A wireless operation and maintenance method and system based on an internet of things

Abstract

The present application relates to the technical field of Internet of Things, in particular to a wireless operation and maintenance method and system based on Internet of Things, comprising S1: collecting original operation data through an Internet of Things device, and converting the original operation data into a current device state vector; S2: combining the current device state vector and a neighbor device vector to obtain an abnormal score and a state tension term of the current device; S3: constructing a response intensity scoring function based on the abnormal score and the state tension term, and outputting a response intensity value through the response intensity scoring function; S4: inputting the response intensity value into a structure mapper, and outputting a control instruction through the structure mapper; S5: issuing the control instruction to the device through a communication channel; and S6: executing an operation according to the control instruction by the Internet of Things device, and feeding back an execution result to the system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things, in particular to a wireless operation and maintenance method and system based on Internet of Things. BACKGROUND

[0002] With the wide application of Internet of Things technology, a large number of wireless terminal devices are deployed in complex scenes such as industrial sites, urban infrastructure, and agricultural environments, and undertake key tasks such as data collection, state monitoring, and control execution. These devices are usually connected to the cloud platform or edge gateway through wireless communication to realize remote operation and maintenance functions such as device state updating, fault repair, and system upgrading. The essence of wireless operation and maintenance is to reliably and timely realize device management and maintenance without interrupting the normal operation of the device. However, as the scale of Internet of Things systems continues to expand, the types of devices tend to diversify, and the deployment environment becomes more complex, existing wireless operation and maintenance technologies face serious challenges.

[0003] First, the wireless communication environment in which Internet of Things devices are located generally has problems such as channel congestion, spectrum interference, and multipath fading, causing unstable transmission of operation and maintenance instructions and high packet loss rate, especially in low-power wide-area communication protocols such as LoRa and Zigbee. Second, the device state sensing mechanism generally uses a timing reporting or simple threshold alarm strategy. Once a device is lost or enters an abnormal state due to communication abnormalities, the system often cannot accurately identify its operating state, and lacks a mechanism to infer its state from surrounding devices. In addition, since most devices are deployed in resource-limited places (such as underground, remote suburbs, and wild fields), they are usually powered by batteries, and have limited communication, computing, and sensing capabilities. Traditional cloud-centric operation and maintenance strategies (such as centralized scheduling and unified control) are not feasible in terms of timeliness and energy consumption. Although existing research has attempted to introduce edge computing, multi-channel polling, and fault-tolerant broadcasting to improve operation and maintenance reliability, these methods often have problems such as heavy models, static strategies, and lack of feedback, and cannot truly achieve high-robustness and high-efficiency wireless operation and maintenance in complex scenarios.

[0004] In summary, current wireless operation and maintenance technologies have significant bottlenecks in three aspects: (1) state unmeasurable: lack of reliable judgment mechanism when devices are lost or have communication abnormalities; (2) instruction unreachable: spectrum congestion and interference cause control instructions to fail; (3) control not closed loop: device execution state is difficult to verify, and the system cannot form an "identification-decision-execution-verification" operation and maintenance closed loop. Therefore, there is an urgent need for a new operation and maintenance method and system with state-aware intelligence, wireless interference sensing capability, and self-adaptive control scheduling mechanism, which can effectively adapt to the heterogeneity, instability, and resource limitations of Internet of Things environments, and improve the stability, self-healing, and intelligent level of the overall system. SUMMARY

[0005] To solve the problems in the background art, the present application provides a wireless operation and maintenance method and system based on the Internet of Things.

[0006] To achieve the above object, the technical scheme adopted by the present application is:

[0007] A wireless operation and maintenance method based on the Internet of Things, comprising:

[0008] S1: Collecting original operation data through an Internet of Things device, and converting the original operation data into a current device state vector;

[0009] S2: Obtaining an abnormal score and a state tension term of the current device in combination with the current device state vector and a neighbor device vector;

[0010] S3: Constructing a response strength scoring function based on the abnormal score and the state tension term, and outputting a response strength value through the response strength scoring function;

[0011] S4: Inputting the response strength value into a structure mapper, and outputting a control instruction through the structure mapper;

[0012] S5: Issuing the control instruction to the device through a communication channel;

[0013] S6: The Internet of Things device performs an operation according to the control instruction, and feeds back an execution result to the system.

[0014] As a preferred, the abnormal score expression in S2 is:

[0015]

[0016] where s i is the abnormal score, z i is the current device state vector, z j is the neighbor device state vector, w ij is the state influence weight of the neighbor device j on the current device i, and λ is a regularization term coefficient; Ω(z i ) is a regularization term.

[0017] As a preferred, the state tension term expression in S2 is:

[0018]

[0019] where Ω(z i ) is the state tension term.

[0020] As a preferred, S1 specifically comprises:

[0021] S1.1: Obtaining original operation data, and periodically sending the original operation data to an edge node through a wireless protocol;​

[0022] S1.2: The edge node standardizes the original operation data;

[0023] S1.3: The original operation data after standardization is subjected to data enhancement to generate a positive sample pair;

[0024] S1.4: The positive sample pair is input into an encoder to generate a current device state vector through the encoder.

[0025] As a preferred, the communication channel generation in S5 comprises:

[0026] The communication module of the Internet of Things device periodically scans available channels and records historical communication state records of each channel;

[0027] Based on the historical communication state records, a plurality of channels are predicted in a sliding average manner;

[0028] The prediction results are input into a comprehensive scheduling scoring function, and the comprehensive scheduling scoring function outputs scoring results of the plurality of channels;

[0029] The channel with the smallest score value is selected as the communication channel.

[0030] As a preferred, the expression of the response strength scoring function in S3 is:

[0031]

[0032] where ψ i is a behavior trust factor, is an average value of device states in the past T time windows, β1, β2, β3 are experience weight coefficients, and σ(·) is a Sigmoid function.

[0033] As a preferred, S6 specifically comprises:

[0034] S6.1: After accepting the control instruction, the Internet of Things device calls the local control module to perform operations through a preset control mapping table;

[0035] S6.2: After the operation is completed, the Internet of Things device constructs a feedback data packet through the same communication channel and returns the feedback data packet to the system;

[0036] S6.3: The system accepts the feedback data packet and extracts an execution result flag bit of the feedback data packet to generate a feedback scalar.

[0037] As a preferred, the feedback data packet of S6.2 comprises an instruction ID, a device ID, an operation type, and an execution result.

[0038] As preferred, the original data in S1 includes voltage, device chip temperature, wireless receiving strength, network response delay and device load.

[0039] A wireless operation and maintenance system based on Internet of Things, applied to the wireless operation and maintenance method based on Internet of Things, comprising:

[0040] An Internet of Things device for collecting original operation data;

[0041] An encoder for generating a state vector of the device;

[0042] A structure mapper for dividing the response strength value into different control levels according to intervals and generating a control instruction

[0043] The present application has the following advantages:

[0044] By constructing a cooperative cognitive mechanism between device states and a dynamic adaptation mechanism of communication links, intelligent and stable wireless remote operation and maintenance control is realized. Specifically, the present application introduces a self-learning state modeling mechanism on the device side in the device state judgment aspect, so that in the case of lack of central control or individual device disconnection, the system can still share state information through the surrounding devices and realize auxiliary judgment, thereby improving the accuracy and coverage of abnormal detection.

[0045] In terms of operation and maintenance instruction execution, the present application designs a communication link self-sensing and dynamic scheduling mechanism, which can sense the current wireless channel state in real time and dynamically select the optimal frequency band or alternative channel based on the current interference, significantly improving the transmission success rate of the instruction.

[0046] In terms of the overall control process, the present application proposes a feedback closed-loop system based on state judgment driving, realizing a continuous closed-loop control chain from state recognition, instruction triggering, path selection to result feedback, truly breaking through the boundary restrictions of the traditional central operation and maintenance model, adapting to the real needs of wide device distribution, uncertain communication and fast response control in the Internet of Things environment. The present application has the characteristics of lightweight structure, flexible deployment, control closed loop and strong robustness, and is particularly suitable for remote, complex and resource-limited wireless operation and maintenance scenarios in the Internet of Things. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 A wireless operation and maintenance method based on Internet of Things in specific embodiments of the present application;

[0048] Figure 2 A process diagram for generating a device state vector in specific embodiments of the present application;

[0049] Figure 3 A process diagram for executing result feedback in specific embodiments of the present application;

[0050] Figure 4 Figure 1 is a block diagram of a wireless operation and maintenance system based on the Internet of Things in specific embodiments of the present application. DETAILED DESCRIPTION

[0051] Referring to Figures 1-4 The present application relates to a wireless operation and maintenance method and system based on the Internet of Things, which specifically includes:

[0052] Step 1: Device state encoding modeling

[0053] This step aims to convert the raw operation data collected by Internet of Things devices into a state vector z i , which serves as the basis for subsequent steps of anomaly identification and control decision-making. Considering the highly heterogeneous device data and the scarcity of annotations in the Internet of Things environment, we use the SimCLR framework in the self-supervised learning method to model the encoding of the original data. SimCLR has a simple structure, stable training, and adapts to edge deployment requirements, which can effectively extract low-dimensional semantic representations of device operation states.

[0054] Input:

[0055] Device raw data x i : refers to the sensor readings uploaded by the device in each sampling period, including:

[0056] - Voltage (such as the real-time battery voltage of a 3.3V powered device);

[0057] - Device chip temperature (such as internal ADC reading);

[0058] - Wireless reception strength RSSI (such as Wi-Fi or LoRa module sampling);

[0059] - Network response delay (obtained from the difference in sending ACK time);

[0060] - Device load (such as CPU usage, memory occupancy), etc.

[0061] Data is periodically sent to edge nodes through wireless protocols such as LoRa, Wi-Fi, Zigbee, etc. Each time the data is sampled to form a set x i , which has a dimension of d, such as

[0062] First, the collected raw data x i is standardized to enable learning of different types of sensor data on the same scale. Standardization uses the Z-score method: each dimension is subtracted by its mean and divided by the standard deviation.

[0063] Subsequently, the self-supervised encoding process is entered. We use the SimCLR framework to generate sample pairs in the training phase: for the original data xi Two views are generated by two different data augmentation methods and as positive sample pairs. The augmentation methods include:

[0064] 1. Gaussian perturbation augmentation: Gaussian white noise with mean 0 and standard deviation σ = 0.05 is added to each sensor dimension;

[0065]

[0066] 2. Time window clipping: sub-segments are cropped from the original time series to disrupt the sampling alignment.

[0067] The two augmented samples are input into the encoder f θ to generate the corresponding state vectors:

[0068]

[0069] where:

[0070] f θ is a self-supervised encoder model with the structure of two layers of 1D convolution (each layer with a kernel size of 3 and a channel number of 64 / 128) + flattening + one layer of fully connected (128 dimensions);

[0071] are state vectors generated from the same device, the same time, but different augmentation methods.

[0072] During training, the model is optimized by the following contrastive loss function to make the positive sample vectors close and the negative sample vectors far apart:

[0073]

[0074] where:

[0075] sim(·,·) represents the cosine similarity calculation;

[0076] τ is the temperature coefficient, which controls the smoothness of the distribution;

[0077] z k is the state vector of other samples, used as negative samples.

[0078] After training, the encoder f θ is deployed to the edge node or device end, only keeping the inference part. In actual operation, each set of device data x i will be directly mapped to the state vector:

[0079] z i = f θ (x i ); (3) ​

[0080] Step output:

[0081] State vector z i : represents the current running state of device i, is a dense embedding vector with dimension 128, as the basis for subsequent anomaly identification.

[0082] Step 2: Neighbor collaborative judgment and state anomaly identification

[0083] In the wireless operation and maintenance system of the Internet of Things, devices are often deployed in complex environments such as industrial workshops, underground shafts, remote sites, etc. There are problems such as unstable communication link, device disconnection, signal drift, etc. In this case, relying on device state upload for anomaly identification is extremely fragile. Therefore, this step proposes a creative neighbor collaborative state judgment mechanism, which uses the state vectors of other communication reachable devices in the network to judge whether the target device is in an abnormal state through state consensus analysis and similarity enhancement modeling. Unlike traditional centralized judgment or rule alarm methods, this mechanism combines the geometric characteristics of the device state space and the network topology characteristics, and still has robust judgment ability under the condition of unstable wireless link and missing device data.

[0084] Input:

[0085] Device state vector z i : output from step one, generated by the encoder f θ , is the current state embedding vector of device i;

[0086] Neighbor device state set The set of state vectors obtained by broadcasting adjacent devices, adjacency relationship Determined dynamically by communication reachability (such as RSSI greater than threshold, signal stability);

[0087] State propagation weight w ij : represents the "consensus force" of neighbor j on device i, determined based on signal strength, historical consistency, and adaptive update mechanism.

[0088] Device i constructs its neighbor set by listening to the broadcast data packets of surrounding devices in the current period Each neighbor device j broadcasts its state vector z j , which comes from the trained state encoder f θ , and is transmitted after quantization and compression (such as 8-bit fixed-point compression). After receiving, device i constructs the neighbor state set {z j} and calculates the consistency score of its own state with the neighbors.

[0089] We first define the anomaly score s i of device i as its state vector zi Weighted dissimilarity with neighbor device status:

[0090] Where:

[0091] z i is the current device status vector;

[0092] z j is the neighbor device status vector;

[0093] w ij is the status influence weight of neighbor j on device i, defined as:

[0094]

[0095] To emphasize the stronger and more stable neighbors have greater impact on the decision result;

[0096] λ is the regularization term coefficient, which controls the strength of bias penalty;

[0097] Ω(z i ) is an innovative designed regularization term, which is used to capture the tension deviation phenomenon in the state space, i.e. whether there is a "pulling" trend (state deviating in multiple directions) in the state space of device i:

[0098]

[0099] It represents the overall pulling force of the device state deviating from the neighbors. If the neighbor state direction is highly dispersed, the Ω value is large, which helps to determine the "state drift" or "edge fault".

[0100] The creativity of the above design lies in:

[0101] The neighbor state "tension potential energy" is introduced as an anomaly regularization term, which overcomes the limitation of traditional similarity mean that it is not sensitive to "boundary drift";

[0102] The state propagation weight w ij is very close to the actual scene of Internet of Things, which is determined dynamically through RSSI to avoid blind averaging.

[0103] The device i will compare the calculated anomaly score s i with the dynamic anomaly threshold δ i to determine whether it is currently abnormal. The threshold δ i is no longer a static value, but a local dynamic adaptive mechanism is introduced:

[0104] δ i = μ i + α · σ i ; (7)

[0105] wherein:

[0106] μ i is the average of the abnormal score in the past T periods;

[0107] σ i is the standard deviation of the abnormal score in the period;

[0108] α is a coefficient for adjusting sensitivity (e.g. α = 2).

[0109] This design can dynamically adjust the threshold according to the historical state fluctuation of the device locally, enhancing the adaptability of the model to different operation modes and device types.

[0110] Step 3: Operation and maintenance strategy generation

[0111] This step aims to further determine whether operation and maintenance is needed based on the abnormal score s i and the state tension term Ω(z i ) calculated in the previous step, and generate structured control instructions Unlike traditional strategies based on a single threshold, this step designs a hierarchical operation and maintenance strategy generation mechanism that combines abnormal degree, neighbor situation tension, and response cost perception, enabling the system to adaptively select the most appropriate control action under various abnormal conditions (e.g. transient disturbance, continuous disconnection, state drift).

[0112] First, we define a comprehensive response intensity scoring function φ i to quantify the "control trigger urgency" of the current state of the device. This function not only integrates the current abnormal score s i and the tension term Ω(z i ), but also adds a behavior trust adjustment term ψ i that captures the deviation of the device's historical behavior from the current abnormality, forming a more stable strategy scoring basis. The scoring function is as follows:

[0113]

[0114] wherein:

[0115] s i is the abnormal score;

[0116] Ω(z i ) is the state tension term;

[0117] ψ i is the behavior trust factor, calculated as wherein represents the average value of the device state in the past T time windows, reflecting the stability of the device behavior;

[0118] β1, β2, β3 are empirical weight coefficients, satisfying β1+β2+β3=1;

[0119] σ(·) is a Sigmoid function, mapping the score to [0,1] for subsequent classification and policy mapping.

[0120] The creativity of this function lies in the introduction of the behavior stability feedback term ψ i , which prevents the device from being misjudged as abnormal when occasional fluctuations occur, while considering the state tension and abnormality degree, and is specially designed for the actual scene of “wireless devices may occasionally lose connection but not necessarily fail”.

[0121] Next, according to the value of φ i , the system will enter the policy mapping process. We designed a three-order response policy structure mapper, which divides φ i into different control levels according to intervals, and introduces an action risk penalty term ρ(t i ) to improve the decision-making prudence in high-risk control scenarios.

[0122] The final policy objective function is:

[0123]

[0124] Where:

[0125] is the control instruction output, t i is the operation type (such as 0=ignore, 1=reconnect,

[0126] 2=restart, 3=enter safe mode), and p i is the additional parameter (such as waiting time, action module);

[0127] · is the system pre-defined policy space;

[0128] is the target response level of each type of policy (such as light policy θ1=0.3, heavy policy θ3=0.8);

[0129] ρ(t i ) is the risk value corresponding to the operation (such as the restart operation brings high system interruption risk);

[0130] λ is the risk control coefficient, which prevents high-risk operations from being triggered unnecessarily;

[0131] Minimizing this function can obtain the most suitable operation policy that best fits the current state.

[0132] The core innovation of this design is:

[0133] The response intensity score φi Instead of simple weighting, historical state stability is considered;

[0134] Strategy selection is a penalized optimization process, rather than a fixed threshold trigger;

[0135] Action risk penalty is introduced as a constraint term in the patent for the first time, embodying the idea that "control is not better the more intense", especially suitable for embedded, low-power, remote device scenarios.

[0136] The system will select the combination from the control strategy library that minimizes the output as the final instruction The instruction content includes operation type t i and parameter p i , which will be the core input for communication scheduling and instruction issuance in the next step.

[0137] Step output:

[0138] Control instruction represents the best operation and parameter combination of the current device, which is used for subsequent steps; this output is the central scheduling information in the system control chain and is an important medium for closed-loop control.

[0139] Step 4: Channel prediction and operation path scheduling

[0140] In the previous step, the system has generated the operation control instruction for device i This control operation needs to be accurately issued to the target device through the wireless communication network. However, the communication link quality in the Internet of Things scenario fluctuates greatly, especially in complex deployment environments such as industry, wild, and underground, which can easily cause channel congestion, short-term interference, or device failure to receive. Therefore, the goal of this step is to dynamically select the best communication channel based on the type of control instruction and the prediction results of network state, and issue the control instruction to device i to maximize the success rate of instruction delivery.

[0141] The communication module of device i (such as SX1276 LoRa chip) will periodically scan available channels and record the historical communication state of each channel. For example, in LoRa, the device can open multiple frequency bands (such as 433MHz, 470MHz, etc.), and record whether the last K times of communication in each frequency band are successful.

[0142] For example, the failure record of channel f is where 1 indicates a communication failure (such as no ACK received), and 0 indicates success.

[0143] The system uses a moving average method to predict the channel:

[0144]

[0145] where:

[0146] • is the availability prediction of device i for channel f in the current period;

[0147] • is whether the last k-th communication failed (1 means failed);

[0148] • The moving average ensures that the prediction is only based on recent history and is not affected by long-term redundancy.

[0149] Considering the different urgency of control operation types t i (e.g. t i = 2 means system restart, high failure cost), we introduce the control operation risk sensitivity p(t i ). Its value is preset by the system: for example, p(1) = 0.2 when t i = 1 is a light control operation, and p(2) = 0.8 when t i = 2 is a heavy control operation.

[0150] We define the following comprehensive scheduling score function:

[0151]

[0152] where:

[0153] • is whether the last communication of this channel failed;

[0154] • λ is the risk modulation factor, which is recommended to be set to λ = 1.0 or adjusted according to the importance of the system;

[0155] • The smaller the overall score, the more reliable the channel, the more suitable it is for the current control instruction to be issued.

[0156] The system selects the channel with the smallest score:

[0157]

[0158] Finally, the control instruction will be sent to device i through channel , and the communication module (such as the LoRa driver stack) will implement actions such as packaging, sending, waiting for ACK, etc. If the channel fails, the system will continue to try in the suboptimal channel, up to M times, until it succeeds or gives up beyond the threshold.

[0159] For example:

[0160] Assume that the control instruction indicates that the current operation type is "system restart"; the parameter pi = {module:3, delay:5} means to act on module 3 with a delay of 5 seconds; the current available channels are The failure rate sliding average of their last 5 communications is as follows

[0161]

[0162] Whether the last communication failed

[0163]

[0164] The operation risk coefficient p(2) = 0.8 (because t i = 2 is a restart, which is a high-risk operation);

[0165] The risk adjustment factor is set to l = 1.0.

[0166] Next, score each channel:

[0167] The score of f1:

[0168]

[0169] The score of f2:

[0170]

[0171] The score of f3:

[0172]

[0173] Therefore, the lowest score is f1, that is:

[0174] The final system selects f1 as the transmission channel of the control command, and sends the command to the target device i through this channel.

[0175] Step 5: Command execution and state feedback update

[0176] In the previous step, the system has successfully sent the operation and maintenance control command to the target device i through the channel This step is responsible for driving the device to actually execute the control operation, and feeding back the execution result (success or failure) to the upper system through wireless means, thereby closing the entire operation and maintenance closed loop.

[0177] After receiving the control command , the device i will execute the control operation according to the control type t i and the parameter pi , the local control module is called for operation through the preset control mapping table.

[0178] For example:

[0179] If t i = 1 (communication reconnection), the device reinitializes the communication stack (such as restarting the LoRa module);

[0180] If t i = 2 (module restart), the specified module p i .module is reset, and after execution, the main loop is reentered;

[0181] If t i = 3 (safety mode switching), a flag bit is written into the register to enter low-power mode.

[0182] After the operation is completed, the device constructs a feedback data packet through the same channel and returns it to the edge node or the host server. The feedback content includes instruction ID, device ID, operation type, execution result, etc., and the system extracts the execution result flag bit to generate a feedback scalar r i :

[0183]

[0184] The feedback mechanism is realized through ACK receipt at the communication protocol layer or heartbeat at the application layer. For important control types (such as t i = 2), the system supports retransmission N times within T seconds if no feedback is received, until r i ≠ 0 or the retry threshold is exceeded.

[0185] The feedback mechanism at the device end is realized through state machines and driver callbacks in software, for example, using a control thread of FreeRTOS to cooperate with a communication thread. All feedback information is sent to the central state database to drive the next round of judgment, control closed loop, or manual intervention reminder.

[0186] The above embodiments only describe the preferred embodiments of the present application, and do not limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements to the technical solutions of the present application made by ordinary engineering technical personnel in the art shall fall within the protection scope determined by the claims of the present application.

Claims

1. A wireless operation and maintenance method based on Internet of Things, characterized in that, The method comprises the following steps: S1: collecting original operation data through an Internet of Things device, and converting the original operation data into a current device state vector; S2: obtaining an abnormal score and a state tension term of the current device by combining the current device state vector and a neighbor device vector; S3: constructing a response intensity score function based on the abnormal score and the state tension term, and outputting a response intensity value through the response intensity score function; S4: inputting the response intensity value into a structure mapper, and outputting a control instruction through the structure mapper; S5: issuing the control instruction to the device through a communication channel; S6: executing an operation according to the control instruction by the Internet of Things device, and feeding back an execution result to the system; The abnormal score expression in S2 is: ； wherein is the abnormal score, is the current device state vector, is the neighbor device state vector, is the neighbor device has an impact on the current device state weight, is the regularization term coefficient; is the regularization term, is the neighbor device set; The state tension term expression in S2 is: ； wherein is a state tension term; The response intensity score function expression in S3 is: ； wherein is the behavior trust factor, calculated as wherein is the average of the device state in the past time windows, is the experience weight coefficient, is the Sigmoid function. 2.The Internet of Things based wireless operation and maintenance method according to claim 1, characterized in that, The S1 specifically comprises: S1.1: obtaining original operation data, and periodically sending the original operation data to an edge node through a wireless protocol; S1.2: performing standardization processing on the original operation data through the edge node; S1.3: performing data enhancement on the original operation data after the standardization processing, and generating a positive sample pair; S1.4: inputting the positive sample pair into an encoder, and generating a current device state vector through the encoder. 3.The Internet of Things based wireless operation and maintenance method according to claim 1, characterized in that, The communication channel generation in S5 comprises: periodically scanning available channels through a communication module of the Internet of Things device, and recording a historical communication state record of each channel; predicting a plurality of channels in a sliding average manner based on the historical communication state record; inputting a prediction result into a comprehensive scheduling score function, and outputting a score result of the plurality of channels through the comprehensive scheduling score function; selecting a channel with the minimum score value as the communication channel. 4.The Internet of Things based wireless operation and maintenance method according to claim 1, characterized in that, The S6 specifically comprises: S6.1: after accepting the control instruction, the Internet of Things device calls a local control module to execute an operation through a preset control mapping table; S6.2: after the operation is completed, the Internet of Things device constructs a feedback data packet through the same communication channel, and returns the feedback data packet to the system; S6.3: the system accepts the feedback data packet, extracts an execution result flag bit of the feedback data packet, and generates a feedback scalar. 5.The wireless operation and maintenance method based on Internet of Things according to claim 4, characterized in that, The feedback data packet of S6.2 comprises an instruction ID, a device ID, an operation type and an execution result. 6.The wireless operation and maintenance method based on Internet of Things according to claim 4, characterized in that, The original data in S1 comprises a voltage, a device chip temperature, a wireless receiving strength, a network response delay and a device load.

7. A wireless operation and maintenance system based on Internet of Things, applied to the wireless operation and maintenance method based on Internet of Things in any one of claims 1-6, characterized in that, The method comprises the following steps: an Internet of Things device for collecting original operation data; an encoder for generating a device state vector; a structure mapper for dividing a response intensity value into different control levels according to intervals, and generating a control instruction.

Images