A bluetooth broadcast one-to-many concentration monitoring method based on end-side electroencephalogram signal processing

Abstract

The application discloses a Bluetooth broadcast one-to-many concentration monitoring method based on end-side electroencephalogram signal processing, which comprises the following steps: S1, original signal acquisition and pretreatment; S2, end-side frequency domain feature calculation; S3, concentration numerical solution; S4, Bluetooth broadcast data packet encapsulation; and S5, connectionless broadcast sending. The application adopts a Bluetooth connectionless broadcast mode, breaks through the limitation of traditional connection number, and can stably collect 50 to 200 terminal data by a single host, which is suitable for large-scale collective monitoring scenes; FFT and concentration solution are completed at the end side, and a random jitter algorithm is used, so that the bandwidth occupation and the packet loss rate are greatly reduced; the terminal does not need to maintain a connection state, the slave power consumption is greatly reduced, and the equipment endurance is prolonged; the end side directly outputs results, so that the steps of original waveform data uploading and cloud computing are omitted, and the user experience, monitoring accuracy and real-time performance are improved; the host is only responsible for receiving and displaying, and does not need complex calculation, so that the hardware threshold and deployment cost are reduced.

Description

Technical Field

[0001] This invention relates to the field of electroencephalogram (EEG) monitoring technology, and in particular to a Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing. Background Technology

[0002] With the development of EEG monitoring technology, the core function of attention monitoring devices (such as EEG headbands) is to collect users' EEG signals and assess their attention levels. Currently, existing attention monitoring technologies mainly adopt two approaches: one is the cloud computing mode, where traditional EEG attention monitoring devices connect to mobile phones or gateways via Bluetooth to upload raw EEG waveform data to the terminal device in real time, where the terminal device performs spectrum analysis and attention algorithm calculations; the other is the Bluetooth connection mode, where some devices use Bluetooth "connection mode" to achieve one-to-one pairing communication.

[0003] The aforementioned existing technologies have significant drawbacks and are insufficient to meet the one-to-many monitoring needs in collective scenarios: 1. Bandwidth bottleneck: The sampling rate of raw EEG data is usually ≥250Hz and the precision is 24 bits. A single device needs to occupy about 10-20kbps of bandwidth. When multiple devices connect to the same host at the same time, the Bluetooth channel quickly becomes saturated, resulting in data packet loss and a sharp increase in latency. 2. Connection limit: In Bluetooth connection mode, mobile phones or ordinary gateways can usually only stably connect to about 7 devices, which cannot meet the needs of "one-to-many" collective monitoring scenarios such as classrooms and conference rooms; 3. Excessive power consumption: Continuous bidirectional connection and large amount of data transmission result in short battery life of the EEG acquisition terminal (slave), usually less than 4 hours, which cannot meet the needs of all-day use; 4. Poor real-time performance: Cloud computing relies on network transmission, which has a latency of hundreds of milliseconds, and cannot meet the needs of real-time feedback on attention. In summary, this application proposes a Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing. Summary of the Invention

[0004] Based on the technical problems existing in the background technology, the present invention proposes a Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing.

[0005] This invention proposes a Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing, comprising the following steps: S1: Raw signal acquisition and preprocessing: The microcontroller unit (MCU) of the EEG acquisition terminal controls the EEG sensor to acquire time-domain EEG signals at a preset sampling rate, and removes interference signals through a built-in digital filter to complete signal preprocessing; S2: End-side frequency domain feature calculation: The microcontroller unit (MCU) performs a fast Fourier transform (FFT) on the preprocessed EEG data frame to convert the time-domain signal into a frequency-domain signal and calculate the power values ​​of the alpha wave band, beta wave band, and background noise band. S3: Attention value calculation: The microcontroller unit (MCU) calculates the attention value based on the frequency domain characteristics calculated in step S2 using a preset attention algorithm model, and quantizes the value into an integer from 0 to 100. S4: Bluetooth broadcast data packet encapsulation: The microcontroller unit (MCU) constructs Bluetooth broadcast data packets, filling the attention value, the unique device ID of the EEG acquisition terminal, and the current battery level information into the manufacturer-specific data field of the broadcast packet, ensuring that the data packet length is less than 31 bytes; S5: Connectionless broadcast transmission: The EEG acquisition terminal enables Bluetooth broadcast mode and sends the data packets encapsulated in step S4 to the surrounding environment at preset broadcast intervals, without needing to establish an ACL connection with the central monitoring device. S6: Host Aggregation and Analysis: The central monitoring device has a built-in Bluetooth module that works in scanning mode. It continuously receives broadcast data packets sent by various EEG acquisition terminals, identifies the devices in this system based on the packet prefix, and parses out the device ID and attention value. It distinguishes different users based on the device ID and displays the attention status in real time.

[0006] Preferably, the specific logical steps of S1 are as follows: S101: The MCU-controlled gel electrode EEG sensor of the EEG acquisition terminal acquires raw time-domain EEG signals from the scalp at a sampling rate of 250Hz. S102: The original signal is amplified and converted from analog to digital by the analog front-end circuit, and then input to the MCU processor; S103: The MCU calls the built-in second-order IIR digital filter to denoise the signal point by point, eliminating 50Hz power frequency interference and ocular artifacts. The specific filtering formula used is as follows: ; In the formula: For filtering the output signal, The raw EEG signal, , , These are the filter coefficients. This is the sampling point number.

[0007] Preferably, the specific logical steps of S2 are as follows: S201: The microcontroller unit (MCU) divides the preprocessed EEG signal into frames of 256 points each. S202: Perform a radix-2 FFT fast Fourier transform on a single frame signal to convert the time-domain signal into a frequency-domain signal; The FFT formula used is as follows: ; In the formula: =256 is the frame length. For rotation factor, The amplitude is in the frequency domain. S203: Calculate the total power of the frequency bands for alpha wave (8-13Hz), beta wave (13-30Hz), and background noise (0-8Hz), respectively; The specific formula for the frequency band power used is as follows: ; In the formula: Total power of the frequency band , These are the start and end frequencies of the frequency band.

[0008] Preferably, the specific logical steps of S3 are as follows: S301: The MCU calculates the original focus score based on alpha wave, beta wave power, and noise power; The original formula for concentration it uses is as follows: ; In the formula: For alpha wave power, For beta wave power, For noise power, , These are weighting coefficients; S302: Normalize and map the original focus scores; S303: Quantizes the result into an integer concentration value of 0-100 by rounding; The specific quantification formula used is as follows: ; In the formula: This is the final concentration value. This is a rounding function.

[0009] Preferably, the specific logical steps of S5 are as follows: S501: The basic broadcast interval of the EEG acquisition terminal is configured to be 500ms; S502: The MCU calls the random jitter module to add ±10% random time offset to the base interval; The specific formula for broadcast interval jitter used is as follows: ; In the formula: This refers to the actual broadcast interval. , The numbers are uniformly random. S503: The terminal sends data packets in connectionless Bluetooth broadcast mode without establishing an ACL link with the host.

[0010] Preferably, the central monitoring device includes, but is not limited to, a tablet computer, a smartphone, or a dedicated PC, which is only responsible for receiving and parsing broadcast data packets and displaying attention status, and does not participate in any EEG signal processing or attention value calculation.

[0011] Preferably, the EEG acquisition terminal further includes an analog front-end circuit, a heart rate and blood oxygen sensor, an IMU sensor, and a Bluetooth radio frequency module; the analog front-end circuit is used to amplify and convert the raw EEG signal into an analog-to-digital signal, the heart rate and blood oxygen sensor is used to acquire the user's heart rate and blood oxygen signal, the IMU sensor is used to acquire the user's posture, and the Bluetooth radio frequency module is integrated into the microcontroller unit (MCU) for sending broadcast data packets.

[0012] Preferably, it further includes an adaptive broadcast frequency control step, the specific logical steps of which are as follows: S801: Terminal calculates the rate of change in concentration values ​​in real time. Or detect the channel packet loss rate; S802: Dynamically adjusts the broadcast interval based on the state of mind; increases the broadcast frequency when the concentration fluctuates drastically. The specific formula for adjusting the broadcast interval used is as follows: ; In the formula: For the adjusted interval, The rate of change of concentration value; S803: Reduces broadcast frequency when the channel is busy, balancing real-time performance and power consumption.

[0013] Preferably, in step S6, after the central monitoring device parses the data packet, it refreshes the attention curves of each user on the interface in real time, which can display a multi-user attention heatmap, suitable for collective monitoring scenarios in classrooms and conference rooms.

[0014] Compared with existing technologies, the beneficial effects of this invention are: 1. By adopting a connectionless Bluetooth broadcast mode, the host only acts as an "observer" to receive data, which gets rid of the handle number limitation of the traditional Bluetooth connection mode. A single host can stably and concurrently collect data from 50 to 200 EEG acquisition terminals, breaking through the connection limit and perfectly adapting to "one-to-many" collective monitoring scenarios such as classrooms and conference rooms, meeting the needs of large-scale collective monitoring. 2. By performing FFT calculation and attention algorithm solution on the device side, only the quantized attention value is broadcast, and the data volume is compressed by more than 95%, which greatly reduces the Bluetooth air interface occupancy rate; combined with the random jitter algorithm, it effectively avoids broadcast conflicts between multiple terminals, and the data reception stability rate is as high as 99% or more, significantly reducing bandwidth consumption and data packet loss rate. 3. In broadcast mode, the EEG acquisition terminal does not need to maintain a complex connection state machine, and the radio frequency transmission time is extremely short. It is in a low-power sleep state most of the time, which reduces power consumption by more than 50% compared with traditional solutions. The device's battery life can reach more than 8 hours, meeting the needs of all-weather use, significantly reducing slave power consumption and extending device battery life. 4. The device directly completes signal processing and attention calculation and broadcasts immediately, eliminating the need to upload raw waveform data and perform cloud computing. The overall system latency is controlled within 500 milliseconds, enabling real-time feedback on attention status and improving user experience, monitoring accuracy and real-time performance. 5. The host is only responsible for receiving and displaying data and does not participate in any complex EEG algorithm calculations. It can run smoothly with ordinary smartphones or low-end gateways, which greatly reduces the deployment cost and hardware threshold of the system and reduces the dependence on the host hardware performance.

[0015] This invention employs a Bluetooth connectionless broadcast mode, breaking through the traditional limitation on the number of connections. A single host can stably collect data from 50 to 200 terminals, adapting to large-scale collective monitoring scenarios. The device performs FFT and attention calculations on the edge, and with the help of a random jitter algorithm, it significantly reduces bandwidth consumption and packet loss rate. The terminal does not need to maintain a connection state, which greatly reduces the power consumption of the slave device and extends the device's battery life. The device outputs results directly, eliminating the need for uploading raw waveform data and cloud computing, thus improving user experience, monitoring accuracy, and real-time performance. The host is only responsible for receiving and displaying data, without the need for complex calculations, reducing hardware barriers and deployment costs. Attached Figure Description

[0016] Figure 1 This is a flowchart of a Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing proposed in this invention; Figure 2 This is a schematic diagram of Bluetooth communication for a Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing proposed in this invention. Figure 3 This is a block diagram of the EEG acquisition terminal for a Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing proposed in this invention. Detailed Implementation

[0017] The present invention will be further explained below with reference to specific embodiments.

[0018] Example Reference Figure 1-3 This embodiment proposes a Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing, including the following steps: S1: Raw signal acquisition and preprocessing: The microcontroller unit (MCU) of the EEG acquisition terminal controls the EEG sensor to acquire time-domain EEG signals at a preset sampling rate, and removes interference signals through a built-in digital filter to complete signal preprocessing; The specific logical steps are as follows: S101: The MCU-controlled gel electrode EEG sensor of the EEG acquisition terminal acquires raw time-domain EEG signals from the scalp at a sampling rate of 250Hz. S102: The original signal is amplified and converted from analog to digital by the analog front-end circuit, and then input to the MCU processor; S103: The MCU calls the built-in second-order IIR digital filter to denoise the signal point by point, eliminating 50Hz power frequency interference and ocular artifacts. The specific filtering formula used is as follows: ; In the formula: For filtering the output signal, The raw EEG signal, , , These are the filter coefficients. The sampling point number; S2: End-side frequency domain feature calculation: The microcontroller unit (MCU) performs a fast Fourier transform (FFT) on the preprocessed EEG data frame to convert the time-domain signal into a frequency-domain signal and calculate the power values ​​of the alpha wave band, beta wave band, and background noise band. The specific logical steps are as follows: S201: The microcontroller unit (MCU) divides the preprocessed EEG signal into frames of 256 points each. S202: Perform a radix-2 FFT fast Fourier transform on a single frame signal to convert the time-domain signal into a frequency-domain signal; The FFT formula used is as follows: ; In the formula: =256 is the frame length. For rotation factor, The amplitude is in the frequency domain. S203: Calculate the total power of the frequency bands for alpha wave (8-13Hz), beta wave (13-30Hz), and background noise (0-8Hz), respectively; The specific formula for the frequency band power used is as follows: ; In the formula: Total power of the frequency band , The start and end frequencies of the frequency band; S3: Attention value calculation: The microcontroller unit (MCU) calculates the attention value based on the frequency domain characteristics calculated in step S2 using a preset attention algorithm model, and quantizes the value into an integer from 0 to 100. The specific logical steps are as follows: S301: The MCU calculates the original focus score based on alpha wave, beta wave power, and noise power; The original formula for concentration it uses is as follows: ; In the formula: For alpha wave power, For beta wave power, For noise power, , These are weighting coefficients; S302: Normalize and map the original focus scores; S303: Quantizes the result into an integer concentration value of 0-100 by rounding; The specific quantification formula used is as follows: ; In the formula: This is the final concentration value. This is a rounding function; S4: Bluetooth broadcast data packet encapsulation: The microcontroller unit (MCU) constructs Bluetooth broadcast data packets, filling the attention value, the unique device ID of the EEG acquisition terminal, and the current battery level information into the manufacturer-specific data field of the broadcast packet, ensuring that the data packet length is less than 31 bytes; S5: Connectionless broadcast transmission: The EEG acquisition terminal enables Bluetooth broadcast mode and sends the data packets encapsulated in step S4 to the surrounding environment at preset broadcast intervals, without needing to establish an ACL connection with the central monitoring device. The specific logical steps are as follows: S501: The basic broadcast interval of the EEG acquisition terminal is configured to be 500ms; S502: The MCU calls the random jitter module to add ±10% random time offset to the base interval; The specific formula for broadcast interval jitter used is as follows: ; In the formula: This refers to the actual broadcast interval. , The numbers are uniformly random. S503: The terminal sends data packets in connectionless Bluetooth broadcast mode without establishing an ACL link with the host; S6: Host Aggregation and Analysis: The central monitoring device has a built-in Bluetooth module that works in scanning mode, continuously receiving broadcast data packets sent by various EEG acquisition terminals. It identifies the devices in this system based on the packet prefix, analyzes the device ID and attention value, distinguishes different users based on the device ID, and displays the attention status in real time; achieving one-to-many real-time monitoring. After the central monitoring equipment parses the data packets, it refreshes the attention curves of each user in real time on the interface, and can display a multi-user attention heatmap, which is suitable for collective monitoring scenarios in classrooms and conference rooms. The central monitoring equipment includes, but is not limited to, tablets, smartphones or dedicated PCs, which are only responsible for receiving and parsing broadcast data packets and displaying attention status, and do not participate in any EEG signal processing or attention value calculation. The EEG acquisition terminal also includes an analog front-end circuit, a heart rate and blood oxygen sensor, an IMU sensor, and a Bluetooth radio frequency module. The analog front-end circuit is used to amplify and convert the raw EEG signals into analog-to-digital signals. The heart rate and blood oxygen sensor is used to acquire the user's heart rate and blood oxygen signals. The IMU sensor is used to acquire the user's posture. The Bluetooth radio frequency module is integrated into the microcontroller unit (MCU) and is used to send broadcast data packets.

[0019] The method also includes an adaptive broadcast frequency control step, the specific logical steps of which are as follows: S801: Terminal calculates the rate of change in concentration values ​​in real time. Or detect the channel packet loss rate; S802: Dynamically adjusts the broadcast interval based on the state of mind; increases the broadcast frequency when the concentration fluctuates drastically. The specific formula for adjusting the broadcast interval used is as follows: ; In the formula: For the adjusted interval, The rate of change of concentration value; S803: Reduces broadcast frequency when the channel is busy, balancing real-time performance and power consumption.

[0020] This invention employs a Bluetooth connectionless broadcast mode, breaking through the traditional limitation on the number of connections. A single host can stably collect data from 50 to 200 terminals, adapting to large-scale collective monitoring scenarios. The device performs FFT and attention calculations on the edge, and with the help of a random jitter algorithm, it significantly reduces bandwidth consumption and packet loss rate. The terminal does not need to maintain a connection state, which greatly reduces the power consumption of the slave device and extends the device's battery life. The device outputs results directly, eliminating the need for uploading raw waveform data and cloud computing, thus improving user experience, monitoring accuracy, and real-time performance. The host is only responsible for receiving and displaying data, without the need for complex calculations, reducing hardware barriers and deployment costs.

[0021] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing, characterized in that, Includes the following steps: S1: Raw signal acquisition and preprocessing: The microcontroller unit (MCU) of the EEG acquisition terminal controls the EEG sensor to acquire time-domain EEG signals at a preset sampling rate, and removes interference signals through a built-in digital filter to complete signal preprocessing; S2: End-side frequency domain feature calculation: The microcontroller unit (MCU) performs a fast Fourier transform on the preprocessed EEG data frame to convert the time-domain signal into a frequency-domain signal and calculate the power values ​​of the alpha wave band, beta wave band, and background noise band. S3: Attention value calculation: The microcontroller unit (MCU) calculates the attention value based on the frequency domain characteristics calculated in step S2 using a preset attention algorithm model, and quantizes the value into an integer from 0 to 100. S4: Bluetooth broadcast data packet encapsulation: The microcontroller unit (MCU) constructs Bluetooth broadcast data packets, filling the attention value, the unique device ID of the EEG acquisition terminal, and the current battery level information into the manufacturer-specific data field of the broadcast packet, ensuring that the data packet length is less than 31 bytes; S5: Connectionless broadcast transmission: The EEG acquisition terminal enables Bluetooth broadcast mode and sends the data packets encapsulated in step S4 to the surrounding environment at preset broadcast intervals, without needing to establish an ACL connection with the central monitoring device. S6: Host Aggregation and Analysis: The central monitoring device has a built-in Bluetooth module that works in scanning mode. It continuously receives broadcast data packets sent by various EEG acquisition terminals, identifies the devices in this system based on the packet prefix, and parses out the device ID and attention value. It distinguishes different users based on the device ID and displays the attention status in real time.

2. The Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing according to claim 1, characterized in that, The specific logical steps of S1 are as follows: S101: The MCU-controlled gel electrode EEG sensor of the EEG acquisition terminal acquires raw time-domain EEG signals from the scalp at a sampling rate of 250Hz. S102: The original signal is amplified and converted from analog to digital by the analog front-end circuit, and then input to the MCU processor; S103: The MCU calls the built-in second-order IIR digital filter to denoise the signal point by point, eliminating 50Hz power frequency interference and ocular artifacts. The specific filtering formula used is as follows: ； In the formula: For filtering the output signal, The raw EEG signal, , , These are the filter coefficients. This is the sampling point number.

3. The Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing according to claim 2, characterized in that, The specific logical steps of S2 are as follows: S201: The microcontroller unit (MCU) divides the preprocessed EEG signal into frames of 256 points each. S202: Perform a radix-2 FFT fast Fourier transform on a single frame signal to convert the time-domain signal into a frequency-domain signal; The FFT formula used is as follows: ； In the formula: =256 is the frame length. For rotation factor, The amplitude is in the frequency domain. S203: Calculate the total power of the α wave, β wave, and background noise in the frequency bands respectively; The specific formula for the frequency band power used is as follows: ； In the formula: Total power of the frequency band , These are the start and end frequencies of the frequency band.

4. The Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing according to claim 3, characterized in that, The specific logical steps of S3 are as follows: S301: The MCU calculates the original focus score based on alpha wave, beta wave power, and noise power; The original formula for concentration it uses is as follows: ； In the formula: For alpha wave power, For beta wave power, For noise power, , These are weighting coefficients; S302: Normalize and map the original focus scores; S303: Quantizes the result into an integer concentration value of 0-100 by rounding; The specific quantification formula used is as follows: ； In the formula: This is the final concentration value. This is a rounding function.

5. The Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing according to claim 4, characterized in that, The specific logical steps of S5 are as follows: S501: The basic broadcast interval of the EEG acquisition terminal is configured to be 500ms; S502: The MCU calls the random jitter module to add ±10% random time offset to the base interval; The specific formula for broadcast interval jitter used is as follows: ； In the formula: This refers to the actual broadcast interval. , The numbers are uniformly random. S503: The terminal sends data packets in connectionless Bluetooth broadcast mode without establishing an ACL link with the host.

6. The Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing according to claim 5, characterized in that, The central monitoring equipment includes, but is not limited to, tablets, smartphones, or dedicated PCs. It is only responsible for receiving and parsing broadcast data packets and displaying attention status, and does not participate in any EEG signal processing or attention value calculation.

7. The Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing according to claim 6, characterized in that, The EEG acquisition terminal also includes an analog front-end circuit, a heart rate and blood oxygen sensor, an IMU sensor, and a Bluetooth radio frequency module. The analog front-end circuit is used to amplify and convert the raw EEG signal into an analog-to-digital signal. The heart rate and blood oxygen sensor is used to acquire the user's heart rate and blood oxygen signal. The IMU sensor is used to acquire the user's posture. The Bluetooth radio frequency module is integrated into the microcontroller unit (MCU) and is used to send broadcast data packets.

8. The Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing according to claim 7, characterized in that, It also includes an adaptive broadcast frequency control step, the specific logical steps of which are as follows: S801: Terminal calculates the rate of change in concentration values ​​in real time. Or detect the channel packet loss rate; S802: Dynamically adjusts the broadcast interval based on the state of mind; increases the broadcast frequency when the concentration fluctuates drastically. The specific formula for adjusting the broadcast interval used is as follows: ； In the formula: For the adjusted interval, The rate of change of concentration value; S803: Reduces broadcast frequency when the channel is busy, balancing real-time performance and power consumption.

9. The Bluetooth broadcast one-to-many attention monitoring method based on end-to-end EEG signal processing according to claim 8, characterized in that, In S6, after the central monitoring device parses the data packet, it refreshes the attention curve of each user in real time on the interface, and can display a multi-user attention heat map, which is suitable for collective monitoring scenarios in classrooms and conference rooms.

Images