A data collection system and method for Internet of Things terminals

Abstract

The present invention discloses a data collection system and method for an Internet of Things terminal, wherein the system comprises: a user sends a request message for data transmission to an Internet of Things terminal, the Internet of Things terminal identifies a unique identifier in the request message, and a server generates a configuration file for data transmission; the Internet of Things terminal detects the configuration file, and verifies the verification code of the user by generating verification information; a first sensor collects the user's behavior data in the configuration file for preprocessing, and the Internet of Things terminal extracts the user's identity for identity authentication; it is determined whether the user's identity is authenticated, and if so, a data collection operation is performed through the user's device, and the second sensor in the device is used to collect environmental information, and the environmental information is transmitted to the Internet of Things terminal. The present invention improves the data collection efficiency of the Internet of Things terminal and ensures the security of data information by reading, detecting and identifying the user's request information.

Description

Technical Field

[0001] The present invention relates to Internet of Things data transmission technology, and in particular to an Internet of Things terminal data collection system and method. Background Art

[0002] With the rapid development of science and technology, the production and management efficiency of various industries has been greatly improved. The use of the Internet of Things for data collection has attracted much attention. However, various difficult-to-solve problems will arise in the process of collecting data. For example, when a large amount of data needs to be collected, the data processing efficiency is low, and data collection will generate a large consumption. In the process of processing network data, it is necessary to collect data through the terminal and transmit the data. After the data is processed by the server, the user can get the server feedback and the required relevant information. However, the traditional Internet of Things data collection method has many limitations. When a large amount of data needs to be collected, more energy is consumed in the collection process, and the security of data collection is difficult to guarantee. For this reason, an Internet of Things terminal data collection system and method are proposed to solve the problems existing in reality. Summary of the invention

[0003] The present invention provides an Internet of Things terminal data acquisition system and method to solve the above problems existing in the prior art.

[0004] An Internet of Things terminal data collection system, comprising:

[0005] Information access module: The user sends a data transmission request message to the IoT terminal, the IoT terminal identifies the unique identifier in the request message, and the server generates a configuration file for data transmission;

[0006] Data encryption module: The IoT terminal detects the configuration file and obtains the user's verification code for verification by generating verification information;

[0007] Identity authentication module: The first sensor collects the user's behavior data in the configuration file for preprocessing, and the IoT terminal extracts the user's identity for identity authentication;

[0008] Data collection module: Determine whether the user identity is authenticated. If so, perform data collection operations through the user's device, use the second sensor in the device to collect environmental information, and transmit the environmental information to the Internet of Things terminal.

[0009] Preferably, the data encryption module comprises:

[0010] Information judgment unit: The configuration file contains data status information, and judges whether the status information is serviceable information that can serve the IoT terminal. If it is, an encryption key generated by the status information is used as the first encryption information to encrypt and transmit the status information;

[0011] Information decryption unit: preset information corresponding to the status information in the configuration file is preset in the server, the server performs key identification on the encryption key, determines whether the encryption key can match the preset information, and if the match is successful, generates a decryption key from the preset information to decrypt the encryption key in the first encrypted information;

[0012] Information verification unit: The IoT terminal generates a verification code as the second encrypted information and sends it to the user. After the user receives the second encrypted information containing the verification code, the user sends the verification code to the IoT terminal to complete the user's information registration on the IoT terminal.

[0013] Preferably, the identity authentication module includes:

[0014] Data preprocessing unit: obtains the configuration file generated by the server to build a data feature model. The IoT terminal obtains the user's behavior data and preprocesses the behavior data; the preprocessing includes smoothing and denoising, outlier processing, and data segmentation operations;

[0015] Feature extraction unit: obtains pre-processed behavior data and builds a user-based convolutional neural network model based on the status information in the configuration file, which is used to extract feature information from user behavior data using a deep neural learning network, and trains the feature information to assist in determining the user's identity;

[0016] User authentication unit: The trained feature information is constructed into a feature set through classification and combination to determine whether the current user is a legitimate user.

[0017] Preferably, the data preprocessing unit includes:

[0018] Data sampling subunit: obtains the user's behavior data, which includes the data signal generated by the user during exercise, and uses a linear interpolation algorithm to uniformly sample the data signal received by the first sensor to obtain a signal sequence;

[0019] Periodic segmentation subunit: obtains the morphological signal in the signal sequence, takes the periodic change of the morphological signal as the morphological period, and evenly segments the morphological signal to obtain sequence units;

[0020] Data denoising subunit: obtains the morphological features in the behavioral data of the sequence unit, uses the built-in filter of the IoT terminal, denoises the sequence unit according to the morphological features, and obtains the user's behavioral data after preprocessing.

[0021] Preferably, the feature extraction unit comprises:

[0022] Feature extraction subunit: obtains the configuration file in the server, builds a convolutional neural network model to extract the feature information of the status information in the configuration file; extracts the segmentation results of the morphological period in the behavior data to train the convolutional neural network model, and finally extracts the user's feature information;

[0023] Identity recognition subunit: obtain the user's behavior data and build a user classification model to determine whether the user is a legitimate user; divide the behavior data into a training set, a validation set, and a test set according to a preset ratio, iterate the behavior data in the training set, and continuously reduce the loss function of the training set until it no longer changes; input the behavior data in the test set into the user classification model, calculate whether the classification accuracy of the test set meets the preset standard, and if so, apply the trained user classification model to the validation set, and determine the legitimacy of the user's identity based on the feature information.

[0024] Preferably, the data acquisition module comprises:

[0025] Model identification unit: builds a device model library to collect the device models currently used by all users. The user sends a request to the IoT terminal to obtain data. The IoT terminal identifies the device model currently used by the user and determines the accuracy of the identification by searching in the device model library.

[0026] Communication establishing unit: if the IoT terminal retrieves the current user's device through the device model library, a communication protocol between the IoT terminal and the second sensor is established through the server for collecting data;

[0027] Connection test unit: Use the communication protocol to perform data collection test on the IoT terminal, read the storage space location after data collection in the second sensor, and if the reading is successful, continue to collect data and store it in the storage space.

[0028] Preferably, the data acquisition module further includes:

[0029] Image acquisition unit: The IoT terminal collects image data generated by the user when using the device. The image data is collected by a movable imaging device preset in the IoT terminal;

[0030] Gesture display unit: The device used by the user is provided with different types of interfaces for collecting different types of environmental data and obtaining feature information in the image data. The feature information includes the user's gesture action, and the type of interface to be called is determined according to the gesture action;

[0031] Interface detection unit: When calling any interface, detect the occupancy status of the current interface. If the current interface is occupied, return the real-time environmental data in the interface. If the current interface is not occupied, extract the feature information in the image data to update the environmental data.

[0032] Preferably, the gesture display unit comprises:

[0033] Image marking subunit: obtains image data collected by the mobile imaging device, visually marks the image data, uses the user's gesture action in the image data as a first mark, and the image background other than the gesture action as a second mark. If multiple first marks appear in the scene, they can be confirmed by the user;

[0034] The user association subunit sets the association between the user's gesture action and the mobile imaging device as a gesture event. When the user triggers the gesture event, the IoT terminal acquires image data and determines the interface object of the user's gesture action according to the first mark in the image data.

[0035] Scene judgment subunit: The movable imaging device also includes locating the current user's position and measuring the moving distance, and judging the scene information of the current user according to the second mark to assist in collecting environmental data.

[0036] A method for collecting data from an Internet of Things terminal, comprising:

[0037] S10: The user sends a data transmission request message to the IoT terminal, the IoT terminal identifies the unique identifier in the request message, and the server generates a configuration file for data transmission;

[0038] S20: The IoT terminal detects the configuration file and obtains the user's verification code for verification by generating verification information;

[0039] S30: The first sensor collects the user's behavior data in the configuration file for preprocessing, and the IoT terminal extracts the user's identity for identity authentication;

[0040] S40: Determine whether the user identity is authenticated. If so, perform data collection operations through the user's device, use the second sensor in the device to collect environmental information, and transmit the environmental information to the Internet of Things terminal.

[0041] Preferably, step S20 includes:

[0042] S21: The configuration file contains status information of the data, and determines whether the status information is serviceable information that can serve the IoT terminal. If it is determined to be yes, an encryption key generated by the status information is used as the first encryption information to encrypt and transmit the status information;

[0043] S22: Preset information corresponding to the status information in the configuration file is preset in the server, and the server performs key identification on the encryption key to determine whether the encryption key can match the preset information. If the match is successful, the preset information is used to generate a decryption key to decrypt the encryption key in the first encrypted information;

[0044] S23: The IoT terminal generates a verification code as the second encrypted information and sends it to the user. After receiving the second encrypted information including the verification code, the user sends the verification code to the IoT terminal to complete the user information registration in the IoT terminal.

[0045] Compared with the prior art, the present invention has the following advantages: the present invention improves the data collection efficiency of the Internet of Things terminal and ensures the security of data information by reading, detecting and identifying the request information for data transmission sent by the user to the Internet of Things terminal.

[0046] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0047] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0049] Figure 1 This is a schematic diagram of the module structure of an Internet of Things data terminal collection system in an embodiment of the present invention;

[0050] Figure 2 This is a structural diagram of user identity authentication in an Internet of Things data terminal collection system in an embodiment of the present invention;

[0051] Figure 3 A flowchart of a data collection test process in a method for collecting data from an Internet of Things terminal in an embodiment of the present invention;

[0052] Figure 4 This is a diagram of the steps of encrypting and transmitting data in a method for collecting data from an Internet of Things terminal in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0054] Reference Figure 1 , an embodiment of the present invention provides an Internet of Things terminal data collection system, including:

[0055] Information access module: The user sends a data transmission request message to the IoT terminal, the IoT terminal identifies the unique identifier in the request message, and the server generates a configuration file for data transmission;

[0056] Data encryption module: The IoT terminal detects the configuration file and obtains the user's verification code for verification by generating verification information;

[0057] Identity authentication module: The first sensor collects the user's behavior data in the configuration file for preprocessing, and the IoT terminal extracts the user's identity for identity authentication;

[0058] Data collection module: Determine whether the user identity is authenticated. If so, perform data collection operations through the user's device, use the second sensor in the device to collect environmental information, and transmit the environmental information to the Internet of Things terminal.

[0059] In another embodiment, a data encryption module includes:

[0060] Information judgment unit: The configuration file contains data status information, and judges whether the status information is serviceable information that can serve the IoT terminal. If it is, an encryption key generated by the status information is used as the first encryption information to encrypt and transmit the status information;

[0061] Information decryption unit: preset information corresponding to the status information in the configuration file is preset in the server, the server performs key identification on the encryption key, determines whether the encryption key can match the preset information, and if the match is successful, generates a decryption key from the preset information to decrypt the encryption key in the first encrypted information;

[0062] Information verification unit: The IoT terminal generates a verification code as the second encrypted information and sends it to the user. After the user receives the second encrypted information containing the verification code, the user sends the verification code to the IoT terminal to complete the user's information registration on the IoT terminal.

[0063] The working principle of the above technical solution is: the solution adopted in this embodiment is, taking a smart wearable device as an example, when a user uses the smart wearable device, he first needs to register his identity information. The smart wearable device requires the Internet of Things terminal to issue a request for data transmission. The Internet of Things terminal identifies the unique identification information in the request information, and the server generates a configuration file for data transmission, wherein the configuration file includes the status information of the smart wearable device, such as the location information of the smart wearable device, the model information of the smart wearable device, the user's registration identification information, and the unique identification in the request information.

[0064] The status information in the configuration file generates an encryption key as the first encryption information, and preset information corresponding to the status information of the configuration file is preset in the server. The server determines whether the first encryption information contains serviceable data information that can serve the IoT terminal. If so, the preset information is used to generate a decryption key to decrypt the first encryption information. In order to ensure the security of encryption and decryption keys during key transmission, it is necessary to generate an algorithm for calling the key generation interface. By using the algorithm to call the interface, the operation process of the encryption key does not need to be saved locally, reducing the risk of encrypted information being cracked or leaked. The IoT terminal generates verification information as the second encryption information that is only sent to the user, where the verification code is a string of characters of a fixed length combined with numbers and letters. When the user receives the second encrypted information containing the verification code, the verification code is sent to the IoT terminal to authenticate the user's registration.

[0065] The beneficial effect of the above technical solution is: adopting the solution provided by this embodiment, by collecting data information in the configuration file and encrypting the user registration request information, it is beneficial to improve the information collection efficiency of the Internet of Things mobile terminal.

[0066] Reference Figure 2 In another embodiment, the identity authentication module includes:

[0067] Data preprocessing unit: obtains the configuration file generated by the server to build a data feature model. The IoT terminal obtains the user's behavior data and preprocesses the behavior data; the preprocessing includes smoothing and denoising, outlier processing, and data segmentation operations;

[0068] Feature extraction unit: obtains pre-processed behavior data and builds a user-based convolutional neural network model based on the status information in the configuration file, which is used to extract feature information from user behavior data using a deep neural learning network, and trains the feature information to assist in determining the user's identity;

[0069] User authentication unit: The trained feature information is constructed into a feature set through classification and combination to determine whether the current user is a legitimate user.

[0070] The working principle of the above technical solution is as follows: the solution adopted in this embodiment is that the smart wearable device collects the user's behavioral feature data through the Internet of Things terminal to verify the user's identity. This authentication method is used as an identity authentication method based on biometrics. When the user performs identity authentication, the identity feature is extracted by determining the unique identity identifier to verify the accuracy of the identity. The configuration file generated by the server is obtained as the basis for data construction. When the user uses the smart wearable device, the first sensor of the Internet of Things terminal, namely the data sensor, collects the user's behavior data, wherein the behavior data includes the behavior operation record generated by the user when using the smart wearable device. The behavior data is preprocessed. Since the data sensor is easily interfered by noise during the data collection process and the data may be abnormal, the behavior data is preprocessed including smoothing denoising, outlier processing and data segmentation operations. From the behavior data after preprocessing, the feature data representing the user's identity is extracted using the extraction method of a deep neural network, and the feature data is constructed into a feature set and trained for user identity determination. The trained feature set is used for user identity authentication to determine whether the current user is a legitimate user.

[0071] The beneficial effects of the above technical solution are: by adopting the solution provided in this embodiment, the user's behavior data is extracted, and the characteristic information of the behavior data is extracted based on the user's behavior data, and the legality of the user identity is judged through the characteristic information, thereby improving the efficiency of identity authentication and ensuring the security and confidentiality of the user's identity.

[0072] In another embodiment, the data preprocessing unit includes:

[0073] Data sampling subunit: obtains the user's behavior data, which includes the data signal generated by the user during exercise, and uses a linear interpolation algorithm to uniformly sample the data signal received by the first sensor to obtain a signal sequence;

[0074] Periodic segmentation subunit: obtains the morphological signal in the signal sequence, takes the periodic change of the morphological signal as the morphological period, and evenly segments the morphological signal to obtain sequence units;

[0075] Data denoising subunit: obtains the morphological features in the behavioral data of the sequence unit, uses the built-in filter of the IoT terminal, denoises the sequence unit according to the morphological features, and obtains the user's behavioral data after preprocessing.

[0076] The working principle of the above technical solution is as follows: the solution adopted in this embodiment is that, due to the performance limitation of data collection, the sensor will be interfered by environmental noise and electronic devices during the collection process, so it is necessary to denoise the user's behavior data and segment it, so as to segment the data of the constructed feature data model, obtain input signals of different components, and obtain the best features of the user. The data input signal received by the data sensor is uniformly sampled using a linear interpolation algorithm to obtain a signal sequence, and a morphological signal in the signal sequence is obtained. The morphological signal refers to the walking state of the feet and the swaying state of the head movement collected by the data sensor when the user is in motion when using the smart wearable device. The morphological change process is a periodic movement, and a morphological period is the time from when the user's heel touches the ground to when the heel touches the ground again. The periodic transformation of the morphological signal determines the change of the waveform amplitude, frequency and jitter of the morphological signal.

[0077] The user's morphological signal is obtained, and the morphological signal is segmented according to the morphological period, so as to complete the denoising of the morphological signal through the built-in filter of the IoT terminal. The specific operation is to obtain the morphological signal and evenly segment the morphological signal into sequence units to facilitate the extraction of human morphological features. Since each morphological period contains similar characteristics, the morphological segmentation is based on a fixed length and the morphological period is used as the segmentation unit for segmentation.

[0078] Feature extraction is performed on the feature data in each morphological period. Since inputting the morphological signal into the data feature model will lead to feature fuzziness, the morphological signal is decomposed into different frequency components, and filters are used to randomly filter the noise of different frequency signals to finally obtain the preprocessed user behavior data.

[0079] The beneficial effect of the above technical solution is: by adopting the solution provided by this embodiment, through noise denoising and collecting user behavior data, compared with traditional denoising processing, the morphological segmentation method reduces process loss and improves denoising efficiency.

[0080] In another embodiment, the feature extraction unit comprises:

[0081] Feature extraction subunit: obtains the configuration file in the server, builds a convolutional neural network model to extract the feature information of the status information in the configuration file; extracts the segmentation results of the morphological period in the behavior data to train the convolutional neural network model, and finally extracts the user's feature information;

[0082] Identity recognition subunit: obtain the user's behavior data and build a user classification model to determine whether the user is a legitimate user; divide the behavior data into a training set, a validation set, and a test set according to a preset ratio, iterate the behavior data in the training set, and continuously reduce the loss function of the training set until it no longer changes; input the behavior data in the test set into the user classification model, calculate whether the classification accuracy of the test set meets the preset standard, and if so, apply the trained user classification model to the validation set, and determine the legitimacy of the user's identity based on the feature information.

[0083] The working principle of the above technical solution is as follows: The solution adopted in this embodiment is that the feature extraction model will affect the performance of user authentication to a certain extent, and the accuracy of manual calculation is limited, so a method based on building a deep neural learning network is needed to learn deep features to assist user authentication. User authentication is completed by a method of scene recognition and building a user classification model. Construct a convolutional neural network model, obtain a configuration file, and input the status information in the configuration file, and train the convolutional neural network model according to the segmentation results of the morphological period. The convolutional neural network model consists of three convolutional layers, and each convolutional layer is used for training, convergence and testing behavior data. The size data of the convolution kernel is an important parameter that affects the extraction of data features. For example, more detailed features can be extracted compared to smaller convolution kernels, so it is necessary to reduce the size of the convolution kernel and increase the number of convolution kernels.

[0084] A user classification model is constructed to determine whether a user is a legitimate user. The behavior data is divided into a training set, a validation set, and a test set in a ratio of 10:3:3. During the training process, the behavior data in the training set is input into the user classification model. The number of iterations is set to 15, and the loss function of the training set is calculated. The loss function in the user classification model is continuously reduced through continuous iterations. In a certain number of rounds, if the currently calculated loss function no longer decreases, the learning performance of the user classification model is guaranteed by the learning rate. After the user classification model is trained, the behavior data in the test set is input into the trained user classification model, and the classification accuracy of the test set is calculated to determine whether it meets the preset standard. If the percentage of accuracy meets the preset accuracy standard, the trained user classification model is applied to the validation set to determine the legitimacy of the user identity.

[0085] The beneficial effects of the above technical solution are: by adopting the solution provided in this embodiment, by constructing a convolutional network neural model and a user classification model, extracting the user's characteristic information, combining the sensor collected data with the behavioral data, continuously training the model's test and recognition accuracy, and segmenting the morphological signals, the efficiency of user identity recognition is improved.

[0086] Reference Figure 3In another embodiment, the data acquisition module comprises:

[0087] Model identification unit: builds a device model library to collect the device models currently used by all users. The user sends a request to the IoT terminal to obtain data. The IoT terminal identifies the device model currently used by the user and determines the accuracy of the identification by searching in the device model library.

[0088] Communication establishing unit: if the IoT terminal retrieves the current user's device through the device model library, a communication protocol between the IoT terminal and the second sensor is established through the server for collecting data;

[0089] Connection test unit: Use the communication protocol to perform data collection test on the IoT terminal, read the storage space location after data collection in the second sensor, and if the reading is successful, continue to collect data and store it in the storage space.

[0090] The working principle of the above technical solution is: the solution adopted in this embodiment is that the Internet of Things terminal in the smart wearable device uses the second sensor, namely the motion sensor, to collect data generated by the user when using the smart wearable device. The user initiates a data collection request to the Internet of Things terminal. The Internet of Things terminal identifies the device model of the smart wearable device used by the current user and searches it in the device model library. If the retrieval is successful, the server constructs a communication protocol between the Internet of Things terminal and the motion sensor to enable the two to be directly connected. The communication protocol is used to test the Internet of Things terminal. After the connection is established, the data location collected by the second sensor is read. If the reading is successful, it means that the test is successful.

[0091] The beneficial effect of the above technical solution is: adopting the solution provided by this embodiment, by searching the device model and establishing a communication protocol, it is beneficial to data collection.

[0092] In another embodiment, the data acquisition module further includes:

[0093] Image acquisition unit: The IoT terminal collects image data generated by the user when using the device. The image data is collected by a movable imaging device preset in the IoT terminal;

[0094] Gesture display unit: The device used by the user is provided with different types of interfaces for collecting different types of environmental data and obtaining feature information in the image data. The feature information includes the user's gesture action, and the type of interface to be called is determined according to the gesture action;

[0095] Interface detection unit: When calling any interface, detect the occupancy status of the current interface. If the current interface is occupied, return the real-time environmental data in the interface. If the current interface is not occupied, extract the feature information in the image data to update the environmental data.

[0096] The working principle of the above technical solution is as follows: the solution adopted in this embodiment is that when the user uses the smart wearable device, the Internet of Things terminal collects the image data generated by the user during exercise. The image data is collected by the movable imaging device in the smart wearable device. Different types of interfaces are provided in the smart wearable device, where different types of interfaces call different environmental data respectively. When calling any interface, the occupancy status of the current interface is detected. If the current interface is occupied, the real-time data returned from the interface is extracted. If the current interface is not occupied, the feature information in the image data is extracted. The feature information includes the gesture features generated by the user during exercise, and the interface type is determined based on the gesture features.

[0097] In another embodiment, the gesture display unit further includes:

[0098] Image marking subunit: obtains image data collected by the mobile imaging device, visually marks the image data, uses the user's gesture action in the image data as a first mark, and the image background other than the gesture action as a second mark. If multiple first marks appear in the scene, they can be confirmed by the user;

[0099] The user association subunit sets the association between the user's gesture action and the mobile imaging device as a gesture event. When the user triggers the gesture event, the IoT terminal acquires image data and determines the interface object of the user's gesture action according to the first mark in the image data.

[0100] Scene judgment subunit: The movable imaging device also includes locating the current user's position and measuring the moving distance, and judging the scene information of the current user according to the second mark to assist in collecting environmental data.

[0101] The beneficial effect of the above technical solution is: by adopting the solution provided in this embodiment, the association between the user's gesture and the movable imaging device is set as a gesture event, the image data collected by the movable imaging device is marked and distinguished, and the current user's location is quickly determined, which is conducive to more efficient acquisition of environmental data.

[0102] In another embodiment, a method for collecting IoT data includes:

[0103] S10: The user sends a data transmission request message to the IoT terminal, the IoT terminal identifies the unique identifier in the request message, and the server generates a configuration file for data transmission;

[0104] S20: The IoT terminal detects the configuration file and obtains the user's verification code for verification by generating verification information;

[0105] S30: The first sensor collects the user's behavior data in the configuration file for preprocessing, and the IoT terminal extracts the user's identity for identity authentication;

[0106] S40: Determine whether the user identity is authenticated. If so, perform data collection operations through the user's device, use the second sensor in the device to collect environmental information, and transmit the environmental information to the Internet of Things terminal.

[0107] Reference Figure 4 In another embodiment, step S20 includes:

[0108] S21: The configuration file contains status information of the data, and determines whether the status information is serviceable information that can serve the IoT terminal. If it is determined to be yes, an encryption key generated by the status information is used as the first encryption information to encrypt and transmit the status information;

[0109] S22: Preset information corresponding to the status information in the configuration file is preset in the server, and the server performs key identification on the encryption key to determine whether the encryption key can match the preset information. If the match is successful, the preset information is used to generate a decryption key to decrypt the encryption key in the first encrypted information;

[0110] S23: The IoT terminal generates a verification code as the second encrypted information and sends it to the user. After receiving the second encrypted information including the verification code, the user sends the verification code to the IoT terminal to complete the user information registration in the IoT terminal.

[0111] In another embodiment, step S30 includes:

[0112] S31: Obtain a configuration file generated by the server for building a data feature model, and the IoT terminal obtains the user's behavior data and preprocesses the behavior data; the preprocessing includes smoothing and denoising, outlier processing, and data segmentation operations;

[0113] S32: Obtaining the pre-processed behavior data, and building a user-based convolutional neural network model according to the status information in the configuration file, for extracting feature information from the user behavior data using a deep neural learning network, and training the feature information to assist in determining the user identity;

[0114] S33: The trained feature information is constructed into a feature set through classification and combination, which is used to determine whether the current user is a legitimate user.

[0115] In another embodiment, step S31 includes:

[0116] S311: Acquire user behavior data, where the behavior data includes data signals generated by the user during exercise, and use a linear interpolation algorithm to uniformly sample the data signals received by the first sensor to obtain a signal sequence;

[0117] S312: acquiring a morphological signal in the signal sequence, and using the periodic change of the morphological signal as the morphological period, uniformly dividing the morphological signal to obtain sequence units;

[0118] S313: Acquire morphological features in the behavior data of the sequence unit, use the built-in filter of the IoT terminal to perform denoising on the sequence unit according to the morphological features, and obtain the pre-processed behavior data of the user.

[0119] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An Internet of Things terminal data collection system, characterized in that: include: Information access module: The user sends a data transmission request message to the IoT terminal, the IoT terminal identifies the unique identifier in the request message, and the server generates a configuration file for data transmission; Data encryption module: The IoT terminal detects the configuration file and obtains the user's verification code for verification by generating verification information; Identity authentication module: The first sensor of the IoT terminal collects the user's behavior data in the configuration file for preprocessing, and the IoT terminal extracts the user's identity for identity authentication; Data collection module: determines whether the user identity is authenticated. If so, performs data collection operations through the user's device, uses the second sensor in the user's device to collect environmental information, and transmits the environmental information to the IoT terminal, including: Model identification unit: builds a device model library to collect the device models currently used by all users. The user sends a data transmission request to the IoT terminal, and the IoT terminal identifies the device model currently used by the user. The IoT terminal determines the accuracy of the identification by searching in the device model library; Communication establishing unit: if the IoT terminal retrieves the current user's device through the device model library, a communication protocol between the IoT terminal and the second sensor is established through the server for collecting data; Connection test unit: Use the communication protocol to perform data collection test on the IoT terminal, read the storage space location after data collection in the second sensor, and if the reading is successful, continue to collect data and store it in the storage space; The data acquisition module further includes: Image acquisition unit: The IoT terminal collects image data generated by the user when using the device. The image data is collected by a movable imaging device preset in the IoT terminal; Gesture display unit: Different types of interfaces are set in the device used by the user to collect different types of environmental data and obtain feature information in the image data. The feature information includes the user's gesture action, and the type of interface to be called is determined according to the gesture action; Interface detection unit: when calling any interface, it detects the occupation status of the current interface. If the current interface is occupied, it returns the real-time environment data in the interface. If the current interface is not occupied, it extracts the feature information in the image data to update the environment data. The gesture display unit comprises: Image marking subunit: obtains image data collected by the mobile imaging device, visually marks the image data, uses the user's gesture action in the image data as a first mark, and uses the image background other than the gesture action as a second mark. If multiple first marks appear in the scene, they can be confirmed by the user; The user association subunit sets the association between the user's gesture action and the mobile imaging device as a gesture event. When the user triggers the gesture event, the IoT terminal acquires the image data and determines the interface object of the user's gesture action according to the first mark in the image data. Scene judgment subunit: The movable imaging device also includes locating the current user's position and measuring the moving distance, and judging the scene information of the current user according to the second mark, which is used to assist in collecting environmental data; The identity authentication module comprises: Data preprocessing unit: obtains the configuration file generated by the server to build a data feature model, and the IoT terminal obtains the user's behavior data and preprocesses the behavior data; Feature extraction unit: obtains the pre-processed behavior data, and builds a user-based convolutional neural network model according to the status information of the user device in the configuration file, which is used to extract feature information from the user behavior data using a deep neural learning network, and trains the feature information to assist in determining the user's identity; User authentication unit: The trained feature information is constructed into a feature set through classification and combination to determine whether the current user is a legitimate user.

2. The Internet of Things terminal data acquisition system according to claim 1, characterized in that: The data encryption module comprises: Information judgment unit: The configuration file contains the status information of the user device, and judges whether the status information is serviceable information that can serve the IoT terminal. If it is judged to be yes, an encryption key generated by the status information is used as the first encryption information to encrypt and transmit the status information; Information decryption unit: preset information corresponding to the status information in the configuration file is preset in the server, the server performs key identification on the encryption key, determines whether the encryption key can match the preset information, and if the match is successful, generates a decryption key from the preset information to decrypt the encryption key in the first encrypted information; Information verification unit: The IoT terminal generates a verification code as the second encrypted information and sends it to the user. After the user receives the second encrypted information containing the verification code, the user sends the verification code to the IoT terminal to complete the user's information registration on the IoT terminal.

3. The Internet of Things terminal data acquisition system according to claim 1, characterized in that: The preprocessing includes smoothing and denoising, outlier processing and data segmentation operations.

4. The Internet of Things terminal data acquisition system according to claim 3 is characterized in that: The data preprocessing unit comprises: Data sampling subunit: obtains the user's behavior data, which includes the data signal generated by the user during exercise, and uses a linear interpolation algorithm to uniformly sample the data signal received by the first sensor to obtain a signal sequence; Periodic segmentation subunit: obtains the morphological signal in the signal sequence, takes the periodic change of the morphological signal as the morphological period, and evenly segments the morphological signal to obtain sequence units; Data denoising subunit: obtains the morphological features in the behavioral data of the sequence unit, uses the built-in filter of the IoT terminal, denoises the sequence unit according to the morphological features, and obtains the user's behavioral data after preprocessing.

5. The Internet of Things terminal data acquisition system according to claim 3 is characterized in that: The feature extraction unit comprises: Feature extraction subunit: obtains the configuration file in the server, builds a convolutional neural network model to extract the feature information of the status information in the configuration file; extracts the segmentation results of the morphological period in the behavior data to train the convolutional neural network model, and finally extracts the user's feature information; Identity recognition subunit: obtain the user's behavior data and build a user classification model to determine whether the user is a legitimate user; divide the behavior data into a training set, a validation set, and a test set according to a preset ratio, iterate the behavior data in the training set, and continuously reduce the loss function of the training set until it no longer changes; input the behavior data in the test set into the user classification model, calculate whether the classification accuracy of the test set meets the preset standard, and if so, apply the trained user classification model to the validation set, and combine the feature information to determine the legitimacy of the user's identity.

6. A method for collecting data from an Internet of Things terminal, characterized in that: include: S10: The user sends a data transmission request message to the IoT terminal, the IoT terminal identifies the unique identifier in the request message, and the server generates a configuration file for data transmission; S20: The IoT terminal detects the configuration file and obtains the user's verification code for verification by generating verification information; S30: The first sensor of the IoT terminal collects the behavior data of the user in the configuration file for preprocessing, and the IoT terminal extracts the identity of the user for identity authentication; S40: Determine whether the user identity is authenticated. If so, perform data collection operations through the user's device, use the second sensor in the user's device to collect environmental information, and transmit the environmental information to the Internet of Things terminal, including: The device model library is constructed to collect the device models currently used by all users. The user sends a data transmission request to the IoT terminal, and the IoT terminal identifies the device model currently used by the user. The IoT terminal determines the accuracy of the identification by searching in the device model library; If the IoT terminal retrieves the current user's device through the device model library, a communication protocol between the IoT terminal and the second sensor is established through the server for collecting data; Use the communication protocol to perform data collection test on the IoT terminal, read the storage space location after data collection in the second sensor, and if the reading is successful, continue to collect data and store it in the storage space; The S40 further includes: the IoT terminal collects image data generated by the user when using the device, and the image data is collected by a movable imaging device preset in the IoT terminal; Different types of interfaces are provided in the device used by the user for collecting different types of environmental data, obtaining feature information in the image data, the feature information including the user's gesture action, and determining the type of interface to be called according to the gesture action, including: obtaining image data collected by the mobile imaging device, visually marking the image data, using the user's gesture action in the image data as the first mark, and the image background other than the gesture action as the second mark, and if multiple first marks appear in the scene, the user can confirm; setting the association between the user's gesture action and the mobile imaging device as a gesture event, when the user triggers the gesture event, the Internet of Things terminal obtains the image data, and determines the interface object of the user's gesture action according to the first mark in the image data; the mobile imaging device also includes locating the current user's position and measuring the moving distance, and determining the current user's scene information according to the second mark, which is used to assist in collecting environmental data; When calling any interface, the occupancy status of the current interface is detected. If the current interface is occupied, the real-time environmental data in the interface is returned. If the current interface is not occupied, the feature information in the image data is extracted to update the environmental data. The S30 includes: The configuration file generated by the server is obtained to build a data feature model. The IoT terminal obtains the user's behavior data and pre-processes the behavior data. Obtain the pre-processed behavior data, and build a user-based convolutional neural network model based on the status information of the user device in the configuration file, which is used to extract feature information from the user behavior data using a deep neural learning network, and train the feature information to assist in determining the user's identity; The trained feature information is constructed into a feature set through classification and combination, which is used to determine whether the current user is a legitimate user.

7. The method for collecting data from an Internet of Things terminal according to claim 6, characterized in that: Step S20 includes: S21: The configuration file includes status information of the user device, and it is determined whether the status information is serviceable information that can serve the IoT terminal. If it is determined to be yes, an encryption key generated by the status information is used as the first encryption information to encrypt and transmit the status information; S22: Preset information corresponding to the status information in the configuration file is preset in the server, and the server performs key identification on the encryption key to determine whether the encryption key can match the preset information. If the match is successful, the preset information is used to generate a decryption key to decrypt the encryption key in the first encrypted information; S23: The IoT terminal generates a verification code as the second encrypted information and sends it to the user. After receiving the second encrypted information including the verification code, the user sends the verification code to the IoT terminal to complete the user information registration in the IoT terminal.

Images