internet of Things. DETAILED DESCRIPTION

[0021] The embodiments of the application will be described in detail below. The embodiments described below are exemplary and are only used to explain the application, and cannot be understood as a limitation of the application. If the specific technology or condition is not indicated in the embodiments, the technology or condition described in the literature in the art or according to the product instruction is used.

[0022] Reference Figure 1 A low-power long-distance communication method based on the Internet of Things, comprising: S1, collecting data information of each Internet of Things node by using the initial configuration parameters and network topology structure of the Internet of Things node, and forming the corresponding communication behavior of each Internet of Things node.

[0023] S2, analyzing the communication quality of each Internet of Things node during the execution of the sleep instruction by obtaining the communication times and associated parameters of each communication behavior, and deducing the state label of each Internet of Things node by using the relative coverage range and relative distance of each Internet of Things node.

[0024] S3, based on the sleep instructions received by each IoT node, performs abnormal behavior analysis on IoT nodes in different status tags and evaluates the indicator deviation of each IoT node under abnormal communication behavior.

[0025] S4, based on the indicator deviation of each IoT node, identifies the path loss and signal-to-noise ratio of the current communication behavior, and generates a parameter processing sequence based on the differences in communication behavior at multiple time points.

[0026] S5, according to the number of abnormal nodes on each IoT node during transmission, perform anomaly detection on the parameter processing sequence, and adjust the control strategy of each IoT node based on the distribution difference of each abnormal node.

[0027] The above communication behavior will include relevant data such as transmission power, communication frequency band, modulation method, data rate, MAC layer parameters, etc. in the long-distance communication scenario.

[0028] Long-distance communication technologies include LoRa (based on CSS spread spectrum modulation), NB-IoT (narrowband IoT over cellular networks), and Sigfox (based on ultra-narrowband modulation). These three methods employ different topologies and processing methods. For example, LoRa's single-hop star topology supports transmissions ranging from several kilometers to tens of kilometers. It operates in the ISM band and supports multi-band polling to avoid interference. It also employs two-way communication and periodic or continuous monitoring to balance transmission costs.

[0029] For example, NB-IoT uses cellular network integration to achieve long-distance transmission at relatively close distances with strong indoor penetration; Sigfox uses uplink and downlink to transmit small data packets for cloud data transmission deployment.

[0030] In the current invention, the IoT nodes corresponding to these technologies are used to traverse the status of each IoT node to find out whether the transmission consumption meets the current acceptable conditions of each IoT node after the IoT nodes are deployed in different locations.

[0031] For example, the devices configured in the current IoT node may be environmental sensors, electricity meter and water meter sensors, parking sensors, equipment operation status sensors, and other sensors used to track multiple process steps of the current IoT node.

[0032] Depending on the technology and topology currently used, the path loss is evaluated in different ways. For example, if signal transmission is performed under a free space model and the transmission is considered to be in an obstacle-free environment, the path loss can be expressed as shown below.

[0033] ; wherein, denotes the path loss of the current transmission, and the value is a relative ideal calculation value used for describing the description mode of the unobstructed scene, denotes the frequency, and represents the frequency of the current transmission signal, denotes the distance, and represents the distance value described in the current long-distance transmission scene, which is the straight-line receiving distance of the transmission and reception.

[0034] Meanwhile, in the scene where the current propagation scene exists a satellite, the distance represented by the distance denotes the distance between the ground base station and the satellite, and the path loss can be further represented as ; denotes the distance between the ground base station and the satellite, and describes the use scene of the current Internet of Things node using the satellite signal transmission.

[0035] Or, in the Okumura-Hata model, the path loss is identified, which mainly identifies the loss of the near-ground electromagnetic wave transmission, that is, ; wherein, denotes the base station height, and denotes the height of the signal sending part; denotes the terminal height, and denotes the height of the signal receiving end, which can affect the signal transmission; denotes the correction factor, which is set according to the current identified corresponding area, for example, the value of the factor is adjusted in different terrains, such as 5 dB in hilly areas, and the correction factor value is also affected in small and medium-sized cities and large cities, and these values are set in the database in advance.

[0036] If there is a conflict between the dimensions and units of the data calculation during the calculation, the parameters are standardized, and the parameters with the dimension are eliminated for calculation.

[0037] As for the signal-to-noise ratio, it is determined by the ratio of the known signal power and the noise power, at this time, the filter can be used to process the data sent by each Internet of Things node, to determine whether the low-power running mode of each Internet of Things node during the execution of the sleep instruction will cause the loss of data.

[0038] In an embodiment of the present application, as shown in Figure 2 the implementation mode of step S1 includes: S11, according to the number of Internet of Things nodes associated with the network topology structure, taking the deployed Internet of Things node as the starting point, after viewing the initial configuration parameters, the Internet of Things nodes are aggregated, and the transmission delay of the Internet of Things nodes in the long-distance scene is obtained according to the preset distance around each Internet of Things node after aggregation.

[0039] S12, in response to the transmission delay of the current Internet of Things node, comparing the communication times of the current Internet of Things node and the associated parameters, and outputting the compared data as the communication behavior corresponding to each Internet of Things node.

[0040] Preferably, the communication times are used to represent the number of times of sending data in the current long-distance scenario, and the associated parameters are used to represent the information saved after receiving the relevant signal, which is used to compare whether there is a lack of transmission strategy currently used, resulting in that the selected Internet of Things node receives incomplete data. The associated parameters include communication duration, data packet size and other stored data recording the communication behavior of the current Internet of Things node.

[0041] Preferably, when the Internet of Things nodes are aggregated, it is necessary to identify whether the multiple Internet of Things nodes in the long-distance scenario can work normally, and the nodes are divided into multiple groups of aggregated nodes according to the distance from the gateway, and the transmission delay of the nodes is used as the output communication behavior. The peripheral preset distance represents the distance threshold in the long-distance scenario, which is beneficial to using the average value of the sensor arranged in the long-distance scenario as the peripheral preset distance at this time, and determining the multiple groups of sensor devices currently configured through the value.

[0042] In an embodiment of the application, in step S2, the overall implementation process of the communication behavior needs to be analyzed and detected, for example, the selected communication behavior, the part compared in step S1 by the communication behavior and the associated parameters, and the transmission of the communication node of the Internet of Things node under the wake-up instruction and the sleep instruction is identified and processed.

[0043] As for the relative coverage range and the relative distance, they represent the communication path used in the transmission strategy currently used, and the communication access authority of the multiple groups of Internet of Things nodes contained in the path, which are used to determine the transmission mode of each Internet of Things node in the long-distance distribution in turn.

[0044] As shown in Figure 3 The implementation mode of step S2 includes: S21, for each selected Internet of Things node, using the peripheral preset range of the Internet of Things node, checking the coverage area of the target node in the peripheral preset range, and determining the relative coverage range and the relative distance of each Internet of Things node in the long-distance transmission to the next node.

[0045] S22, using the relative distance of different Internet of Things nodes, querying the processing order of the Internet of Things node by the sleep instruction, and setting the state label of each Internet of Things node completing data transmission under the time window translation sliding.

[0046] Preferably, the relative coverage range indicates the intersection of the coverage area of the current Internet of Things node and the next Internet of Things node after the coverage area is obtained according to the preset range of the periphery, and the relative distance indicates the distance between the current Internet of Things node and the next Internet of Things node.

[0047] Preferably, when the state label is set by using the relative distance of different Internet of Things nodes, the implementation manner further includes: sorting the relative distances between each Internet of Things node, obtaining the time deviation of each Internet of Things node during the receiving of the sleep instruction in the time window translation sliding, and setting the state label of each Internet of Things node by using the time deviation.

[0048] Preferably, when the state label is obtained by using the relative distance of different Internet of Things nodes, the state label can also be set based on the relative coverage range of each Internet of Things node for long-distance transmission of the next node.

[0049] For example, when the relative coverage range is used for analysis, the response time corresponding to the communication times reported by the Internet of Things node and the communication duration corresponding to the associated parameters are used to set the state label of each Internet of Things node according to the response time and the communication duration of each Internet of Things node.

[0050] At this time, when the time deviation and the relative distance are used for expression, the Internet of Things node is used for accurate identification at the location and the time point, so as to avoid data missing caused by different time synchronization; as for the response time and the communication duration, the sleep interval of the Internet of Things node is used for identification, so as to identify the relative wake-up times, which are used for explaining the use interval of each Internet of Things node under dynamic power consumption; and the relative coverage range obtained can be linked with the communication times to explain the frequency of data collection of different Internet of Things nodes, and finally the time deviation can be used to determine whether there is an overtime reaction problem of the current Internet of Things node, so as to identify whether there is an abnormal damage problem of each Internet of Things node in the multiple states of sleep and wake-up.

[0051] The finally output state label can explain that the time synchronization state contains normal, clock drift and serious step-out; the communication quality contains high, medium and low states; the power consumption level contains low power consumption, balance and high power consumption; and the abnormal warning contains time correction failure, coverage contraction, response timeout, communication congestion and other state labels described in various combinations.

[0052] At this time, the state label explained can be set in the database for the sensor represented by the current Internet of Things node, and then the state labels are summarized to complete the setting of the wake-up strategy, so as to improve the sleep proportion after the execution of the sleep instruction, and to reduce the power consumption processing mode.

[0053] Preferably, the content of the processing order query is the initial priority of the current Internet of Things node or the processing order of the current Internet of Things node in the last cycle, and the current Internet of Things node is divided into a time window in the corresponding order, so as to process the Internet of Things data collected in each time window.

[0054] It should be noted that the execution of the sleep instruction indicates that the current configuration sensor only performs the collection task of the fixed time interval, and only works under the condition of completing the minimum task requirement. If there are various data problems such as slow data response and data upload congestion during the execution of the sleep instruction, the working content will be adjusted to a balanced running state or a performance state under high power consumption according to the form of the problem, so as to adjust the working condition of each Internet of Things node, and finally realize the work and identification of the Internet of Things covering various regions.

[0055] In an embodiment of the present application, after receiving the corresponding sleep instruction, the Internet of Things nodes in step S3 perform the wake-up strategy, monitor and evaluate the network communication of the Internet of Things nodes in these ways, and identify the abnormal behavior, and complete the regional analysis of the Internet of Things nodes.

[0056] As shown in Figure 4 The implementation mode of step S3 includes: S31, based on the state label of the Internet of Things node, performing state detection on the Internet of Things node, and performing element description mapping based on the node state, communication link and signal strength corresponding to the Internet of Things node, to form an element relationship network.

[0057] S32, capturing the network channel associated with the Internet of Things node based on the feature association of each Internet of Things node in the element relationship network, and statistically analyzing the interference elements in each network channel based on the communication interval and data volume change corresponding to the network channel.

[0058] S33, based on the time window in which each interference element occurs, locally tracing and displaying the interference element, and quantifying the index deviation of each Internet of Things node.

[0059] The interference elements include heartbeat packet timeout, sudden increase / decrease of traffic, etc. These abnormalities will cause the corresponding Internet of Things node data transmission to be interrupted, and the network bandwidth to be congested, and the node processing capacity to be overloaded, etc. The interference elements are used to identify whether the low-power mode under the execution of the sleep instruction can meet the requirements configured in the current Internet of Things node. According to these interference elements, part of the equipment can be awakened in time to adjust its state to reduce the resources consumed by each node in the high-energy consumption mode, and improve the utilization efficiency of the configuration sensor of the Internet of Things node.

[0060] Preferably, when the state of the Internet of Things node is detected, the node state represents active, dormant and the like; the communication link represents the current connection mode, such as TCP / UDP connection, LoRa channel, and signal strength is a value directly reflecting the current connection state of the Internet of Things node, such as RSSI value; finally, according to the node state, the communication link and the signal strength, the communication relationship between each Internet of Things node is used to connect, and the connected Internet of Things node is composed of element relationship network.

[0061] Preferably, when the element description mapping is performed, the mapping can also be performed based on the content identified by each state label to describe the relative state of each Internet of Things node under the corresponding communication relationship.

[0062] Preferably, when the network channel associated with the Internet of Things node is captured, the network channel associated with the Internet of Things node under the conditions of communication dependence, physical proximity, functional cooperation and state similarity is extracted according to the type of device connected by the current Internet of Things node; Then the value of the communication interval and the data volume of the Internet of Things node in the corresponding network channel is taken, and the part exceeding the average value plus three times the standard deviation is regarded as the interference element; Finally, the deviation degree of the interference element under the associated communication behavior is identified, and the calculation of the index deviation degree is finally completed.

[0063] Preferably, the implementation mode of step S32 comprises: identifying the associated characteristics of each Internet of Things node after the connection of the element relationship network, the associated characteristics being represented as whether there are Internet of Things nodes that meet the conditions of communication dependence, physical proximity, functional cooperation and state similarity when working, and if there are, the Internet of Things nodes and the corresponding conditions are coded as the current associated characteristics.

[0064] The associated characteristics of each Internet of Things node are used to determine the relationship subnetwork of the element relationship network, and the network channel associated with the Internet of Things node is selected according to the continuous number of abnormal changes of the communication interval and the data volume of each Internet of Things node in the relationship subnetwork.

[0065] At this time, when the communication interval is abnormal, the confidence interval of normal communication is used as the judgment condition. When the value exceeds the upper and lower limit values of the 95% confidence level, it is considered that the communication interval is abnormal. For data volume changes, the current value is subtracted from the moving average value of the time window, and divided by the standard deviation corresponding to the data volume. When the calculated value is greater than three times the standard deviation, it is considered that the data volume change is abnormal. Then, the number of consecutive abnormal times of the communication interval and the data volume change is counted. When the number of consecutive abnormal times is greater than three, the corresponding network channel is marked as the current main recognition processing network channel. The interference elements are set based on the description of the combination of the communication interval and the data volume change. For example, interval surge + data volume sharp decrease may indicate device failure, interval fluctuation + data volume stability may indicate electromagnetic interference, and interval gradual increase + data volume gradual increase may indicate network congestion. The time points at which the current communication interval and data volume change are abnormal are traversed, and the data of the corresponding time points are combined in turn. The interference elements under the corresponding conditions are queried from the database, and these interference elements are added to each Internet of Things node.

[0066] Preferably, when local traceability display is performed, the time window in which the interference elements appear is marked. The associated features and state labels corresponding to the interference elements are used as their weights to obtain their index deviation degrees.

[0067] For example, the index deviation degree can be represented as ; wherein, represents the index deviation degree, indicating the degree value of the current value exceeding the normal situation; represents the number of time points, used to indicate the length of the corresponding time window when the anomaly occurs; represents the weight of the current interference element at time j. The value is set by using the ratio of the occurrence frequency of the combination of the associated features and the state labels at the corresponding time point to the total data occurrence frequency; represents the value of the current interference element at time j. The communication interval, data volume, and signal strength are used as the values calculated by the interference elements. The current calculation is based on time points. The interference elements with abnormalities are input. If only one of the three values is abnormal, the above formula is used for calculation, and the calculated value is regarded as the index deviation degree of the corresponding Internet of Things node. If there are more than two, they are input into the above index deviation degree formula according to the corresponding time points. The sum of the calculation data is used as the output index deviation degree; represents the average value of the current interference element. At this time, the average value and the standard deviation are calculated based on all sampled data; represents the standard deviation of the current interference element. After the index deviation degree is calculated, it can be known whether the current Internet of Things node can complete the basic monitoring operation after receiving the sleep instruction. If not, each node needs to be adjusted according to its index deviation degree to complete the control processing of the overall sensor.

[0068] In one embodiment of the present application, when generating the parameter processing sequence, the relative distance of the nodes is used to arrange the positions of each material web node at the relative distance to form a sequence, the obtained index deviation is filled in the sequence according to the distance, and the specific parameters related to the actual communication such as the communication frequency band, the modulation mode, the data rate and the like contained in the current communication behavior are marked. The parameter processing sequence is analyzed in different time periods, and based on the plurality of Internet of Things nodes extracted by the parameter processing sequence, the main loss parameters and the conditions shown by these parameters when the path transmission loss occurs are recorded to describe the data sequence in the communication processing scene, so as to facilitate subsequent adjustment of the initial configuration parameters of each Internet of Things node and the communication range of each Internet of Things node to complete the relative loss reduction processing.

[0069] As shown in Figure 5 S41, the path loss and the signal-to-noise ratio before and after the end of each data cycle are used to traverse each Internet of Things node, and the index deviation of each Internet of Things node is used as the constraint condition of the corresponding Internet of Things node.

[0070] S42, when the constraint condition corresponds to the same state label, the difference values of the path loss and the signal-to-noise ratio are compared, and if the difference values are both less than a preset threshold, the corresponding constraint condition is merged, and the value of the merged constraint condition on the time sequence is regarded as the parameter processing sequence.

[0071] S43, if the constraint condition corresponds to no same state label and the difference values of the path loss and the signal-to-noise ratio are both greater than the preset threshold, the data of the time window corresponding to the maximum difference value is taken as the parameter processing sequence; when the maximum difference value is output, the time window is queried according to the difference values of the path loss and the signal-to-noise ratio, respectively, and the data in the time window corresponding to the two values is output.

[0072] S44, if neither of the above descriptions is satisfied, the data with the index deviation greater than a preset deviation is selected as the output parameter processing sequence. In step S44, the conditions covered in steps S42 and S43 are not satisfied, at this time, the output data is output in the form of a plurality of time windows combined according to the time sequence by relying on the loop judgment of each Internet of Things node, so as to facilitate subsequent statistics of the number of abnormal nodes and the distribution of the abnormal nodes to adjust the control strategy set for each Internet of Things node and other maintenance and processing schemes for the Internet of Things nodes.

[0073] Before and after the end of each data cycle, the system will traverse all the Internet of Things nodes, calculate their path loss, signal-to-noise ratio and corresponding index deviation of the Internet of Things nodes; according to the state label corresponding to the index deviation value, it can be known whether the data sorted by path loss and signal-to-noise ratio can be merged, if it is less than the preset threshold, the related data of these nodes will be merged to obtain a parameter sequence that needs to be concentrated. For example, when the state labels of multiple Internet of Things nodes all contain high power consumption mode and low signal strength, and the difference values of their path loss and signal-to-noise ratio are less than the preset threshold, the system will merge the constraint conditions of these nodes, and the data of the merged constraint conditions in the continuous time period is abstracted as a unified parameter processing sequence. Merging the constraint conditions can reduce redundant calculation and simplify the subsequent processing logic, the nodes under the same state label have similar behaviors, and after merging, the control strategy can be uniformly executed to complete the control and processing of the Internet of Things nodes.

[0074] When the state labels of the Internet of Things nodes are inconsistent, and the difference values of the path loss and signal-to-noise ratio are greater than the preset threshold: select the time window corresponding to the maximum difference value as the parameter processing sequence. These maximum difference values will represent the most serious problem in multiple Internet of Things nodes with inconsistent labels, and then by taking the index offset degree as the condition, the path loss and signal-to-noise ratio difference of the current Internet of Things node according to the value in multiple time windows is determined, and the maximum difference is extracted, which helps to identify the root cause of the current communication problem, triggers the automatic warning of each Internet of Things node or triggers the alarm to process each Internet of Things node.

[0075] When neither the constraint merging condition is met nor it belongs to the significant difference scenario, select the data with an index deviation greater than the preset deviation as the parameter processing sequence, which is used to capture slight abnormalities in multiple time windows, and fill the extracted data according to these slight or other abnormalities, and complete the differential processing of the overall data.

[0076] Preferably, the preset threshold value set for the difference values of the path loss and signal-to-noise ratio is based on the average value of the difference between the upper limit value and the lower limit value of the path loss and signal-to-noise ratio under normal transmission, which is used as the preset threshold for the current screening. As for the preset deviation, the average value of the index offset degrees of all Internet of Things nodes is selected to set, to identify data that exceeds a specific condition.

[0077] In an embodiment of the present application, in step S5, the abnormal events output in the parameter processing sequence are processed according to the number of abnormal nodes existing on each Internet of Things node, and abnormal event detection is performed according to these abnormal Internet of Things nodes, the distribution difference of each abnormal event is viewed according to spatial correlation and temporal correlation, respectively, and then the difference is used for transmission strategy reorganization, the control strategy of each Internet of Things node is adjusted, and adaptive control for long-distance transmission in multiple positions is realized; the working time of each device is prolonged, and the power consumption of long-distance transmission is reduced.

[0078] As shown in Figure 6 , the implementation mode of step S5 includes: S51, according to the control requirements of each Internet of Things node, collecting the initial abnormal node number on each Internet of Things node, and detecting the type of abnormal nodes according to the change value of the number of abnormal nodes.

[0079] S52, in response to the number of current abnormal nodes, determining whether the current abnormal nodes satisfy the target condition according to the association relationship between each abnormal node, if yes, determining that the data set associated with the current abnormal node is the target parameter set.

[0080] S53, reading the abnormal type of the target parameter set, and according to the record information between the execution of the sleep instruction and the wake-up instruction, recovering the Internet of Things node corresponding to the target parameter set, and taking the data of the recovered Internet of Things node as the control strategy of the Internet of Things node.

[0081] When processing the initial abnormal node number on each Internet of Things node, the abnormal nodes in each time window are described after collecting the abnormal nodes through a fixed time window, and the type of the current abnormal event is detected according to the change value of the number of abnormal nodes, so as to explain that the currently identified abnormality belongs to any form of abnormal type of diffusion type, local abnormality, and periodic fluctuation abnormality, then the communication behavior associated with the corresponding abnormal type and other parameters are viewed to modify the description range of the abnormal type, and the control strategy of the current Internet of Things node is generated.

[0082] That is, the implementation mode of step S51 includes: reading the state label of the initial abnormal node number, and verifying the abnormal nodes retained by the initial abnormal nodes under the change of the number of abnormal nodes for two consecutive times according to the data read by the initial abnormal nodes under the flag abnormality. Here, the node number and state label of the part marked as an abnormal node are read, the nodes with continuous abnormality are quickly analyzed to exclude the abnormal nodes existing under the influence of transient fluctuation or noise, the invalid processing of short-time abnormality is reduced, and the node number with continuous abnormality is concentratedly analyzed.

[0083] Determine whether the number of reserved abnormal nodes is equal in the continuous time window; if equal, the reserved abnormal nodes are used for abnormal type detection. If the number of abnormal nodes that are continuously abnormal in the continuous time window is stable at this time, it means that there is a systematic abnormality, such as network attack, device aging, etc. Type of abnormality, then process these stable abnormal nodes to locate the type of abnormality existing on the current Internet of Things node, and reduce the processing of other transient abnormal nodes.

[0084] If not equal, the union set of abnormal nodes in each time window is used for abnormal type detection to determine the current abnormal type. If the number of abnormal nodes changes, it means that the abnormality is scattered or diverse, and all time window data needs to be combined for global analysis to find out potential abnormal points and avoid missing relevant dynamic abnormal point changes.

[0085] Through the processing of the number of abnormal nodes, the current system can identify transient and persistent abnormalities in different time windows, balance the accuracy and real-time of processing, determine the stable abnormal point for processing first, and implement the dynamic node set processing mode to support the abnormal response mode of different Internet of Things nodes under long-distance transmission.

[0086] In the abnormal type detection, the abnormal nodes on multiple Internet of Things nodes in the corresponding time window are clustered respectively, and the clustering of the abnormal nodes in a single Internet of Things node in multiple time windows is used to complete the clustering of the current abnormal nodes. The data after clustering is set as the type of abnormality to complete the type detection of abnormal nodes.

[0087] Preferably, the target condition of the abnormal node is further described by using the time interval of the occurrence of the abnormal node and the distance between different Internet of Things nodes when the abnormal node occurs.

[0088] The implementation mode of the target condition in step S52 includes: judging whether the time interval of the occurrence of the current abnormal node is less than the preset time interval, which is recorded as the first condition; the first condition is used to indicate whether the abnormality occurs frequently, whether there is periodic fluctuation or sudden concentration of abnormality, when it is less than the preset time interval, it may indicate that the abnormality occurs frequently, if the abnormality can also occur continuously, it means that the abnormality is persistent, and if the same length of abnormality occurs according to the current preset time interval, it also indicates that the current abnormality is periodic. The preset time interval is set according to the average time interval of the abnormal nodes on the Internet of Things node, and the time interval of the current judgment is input to the corresponding data to process the relatively frequent abnormality.

[0089] whether the distance between the adjacent IoT nodes with simultaneous abnormality is less than a preset distance, denoted as a second condition; the second condition is used to identify whether there is a local abnormality aggregation situation, for example, the current IoT nodes are prone to abnormality in a certain area, at this time, these distributed concentrated IoT nodes are processed in a centralized manner, at this time, the preset distance can be based on the average value of the IoT node configuration distance to identify whether the adjacent IoT nodes will synchronously appear abnormality to identify the scene prone to high risk in a centralized manner.

[0090] When at least one of the first condition and the second condition is met, the corresponding data is output as the target parameter set. As long as one of the first condition and the second condition is met, it means that the abnormal nodes on the problem IoT nodes are associated and need to be processed in a centralized manner.

[0091] Preferably, in the implementation of the control strategy, the control strategy used is selected according to the current abnormal type, and the communication parameters and energy consumption state of the IoT nodes are restored by using the historical data in the sleep and wake-up records, to ensure the stability of the running state of the nodes after recovery, and then the recovered node data is converted into a reusable control strategy to improve the robustness of the system in long-term operation.

[0092] Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application, which are still covered by the protection scope of the present application.

Claims

1. A low-power long-distance communication method based on the Internet of Things, characterized in that: include: S1, based on the initial configuration parameters and network topology of the IoT nodes, collects data information of each IoT node and forms the corresponding communication behavior of each IoT node; S2, by obtaining the communication times and associated parameters of each communication behavior, analyze the communication quality of each IoT node during the execution of the sleep instruction, and derive the status label of each IoT node based on the relative coverage and relative distance of each IoT node; S3, based on the sleep instructions received by each IoT node, analyzes the abnormal behavior of IoT nodes in different status tags and evaluates the indicator deviation of each IoT node under abnormal communication behavior; S4, based on the indicator deviation of each IoT node, identifies the path loss and signal-to-noise ratio of the current communication behavior, and generates a parameter processing sequence based on the differences in communication behavior at multiple time points; S5, according to the number of abnormal nodes on each IoT node during transmission, perform anomaly detection on the parameter processing sequence, and adjust the control strategy of each IoT node based on the distribution difference of each abnormal node.

2. The low-power long-distance communication method based on the Internet of Things according to claim 1, characterized in that: The implementation of step S1 includes: S11, based on the number of IoT nodes associated with the network topology, starting from the deployed IoT nodes, checking the initial configuration parameters, and performing node aggregation on the IoT nodes. Based on the preset distance around each IoT node after aggregation, the transmission delay of the IoT nodes in the long-distance scenario is obtained; S12, in response to the transmission delay of the current IoT node, comparing the communication times and associated parameters of the current IoT node, and outputting the compared data as the communication behavior corresponding to each IoT node.

3. The low-power long-distance communication method based on the Internet of Things according to claim 1, characterized in that: The implementation of step S2 includes: S21, for each selected IoT node, checking the coverage area of ​​the target node within the preset range around the IoT node, and determining the relative coverage range and relative distance of each IoT node for long-distance transmission to the next node; S22, using the relative distances of different IoT nodes, querying the processing order of IoT nodes with the issued sleep instructions, and setting the status label of each IoT node completing data transmission under the time window translation sliding.

4. The low-power long-distance communication method based on the Internet of Things according to claim 3, characterized in that: The implementation of status labels also includes: The IoT nodes are sorted according to their relative distances, and the time deviation of each IoT node receiving the sleep instruction under the sliding time window is obtained. The time deviation is used to set the status label of each IoT node.

5. The low-power long-distance communication method based on the Internet of Things according to claim 1, characterized in that: The implementation of step S3 includes: S31, based on the status tag of the IoT node, performing status detection on the IoT node, performing element description mapping based on the node status, communication link, and signal strength corresponding to the IoT node, and forming an element relationship network; S32, capturing the network channels associated with the IoT nodes based on the feature associations of the IoT nodes in the element relationship network, and counting the interference factors in each network channel based on the communication interval and data volume changes corresponding to the network channel; S33, based on the time window in which each interference factor occurs, the interference factor is locally traced and displayed, and the indicator deviation of each IoT node is quantified.

6. The low-power long-distance communication method based on the Internet of Things according to claim 5, characterized in that: The implementation of step S32 includes: identifying the associated features of each IoT node after the element relationship network is connected; The relational sub-network of the element relational network is determined by using the association characteristics of each IoT node. The network channel associated with the IoT node is selected based on the communication interval of each IoT node in the relational sub-network and the number of consecutive abnormal changes in data volume.

7. The low-power long-distance communication method based on the Internet of Things according to claim 1, characterized in that: The implementation of step S4 further includes: S41, traverse each IoT node based on the path loss and signal-to-noise ratio before and after the end of each data cycle, and use the indicator deviation of each IoT node as a constraint condition for the corresponding IoT node; S42: When the state labels corresponding to the constraints have the same state label, compare the difference values ​​of the path loss and the signal-to-noise ratio. If the difference values ​​are both less than a preset threshold, merge the corresponding constraints, and regard the values ​​of the merged constraints in the time series as the parameter processing sequence. S43, if the state labels corresponding to the constraint conditions do not have the same state label and the difference values ​​of the path loss and the signal-to-noise ratio are both greater than a preset threshold, using the data of the time window corresponding to the maximum difference value as the parameter processing sequence; S44: If none of the above descriptions are met, data with an indicator deviation greater than a preset deviation is selected as the output parameter processing sequence.

8. The low-power long-distance communication method based on the Internet of Things according to claim 1, characterized in that: The implementation of step S5 includes: S51, according to the control requirements of each IoT node, collect the initial number of abnormal nodes on each IoT node, and perform type detection on the abnormal nodes based on the change in the number of abnormal nodes; S52, in response to the number of current abnormal nodes, determining whether the current abnormal node meets the target condition based on the association relationship between the abnormal nodes; if so, determining the data set associated with the current abnormal node as the target parameter set; S53, read the exception type of the target parameter set, and restore the IoT node corresponding to the target parameter set according to the recorded information between executing the sleep instruction and the wake-up instruction of the exception type, and use the data of the restored IoT node as the control strategy of the IoT node.

9. The low-power long-distance communication method based on the Internet of Things according to claim 8, characterized in that: The implementation of step S51 includes: Read the status tag of the initial abnormal node number, and verify the abnormal nodes retained by the initial abnormal node under two consecutive changes in the number of abnormal nodes according to the data read when the initial abnormal node is marked abnormal; Determine whether the number of retained abnormal nodes is equal in consecutive time windows; if equal, perform abnormal type detection on the retained abnormal nodes; If they are not equal, the anomaly type detection is performed based on the union of the anomaly nodes in each time window to determine the current anomaly type.

10. The low-power long-distance communication method based on the Internet of Things according to claim 8, characterized in that: The implementation of the target condition in step S52 includes: Determine whether the time interval between the occurrence of the current abnormal node is less than the preset time interval, which is recorded as the first condition; Determine whether the distance between adjacent IoT nodes where abnormalities occur simultaneously is less than a preset distance, which is recorded as the second condition; When at least one of the first condition and the second condition is satisfied, the corresponding data is output as the target parameter set.

Images