Smart home interaction method and system based on non-intrusive electroencephalogram / brain magnetic signal identity authentication and storage medium

Abstract

The invention discloses a smart home interaction method and system based on non-intrusive electroencephalogram / brain magnetic signal identity authentication and a storage medium, and relates to the technical field of smart home. Constructing a smart home integrated interaction model; the method comprises the following steps: acquiring a control instruction signal of a current user in a multi-modal fusion non-invasive mode, converting the control instruction signal into a digital signal, extracting electroencephalogram / electroencephalogram signal features, extracting an electroencephalogram / electroencephalogram signal feature template of an authorized user from an electroencephalogram / electroencephalogram signal database, and comparing the electroencephalogram / electroencephalogram signal features with the electroencephalogram / electroencephalogram signal feature template; if the comparison result is higher than a preset threshold value, marking that the identity authentication of the current user is passed, receiving electroencephalogram / brain magnetic signal characteristics of the current user by a brain control instruction structure, and converting the electroencephalogram / brain magnetic signal characteristics into a smart home interaction instruction; and after receiving the instruction, the smart home integrated interaction model controls the smart home to start and stop, and feeds back an operation result to the user to complete a brain control operation process. According to the invention, safe, convenient and accurate non-invasive brain control of the smart home is realized.

Description

Technical Field

[0001] This invention relates to the field of smart home technology, and more specifically to a smart home interaction method, system, and storage medium based on non-invasive EEG / MEG signal authentication. Background Technology

[0002] Biometric-based control technologies are gradually becoming a research hotspot, among which EEG / MEG signal interaction technology has attracted much attention due to its unique advantages. EEG / MEG signals are electromagnetic signals generated by the activity of neurons in the brain, which can directly reflect human consciousness. By analyzing EEG / MEG signals, "thought control" without physical movements can be achieved, providing a brand-new interaction method for smart home control.

[0003] However, non-invasive EEG / MEG smart home interaction still faces many challenges in practical applications, among which identity authentication is particularly critical. On the one hand, there are significant individual differences in EEG / MEG signals among different individuals. The same command will exhibit different EEG / MEG signal characteristics on different people. Without effective identity authentication, it may lead to misidentification of control commands, affecting the accuracy of control. On the other hand, EEG / MEG signals contain personal physiological and consciousness information, which are sensitive and private data. Once illegally obtained or misused, they may cause security risks. For example, unauthorized personnel could control smart home devices by forging EEG / MEG signals, threatening users' property and personal safety.

[0004] Furthermore, non-invasively acquired EEG / MEG signals typically have a low signal-to-noise ratio and are susceptible to physiological interference such as blinking and electromyography (EMG), as well as environmental electromagnetic interference. This poses challenges to identity authentication and command recognition based on EEG / MEG signals. How to accurately extract individual EEG / MEG signal features in complex and interfering environments to achieve reliable identity authentication, and how to combine this with effective command parsing algorithms to ensure the security and accuracy of smart home interactions, has become a pressing issue that needs to be addressed. Summary of the Invention

[0005] In view of this, the present invention provides a smart home interaction method, system and storage medium based on non-invasive EEG / MEG signal authentication, which realizes safe, convenient and accurate non-invasive brain control of smart homes, so as to solve the problems existing in the background technology.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A smart home interaction method based on non-invasive EEG / MEG signal authentication includes the following steps:

[0008] Construct an integrated interactive model for smart homes, including a database of authorized users' EEG / MEG signals, a smart home network topology, and a brain-controlled command parsing structure;

[0009] The system uses a non-invasive method of multimodal fusion to collect the current user's control command signals, converts them into digital signals to extract EEG / MEG signal features, extracts the authorized user's EEG / MEG signal feature templates from the EEG / MEG signal database, and compares the EEG / MEG signal features with the EEG / MEG signal feature templates.

[0010] If the comparison result is higher than the preset threshold, it is recorded as the current user's identity authentication is successful. The brain control command structure receives the current user's EEG / MEG signal characteristics, identifies the EEG / MEG signal patterns in the current user's brain related to smart home interaction actions based on the trained neural network model, and converts them into smart home interaction commands.

[0011] After receiving the instruction, the smart home integrated interaction model transmits the instruction to the smart home network topology, controls the start and stop of the smart home, and feeds back the operation result to the user to complete the brain-controlled operation process.

[0012] Optional methods for constructing smart home network topologies are as follows:

[0013] Collect business information and switch control information of smart home devices in the current home, where business information refers to the functions of smart home devices;

[0014] Using smart home devices as nodes, and each node's business information and switch control information as its initial features, a graph structure is constructed.

[0015] Based on the business interaction relationships between devices, data transmission paths, and physical location correlations, the connection edges between nodes are determined, and the current smart home network topology is constructed.

[0016] Optionally, it also includes real-time monitoring of device status changes, including the addition of new devices, the removal of old devices, device function upgrades, or device malfunctions; when a new device is connected, its business information and switch control information are automatically collected to determine the connection relationship between the new device and existing nodes and update the graph structure; if a device is removed or malfunctions, the corresponding node and related connection edges are deleted in a timely manner to ensure that the topology structure can accurately reflect the actual situation of the current smart home system.

[0017] Optionally, the training process for the neural network model is as follows:

[0018] A generative adversarial neural network architecture was constructed, which takes labeled EEG / MEG signals and smart home start / stop control signals as inputs, with different labels corresponding to different smart home start / stop signals;

[0019] Using EEG / MEG signals as the source domain and smart home start / stop interaction signals as the target domain, an adversarial neural network is used to migrate the source domain to the target domain until the mapping relationship between EEG / MEG signals and smart home start / stop control signals meets a preset threshold.

[0020] Optionally, non-invasive methods using multimodal fusion can be used to acquire the current user's EEG / MEG signals, specifically including:

[0021] Millimeter-scale radar technology and quantum sensors are used to collect the current user's EEG / MEG signals, and the collected EEG and MEG signals are fused to obtain the current user's control command signals.

[0022] Optionally, multimodal fusion can use the Kalman filter algorithm to perform spatiotemporal alignment and state estimation of the magnetoencephalogram (MEG) signal and electroencephalogram (EEG), then extract multimodal features through a convolutional neural network in deep learning, and use an attention mechanism to weightedly fuse and generate control command signals.

[0023] A smart home interaction system based on non-invasive EEG / MEG signal authentication includes:

[0024] Smart Home Integration and Interaction Model Construction Module: Used to construct a smart home integration and interaction model, including an authorized user's EEG / MEG signal database, a smart home network topology, and a brain control command parsing structure;

[0025] Brain-controlled signal identity authentication module: Used to collect the current user's EEG / MEG signals in a non-invasive way using multimodal fusion, convert them into digital signals to extract EEG / MEG signal features, extract the authorized user's EEG / MEG signal feature template from the EEG / MEG signal database, and compare the EEG / MEG signal features with the EEG / MEG signal feature template;

[0026] Smart home interaction command generation module: If the comparison result is higher than the preset threshold, it is recorded as the current user's identity authentication is successful. The brain control command structure receives the current user's EEG / MEG signal characteristics, identifies the EEG / MEG signal patterns in the current user's brain related to smart home interaction actions based on the trained neural network model, and converts them into smart home interaction commands.

[0027] Smart Home Start / Stop Interaction Module: After receiving instructions from the smart home integrated control model, the module transmits the instructions to the smart home network topology, controls the start and stop of the smart home, and feeds back the operation results to the user to complete the brain-controlled operation process.

[0028] A computer storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of any one of the following smart home interaction methods based on non-invasive EEG / MEG signal authentication.

[0029] As can be seen from the above technical solutions, compared with the prior art, the present invention discloses a smart home interaction method, system, and storage medium based on non-invasive EEG / MEG signal authentication, which has the following beneficial effects:

[0030] 1. Enhanced security and privacy: Identity authentication is achieved through comparison of EEG / MEG signal feature templates, deeply binding control permissions with user biometrics. This prevents unauthorized personnel from illegally controlling devices by imitating operations or cracking commands, solving the problems of easy leakage and forgery of traditional password and voice authentication methods. It is especially suitable for privacy and security sensitive scenarios in the home environment.

[0031] 2. Achieve contactless and natural interaction: Adopting a non-invasive EEG / MEG signal acquisition method with multimodal fusion, it eliminates the need for users to wear electrode caps, thus getting rid of the limitations of physical contact or manual operation. It is more user-friendly for people with mobility impairments (such as the elderly and disabled), which is in line with the development trend of "seamless interaction" in smart homes and improves ease of use.

[0032] 3. Enhance the accuracy and robustness of command recognition: Combine neural network models to recognize EEG / MEG signal patterns. Through training and optimization, it can accurately match EEG / MEG signal features related to control actions. At the same time, multimodal signal fusion reduces noise interference from single acquisition methods, making command parsing more stable and reducing the probability of misoperation.

[0033] 4. Strong system compatibility and scalability: By constructing a smart home network topology, brain control commands can be adapted to different brands and types of devices without the need for large-scale modifications to the existing home system; the modular design of the integrated control model (separation of database, topology, and parsing structure) facilitates the addition of new devices or expansion of control functions in the future.

[0034] 5. Feedback mechanism enhances user experience: Real-time feedback on operation results allows users to promptly confirm the execution status of instructions, forming a closed loop of "intent-instruction-feedback" and reducing operational uncertainty; In long-term use, the model can be continuously optimized based on feedback data, gradually adapting to changes in individual user EEG characteristics and improving the smoothness of interaction. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0036] Figure 1This is a schematic diagram of the method flow provided by the present invention;

[0037] Figure 2 This is a schematic diagram of the system structure provided by the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Example 1

[0040] This invention discloses a smart home interaction method based on non-invasive EEG / MEG signal authentication, such as... Figure 1 As shown, it includes the following steps:

[0041] Step 1: Construct a smart home integrated interaction model, including an authorized user's EEG / MEG signal database, a smart home network topology, and a brain control command parsing structure;

[0042] Step 2: Collect the current user's control command signals using a non-invasive method of multimodal fusion, convert them into digital signals to extract EEG / MEG signal features, extract the authorized user's EEG / MEG signal feature template from the EEG / MEG signal database, and compare the EEG / MEG signal features with the EEG / MEG signal feature template.

[0043] Step 3: If the comparison result is higher than the preset threshold, it is recorded as the current user's identity authentication is successful. The brain control command structure receives the current user's EEG / MEG signal characteristics, identifies the EEG / MEG signal patterns in the current user's brain related to smart home interaction actions based on the trained neural network model, and converts them into smart home interaction commands.

[0044] Step 4: After receiving the instruction, the smart home integrated interaction model transmits the instruction to the smart home network topology, controls the start and stop of the smart home, and feeds back the operation result to the user to complete the brain control operation process.

[0045] Furthermore, in step one, the smart home network topology is constructed as follows:

[0046] Collect business information and switch control information of smart home devices in the current home, where business information refers to the functions of smart home devices;

[0047] Using smart home devices as nodes, and each node's business information and switch control information as its initial features, a graph structure is constructed.

[0048] Based on the business interaction relationships between devices, data transmission paths, and physical location correlations, the connection edges between nodes are determined, and the current smart home network topology is constructed.

[0049] Furthermore, in the above technical solution, the connection edges between nodes can be determined based on the business interaction relationships between devices (such as the linkage control between smart door locks and indoor lighting), data transmission paths (such as sensors sending monitoring data to controllers), and physical location correlations (such as devices in the same room). Simultaneously, each edge is assigned a weight, which can be calculated comprehensively based on factors such as interaction frequency, data transmission volume, and response timeliness requirements, thereby more accurately reflecting the degree of connection between devices.

[0050] Furthermore, in step one, the current user's EEG / MEG signals are collected using a non-invasive method of multimodal fusion. Specifically, this includes: collecting the current user's EEG / MEG signals using millimeter radar technology and quantum sensors, and fusing the collected EEG and MEG signals to obtain the current user's control command signals.

[0051] Furthermore, multimodal fusion can use the Kalman filter algorithm to perform spatiotemporal alignment and state estimation of EEG and MEG signals, then extract multimodal features through convolutional neural networks in deep learning, and use attention mechanisms to weightedly fuse and generate control command signals.

[0052] A dynamic update mechanism is introduced to monitor device status changes in real time, including the addition of new devices, the removal of old devices, device function upgrades, or malfunctions. When a new device is connected, its business information and switch control information are automatically collected to determine its connection relationship with existing nodes and update the graph structure. If a device is removed or malfunctions, the corresponding node and related connection edges are deleted in a timely manner to ensure that the topology accurately reflects the actual situation of the current smart home system.

[0053] A hierarchical topology is constructed, categorized according to the functional type and hierarchical relationship of the devices. For example, it can be divided into a sensing layer (various sensors), a control layer (intelligent controllers), an execution layer (smart switches, home appliances, etc.), and an application layer (user terminal APP). Different layers interact with each other through specific communication protocols and interfaces, clarifying the responsibilities of each layer and the data flow, making the topology more logical and manageable.

[0054] By adding security protection nodes and links, security components such as firewalls and intrusion detection devices are integrated into the topology as special nodes. These nodes establish secure connection edges with other device nodes and are responsible for monitoring, filtering, and encrypting communication data between devices to prevent unauthorized access and data leakage, thus ensuring the security of the entire smart home network.

[0055] Conduct topology optimization and regularly analyze topology performance metrics such as network latency, data transmission rate, and node load balancing. Based on the analysis results, adjust node connection methods, edge weights, and hierarchical structures. For example, when a control node is overloaded, the load can be balanced by reallocating control permissions for some execution devices, thereby improving the overall efficiency and stability of the smart home network.

[0056] Furthermore, in step three, the training process of the neural network model is as follows:

[0057] A generative adversarial neural network architecture was constructed, which takes labeled EEG / MEG signals and smart home start / stop interaction signals as inputs, with different labels corresponding to different smart home start / stop operations;

[0058] Using EEG / MEG signals as the source domain and smart home start / stop interaction signals as the target domain, an adversarial neural network is used to migrate the source domain to the target domain until the mapping relationship between EEG / MEG signals and smart home start / stop interaction signals meets a preset threshold.

[0059] and Figure 1 Corresponding to the method shown, this invention also discloses a smart home interaction system based on non-invasive EEG / MEG signal authentication, used for... Figure 1 The implementation of the method, specifically its structure, is as follows: Figure 2 As shown, it includes:

[0060] Smart Home Integration and Interaction Model Construction Module: Used to build a smart home integration and control interaction model, including an authorized user's EEG / MEG signal database, a smart home network topology, and a brain control command parsing structure;

[0061] Brain-controlled signal authentication module: Used to collect the current user's control command signals in a non-invasive way using multimodal fusion, convert them into digital signals to extract EEG / MEG signal features, extract the authorized user's EEG / MEG signal feature template from the EEG / MEG signal database, and compare the EEG / MEG signal features with the EEG / MEG signal feature template;

[0062] Smart home interaction command generation module: If the comparison result is higher than the preset threshold, it is recorded as the current user's identity authentication is successful. The brain control command structure receives the current user's EEG / MEG signal characteristics, identifies the EEG / MEG signal patterns in the current user's brain related to smart home interaction actions based on the trained neural network model, and converts them into smart home interaction commands.

[0063] Smart Home Start / Stop Interaction Module: After receiving instructions from the smart home integrated interaction model, the module transmits the instructions to the smart home network topology, controls the start and stop of the smart home, and feeds back the operation results to the user to complete the brain-controlled operation process.

[0064] This embodiment also discloses a computer storage medium storing a computer program, which, when executed by a processor, implements the steps of any one of the smart home interaction methods based on non-invasive EEG / MEG signal authentication.

[0065] Example 2

[0066] This embodiment uses a lighting fixture as an example to illustrate the technical solution of the present invention in detail, as follows:

[0067] Build an integrated interactive model for smart homes (specifically for lighting fixtures).

[0068] Authorized User EEG / MEG Signal Database: Pre-collect EEG / MEG signal samples of authorized users (such as family members) when controlling lighting (such as EEG / MEG signals corresponding to intentions such as "turn on the light", "turn off the light", "brighten", "dimen"), extract features and establish a dedicated template library, while storing basic user information and lighting control permissions (such as child users can only control bedroom lights, while adults can control the lighting of the whole house).

[0069] Lighting network topology: Organize the connection relationships of home lighting devices, including the location of the lights (living room main light, bedroom table lamp, etc.), communication protocols (Wi-Fi, ZigBee, etc.) and control logic (such as linkage scenarios: "bedtime mode" turns off the living room light and dims the bedroom light simultaneously), forming a visual topology map embedded in the interactive model.

[0070] Brain-controlled command parsing structure: Preset command set for lighting control scenarios (such as switching, brightness adjustment, color temperature switching, etc.), define the EEG / MEG signal mode characteristics corresponding to each command (such as "brighten" corresponding to alpha wave frequency band energy enhancement, "turn off" corresponding to specific event-related potential), and associate it with the control interface of each lamp in the topology structure.

[0071] EEG / MEG signal acquisition: The device acquires EEG signals in real time when the user intends to control the lighting—the radar module captures the micro-vibrations of the scalp caused by the activity of brain neurons, and then uses a biomechanical-electrophysiological coupling model to invert the original neural electrical activity, ultimately reconstructing it into an EEG signal. The quantum sensor captures the MEG signals generated by neuronal activity, and uses a multimodal fusion algorithm (such as CNN + attention mechanism) to generate a complete digital signal of brain neuronal activity, extracting feature parameters (such as frequency band energy and waveform complexity).

[0072] Identity verification: The control model calls the lighting control EEG / MEG signal feature templates of authorized users in the database and calculates the matching degree between the current feature and the template through cosine similarity or Euclidean distance. If the matching degree is greater than or equal to a preset threshold (e.g., 85%), the user is determined to be an authorized user and is allowed to enter the command recognition stage; otherwise, the control request is rejected and a reminder is triggered (e.g., flashing lights indicating "Identity not verified").

[0073] Command parsing: After successful identity authentication, the brain control command parsing structure receives the current user's EEG / MEG signal characteristics and inputs them into the trained neural network model (the model has been trained with a large number of "lighting control intention - EEG pattern" samples). For example, when the user has the intention to "turn on the living room light", the model identifies the specific combination pattern of theta waves and beta waves in the corresponding EEG and matches it with the "living room light - turn on" command in the command set.

[0074] Command transmission and execution: The integrated control model transmits the parsed commands to the target device (such as the control node of the main living room light) through the lighting network topology, triggering the lights to turn on; at the same time, the topology provides real-time feedback on the execution status (such as "the living room light is on, brightness is 100%)", informing the user through visual (lights flash briefly) or auditory (related speaker broadcast).

[0075] If the user's intention does not match the execution result (e.g., wanting to "dim" but actually "turning off"), it can be corrected through secondary EEG commands (e.g., the EEG / MEG signal pattern corresponding to "cancel"). The system records this deviation and updates the user's EEG / MEG signal feature template, making subsequent recognition more accurate. For example, children's EEG / MEG signals have lower stability, and the model will learn their characteristic fluctuation patterns through multiple interactions, gradually improving control accuracy.

[0076] Through the above process, a closed loop of "identity security + mind control" is achieved for smart home devices represented by lighting fixtures, taking into account both convenience and security, and is especially suitable for scenarios where both hands are occupied (such as adjusting lights while cooking) or where mobility is limited.

[0077] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0078] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A smart home interaction method based on non-invasive EEG / MEG signal authentication, characterized in that, Includes the following steps: Construct an integrated control model for smart homes, including a database of authorized users' EEG / MEG signals, a smart home network topology, and a brain-controlled command parsing structure; The system uses a non-invasive method of multimodal fusion to collect the current user's control command signals, converts them into digital signals to extract EEG / MEG signal features, extracts the authorized user's EEG / MEG signal feature template from the EEG / MEG signal database, and compares the EEG / MEG signal features with the EEG / MEG signal feature template. If the comparison result is higher than the preset threshold, it is recorded as the current user's identity authentication is successful. The brain control command structure receives the current user's EEG / MEG signal characteristics, identifies the EEG / MEG signal patterns in the current user's brain related to smart home interaction actions based on the trained neural network model, and converts them into smart home interaction commands. After receiving the instruction, the smart home integrated control model transmits the instruction to the smart home network topology, controls the start and stop of the smart home, and feeds back the operation result to the user to complete the brain-controlled operation process.

2. The smart home interaction method based on non-invasive EEG / MEG signal authentication according to claim 1, characterized in that, The smart home network topology is constructed as follows: Collect business information and switch control information of smart home devices in the current home, where business information refers to the functions of smart home devices; Using smart home devices as nodes, and each node's business information and switch control information as its initial features, a graph structure is constructed. Based on the business interaction relationships between devices, data transmission paths, and physical location correlations, the connection edges between nodes are determined, and the current smart home network topology is constructed.

3. The smart home interaction method based on non-invasive EEG / MEG signal authentication according to claim 2, characterized in that, It also includes real-time monitoring of device status changes, including the addition of new devices, the removal of old devices, device function upgrades, or device malfunctions; when a new device is connected, its business information and switch control information are automatically collected to determine the connection relationship between the new device and existing nodes and update the graph structure; if a device is removed or malfunctions, the corresponding node and related connection edges are deleted in a timely manner to ensure that the topology structure can accurately reflect the actual situation of the current smart home system.

4. The smart home interaction method based on non-invasive EEG / MEG signal authentication according to claim 1, characterized in that, The training process of a neural network model is as follows: A generative adversarial neural network architecture was constructed, which takes labeled EEG / MEG signals and smart home start / stop control signals as inputs, with different labels corresponding to different smart home start / stop signals; Using EEG / MEG signals as the source domain and smart home start / stop control signals as the target domain, an adversarial neural network is used to migrate the source domain to the target domain until the mapping relationship between EEG / MEG signals and smart home start / stop control signals meets a preset threshold.

5. The smart home interaction method based on non-invasive EEG / MEG signal authentication according to claim 1, characterized in that, The non-invasive acquisition of the current user's control command signals using multimodal fusion specifically includes: acquiring the current user's EEG / MEG signals using millimeter radar technology and quantum sensors, and fusing the acquired EEG and MEG signals to obtain the current user's control command signals.

6. A smart home interaction method based on non-invasive EEG / MEG signal authentication according to claim 5, characterized in that, Multimodal fusion can use the Kalman filter algorithm to perform spatiotemporal alignment and state estimation of EEG and MEG signals, then extract multimodal features through convolutional neural networks in deep learning, and use attention mechanism to weighted fuse and generate control command signals.

7. A smart home interaction system based on non-invasive EEG / MEG signal authentication, characterized in that, include: Smart Home Integration and Interaction Model Construction Module: Used to construct a smart home integration and interaction model, including an authorized user's EEG / MEG signal database, a smart home network topology, and a brain control command parsing structure; Brain-controlled signal authentication module: Used to collect the current user's EEG / MEG signals in a non-invasive manner using multimodal fusion, convert them into digital signals to extract EEG / MEG signal features, extract the authorized user's EEG / MEG signal feature template from the EEG / MEG signal database, and compare the EEG / MEG signal features with the EEG / MEG signal feature template; Smart home interaction command generation module: If the comparison result is higher than the preset threshold, it is recorded as the current user's identity authentication is successful. It is used to receive the current user's EEG / MEG signal characteristics in the brain control command structure, identify the EEG / MEG signal patterns in the current user's brain related to smart home control actions based on the trained neural network model, and convert them into smart home interaction commands. Smart Home Start / Stop Interaction Module: After receiving instructions from the smart home integrated interaction model, the module transmits the instructions to the smart home network topology, controls the start and stop of the smart home, and feeds back the operation results to the user to complete the brain-controlled operation process.

8. A computer storage medium, characterized in that, The computer storage medium stores a computer program, which, when executed by a processor, implements the steps of a smart home interaction method based on non-invasive EEG / MEG signal authentication as described in any one of claims 1-6.

Images