System and method for high-density electromyographic signal synchronous acquisition

Abstract

The present disclosure relates to the field of signal acquisition, and in particular, to a system and method for high-density electromyographic signal synchronous acquisition. In the system, a signal acquisition module and a base station are electrically connected via a pogo pin connector and are mechanically connected via an electromagnetic lock. The pogo pin connector comprises various communication protocols and functional ports and supports various synchronous operations. Module charging, data export, state reading, parameter configuration, firmware upgrading, time synchronization, and other functions can be completed by simply inserting the module into the base station, featuring ease to use and compactness. The electromagnetic lock locks the module inserted into the base station to ensure reliable communication between the module and the base station and the safety and integrity of data files. The method is automatically implemented in the condition that the module is inserted into the base station, so as to ensure the real-time synchronization of all modules inserted in the base station with the base station. The electromyographic signal acquisition process does not depend on wireless networks or limit the scene of use to areas with network coverage, thus reducing the wireless connection power consumption, increasing the cost-efficiency, and greatly improving the synchronization reliability.

Description

A high-density electromyographic signal synchronous acquisition system and method

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on February 2, 2024, with application number 202410153596.2 and invention name “A High-density Electromyography Signal Synchronous Acquisition System and Method”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of signal acquisition technology, and in particular to a high-density electromyography signal synchronous acquisition system and method. Background Art

[0003] High-density EMG (High-density EMG) captures the bioelectrical changes generated by the activity of single or multiple muscle cells or parts of muscle tissue. Electrodes are used to guide, amplify, record, and display the resulting multidimensional time series of voltage changes. This accurately reflects the state of motion of the human nervous, skeletal, and muscular systems. High-density EMG captures array-based surface EMG signals, transforming traditional single- or two-point EMG detection into surface-based detection. Medical signal processing techniques are then used to comprehensively process the collected signals. Because of its use of arrayed surface electrodes, it boasts dozens of times more channels than traditional single- or dual-channel surface EMG. This allows for a more comprehensive recording of the electrical signals from the muscles at the measured location, preserving the maximum amount of muscle activity information contained within the signals. Furthermore, methods such as independent component analysis and deconvolution can be used to decompose multi-channel high-density EMG signals into the microscopic firing behavior of individual neurons, enabling the observation of individual neuron firing and enabling applications such as human-computer interaction, sports injury analysis, and muscle fatigue monitoring. In some application scenarios (such as whole-body motion monitoring), all muscle groups at different body positions of the same subject need to be monitored, and the monitoring equipment cannot affect the subject's normal movements and needs to be fixed at various positions of the body in a wireless rather than wired form. Therefore, a distributed high-density electromyography acquisition solution is required.

[0004] For wireless, distributed, multi-channel physiological signal acquisition systems like distributed high-density electromyography (EMG) acquisition systems, time synchronization between distributed sensing modules is crucial. Since each module inevitably uses a separate high-frequency sampling crystal oscillator (2.048 MHz) for sampling, sampling occurs asynchronously between modules, meaning that all samples cannot be guaranteed to be generated at the same absolute instant. A more serious problem is that the commonly used high-frequency MHz crystal oscillators used for sampling have absolute sampling rate errors. This can lead to inconsistent numbers of sample points over extended sampling periods (over 10 seconds). For example, module A might actually acquire 11,000 samples at a sampling rate of 1 ksps over 10 seconds, while module B might only acquire 9,950. This inconsistency increases over time, making it impossible to align the sample points between modules on the absolute time axis, leading to a loss of synchronization between modules. With a typical 2.048 MHz sampling clock of 50 ppm, the inter-module timestamp error can reach 1.44 seconds after synchronizing the modules and conducting a 4-hour experiment. In many experimental scenarios (such as multi-muscle synchronous force reaction testing), the timestamp synchronization error between modules measuring each muscle group must be less than 8ms. The above sampling clock accuracy obviously cannot meet the requirements.

[0005] To address the synchronization issues mentioned above, there are currently two main approaches: using very-high-precision clock crystal oscillators or employing wireless synchronization. Very-high-precision clocks typically employ oven-controlled crystal oscillators (OCXOs), which are large, power-hungry, and expensive. They are primarily used in large-scale facilities such as defense, satellite navigation, remote sensing and mapping, and communication base stations, and are unsuitable for wearable, low-power devices. Wireless synchronization is the predominant approach currently employed. This approach uses a wireless base station to connect all modules to be synchronized in real time and uses time synchronization protocols such as the Network Time Protocol (NTP) to synchronize module sampling times at regular intervals. However, its main drawbacks are: 1) it requires all modules to be connected to a specific wireless network in real time, limiting its use in scenarios such as outdoor environments where space is not a constraint. 2) wireless signals are susceptible to electromagnetic environmental factors, resulting in uncontrollable synchronization delays that vary between modules due to transmission delays or interruptions, such as electromagnetic wave reflections and disconnections. 3) it requires a constant wireless connection, which consumes power, impacting battery life and shortening the maximum experimental time. Summary of the Invention

[0006] In response to the problems raised in the above background technology, the present application provides a high-density electromyography signal synchronous acquisition system and method to reduce the device size and power consumption, reduce signal acquisition costs, expand usage scenarios and improve synchronization reliability.

[0007] To achieve the above objectives, this application provides the following solutions.

[0008] On the one hand, the present application provides a high-density electromyographic signal synchronous acquisition system, comprising: a host computer, a base station, and multiple signal acquisition modules;

[0009] The signal acquisition module and the host computer are respectively connected to a wireless router node via Wi-Fi. Under the routing forwarding of the wireless router, the host computer establishes a TCP server, and the signal acquisition module connects to the TCP server as a client to realize socket communication; the host computer sends instructions to the signal acquisition module to start and end the experiment;

[0010] The signal acquisition module is connected to the flexible electrode during the experiment and fixed to the position of the muscle to be tested on the human body, and is used to collect, process, store and send high-density surface electromyography signals in real time;

[0011] After the experiment, multiple signal acquisition modules were electrically connected to the base station through a spring pin interface and mechanically connected through an electromagnetic lock; the base station was connected to the host computer through a USB 2.0 bus;

[0012] The base station is used to store multiple signal acquisition modules and communicate with the host computer, charge the signal acquisition modules, and read, configure, synchronize time, and eject the signal acquisition modules according to the host computer's instructions. The base station also has a multi-channel analog signal input interface for collecting analog signals from external devices and digitizing them for transmission to the host computer.

[0013] The host computer provides a user graphical operation interface for managing and configuring the signal acquisition module, and receiving, storing and processing the offline sampling data uploaded in real time to complete the synchronization of high-density surface electromyography signals.

[0014] In an exemplary embodiment, the pin interface includes an SPI bus, a USB bus, a charging interface, a hot-swap interface, a synchronization interface and a firmware upgrade interface; the SPI bus connects the base station and all signal acquisition modules, and is used by the base station to read the signal acquisition module status, configuration module, and synchronization module time; the USB bus is connected to the USB HUB in the base station, so that the signal acquisition module can transmit offline saved data to the host computer; the charging interface is used to charge the signal acquisition module, and the signal acquisition module automatically charges after being inserted into the base station, and automatically stops charging after charging is completed; the hot-swap interface is used to bidirectionally identify whether the signal acquisition module is connected to the base station. When the signal acquisition module is inserted or ejected, the signal acquisition module and the base station will receive hot-swap feedback; the synchronization interface is used to synchronize the real-time clock system between the signal acquisition module and the base station and each signal acquisition module, and cooperate with SPI bus instructions, software and algorithms to achieve microsecond time synchronization; the firmware upgrade interface is used to upgrade the internal firmware of the module online.

[0015] In an exemplary embodiment, the electromagnetic lock consists of a lock body and a latch; the lock body is fixed to the base station by screws, and the latch cooperates with the lock hole at the corresponding position of the signal acquisition module to fix the signal acquisition module in the base station; the electromagnetic lock includes two states: locked and unlocked; when the signal acquisition module is inserted into the base station, the electromagnetic lock latch slides into the lock hole of the signal acquisition module, locking the signal acquisition module and sending insertion feedback to the base station; when the electromagnetic lock receives the pop-up command from the base station, the lock body is powered on, the latch is withdrawn from the lock hole to unlock and pop out the signal acquisition module.

[0016] In an exemplary embodiment, the host computer sends instructions to the base station through the CDC protocol based on the USB 2.0 bus, controlling the base station to complete status reading and writing, experimental configuration, firmware upgrade and pop-up control operations of the signal acquisition module inserted therein; the host computer is also connected to the signal acquisition module through the USB HUB in the base station via the USB 2.0 bus, and reads the offline sampling data stored in the SD card of the signal acquisition module through the USB mass storage device protocol.

[0017] On the other hand, the present application also provides a high-density electromyographic signal synchronous acquisition method based on the high-density electromyographic signal synchronous acquisition system, comprising:

[0018] Use the coherent calibration method to calibrate the extremely high-precision RTC clocks in each signal acquisition module and base station using an oscilloscope at room temperature;

[0019] After coherent calibration, the current timestamps of the RTC clocks of all devices are synchronized using the software and hardware of the base station, signal acquisition module, and host computer.

[0020] Each signal acquisition module and base station combines the input signal sampling point generated under the command of the low-precision sampling clock with the high-precision RTC timestamp;

[0021] Based on the combined timestamps, the host computer normalizes the time axis of the recorded sampling data of each signal acquisition module and base station to form unified time sampling data.

[0022] In an exemplary embodiment, the coherent calibration method is used to calibrate each signal acquisition module and the extremely high-precision RTC clock in the base station using an oscilloscope at room temperature, specifically including:

[0023] At room temperature, connect the output of the base station RTC crystal oscillator to channel 1 of a dual-channel oscilloscope, and connect the output of the RTC crystal oscillator in the signal acquisition module to channel 2 of the dual-channel oscilloscope. Use synchronous sampling within the oscilloscope, set channel 1 to trigger on the rising edge or intermediate level value, and display the signals of channels 1 and 2 simultaneously. When the channel 2 signal moves to the left relative to the channel 1 signal, gradually increase the aging parameter until the channel 2 signal stops moving. When the channel 2 signal moves to the right relative to the channel 1 signal, gradually decrease the aging parameter until the channel 2 signal stops moving.

[0024] In an exemplary embodiment, synchronizing the current timestamps of the RTC clocks of all devices using the software and hardware of the base station, the signal acquisition module, and the host computer specifically includes:

[0025] When the signal acquisition module is inserted into the base station, it communicates with the base station via the SPI bus and is connected to the RTC synchronization line; the RTC chip of the base station sends a second pulse every second as an interrupt transmission to the base station main control board; when a time synchronization needs to be completed, the base station main control board first sends the timestamp that should correspond to the next synchronization to each signal acquisition module through the SPI bus in accordance with the communication protocol. This timestamp is the timestamp corresponding to the next arrival of the base station RTC pulse. The SPI bus communication needs to be completed before the next arrival of the base station RTC pulse; when the base station RTC pulse arrives, the base station main control board receives the pulse interrupt and immediately sends a pulse on the RTC synchronization line. After receiving the pulse, each signal acquisition module immediately writes the timestamp previously transmitted to it through the SPI bus into the RTC chip, thereby synchronizing the RTCs of all devices at once.

[0026] In an exemplary embodiment, each signal acquisition module and the base station combines the input signal sampling points generated under the command of the low-precision sampling clock with the high-precision RTC timestamp, specifically including:

[0027] When the signal acquisition module is popped up by the base station for experiment, the signal acquisition module pulse performs a signal sampling when each sampling pulse is generated; the RTC clock pulse of the signal acquisition module is generated once at a preset time and carries a timestamp that is precisely synchronized with the base station and other signal acquisition modules; when the RTC clock pulse arrives, the signal acquisition module combines the timestamp T carried by the RTC clock pulse with the sampling pulse closest to the arrival time of the RTC clock pulse, and determines that the timestamp closest to the moment when the sampling pulse is generated is T.

[0028] In an exemplary embodiment, the host computer normalizes the time axis of the recorded sampling data of each signal acquisition module and base station based on the combined timestamp to form unified time sampling data, specifically including:

[0029] When the experiment is over, the data acquisition module is inserted into the base station and the local files are imported into the database. For all the imported sampling data files, the sampling rate f sample , the sampling number of each signal acquisition module and base station in each timestamp interval is resampled to the standard sampling number N through the signal resampling algorithm; then the sampling data of the same number of signal acquisition modules and base stations in the same timestamp interval are stored in the same standard sampling matrix to achieve time axis normalization.

[0030] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0031] In the high-density electromyographic signal synchronous acquisition system provided by the present application, the signal acquisition module and the base station are electrically connected through a spring pin interface and mechanically connected through an electromagnetic lock. The spring pin interface includes multiple communication protocols and functional ports, which can support multiple operations at the same time. The module's charging, data export, status reading, parameter configuration, firmware upgrade, and time synchronization functions can be completed by simply inserting the signal acquisition module into the base station. The use process is simple and the device size is reduced. The electromagnetic lock structure can lock the module inserted into the base station to ensure reliable communication between the module and the base station and the security and integrity of the data file. The high-density electromyographic signal synchronous acquisition method of the present application, data transmission, time synchronization, charging and other operations are automatically completed when the module is inserted into the base station. All modules inserted in the base station can be synchronized with the base station in real time. It does not rely on wireless networks when collecting electromyographic signals. Therefore, the usage scenario is not limited to areas with network coverage, which reduces the power consumption of wireless connections and the cost of signal acquisition, and greatly improves the synchronization reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] FIG1 is a schematic diagram of the overall structure of the high-density electromyographic signal synchronous acquisition system of the present application;

[0034] Figure 2 is a schematic diagram of the experimental parameter configuration process;

[0035] Figure 3 is a schematic diagram of the data acquisition process;

[0036] Figure 4 is a schematic diagram of the data import process;

[0037] Figure 5 is a schematic diagram of the coherent calibration principle;

[0038] Figure 6 is a schematic diagram of the synchronous timing principle;

[0039] Figure 7 is a schematic diagram of the sampling time joint principle;

[0040] FIG8 is a schematic diagram of the time axis normalization principle. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0042] The purpose of this application is to provide a high-density electromyography signal synchronous acquisition system and method to reduce the device size and power consumption, reduce signal acquisition costs, expand usage scenarios and improve synchronization reliability.

[0043] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0044] The present application provides a high-density electromyographic signal synchronous acquisition system, as shown in Figure 1, including: a host computer, a base station and multiple signal acquisition modules (also referred to as modules); it is a distributed wearable high-density electromyographic signal synchronous acquisition system.

[0045] The signal acquisition module and the host computer are each connected to a wireless router node via Wi-Fi. Under the routing forwarding of the wireless router, the host computer establishes a TCP server, and the signal acquisition module connects to the TCP server as a client, achieving socket communication. The host computer sends commands to the signal acquisition module to start and end the experiment. The signal acquisition module transmits its status information and real-time sampling data to the host computer.

[0046] The signal acquisition module is connected to the flexible electrode during the experiment and fixed to the position of the muscle to be tested in the human body, and is used for real-time acquisition, processing, storage and transmission of high-density surface electromyographic signals.

[0047] After the experiment, the signal acquisition modules were electrically connected to the base station via a spring-loaded interface and mechanically connected via an electromagnetic lock. The base station was connected to the host computer via a USB 2.0 bus.

[0048] The base station is used to store multiple signal acquisition modules and communicate with a host computer. It also charges the modules and, according to host computer instructions, reads status and data, configures, synchronizes time, and ejects the modules. The base station also has a multi-channel analog signal input interface for collecting analog signals from external devices (such as ergometers, electrocardiogram electrodes, and single-channel electromyography electrodes) and digitizing them for transmission to the host computer.

[0049] The host computer provides a user graphical operation interface for managing and configuring the signal acquisition module, and receiving, storing and processing the offline sampling data uploaded in real time to complete the synchronization of high-density surface electromyography signals.

[0050] Specifically, the pin interface includes an SPI (Serial Peripheral Interface) bus, a USB bus, a charging interface, a hot-swap interface, a synchronization interface, and a firmware upgrade interface. Among them, the SPI bus connects the base station with all signal acquisition modules, and is used by the base station to read the status of the signal acquisition module, the configuration module, and the synchronization module time. The USB bus is connected to the USB HUB in the base station to enable the signal acquisition module to transmit offline saved data to the host computer. The charging interface is used to charge the signal acquisition module. After the signal acquisition module is inserted into the base station, it automatically charges and automatically stops charging after charging is completed. The hot-swap interface is used to bidirectionally identify whether the signal acquisition module is connected to the base station. When the signal acquisition module is inserted or ejected, the signal acquisition module and the base station will receive hot-swap digital level feedback. The synchronization interface is used to synchronize the real-time clock system between the signal acquisition module and the base station and each signal acquisition module, and cooperate with the SPI bus instructions, software and algorithms to achieve microsecond time synchronization. The firmware upgrade interface is used to upgrade the internal firmware of the module online.

[0051] The electromagnetic lock consists of a lock body and a latch. The lock body is fixed to the base station with screws, and the latch engages with a keyhole corresponding to the signal acquisition module to secure the signal acquisition module in the base station. The electromagnetic lock has two states: locked and unlocked. When the signal acquisition module is inserted into the base station, the electromagnetic lock latch slides into the keyhole, locking the module and sending an insertion feedback level to the base station. When the electromagnetic lock receives an eject command from the base station, the lock body is powered on, the latch withdraws from the keyhole, unlocking the module, and ejecting the signal acquisition module.

[0052] The base station and host computer are connected via a USB 2.0 bus, with the base station's internal USB hub enabling communication between the base station and host computer, as well as between the signal acquisition module and the host computer. The host computer issues commands to the base station via the USB 2.0 bus-based Communication Device Class (CDC) protocol, controlling the base station to perform operations such as reading and writing status, experimental configuration, firmware upgrades, and ejection control of the inserted signal acquisition module. The host computer also connects to the signal acquisition module via the USB 2.0 bus via the base station's internal USB hub, accessing offline sampling data stored on the signal acquisition module's SD card using the USB mass storage device protocol.

[0053] The working process of the high-density electromyographic signal synchronous acquisition system shown in Figure 1 is as follows:

[0054] The first step is to insert the module into the base station: insert the signal acquisition module into the base station cabin. At this time, after the spring pin seat at the module insertion end contacts the spring pin inside the base station cabin, the circuit is connected and the signal is detected, the base station pushes the module information to the host computer, and the base station main control board controls the electromagnetic lock pin to insert the module lock hole. At this time, the module is locked and cannot be removed; the base station automatically charges the locked module.

[0055] The second step is for the host computer to read and update the information: after the module is inserted into the base station and locked, the host computer reads the information and real-time status of the module in the base station through the USB Type-C cable and displays it on the interface; the module information includes remaining power, remaining memory, firmware information, serial number, network status, sampling rate, experiment ID, and experiment notes; the module real-time status includes configurable status (in this state, the module can be selected for new experiment configuration, data export, firmware upgrade, etc., the light ring is blue and the pop-up button is operable), read / write status (in this state, the module cannot be selected, the light ring is red and breathing, and the host computer pop-up button and the base station pop-up button cannot pop out the module to prevent interruptions in the data reading and writing process, resulting in data loss, configuration errors, and other problems), and cabin no module status (in this state, no module is displayed in the cabin, the light ring is gray, and there is no pop-up word).

[0056] The third step is to configure the experimental parameters: as shown in Figure 2, first click the configuration button on the host computer homepage to jump to the parameter configuration interface, then select the configurable module to set the experiment ID, sampling rate and experimental notes, and click the confirmation button to start writing the configuration. If the writing is successful, a pop-up window will pop up to remind you and jump back to the parameter configuration interface, and reset the parameters at the same time. If the writing fails, a pop-up window will report an error and jump back to the parameter configuration interface. In this case, the parameters will not be reset. Click the Cancel button to cancel the configuration and return to the homepage.

[0057] Step 4. Pop out the configured module: After the module completes the experimental parameter configuration, click the pop-up button on the host computer homepage or press the pop-up button on the base station. The host computer will send a pop-up command to the base station main control board. The main control board controls the electromagnetic lock pin to withdraw from the module lock hole, pop out and remove the module.

[0058] Step 5: Start data acquisition: As shown in Figure 1, the data acquisition module is equipped with an OS button 101, a user event trigger button 102, a Wi-Fi button 103, and a display screen 104. The module has a dedicated silicone shell. First, install the silicone shell onto the module, then insert the flexible electrodes into the module's electrode interface. Then, secure the strap to the muscle to be measured on the human torso. Finally, secure the module to the strap and the flexible electrodes to the skin surface of the test area, completing the securement of the flexible electrodes, module, and strap. As shown in Figure 3, after wearing the device, briefly press the module's OS button 101 or click the experiment start button on the host computer. The device will first perform a self-test. After the self-test is successful, the module's OLED display screen 104 will display the module status and experiment configuration information. After confirming that the information is correct, briefly press the OS button 101 to start the experiment. During the experiment, if you need to mark the experimental data, press the user event trigger button 102, and to turn the network on or off, press the Wi-Fi button 103. To end the experiment, press the OS button 101 or click the experiment end button on the host computer software.

[0059] Step 6: Data Import: As shown in Figure 4, after completing the experiment, remove the module and bandage from the test area and clean the skin with alcohol to prevent allergies. Remove the flexible electrode and insert the module back into the base station. Now, click the Import button on the host computer homepage to enter the data import interface. Users can view and select files based on the experiment ID and module serial number. After selecting the target module and file, click the Confirm button to begin importing.

[0060] Step 7. Data export: After the import is completed, return to the host computer homepage and click the database management button to enter the database management interface. At this time, you can operate the experimental files that have been imported into the database. If you need to export the experimental files to a local computer, select the target file first and then click Export; the same applies to deleting, copying, and replaying data.

[0061] To solve the problem of synchronizing sampling data timestamps between modules and between modules and the base station, this application also proposes a high-density electromyographic signal synchronization acquisition method, which is collaboratively completed by the host computer software, base station, and module hardware. It includes four steps: coherence calibration, synchronous timing, sampling time combination, and time axis normalization. The advantages of this high-density electromyographic signal synchronization acquisition method are that it does not require a wireless network connection, can be applied to offline acquisition scenarios, is not restricted by time and space, has low operating power consumption, low implementation cost, and high synchronization accuracy (up to microseconds).

[0062] Specifically, the present application is a high-density electromyographic signal synchronous acquisition method based on the high-density electromyographic signal synchronous acquisition system, comprising:

[0063] Step 1: Coherent Calibration. Using an oscilloscope, calibrate each signal acquisition module and the high-precision RTC clock in the base station at room temperature using the coherent calibration method. This coherent calibration applies aging calibration to each module and the high-precision RTC clock in the base station at the same temperature, improving the crystal oscillator accuracy to above 0.5ppm.

[0064] The basic principle of coherent calibration is shown in Figure 5. Although civilian-grade extremely high-precision RTC clock chips (such as DS3231SN) are used in both the module and the base station, their factory accuracy can usually only reach 2ppm, which cannot meet the synchronization error requirements of the experiment (±8ms / 4h). For this reason, the 32.768kHz crystal oscillators of all RTC clocks in the same set of equipment are calibrated when the equipment leaves the factory and during annual inspection. Taking into account that the module and the base station are generally used in the same temperature environment, and the after-sales annual inspection environment usually cannot use a high-precision frequency meter, this application uses the coherent calibration method shown in Figure 5 to perform calibration using an oscilloscope at room temperature.

[0065] More specifically, at room temperature, it is assumed that the RTC clock crystal frequency f inside the base station is base To get an accurate 32.768kHz, connect the output of the base station RTC crystal oscillator to channel 1 of a dual-channel oscilloscope; consider the RTC clock crystal frequency f in the module to be moduleThe frequency is approximately equal to 32.768kHz, but fine-tuning is required. Connect the output of its crystal oscillator to Channel 2 of a dual-channel oscilloscope. Use the oscilloscope's internal synchronous sampling, set the trigger on Channel 1 at the rising edge or mid-level, and display the signals on Channels 1 and 2 simultaneously. Due to the slight difference in the period of the two input signals, the Channel 1 signal will remain stationary, while the Channel 2 signal may become unstable and shift left or right relative to Channel 1. If the Channel 2 signal shifts leftward, its frequency is greater than 32.768kHz. Gradually increase the aging parameter to reduce the frequency until the Channel 2 signal barely shifts. If the Channel 2 signal shifts rightward, its frequency is less than 32.768kHz. Gradually decrease the aging parameter to achieve virtually no shift. Repeat this process for each module until the RTC crystal oscillator frequency of all modules is calibrated to be nearly identical to that of the base station. Ideally, after this calibration, the module's RTC frequency stability can reach 0.1ppm.

[0066] Step 2: Synchronize timing: After coherent calibration, use the base station, signal acquisition module, and host computer hardware and software to synchronize the current timestamps of the RTC clocks of all devices.

[0067] The principle of synchronous timing is shown in Figure 6, where the hardware connection principle of synchronous timing is shown on the left side of Figure 6, and the corresponding synchronous timing sequence diagram is shown on the right side of Figure 6. When the module is inserted into the base station, the 14-pin spring pin interface contacts, and the module communicates with the base station via the SPI bus and is connected to the RTC synchronization line. As shown in the right figure of Figure 6, the base station's RTC chip sends a second pulse every second as an interrupt transmission to the base station main control board. When time synchronization needs to be completed, the main control board performs the following steps: (2.1) First, the timestamp corresponding to the next synchronization is sent to each module via the SPI bus according to the communication protocol. This timestamp is the timestamp corresponding to the next arrival of the base station RTC pulse. SPI bus communication must be completed before the next arrival of the base station RTC pulse; (2.2) When the base station RTC pulse arrives, the main control board receives the pulse interrupt and immediately sends a pulse on the RTC synchronization line. After receiving the pulse, the module immediately writes the timestamp previously transmitted to it via the SPI bus into the RTC chip. After the above two steps (2.1) and (2.2), ignoring the interrupt processing time and the RTC chip setting write time (these times are fixed and can be compensated by software later), the RTCs of all devices can be synchronized at once. This synchronization can be performed at regular intervals. For example, the timing diagram on the right side of Figure 6 shows synchronization every 2 seconds.

[0068] The base station is set to perform the above synchronization every few seconds. Therefore, as long as the module is plugged into the base station, it will remain accurately synchronized with the base station's time system.

[0069] Step 3: Sampling time combination: Each signal acquisition module and base station combines the input signal sampling point generated under the command of the low-precision sampling clock with the high-precision RTC timestamp.

[0070] Sampling time integration is performed independently by each module and base station. It combines the input signal sampling points generated by the lower-precision sampling clock with the high-precision RTC timestamp. The principle of sampling time integration is shown in Figure 7. The left side of Figure 7 shows the hardware principle of sampling time integration, and the right side shows the corresponding sampling time integration timing diagram.

[0071] As described in step 2, when the module is plugged into the base station, the base station synchronizes timestamps every few seconds, ensuring that the module's internal RTC time remains highly consistent with the base station's. When the module is removed and used in experiments to record physiological signals such as electromyography, the sampling trigger clock used is an internal 2.048MHz active crystal oscillator, as shown on the left side of Figure 7. The frequency stability of civilian-grade MHz crystal oscillators is typically 20ppm or worse, making it impossible to use the time pulses generated by MHz crystal oscillators as absolute timestamps for synchronizing with other devices. In this case, the module's RTC clock, which has been calibrated and synchronized with the base station and boasts a frequency stability of 0.1ppm, is required to provide reliable absolute timestamps. Since this absolute timestamp is generated once per second, while the sampling pulse can range from 200 to 2000Hz depending on user configuration, and the RTC clock pulses and sampling pulses are generated asynchronously, it is necessary to align the RTC timestamps with the sampling clock pulses.

[0072] More specifically, in a typical example shown on the right side of Figure 7, the sampling clock pulse is generated at a frequency of 1kHz, and a signal sample is performed once each pulse is generated. The RTC clock pulse is generated once every 1s and carries a timestamp that is precisely synchronized with the base station and other modules. The arrival time of the pulse is asynchronous with the sampling clock pulse, that is, it may be generated between two sampling clock pulses. When the RTC clock pulse arrives, the module combines the timestamp T carried by the RTC clock pulse with the sampling pulse closest to the arrival time of the RTC clock pulse (for example, the n+2th sampling in Figure 7), that is, the timestamp T is recorded to the time closest to the sampling pulse (for example, the n+2th sampling in Figure 7), and the timestamp closest to the time when the sampling pulse (the n+2th sampling in the figure) is generated is T. It can be foreseen that after such combination, the maximum error between the timestamp of the actual generation time of the sampling pulse (the n+2th sampling in Figure 7) and T is where f sample The sampling rate is set, and the error does not accumulate over time and is within an acceptable range.

[0073] After the above operation, at every f sampleEach sampling point has one sampling point associated with it by the RTC and carries an RTC timestamp of full seconds. These timestamps and their associated sampling point positions are recorded for use in subsequent steps.

[0074] Step 4: Time axis normalization: Based on the combined timestamp, the host computer normalizes the time axis of the recorded signal acquisition modules and base station sampling data to form unified time sampling data.

[0075] The time axis normalization process uses an algorithm within the host computer to unify the time axes of the recorded sampling data from each module and base station, forming a unified time sampling for subsequent analysis. The principle of time axis normalization is shown in Figure 8. The top part of Figure 8 shows the principle of multi-module time axis normalization, and the bottom part of Figure 8 shows the details of the resampling operation.

[0076] During the experiment, the signals collected by each module and the base station were timestamped using the method shown in step 3, but the sampling rates and sampling data lengths of each device were still different. When the experiment is completed, the module is inserted into the base station and the local file is imported into the database, it is necessary to normalize the time axis of all imported sampling data files so that the sampling data of all modules and base stations have the same ideal sampling rate f sample , have the same number of sampling points N = f within 1s sample ×1s, and align the same timestamps between different modules and between modules and base stations to store them in the database for subsequent playback, display, and export of multi-module data merging.

[0077] The specific principle is shown in the upper part of Figure 8. The position corresponding to each time stamp dotted line (for example, the dotted line position corresponding to time T) is the sampling point associated with that time in step 3. In a typical example shown in the upper part of Figure 8, one base station corresponds to eight modules, then a n ,b n ,…,h n ,i n The base station sampling clock represents the actual number of samples taken by the first to eighth modules and the base station in the nth timestamp interval, respectively, based on the sampling clocks of the first to eighth modules and the base station. Generally speaking, these numbers deviate from the standard number of samples, N. The base station sampling clock is used to collect external analog signal inputs. In addition to functions related to module management and host computer communication, the base station can also collect and digitize eight channels of analog signal inputs from external devices (such as ergometers, ECG electrodes, single-channel EMG electrodes, etc.). These analog signals are sampled using the base station sampling clock described here. It is a higher-frequency, lower-precision clock with similar parameters to the module sampling clock, so the same method is used to process it in steps 1 to 4.

[0078] According to the sampling rate f sample, select a suitable signal resampling algorithm, resample the sampling number of each module and base station in each timestamp interval to the standard sampling number N, and then store the same number of sampling data of each module and base station in the same timestamp interval in the same standard sampling matrix to achieve time axis normalization. Specifically, for any module or base station x∈[1,9] (the first to eighth modules and base stations are numbered 1-9 respectively), its original sampling data matrix in the nth timestamp interval is It is an N'×M matrix, where N' is the actual number of samples and M is the total number of sampled channels. In a specific embodiment, the module has 73 channels and the base station has 8 channels. The following signal resampling algorithm is used:

[0079] Get an N×M standard sampling data matrix For all x∈[1,9] Perform the following splicing to obtain:

[0080] In the above formula, f(x) is a multidimensional resampling algorithm, which can be implemented using polynomial interpolation, Lagrange interpolation, sinusoidal interpolation and other algorithms according to actual needs. This is the final normalized output data matrix.

[0081] The lower portion of Figure 8 visually illustrates the details of a resampling operation. In this example, the standard sampling rate is 20 sps, and N is 20. Given the same input signal, the two modules have different sampling times and sampling rates, resulting in different numbers of sampling points within the T to T+1 timestamp range. The proposed method first combines the sampling points closest to the T and T+1 timestamps of the two modules into the corresponding timestamps. Then, using a resampling algorithm, the a and b sampling points within the two timestamps are resampled to N for concatenation.

[0082] To evaluate the synchronization error of the high-density EMG signal synchronization acquisition method under different conditions, an experiment was conducted using a base station and three modules (M1-M3) to evaluate the synchronization effect. First, calibration and timing were performed according to the methods described in steps 1 and 2 above, and the following tests were performed respectively:

[0083] 1. Test in charging state

[0084] While charging, the modules heat up, affecting the RTC temperature and operational stability. After one time synchronization, disable the base station's automatic timing and plug the three modules into the charger. After four hours of operation, use a logic analyzer to compare the RTC pulse-per-second arrival time errors (M1-M3 relative to the base station's error), as shown in Table 1.

[0085] Table 1

[0086] 2. Test in static running state

[0087] After one time synchronization, the three modules were ejected and placed at room temperature. After running for 4 hours, the RTC second pulse arrival time error was compared using a logic analyzer, as shown in Table 2.

[0088] Table 2

[0089] 3. Vibration test

[0090] Vibration can affect the stability of the RTC's internal quartz crystal oscillation. After synchronizing the four RTCs indoors, disconnect the three modules from the base station. After leaving them for 5 minutes, shake each module vigorously for 1 minute. Repeat this process five times. Using a logic analyzer, measure the RTC second pulse arrival time error, as shown in Table 3.

[0091] Table 3

[0092] 4. High and low temperature test

[0093] High temperature test: After synchronizing and removing the modules, the three modules were fixed together in the same box, heated at 40 degrees Celsius for 5 minutes, and naturally cooled for 5 minutes. The above cycle was repeated 4 times, and then waited for 10 minutes at room temperature. The data was tested using a logic analyzer.

[0094] Low-temperature test: After RTC synchronization, place the box in the refrigerator, close the door and let it cool for 20 minutes. Then take the box out and warm it up at room temperature for 20 minutes. Repeat this cycle four times and test the data using a logic analyzer.

[0095] The high temperature test results are shown in Table 4.

[0096] Table 4

[0097] The low temperature test results are shown in Table 5.

[0098] Table 5

[0099] 5. Test conclusion

[0100] The test data in Tables 1 to 5 show that under the test environments of normal charging, normal use, high temperature, low temperature, and severe vibration, the method of this application can keep the RTC synchronization error less than 3ms within 4 hours, which is far less than the synchronization error requirement in industry usage scenarios.

[0101] Compared with traditional electromyographic signal acquisition systems, this application has at least the following advantages.

[0102] In traditional technical solutions, the module's charging, data export, status reading, parameter configuration, firmware upgrade, and time synchronization functions are completed using separate processes. For example, charging uses a charging cable, data export uses a TF card and card reader, status reading, parameter configuration, and firmware upgrade use OTA, and time synchronization uses wireless network synchronization. The steps are scattered, the process is complicated, time-consuming, and prone to errors. The device usage process in this application is simple, and all the above operations can be completed by simply inserting the module into the base station. This advantage comes from the setting of the multi-function pin interface of this application, which contains multiple communication protocols and function ports. This heterogeneous multi-function port can support multiple operations simultaneously.

[0103] In conventional technical solutions, if there is a base station for connecting modules, these base stations only electrically connect the modules to them without a mechanical locking structure. Therefore, critical communication points between the module and the base station (for example, when data is being imported into the module or when a configuration file is being written to the module) may be interrupted by user operations or unexpected events (for example, a user accidentally unplugs the module being read or written, or an external force causes the module to accidentally detach from the base station), which will cause damage to the module and the data files in the base station. The electromagnetic lock structure proposed in this application can lock the module inserted into the base station and continuously lock the module at critical communication points between the module and the base station to prevent it from accidentally popping out, thereby ensuring reliable communication between the module and the base station and the security and integrity of data files.

[0104] Traditional EMG signal acquisition systems generally collect single-channel or dual-channel macro EMG signals. Traditional wearable high-density EMG acquisition systems can only support high-density EMG acquisition with a maximum of two acquisition modules and a total of 128 channels (64 channels per module), covering only two muscle groups at most. However, this application supports high-density EMG acquisition with at least eight modules, 64 channels per module, for a total of 512 channels, covering more muscle groups and collecting more information.

[0105] Traditional distributed electromyography acquisition systems (whether traditional low-density electromyography or high-density electromyography) can only operate multiple modules simultaneously when there is wireless network coverage. If the wireless connection is interrupted, the signal is unstable, the subject moves beyond the coverage range of the wireless signal, etc., time synchronization cannot be achieved, so that the signal is unavailable. The time synchronization between modules and between modules and base stations in this application does not rely on wireless networks. In the absence of a network connection, electromyographic data and its corresponding accurate absolute timestamp can still be recorded offline. Therefore, the usage scenario is not limited to areas with network coverage and the synchronization reliability is high. In theory, the method of this application can be extended to any number of modules and has broad application prospects.

[0106] Traditional wireless wearable high-density electromyography acquisition equipment has poor reliability and is susceptible to electromagnetic interference due to its reliance on real-time wireless connections. When using wireless synchronization, all modules are required to be connected to the specified wireless network in real time, which limits their use and cannot be used in scenarios where there are no spatial constraints, such as outdoors. Wireless signals are easily affected by electromagnetic environmental factors, and in the event of transmission delays or interruptions such as electromagnetic wave reflection and disconnection, uncontrollable synchronization delays will occur that vary between modules. The need to maintain a wireless connection at all times consumes power, affects battery life, and shortens the maximum experiment time. In addition, sampling cannot be synchronized in an offline state without a network connection. The multiple modules of this application (theoretically, any number is supported) can be fixed at various positions on the body to measure various muscle groups at the same time; the self-locking base station locks the inserted modules, which can ensure the stability of operations such as charging, data transmission, configuration reading and writing, and firmware upgrades; the base station and the module use a 14-pin probe interface, which includes a variety of bus interfaces and functional ports. You only need to insert the module into the base station to complete charging, data transmission, configuration reading and writing, firmware upgrades, time synchronization and other operations; and this application realizes offline synchronization between multiple modules and between modules and base stations based on the coordination of software and hardware. The synchronization acquisition method of this application does not rely on wireless networks. It automatically completes the operation when the module is inserted into the base station, keeps all modules inserted in the base station synchronized with the base station in real time, and ensures that the module is still reliably synchronized with the base station within 4 hours after leaving the base station, greatly improving the synchronization reliability.

[0107] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A high-density electromyographic signal synchronous acquisition system, characterized in that: include: Host computer, base station and multiple signal acquisition modules; The signal acquisition module and the host computer are respectively connected to a wireless router node via Wi-Fi. Under the routing forwarding of the wireless router, the host computer establishes a TCP server, and the signal acquisition module connects to the TCP server as a client to realize socket communication; the host computer sends instructions to the signal acquisition module to start and end the experiment; The signal acquisition module is connected to the flexible electrode during the experiment and fixed to the position of the muscle to be tested on the human body, and is used to collect, process, store and send high-density surface electromyography signals in real time; After the experiment, multiple signal acquisition modules were electrically connected to the base station through a spring pin interface and mechanically connected through an electromagnetic lock; the base station was connected to the host computer through a USB 2.0 bus; The base station is used to store multiple signal acquisition modules and communicate with the host computer, charge the signal acquisition modules, and read, configure, synchronize time, and eject the signal acquisition modules according to the host computer's instructions. The base station also has a multi-channel analog signal input interface for collecting analog signals from external devices and digitizing them for transmission to the host computer. The host computer provides a user graphical operation interface for managing and configuring the signal acquisition module, and receiving, storing and processing the offline sampling data uploaded in real time to complete the synchronization of high-density surface electromyography signals.

2. The high-density electromyographic signal synchronous acquisition system according to claim 1, characterized in that: The pin interface includes an SPI bus, a USB bus, a charging interface, a hot-swap interface, a synchronization interface and a firmware upgrade interface; the SPI bus connects the base station and all signal acquisition modules, and is used by the base station to read the signal acquisition module status, configuration module, and synchronization module time; the USB bus is connected to the USB HUB in the base station to enable the signal acquisition module to transmit offline saved data to the host computer; the charging interface is used to charge the signal acquisition module, and the signal acquisition module automatically charges after being inserted into the base station, and automatically stops charging after charging is completed; the hot-swap interface is used to bidirectionally identify whether the signal acquisition module is connected to the base station. When the signal acquisition module is inserted or ejected, the signal acquisition module and the base station will receive hot-swap feedback; the synchronization interface is used to synchronize the real-time clock system between the signal acquisition module and the base station and each signal acquisition module, and cooperate with SPI bus instructions, software and algorithms to achieve microsecond time synchronization; the firmware upgrade interface is used to upgrade the internal firmware of the module online.

3. The high-density electromyographic signal synchronous acquisition system according to claim 1, characterized in that: The electromagnetic lock consists of a lock body and a latch; the lock body is fixed to the base station by screws, and the latch cooperates with the lock hole at the corresponding position of the signal acquisition module to fix the signal acquisition module in the base station; the electromagnetic lock has two states: locked and unlocked; when the signal acquisition module is inserted into the base station, the electromagnetic lock latch slides into the lock hole of the signal acquisition module, locking the signal acquisition module and sending insertion feedback to the base station; when the electromagnetic lock receives the pop-up command from the base station, the lock body is powered on, the latch is withdrawn from the lock hole to unlock and pop out the signal acquisition module.

4. The high-density electromyographic signal synchronous acquisition system according to claim 1, characterized in that: The host computer sends instructions to the base station through the CDC protocol based on the USB 2.0 bus, controlling the base station to complete the status reading and writing, experimental configuration, firmware upgrade and pop-up control operations of the signal acquisition module inserted therein; the host computer is also connected to the signal acquisition module through the USB HUB in the base station through the USB 2.0 bus, and reads the offline sampling data stored in the SD card of the signal acquisition module through the USB mass storage device protocol.

5. A high-density electromyographic signal synchronous acquisition method based on the high-density electromyographic signal synchronous acquisition system according to claim 1, characterized in that: include: Use the coherent calibration method to calibrate the extremely high-precision RTC clocks in each signal acquisition module and base station using an oscilloscope at room temperature; After coherent calibration, the current timestamps of the RTC clocks of all devices are synchronized using the software and hardware of the base station, signal acquisition module, and host computer. Each signal acquisition module and base station combines the input signal sampling point generated under the command of the low-precision sampling clock with the high-precision RTC timestamp; Based on the combined timestamps, the host computer normalizes the time axis of the recorded sampling data of each signal acquisition module and base station to form unified time sampling data.

6. The method for synchronously collecting high-density electromyographic signals according to claim 5, wherein: The coherent calibration method is used to calibrate each signal acquisition module and the extremely high-precision RTC clock in the base station using an oscilloscope at room temperature, specifically including: At room temperature, connect the output of the base station RTC crystal oscillator to channel 1 of a dual-channel oscilloscope, and connect the output of the RTC crystal oscillator in the signal acquisition module to channel 2 of the dual-channel oscilloscope. Use synchronous sampling within the oscilloscope, set channel 1 to trigger on the rising edge or intermediate level value, and display the signals of channels 1 and 2 simultaneously. When the channel 2 signal moves to the left relative to the channel 1 signal, gradually increase the aging parameter until the channel 2 signal stops moving. When the channel 2 signal moves to the right relative to the channel 1 signal, gradually decrease the aging parameter until the channel 2 signal stops moving.

7. The method for synchronously collecting high-density electromyographic signals according to claim 5, wherein: The method of synchronizing the current timestamps of the RTC clocks of all devices using the software and hardware of the base station, the signal acquisition module, and the host computer specifically includes: When the signal acquisition module is inserted into the base station, it communicates with the base station via the SPI bus and is connected to the RTC synchronization line; the RTC chip of the base station sends a second pulse every second as an interrupt transmission to the base station main control board; when a time synchronization needs to be completed, the base station main control board first sends the timestamp that should correspond to the next synchronization to each signal acquisition module through the SPI bus in accordance with the communication protocol. This timestamp is the timestamp corresponding to the next arrival of the base station RTC pulse. The SPI bus communication needs to be completed before the next arrival of the base station RTC pulse; when the base station RTC pulse arrives, the base station main control board receives the pulse interrupt and immediately sends a pulse on the RTC synchronization line. After receiving the pulse, each signal acquisition module immediately writes the timestamp previously transmitted to it through the SPI bus into the RTC chip, thereby synchronizing the RTCs of all devices at once.

8. The method for synchronously collecting high-density electromyographic signals according to claim 5, wherein: Each signal acquisition module and the base station combines the input signal sampling point generated under the command of the low-precision sampling clock with the high-precision RTC timestamp, specifically including: When the signal acquisition module is popped up by the base station for experiment, the signal acquisition module pulse performs a signal sampling when each sampling pulse is generated; the RTC clock pulse of the signal acquisition module is generated once at a preset time and carries a timestamp that is precisely synchronized with the base station and other signal acquisition modules; when the RTC clock pulse arrives, the signal acquisition module combines the timestamp T carried by the RTC clock pulse with the sampling pulse closest to the arrival time of the RTC clock pulse, and determines that the timestamp closest to the moment when the sampling pulse is generated is T.

9. The method for synchronously collecting high-density electromyographic signals according to claim 5, wherein: The host computer normalizes the time axis of the recorded sampling data of each signal acquisition module and base station based on the combined timestamp to form unified time sampling data, specifically including: When the experiment is over, the data acquisition module is inserted into the base station and the local files are imported into the database. For all the imported sampling data files, the sampling rate f sample , the sampling number of each signal acquisition module and base station in each timestamp interval is resampled to the standard sampling number N through the signal resampling algorithm; then the sampling data of the same number of signal acquisition modules and base stations in the same timestamp interval are stored in the same standard sampling matrix to achieve time axis normalization.

Images