Smart home regulation and control method based on Internet of Things

Abstract

The invention relates to the technical field of smart home, and particularly discloses a smart home regulation and control method based on the Internet of Things, and the method comprises the following steps: S1, an environment sensing module collects environment parameters through a distributed sensor network; s2, a user behavior analysis module constructs a user behavior probability model based on historical operation data and real-time action recognition, and outputs a behavior preference weight; s3, the cloud decision engine generates an equipment control instruction set by fusing the environmental parameters and the behavior preference weights, and calculates regulation and control priorities; and S4, the equipment execution network drives the home equipment according to the instruction set, and optimizes the regulation and control parameters through a feedback mechanism. According to the smart home regulation and control method based on the Internet of Things, intelligent and automatic regulation and control of home equipment are achieved by means of the advanced Internet of Things technology and fusion of multiple modules such as environment perception, user behavior analysis, cloud decision and equipment execution, and a comfortable, convenient and safe home environment is created for a user.

Description

Technical Field

[0001] The present invention relates to the technical field of smart home, and in particular to a smart home control method based on the Internet of Things. Background Art

[0002] With the rapid progress of technology and the continuous improvement of people's living standards, the smart home market has shown a rapid development trend. The vision of a comfortable and convenient home life it depicts attracts many consumers. However, behind its booming development, existing problems have gradually emerged, severely restricting the further expansion of the market.

[0003] In the traditional home system, the independent state of each device has become a major obstacle to users' convenient life. Devices such as air conditioners, lighting, and curtains are like individual "islands". Users have to operate them separately and need to perform cumbersome settings on different devices. This not only wastes time and energy but also lacks coordination between devices, making it difficult to form an organic whole and thus unable to build an ideal comfortable and convenient home environment for users.

[0004] Even if some smart home systems have been connected to the network, their degree of intelligence is still unsatisfactory. In terms of data collection, a single sensor has limited functions and cannot comprehensively perceive multiple environmental factors. For example, a temperature sensor can only obtain temperature information and cannot take into account other important indicators such as humidity and air quality; moreover, the data collection frequency is too low to capture the dynamic changes of the environment in a timely manner. In a smart home system, this lag will cause device control to fail to meet actual needs, greatly reducing the user experience.

[0005] In the control decision-making stage, the problems are equally prominent. Due to the lack of efficient data processing and in-depth analysis means, the system is difficult to make scientific and intelligent decisions based on the collected data. In addition, traditional home control systems rely too much on network stability, which brings many hidden dangers in actual use. In remote areas with weak network signals or in urban environments with signal congestion, device control is prone to delays or even out-of-control situations. This will not only affect the normal use of users but also pose a threat to home security.

[0006] The rise of the Internet of Things, embedded, and machine learning technologies provides new possibilities for breaking through these bottlenecks. The present invention aims to use advanced technologies to solve existing problems, promote the smart home market to a new height, and bring users a truly intelligent, convenient, and safe home experience. Summary of the Invention

[0007] The objective of the present invention is to provide an Internet of Things-based intelligent home control method, which solves the problems of traditional intelligent home devices being difficult to collaborate independently, incomplete and lagging data collection, lack of intelligence in control decision-making, and excessive dependence on the network, improves the level of home intelligence, and creates a comfortable, convenient and safe home environment.

[0008] To achieve the above objective, the present invention provides an Internet of Things-based intelligent home control method, including the following steps:

[0009] S1. The environmental perception module collects environmental parameters through a distributed sensor network;

[0010] S2. The user behavior analysis module constructs a user behavior probability model based on historical operation data and real-time action recognition, and outputs the behavior preference weight W(u);

[0011] S3. The cloud decision-making engine generates a device control instruction set C by fusing environmental parameters and behavior preference weights, and calculates the regulation priority;

[0012] S4. The device execution network drives home devices according to the instruction set and optimizes the regulation parameters through a feedback mechanism.

[0013] Preferably, in S1, the environmental parameters include temperature T, humidity H, light intensity L, air quality index A, and sound decibel D. The environmental perception module uses multi-sensor fusion technology to fuse data, and the fusion formula is:

[0014]

[0015] where S is the fused environmental state value, α i is the weight of the i-th type of sensor, S i is the measured value of the i-th type of sensor, β is the noise correction coefficient, and σ is the environmental disturbance factor.

[0016] Preferably, in S1, the sampling frequency range of the environmental perception module is 1 - 60Hz.

[0017] Preferably, in S1, the sensor weight α i is updated through a covariance matrix adaptive strategy:

[0018]

[0019] where S avg is the mean value of the same type of sensor, S max and S min are the maximum and minimum values of the sensor range.

[0020] Preferably, in S1, the collected environmental parameter data uses the AES-256 encryption transmission protocol, and the encryption strength E satisfies:

[0021]

[0022] Among them, B represents the data packet size, K represents the dynamic key length, and M represents the security redundancy coefficient.

[0023] Preferably, in S2, the user behavior analysis module uses the Hidden Markov Model (HMM) to calculate the user behavior probability P(u):

[0024]

[0025] Among them, O t is the observed behavior at time t, s t is the hidden state at time t, is the observation probability, and ψ is the state transition function.

[0026] Preferably, in S3, the device control instruction set C = f(T, H, L, A, D, W), and the regulation priority Q is calculated through the following optimization equation:

[0027]

[0028] Among them, k1, k2, and k3 are environmental parameter normalization coefficients, and k1 + k2 + k3 = 1.

[0029] Preferably, in S3, the normalization coefficients k1, k2, and k3 are dynamically adjusted through the user physiological comfort model:

[0030]

[0031] Among them, θ i represents the user personalized preference parameter, and C p represents the real-time physiological index.

[0032] Preferably, in S4, the device execution network uses a fuzzy PID controller to optimize the regulation parameters, and the response time R satisfies:

[0033]

[0034] Among them, represents the device basic response threshold, and N represents the current number of online devices.

[0035] Therefore, the present invention adopts the above-mentioned intelligent home control method based on the Internet of Things, and the beneficial effects are as follows:

[0036] (1) Through the comprehensive analysis of environmental parameters and user behavior, the system of the present invention can automatically and intelligently control household devices. Without manual operation by the user, the system can adjust the air conditioner temperature, turn on and off the lights, etc. in advance according to environmental changes and user habits, greatly improving the convenience of life and allowing users to enjoy the comfortable experience brought by smart home.

[0037] (2) The multi-source sensors of the present invention collect rich environmental parameters and process them through advanced fusion technology. Combined with the scientific calculation of control priorities, the system can accurately control household devices and maintain the indoor environment in the best state. For example, it can accurately adjust the working mode of the air purifier according to the indoor air quality index to provide a healthy living environment for users.

[0038] (3) The present invention uses the AES-256 encryption transmission protocol to ensure the secure transmission of data and reduce the risk of data being stolen or tampered with. At the same time, the device execution network optimizes the control parameters through a fuzzy PID controller and adjusts the response time according to the number of online devices, enhancing the stability and reliability of the system in different network environments and device load conditions and ensuring the stable operation of household devices.

[0039] The technical solution of the present invention will be further described in detail below through the accompanying drawings and embodiments. Description of the Drawings

[0040] Figure 1 is the overall process schematic diagram of an embodiment of a smart home control method based on the Internet of Things of the present invention;

[0041] Figure 2 is the schematic diagram of the improved CNN model structure of an embodiment of a smart home control method based on the Internet of Things of the present invention;

[0042] Figure 3 is the schematic diagram of the device execution network implementation of an embodiment of a smart home control method based on the Internet of Things of the present invention. Detailed Embodiments

[0043] The technical solution of the present invention will be further described below through the accompanying drawings and embodiments.

[0044] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meaning as understood by those of ordinary skill in the field to which the present invention belongs.

[0045] A smart home control method based on the Internet of Things includes the following steps:

[0046] S1. The environmental perception module collects environmental parameters through a distributed sensor network;

[0047] The environmental parameters include temperature T, humidity H, light intensity L, air quality index A, and sound decibel D. The environmental perception module uses multi-sensor fusion technology to fuse data, and the fusion formula is:

[0048]

[0049] where S is the fused environmental state value, α i is the weight of the i-th type of sensor, S i is the measured value of the i-th type of sensor, β is the noise correction coefficient, and σ is the environmental disturbance factor.

[0050] The noise correction coefficient β is used to eliminate the random noise interference that may be received during the sensor measurement process, including the thermal noise of electronic components, external electromagnetic interference, etc. The environmental disturbance factor σ takes into account the sudden and short-term interference factors in the environment, such as the airflow change caused by suddenly opening the doors and windows, and the short-term strong light irradiation.

[0051] The sampling frequency range of the environmental perception module is 1 - 60 Hz. Different environmental parameters can be set with different sampling frequencies according to their changing speed characteristics. The changes in temperature and humidity in the environmental parameters are relatively slow, and a lower sampling frequency can be set, such as 1 - 5 Hz; while the light intensity and sound decibel may change rapidly, and a higher sampling frequency can be set, such as 10 - 60 Hz. This can not only ensure the accuracy of the data but also avoid data redundancy and resource waste caused by over-sampling.

[0052] The sensor weight α i is updated through the covariance matrix adaptive strategy, and the formula is:

[0053]

[0054] where S avg is the mean value of the same type of sensors, S max and S min are the maximum and minimum values of the sensor range. As time goes by and the environment changes, the performance of the sensors may drift. Through the covariance matrix adaptive strategy, the weights of the sensors can be dynamically adjusted to make the fused data more accurate and reliable. When the measured value of a certain temperature sensor deviates greatly from the mean value of the same type of sensors, its weight will be correspondingly reduced to reduce the influence of this sensor on the fusion result.

[0055] At the same time, the collected environmental parameter data uses the AES-256 encryption transmission protocol, and the encryption strength E satisfies:

[0056]

[0057] Among them, B represents the data packet size, K represents the dynamic key length, and M represents the security redundancy factor. AES-256 is an advanced encryption standard with high security. The dynamic key length K is dynamically adjusted according to the security requirements of data transmission, and the security redundancy factor M further enhances the reliability of encryption, preventing data from being stolen or tampered with during transmission.

[0058] S2. The user behavior analysis module constructs a user behavior probability model based on historical operation data and real-time action recognition, and outputs the behavior preference weight W(u).

[0059] Collect the operation data of users on various household devices at home in the past year. This data includes operation time and operation type (such as turning on / off the light, adjusting the air conditioner temperature, controlling the opening and closing of the curtain, etc.). The operation time reflects the user's daily routine, and the operation type reflects the user's usage habits of different devices. At the same time, install action recognition cameras in the main activity areas such as the living room and bedroom to capture the user's actions in real time. The action recognition camera can adopt a convolutional neural network (CNN) improved by deep learning algorithms to accurately recognize the user's actions, such as recognizing the user's gestures, postures, etc., so as to more comprehensively understand the user's intentions.

[0060] The overall structure of the improved CNN model is as follows:

[0061] Input layer:

[0062] The action recognition camera in the smart home captures video frame image data containing the user's actions. Considering the large amount of video data in the actual scenario, in order to reduce the subsequent calculation burden, the input layer first preprocesses the original video frames. The downsampling technology is adopted to reduce the size of the high-resolution video frames according to a certain ratio. For example, the image with a resolution of 1920×1080 is downsampled to a resolution of 224×224. This can not only retain the key features of the actions but also significantly reduce the amount of data.

[0063] The image data is normalized, and the pixel value range is uniformly mapped to the [0,1] interval, making the model easier to converge during training and improving the training efficiency.

[0064] Convolutional layer group:

[0065] Parallel convolutional layers with different sizes of convolutional kernels are introduced to capture multi-scale features, and convolutional kernels of 3×3, 5×5, and 7×7 are used at the same time. The 3×3 convolutional kernel can capture local detail features, the 5×5 convolutional kernel can obtain features in a slightly larger range, and the 7×7 convolutional kernel can capture context information in a larger range. The output results of these parallel convolutional layers are concatenated in the channel dimension to enrich the feature expression.

[0066] To alleviate the vanishing gradient problem and enhance the network's learning ability for deep features, residual connections are added between convolutional layers. Taking two consecutive convolutional layers as an example, the input is processed by the first convolutional layer and the ReLU activation function, then added to the output of the second convolutional layer, and then input into the next convolutional layer. This structure enables the network to learn complex action features more effectively and makes the training process more stable.

[0067] The dilated convolution technique is adopted to expand the receptive field of the convolutional kernel without increasing the number of parameters and computational complexity. For a specific convolutional layer, an appropriate dilation rate is set. For example, when the dilation rate is 2, the actual effective range of the convolutional kernel is equivalent to that of a 5×5 convolutional kernel; when the dilation rate is 3, the actual effective range of the convolutional kernel is equivalent to that of a 7×7 convolutional kernel. This enhances the model's ability to capture distant features and comprehensively understand user actions.

[0068] Pooling layer:

[0069] After the convolutional layer group, an improved deformable pooling layer is used to replace the traditional max pooling or average pooling. The deformable pooling layer can adaptively adjust the pooling region according to the distribution of the input features, better retaining the key features related to user actions. For example, when recognizing a waving action, the deformable pooling can more accurately locate the hand region and perform the pooling operation, avoiding the loss of important information due to a fixed pooling region.

[0070] The output dimension of the pooling layer is reasonably adjusted according to the model complexity and computing resources. For example, the size of the feature map is reduced by half while keeping the number of channels unchanged to balance the computational complexity and the retention of feature information.

[0071] Attention mechanism module:

[0072] A channel attention mechanism (such as the Squeeze-and-Excitation module) and a spatial attention mechanism (such as the SpatialAttentionModule) are introduced after the pooling layer. The channel attention mechanism obtains the global features of each channel through global average pooling, then calculates the channel weights through a fully connected layer and an activation function, and re-weights the features of different channels to highlight the important channels related to action recognition. The spatial attention mechanism generates a spatial attention map through convolutional operations, weights different spatial positions of the feature map, and focuses on the key regions where user actions occur. The present invention combines the two to enable the model to more effectively focus on the key features of actions and improve the recognition accuracy.

[0073] Fully connected layer:

[0074] Unfold the feature map processed by the attention mechanism into a one-dimensional vector and input it into the fully connected layer. To prevent overfitting, add a Dropout layer between the fully connected layers, randomly discarding a certain proportion (such as 0.5) of neuron connections to enhance the generalization ability of the model.

[0075] Adjust the number of neurons in the fully connected layer and optimize it according to the actual requirements and the size of the dataset. For the smart home action recognition task, the number of neurons can be appropriately reduced to reduce the computational complexity while ensuring the model performance, so as to adapt to the limited computing resources of smart home devices.

[0076] Output layer:

[0077] The output layer uses a Softmax classifier to output the probability distribution of various user actions. Suppose the model needs to recognize 10 common user actions. The output layer will have 10 neurons, each neuron corresponding to an action category, and output the predicted probability of this category of action. The category with the highest probability value is the user action recognized by the model.

[0078] The user behavior analysis module uses the Hidden Markov Model (HMM) to calculate the user behavior probability P(u):

[0079]

[0080] where O t is the observed behavior at time t, s t is the hidden state at time t, is the observation probability, and ψ is the state transition function.

[0081] The Hidden Markov Model is suitable for processing data with sequence characteristics. By taking historical operation data and real-time action recognition data as inputs, the Hidden Markov Model is trained to learn the patterns and rules of user behavior. The model can learn that users usually turn off the living room lights and turn on the bedroom lights at a specific time in the evening, so as to predict the behavior probability of users at the same time in the future.

[0082] The trained Hidden Markov Model can accurately output the behavior preference weights of users' operations on various devices in different scenarios, which are used to reflect the likelihood of users' operations on different devices and provide an important basis for subsequent cloud decisions.

[0083] S3. The cloud decision-making engine generates a device control instruction set C by fusing environmental parameters and behavior preference weights and calculates the regulation priority;

[0084] The cloud decision-making engine receives in real time the environmental parameters uploaded by the environmental perception module and the behavior preference weights output by the user behavior analysis module, and uses a preset algorithm to fuse the two, and generates a device control instruction set according to the fusion result. When the environmental perception module detects that the indoor temperature rises and the user behavior preference weights show that the user often lowers the air conditioner temperature in similar situations, a corresponding air conditioner temperature adjustment instruction is generated. At the same time, if the light intensity is too strong and the user has often drawn the curtains in this situation in the past, then an instruction to control the curtains to close will be generated.

[0085] The device control instruction set C = f(T, H, L, A, D, W), and the regulation priority Q is calculated by the following optimization equation:

[0086]

[0087] where k1, k2, k3 are environmental parameter normalization coefficients, and k1 + k2 + k3 = 1.

[0088] The normalization coefficients k1, k2, k3 are dynamically adjusted through the user physiological comfort model:

[0089]

[0090] where θ i represents the user's personalized preference parameter, and C p represents the real-time physiological index. The user's personalized preference parameter can be set by the user in the system, such as the user's preference range for temperature and humidity. The real-time physiological index can be obtained through wearable devices (such as smart bracelets, smart watches, etc.), such as the user's heart rate, body temperature, etc. By comprehensively considering the environmental parameters and the user behavior preference weights, the regulation priority is calculated, and the device regulation instructions that have a greater impact on the user's comfort are processed first to ensure that the system can respond quickly and meet the user's needs.

[0091] S4. The device execution network drives the home devices according to the instruction set and optimizes the regulation parameters through a feedback mechanism.

[0092] The device execution network receives the instruction set sent by the cloud decision-making engine and drives the corresponding home devices to perform operations. For example, controlling the air conditioner to adjust the temperature, controlling the light switch or adjusting the brightness, controlling the curtains to open and close, etc. The device execution network can communicate with the home devices through wireless communication protocols (such as Wi-Fi, Bluetooth, ZigBee, etc.) to ensure that the instructions can be transmitted to the device side accurately.

[0093] After the device performs the operation, the device operation status data is collected in real time through feedback sensors (such as the temperature feedback sensor of the air conditioner, the brightness feedback sensor of the light, the position feedback sensor of the curtains, etc.) and fed back to the device execution network.

[0094] The device execution network uses a fuzzy PID controller to optimize the control parameters, adjusts the device response time according to the current number of online devices, ensures stable and efficient operation of the device, and the corresponding time R satisfies:

[0095]

[0096] Among them, represents the basic response threshold of the device, and N represents the current number of online devices. When the number of online devices is large, the device response time is appropriately extended to avoid failure of instruction execution caused by network congestion. The fuzzy PID controller combines the advantages of fuzzy control and PID control, can dynamically adjust the control parameters according to the actual operation of the device, and improves the control accuracy and stability of the system.

[0097] Embodiment

[0098] Deployment of the environmental perception module and data collection:

[0099] In the living room, install the temperature sensor on the wall at least 1.5 meters away from the air-conditioning outlet and the window. Since the temperature at the air-conditioning outlet is affected by air-conditioning cooling or heating, and the temperature near the window is affected by the outdoor temperature, installing the temperature sensor in this position can ensure that the measured data is not disturbed by these factors and more accurately reflects the actual temperature of the living room.

[0100] Install the humidity sensor in a corner far from the water source. For example, the water vapor in the bathroom will directly affect the measurement result of the humidity sensor, and keeping away from the water source can avoid this influence.

[0101] Install the light sensor on the edge of the window glass, facing the direction of the main outdoor light source, so that the change of outdoor light intensity can be accurately captured.

[0102] Install the air quality sensor at a height of 1.5 - 2 meters from the ground and in a well-ventilated position, such as the central area of the living room. This height and position can better reflect the overall indoor air quality.

[0103] Install the sound decibel sensor in a place far from noise sources such as electrical appliances and doors and windows, such as the corner of the living room, to ensure accurate collection of environmental noise.

[0104] After all types of sensors are installed, use professional calibration equipment for calibration. The calibration process includes comparing the sensors with standard measurement devices and adjusting the measurement values of the sensors to make them consistent with the standard values. Set the temperature and humidity sensors to collect data every 5 minutes, and the light, air quality, and sound decibel sensors to collect data every 1 minute. The sensors transmit the data to the receiving end of the environmental perception module through wireless communication technologies such as Wi-Fi, Bluetooth, or ZigBee. During the transmission process, use the AES-256 encryption transmission protocol to encrypt the data to ensure data security.

[0105] Construction of the user behavior analysis module:

[0106] Collect the operation data of users on various home appliances at home in the past year. This data can be obtained from the log files of the smart home system. Clean and preprocess the operation data to remove outliers and error data.

[0107] At the same time, install action recognition cameras in the main activity areas such as the living room and bedroom to capture the actions of users in real time. The action recognition cameras can use high-definition cameras to improve the accuracy of action recognition. Use the Python language combined with machine learning libraries (such as Scikit-learn, TensorFlow, etc.) to build a Hidden Markov Model (HMM). Take the historical operation data and real-time action recognition data as inputs, and train the model to build a user behavior probability model.

[0108] During the training process, use the cross-validation method to evaluate and optimize the model, and continuously adjust the parameters of the model to enable it to accurately output the behavior preference weights of users for operating various devices in different scenarios.

[0109] Operation of the cloud decision-making engine:

[0110] The cloud decision-making engine receives the environmental parameters uploaded by the environmental perception module and the behavior preference weights output by the user behavior analysis module in real time, uses a preset algorithm to fuse the two, and generates a device control instruction set according to the fusion result.

[0111] When the environmental perception module detects that the indoor temperature rises and the user behavior preference weights show that the user often lowers the air conditioner temperature in similar situations, generate the corresponding air conditioner temperature adjustment instruction. Calculate the regulation priority through an optimization equation, and give priority to processing the device regulation instructions corresponding to the changes in environmental parameters such as temperature and air quality that have a greater impact on user comfort. The cloud decision-making engine adopts a distributed computing architecture to improve the speed and efficiency of data processing and ensure that it can respond to environmental changes and user needs in a timely manner.

[0112] Implementation of the device execution network:

[0113] The device executes the instruction set sent by the cloud decision engine through the network, driving the corresponding home devices to perform operations, including controlling the air conditioner to adjust the temperature, controlling the light switch or adjusting the brightness, etc. After the device performs the operation, the operation state data of the device is collected in real time through feedback sensors, including the temperature feedback sensor of the air conditioner, the brightness feedback sensor of the light, etc., and fed back to the device execution network.

[0114] The device execution network uses a fuzzy PID controller to optimize the control parameters, adjusts the device response time according to the current number of online devices, and ensures the stable and efficient operation of the device. For example, when the number of online devices is large, the device response time is appropriately extended to avoid the failure of instruction execution caused by network congestion; the device execution network can adopt multi-threading technology to process the instructions of multiple devices simultaneously, improving the concurrent processing ability of the system.

[0115] Therefore, the present invention adopts the above-mentioned smart home control method based on the Internet of Things. Through comprehensive and detailed environmental perception, accurate user behavior analysis, intelligent cloud decision-making, and efficient device execution, it can realize the intelligent and automated control of home devices, providing users with a more comfortable, convenient, and safe home life experience.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A smart home control method based on the Internet of Things, characterized in that, It includes the following steps: S1. The environmental perception module collects environmental parameters through a distributed sensor network; S2. The user behavior analysis module constructs a user behavior probability model based on historical operation data and real-time action recognition, and outputs the behavior preference weight W(u); S3. The cloud decision engine generates a device control instruction set C by fusing environmental parameters and behavior preference weights, and calculates the regulation priority; S4. The device execution network drives home appliances according to the instruction set and optimizes the regulation parameters through a feedback mechanism.

2. The smart home control method based on the Internet of Things according to claim 1, wherein In S1, the environmental parameters include temperature T, humidity H, light intensity L, air quality index A, and sound decibel D. The environmental perception module uses multi-sensor fusion technology to fuse data, and the fusion formula is: Among them, S is the fused environmental state value, and α i is the weight of the i-th type of sensor, and S i is the measured value of the i-th type of sensor, β is the noise correction coefficient, and σ is the environmental disturbance factor.

3. The smart home control method based on the Internet of Things according to claim 2, wherein In S1, the sampling frequency range of the environmental perception module is 1 - 60Hz.

4. A smart home control method based on the Internet of Things according to claim 3, characterized in that In S1, the sensor weight α i is updated by the covariance matrix adaptation strategy: Among them, S avg is the average value of sensors of the same type, S max , S min are the maximum and minimum values of the sensor range.

5. The smart home control method based on the Internet of Things according to claim 4, characterized in that, In S1, the collected environmental parameter data uses the AES-256 encryption transmission protocol, and the encryption strength E satisfies: where B represents the data packet size, K represents the dynamic key length, and M represents the security redundancy factor.

6. The smart home control method based on the Internet of Things according to claim 5, characterized in that, In S2, the user behavior analysis module uses the Hidden Markov Model HMM to calculate the user behavior probability P(u): Among them, O t is the observation behavior at time t, s t is the hidden state at time t, is the observation probability, and ψ is the state transition function.

7. The smart home control method based on the Internet of Things according to claim 6, wherein, In S3, the device control instruction set C = f(T, H, L, A, D, W), and the regulation priority Q is calculated through the following optimization equation: where k1, k2, and k3 are environmental parameter normalization coefficients, and k1 + k2 + k3 = 1.

8. The smart home control method based on the Internet of Things according to claim 7, characterized in that, In S3, the normalization coefficients K1, K2, and K3 are dynamically adjusted through the user physiological comfort model: Among them, θ i represents the user's personalized preference parameter, and C p represents the real-time physiological index.

9. The smart home control method based on the Internet of Things according to claim 8, characterized in that, In S4, the device execution network uses a fuzzy PID controller to optimize the regulation parameters, and the response time R satisfies: Among them, represents the response threshold of the equipment foundation, and N represents the current number of online devices.

Images