Internet of Things remote sensing and monitoring system and method based on big data

Abstract

The invention discloses an Internet of Things remote sensing and monitoring system and method based on big data, and relates to the technical field of Internet of Things communication, the remote sensing and monitoring system comprises a sensing layer, a network layer, a data layer and an application layer, the sensing layer is provided with a data acquisition module, and the data acquisition module is provided with different types of sensors; and the data acquisition module acquires different types of sensor energy losses, marks the different types of sensor energy losses as a class A, and marks data acquired by the different types of sensors as a class B. The system has the advantages that the class B data is classified and stored through the network layer and the data layer, the application layer calls the classified and stored data, and the data acquisition efficiency is improved. The energy loss database is constructed according to the A-type data and the B-type data, and then the working states of different sensors are adjusted by using the energy loss database, so that the energy loss of the sensors is optimized, and the energy loss is optimized while the working state of the monitoring system is ensured.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things communication technology, and in particular to an Internet of Things remote sensing monitoring system and method based on big data. Background Art

[0002] The IoT system generally includes a large number of IoT nodes, such as terminal device nodes and server nodes. The terminal device nodes are used to collect various data, and the server nodes and terminal device nodes transmit data by establishing communication channels and perform data analysis and processing. However, the existing Internet of Things remote perception monitoring system needs to collect target data in real time to ensure the accuracy of the perception monitoring system. However, this approach also increases the energy loss of the perception system. To this end, we propose an Internet of Things remote perception monitoring system and method based on big data. Summary of the invention

[0003] The purpose of the present invention is to provide a big data-based remote sensing monitoring system and method for the Internet of Things.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a remote sensing monitoring system and method of the Internet of Things based on big data, wherein the remote sensing monitoring system comprises a sensing layer, a network layer, a data layer and an application layer; The perception layer is provided with a data acquisition module, in which different types of sensors are provided, and the data acquisition module obtains different types of sensor energy losses, marks different types of sensor energy losses as Class A, and marks data collected by different types of sensors as Class B; The network layer is provided with a data cleaning module, a transmission monitoring module and a data encryption module. The data cleaning module is used to clean the Class B data. The data encryption module uses the AES algorithm to encrypt the cleaned Class B data and then transmits the encrypted Class B data to the data layer. The transmission monitoring module is used to monitor the encrypted Class B data and transmit the Class A data marked in the perception layer. The data layer is provided with a data extraction module, a local storage module and a cloud storage module. The cloud storage module establishes various types of storage areas according to different types of sensors, transmits Class B data to the cloud storage module, and stores Class B data according to the corresponding storage area. The local storage module stores Class A data after receiving it, and the data extraction module extracts feature data from Class B data; The application layer is provided with a data processing module and an energy management module. The data processing module is used to analyze and predict the development status of Class B data, and the energy management module is used to manage Class A data. The data processing module uses digital twin technology to analyze and predict Class B data.

[0005] As a further solution of the present invention: the energy management module is provided with an energy loss database, an energy optimization unit and a dynamic adjustment unit, the energy loss database is used to store historical energy loss data and Class A data of different types of sensor nodes, wherein the Class A data further includes data of sensor type, operating frequency and environmental parameters associated with the energy loss data, the energy optimization unit constructs an energy loss model through the energy loss database, and the dynamic adjustment unit adjusts the energy loss of different types of sensors according to the energy loss model.

[0006] As a further solution of the present invention: the energy optimization unit expresses the energy loss through a formula, and the specific formula is as follows: ; in, Indicates The energy loss of each sensor is It represents the square term of the operating frequency, reflecting the nonlinear energy loss. Indicates The operating frequency of the sensor is Indicates The ambient temperature of each sensor, Indicates The interaction term between the operating frequency of each sensor and the ambient temperature reflects the impact of the two on energy loss. Indicates The energy loss constant of each sensor is , as well as They respectively represent the weight coefficients of the operating frequency and ambient temperature obtained by fitting the historical data, thereby optimizing the energy loss of different types of sensors.

[0007] As a further solution of the present invention: the data accuracy of the perception layer With the operating frequency Proportional, introduce the following relationship: ; in, represents the proportionality coefficient, is expressed as a constant, and the energy loss model is brought into the dynamic adjustment unit to generate the sensor frequency optimization strategy through the formula. The specific formula is as follows: ; in, Indicates The energy loss function of each sensor is expressed as: ; By adjusting the sensor operating frequency , making the total energy loss of all sensors Minimize, where Indicates The optimization strategy of the sensor operating frequency is It represents the variable value when the objective function takes the minimum value, and then the dynamic adjustment unit adjusts the working frequency of different sensors in real time; The constraints are: ; in, Indicates the lower threshold of data accuracy, Indicates The data accuracy of each sensor.

[0008] As a further solution of the present invention: the data processing module analyzes Class B data, obtains the accuracy of Class B data collection by different sensors, transmits the Class B data and the accuracy data of different sensors to the energy management module, and then the energy management module adjusts the working status data of different sensors according to the accuracy of Class B data.

[0009] As a further solution of the present invention: a prediction model is provided in the data processing module, the prediction model retrieves historical Class B data, and the prediction model analyzes the prediction data for the next 24 hours according to the historical trends of different types of Class B data, and the prediction model adopts a three-level progressive prediction of 4h+8h+12h, and each level of prediction is based on the results of the previous level: Level 1: 0-4h forecast is short-term with high accuracy; Level 2: 4-12h prediction is an 8h extension based on the 4h result; Level 3: 12-24h prediction is a 12h extension based on the 12h result; The prediction model obtains the result data of the historical trend of Class B data in the next 24 hours through the formula. The specific formula is as follows: ; ; in, It represents the result data of the prediction model for Class B data for 12 hours. Indicates the prediction result data of the prediction model based on the first level. It represents the 12-hour result data directly predicted by the prediction model. It represents the result data of the prediction model for Class B data for 24 hours. Indicates the prediction result data based on the prediction model at level 2. It represents the 24-hour result data directly predicted by the prediction model. and Represent the dynamic weight coefficients respectively.

[0010] As a further solution of the present invention: a backup sensor unit is provided in the data acquisition module, and the backup sensor unit is used to temporarily obtain Class B data; an energy loss threshold is provided in the energy management module; when the transmission monitoring module detects that the sensor is abnormal and When the energy loss threshold is reached, the data acquisition module switches to the backup sensor unit to obtain Class B data.

[0011] As a further solution of the present invention: an early warning unit is set in the application layer, and the early warning unit is used to obtain data in the data processing module and the energy management module, and perform early warning processing when abnormal data appears in the data processing module and the energy management module.

[0012] In addition, the present invention also provides a remote sensing monitoring method based on big data, and the specific steps are as follows: S1. Construct multiple sensors of different types to collect the perception data of the target object, then mark the data about the sensor itself as Class A data, and mark the data collected by the sensor as Class B data; S2. The network layer then cleans and classifies the Class B data for transmission, and then classifies and transmits the Class B data and Class A data, and encrypts and stores the Class B data; S3. After receiving the A-type data and the B-type data, extract features from the B-type data, and classify and store the A-type data and the B-type data; S4. Construct an energy loss database, which records the historical energy loss data of each sensor node and associates it with different types of sensors, operating frequencies and environmental parameters, thereby adjusting the energy loss between different types of sensors and optimizing the energy loss of the overall monitoring system.

[0013] By adopting the above technical solution, compared with the prior art, the beneficial effects of the present invention are: 1. The present invention obtains perception data through the perception layer, marks the perception target data collected by the sensor as Class B data, that is, the perception target data is the data collected by the sensor, and then uses the network layer and the data layer to classify and store the Class B data, the application layer calls the classified and stored data, and constructs an energy loss database based on Class A data and Class B data, and then uses the energy loss database to adjust the working status of different sensors, thereby optimizing the energy loss of the sensor, so that the monitoring system optimizes the energy loss while ensuring the working status; 2. The present invention uses an energy optimization unit to analyze the Class A data and Class B data to obtain energy loss data, and adjusts the energy loss of different sensors according to the energy loss data, so that the monitoring system can adjust the working status of different sensors in real time; 3. The present invention uses a dynamic adjustment unit to analyze the operating frequencies of different sensors in real time, optimize the best working state of the sensor, and effectively avoid the data redundancy problem caused by excessively high operating frequency; 4. The present invention obtains Class B data for future time periods by predicting and processing data from different time periods, so that the monitoring system can generate corresponding processing solutions according to different time periods, which is convenient for timely adjustment when abnormalities occur in the perception layer, network layer, data layer and application layer.

[0014] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a remote sensing monitoring system in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0017] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0018] Embodiment 1: Please refer to the attached Figure 1 The present invention provides a remote sensing monitoring system and method of the Internet of Things based on big data. The remote sensing monitoring system includes a sensing layer, a network layer, a data layer and an application layer; The perception layer is provided with a data acquisition module, in which different types of sensors are provided. The data acquisition module obtains different types of sensor energy losses, marks different types of sensor energy losses as Class A, and marks data collected by different types of sensors as Class B; The network layer is equipped with a data cleaning module, a transmission monitoring module and a data encryption module. The data cleaning module is used to clean Class B data. The data encryption module uses the AES algorithm to encrypt the cleaned Class B data and then transmits the encrypted Class B data to the data layer. The transmission monitoring module is used to monitor the encrypted Class B data and transmit the Class A data marked in the perception layer. The data layer is provided with a data extraction module, a local storage module and a cloud storage module. The cloud storage module establishes various types of storage areas according to different types of sensors, transmits Class B data to the cloud storage module, and stores Class B data according to the corresponding storage area. The local storage module stores Class A data after receiving it, and the data extraction module extracts feature data from Class B data; The application layer is equipped with a data processing module and an energy management module. The data processing module is used to analyze and predict the development status of Class B data, and the energy management module is used to manage Class A data. The data processing module uses digital twin technology to analyze and predict Class B data.

[0019] Specific workflow: Build multiple sensors of different types to collect perception data of the target object, then mark the data about the sensor itself as Class A data, mark the data collected by the sensor as Class B data, and then the network layer cleans and classifies the Class B data for transmission, and then classifies and transmits Class B data and Class A data, encrypts and stores Class B data, and after receiving Class A data and Class B data, extracts features from Class B data, and classifies and stores Class A data and Class B data, builds an energy loss database, which records the historical energy loss data of each sensor node, and associates it with the sensor type, operating frequency, and environmental parameters, and then adjusts the energy loss between each sensor, thereby optimizing the energy loss of the overall monitoring system.

[0020] Furthermore, the perception data is obtained through the perception layer, and the perception target data collected by the sensor is marked as Class B data, that is, the perception target data is the data collected by the sensor, and then the Class B data is classified and stored using the network layer and the data layer. The application layer retrieves the classified stored data and builds an energy loss database based on Class A data and Class B data. The energy loss database is then used to adjust the working status of different sensors, thereby optimizing the energy loss of the sensor, thereby achieving the goal of optimizing the energy loss while ensuring the working status of the monitoring system.

[0021] Embodiment 2: Based on the first embodiment, please refer to the attached Figure 1 As shown, the energy management module is provided with an energy loss database, an energy optimization unit and a dynamic adjustment unit. The energy loss database is used to store historical energy loss data and Class A data of different types of sensor nodes, wherein the Class A data further includes data of sensor type, operating frequency and environmental parameters associated with the energy loss data. The energy optimization unit constructs an energy loss model through the energy loss database. The dynamic adjustment unit adjusts the energy loss of different types of sensors according to the energy loss model. The energy optimization unit expresses the energy loss through a formula. The specific formula is as follows: ; in, Indicates The energy loss of each sensor is It represents the square term of the operating frequency, reflecting the nonlinear energy loss. Indicates The operating frequency of the sensor is Indicates The ambient temperature of each sensor, Indicates The interaction term between the operating frequency of each sensor and the ambient temperature reflects the impact of the two on energy loss. Indicates The energy loss constant of each sensor is , as well as They respectively represent the weight coefficients of the operating frequency and ambient temperature obtained by fitting the historical data, thereby optimizing the energy loss of different types of sensors.

[0022] Specific workflow: Sensor data acquisition: Heterogeneous sensor arrays collect sensory data of target objects such as temperature, humidity, vibration, light, etc. in real time, and simultaneously record the operating frequency, ambient temperature and sensor type identification of each sensor node. Environmental parameter monitoring: Obtain real-time environmental parameters (such as temperature, humidity, electromagnetic interference intensity) of the sensor deployment area through an independent environmental monitoring module to ensure the correlation between data and sensor energy consumption. Data cleaning and standardization: Outlier removal: Use the 3σ principle or box plot method to filter abnormal energy consumption data (such as instantaneous current spikes). Data normalization: Min-Max normalization of operating frequency and ambient temperature to eliminate dimensional effects. Missing value filling: Use time series interpolation (such as linear interpolation) or the mean of similar sensor data to fill missing values, and use multivariate linear regression to fit historical data to obtain the formula. , as well as data.

[0023] Furthermore, the energy optimization unit analyzes the Class A data and Class B data to obtain energy loss data, and adjusts the energy loss of different sensors based on the energy loss data, so that the monitoring system can adjust the working status of different sensors in real time.

[0024] Embodiment three: Based on the second embodiment, please refer to the attached Figure 1 As shown, the data accuracy of the perception layer With the operating frequency Proportional, introduce the following relationship: ; in, represents the proportionality coefficient, It is expressed as a constant, and the energy loss model is brought into the dynamic adjustment unit to generate the sensor frequency optimization strategy through the formula. The specific formula is as follows: ; in, Indicates The energy loss function of each sensor is expressed as: ; By adjusting the sensor operating frequency , making the total energy loss of all sensors Minimize, where Indicates The optimization strategy of the sensor operating frequency is It represents the variable value when the objective function takes the minimum value, and then the dynamic adjustment unit adjusts the working frequency of different sensors in real time; The constraints are: ; in, Indicates the lower threshold of data accuracy, Indicates The data processing module analyzes the Class B data to obtain the accuracy of Class B data collected by different sensors, and transmits the Class B data and the accuracy data of different sensors to the energy management module. Then the energy management module adjusts the working status data of different sensors according to the accuracy of Class B data.

[0025] Specific workflow: Data collection: Collect historical data, including the operating frequency of the sensor , Ambient temperature , Energy loss and data accuracy , using historical data to fit the energy loss model , as well as , using historical data to fit the data accuracy model and , using an optimization algorithm to solve , ensuring that energy loss is minimized while meeting data accuracy constraints, and dynamically adjusting the operating frequency of the sensor based on real-time data ,Ensure that the system is always in the optimal state, by building the minimum target threshold,,ensuring the sensor working state data when the perception layer obtains the target data,,thus ensuring the sensor working state.

[0026] Furthermore, the operating frequencies of different sensors are analyzed in real time through the dynamic adjustment unit, thereby obtaining the optimal working states of different sensors, effectively preventing data duplication caused by high operating frequencies of sensors.

[0027] Embodiment 4: Based on the third embodiment, please refer to the attached Figure 1 As shown, a prediction model is set in the data processing module. The prediction model retrieves historical Class B data, and the prediction model analyzes the prediction data for the next 24 hours based on the historical trends of different types of Class B data. The prediction model adopts a three-level progressive prediction of 4h+8h+12h, and each level of prediction is based on the results of the previous level: Level 1: 0-4h forecast is short-term with high accuracy; Level 2: 4-12h prediction is an 8h extension based on the 4h result; Level 3: 12-24h prediction is a 12h extension based on the 12h result; The prediction model uses a formula to obtain the result data of the historical trend of Class B data in the next 24 hours. The specific formula is as follows: ; ; in, It represents the result data of the prediction model for Class B data for 12 hours. Indicates the prediction result data of the prediction model based on the first level. It represents the 12-hour result data directly predicted by the prediction model. It represents the result data of the prediction model for Class B data for 24 hours. Indicates the prediction result data based on the prediction model at level 2. It represents the 24-hour result data directly predicted by the prediction model. and They represent dynamic weight coefficients respectively. A backup sensor unit is set in the data acquisition module. The backup sensor unit is used to temporarily obtain Class B data. An energy loss threshold is set in the energy management module. When the transmission monitoring module detects that the sensor is abnormal and When the energy loss threshold is reached, the data acquisition module switches to the backup sensor unit to obtain Class B data. An early warning unit is set in the application layer. The early warning unit is used to obtain data from the data processing module and the energy management module, and perform early warning processing when abnormal data appears in the data processing module and the energy management module.

[0028] Furthermore, by predicting and processing data from different time periods, Class B data at different times can be obtained, and the monitoring system can generate corresponding processing plans based on different time periods, so that the monitoring system can make timely adjustments when anomalies occur in the perception layer, network layer, data layer and application layer.

[0029] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A remote sensing monitoring system of the Internet of Things based on big data, characterized by: The remote sensing monitoring system includes a sensing layer, a network layer, a data layer and an application layer; The perception layer is provided with a data acquisition module, in which different types of sensors are provided, and the data acquisition module obtains different types of sensor energy losses, marks different types of sensor energy losses as Class A, and marks data collected by different types of sensors as Class B; The network layer is provided with a data cleaning module, a transmission monitoring module and a data encryption module. The data cleaning module is used to clean the Class B data. The data encryption module uses the AES algorithm to encrypt the cleaned Class B data and then transmits the encrypted Class B data to the data layer. The transmission monitoring module is used to monitor the encrypted Class B data and transmit the Class A data marked in the perception layer. The data layer is provided with a data extraction module, a local storage module and a cloud storage module. The cloud storage module establishes various types of storage areas according to different types of sensors, transmits Class B data to the cloud storage module, and stores Class B data according to the corresponding storage area. The local storage module stores Class A data after receiving it, and the data extraction module extracts feature data from Class B data; The application layer is provided with a data processing module and an energy management module. The data processing module is used to analyze and predict the development status of Class B data, and the energy management module is used to manage Class A data. The data processing module uses digital twin technology to analyze and predict Class B data.

2. According to the big data-based IoT remote sensing monitoring system of claim 1, it is characterized by: The energy management module is provided with an energy loss database, an energy optimization unit and a dynamic adjustment unit. The energy loss database is used to store historical energy loss data and Class A data of different types of sensor nodes, wherein the Class A data further includes data of sensor type, operating frequency and environmental parameters associated with the energy loss data. The energy optimization unit constructs an energy loss model through the energy loss database, and the dynamic adjustment unit adjusts the energy loss of different types of sensors according to the energy loss model.

3. The IoT remote sensing monitoring system based on big data according to claim 2 is characterized in that: The energy optimization unit expresses energy loss through a formula, and the specific formula is as follows: ； in, Indicates The energy loss of each sensor is It represents the square term of the operating frequency, reflecting the nonlinear energy loss. Indicates The operating frequency of the sensor is Indicates The ambient temperature of each sensor, Indicates The interaction term between the operating frequency of each sensor and the ambient temperature reflects the impact of the two on energy loss. Indicates The energy loss constant of each sensor is , as well as They respectively represent the weight coefficients of the operating frequency and ambient temperature obtained by fitting the historical data, thereby optimizing the energy loss of different types of sensors.

4. The IoT remote sensing monitoring system based on big data according to claim 3 is characterized in that: The data accuracy of the perception layer With the operating frequency Proportional, introduce the following relationship: ； in, represents the proportionality coefficient, is expressed as a constant, and the energy loss model is brought into the dynamic adjustment unit to generate the sensor frequency optimization strategy through the formula. The specific formula is as follows: ； in, Indicates The energy loss function of each sensor is expressed as: ； By adjusting the sensor operating frequency , making the total energy loss of all sensors Minimize, where Indicates The optimization strategy of the sensor operating frequency is It represents the variable value when the objective function takes the minimum value, and then the dynamic adjustment unit adjusts the working frequency of different sensors in real time; The constraints are: ； in, Indicates the lower threshold of data accuracy, Indicates The data accuracy of each sensor.

5. The IoT remote sensing monitoring system based on big data according to claim 4 is characterized in that: The data processing module analyzes Class B data, obtains the accuracy of Class B data collected by different sensors, and transmits Class B data and accuracy data of different sensors to the energy management module, which then adjusts the working status data of different sensors according to the accuracy of Class B data.

6. The IoT remote sensing monitoring system based on big data according to claim 5 is characterized by: The data processing module is provided with a prediction model, which retrieves historical Class B data and analyzes the prediction data for the next 24 hours based on the historical trends of different types of Class B data. The prediction model adopts a three-level progressive prediction of 4h+8h+12h, and each level of prediction is based on the results of the previous level: Level 1: 0-4h forecast is short-term with high accuracy; Level 2: 4-12h prediction is an 8h extension based on the 4h result; Level 3: 12-24h prediction is a 12h extension based on the 12h result; The prediction model obtains the result data of the historical trend of Class B data in the next 24 hours through the formula. The specific formula is as follows: ； ； in, It represents the result data of the prediction model for Class B data for 12 hours. Indicates the prediction result data of the prediction model based on the first level. It represents the 12-hour result data directly predicted by the prediction model. It represents the result data of the prediction model for Class B data for 24 hours. Indicates the prediction result data based on the prediction model at level 2. It represents the 24-hour result data directly predicted by the prediction model. and Represent the dynamic weight coefficients respectively.

7. The IoT remote sensing monitoring system based on big data according to claim 6 is characterized by: The data acquisition module is provided with a backup sensor unit, which is used to temporarily obtain Class B data. The energy management module is provided with an energy loss threshold. When the transmission monitoring module detects that the sensor is abnormal and When the energy loss threshold is reached, the data acquisition module switches to the backup sensor unit to obtain Class B data.

8. The IoT remote sensing monitoring system based on big data according to claim 7 is characterized by: An early warning unit is provided in the application layer, and the early warning unit is used to obtain data in the data processing module and the energy management module, and to perform early warning processing when abnormal data appears in the data processing module and the energy management module.

9. A remote sensing monitoring method for a big data-based Internet of Things remote sensing monitoring system applicable to any one of claims 1-8, characterized in that: The specific steps are as follows: S1. Construct multiple sensors of different types to collect the perception data of the target object, then mark the data about the sensor itself as Class A data, and mark the data collected by the sensor as Class B data; S2. The network layer then cleans and classifies the Class B data for transmission, and then classifies and transmits the Class B data and Class A data, and encrypts and stores the Class B data; S3. After receiving the A-type data and the B-type data, extract features from the B-type data, and classify and store the A-type data and the B-type data; S4. Construct an energy loss database, which records the historical energy loss data of each sensor node and associates it with different types of sensors, operating frequencies and environmental parameters, thereby adjusting the energy loss between different types of sensors and optimizing the energy loss of the overall monitoring system.

Images