Multi-modal biosensor monitoring sock system

Abstract

The invention relates to the technical field of intelligent wearable equipment, and particularly discloses a multi-mode biosensor monitoring sock system. The system comprises a flexible sock body, a multi-mode sensor array integrated on the flexible sock body, a master control synchronization module and a data processing module. The master control synchronization module generates global hardware synchronization pulses based on a single high-precision crystal oscillator, directly triggers analog-to-digital converters of all pressure, inertia and myoelectricity sensor channels, and realizes synchronous sampling of a hardware level. And the data processing module gives a uniform timestamp to the synchronous data stream and then sends the synchronous data stream. According to the method, high-precision time alignment of multi-modal signals is ensured, a reliable basis is provided for subsequent data fusion and gait analysis, and the power consumption and complexity of the system are reduced.

Description

Technical Field

[0001] This invention belongs to the field of smart wearable device technology, specifically relating to a multimodal biosensor monitoring sock system. Background Technology

[0002] In the field of smart wearables and remote health monitoring, the continuous and non-invasive monitoring of human physiological signals by integrating multiple biosensors has become an important development direction. As a key part of the human body for weight-bearing and movement, the biomechanical parameters and gait information of the feet are of great value for movement analysis, rehabilitation assessment, and chronic disease management.

[0003] The biosensor monitoring sock system based on smart textiles aims to flexibly integrate sensors of multiple modalities, such as pressure, inertial measurement units, and electromyography, into the sock body to achieve simultaneous acquisition and analysis of plantar pressure distribution, joint motion angles, and muscle activity. The system's fundamental goal is to provide a more comprehensive and accurate assessment of gait and physiological status through multi-source information fusion.

[0004] In existing technologies, multimodal biosensor monitoring systems face technical bottlenecks. To achieve wireless data transmission, each sensor module typically employs independent analog-to-digital converters and Bluetooth communication units. Due to individual differences in hardware crystal oscillators and wireless channel competition, the data sampling times of different sensor nodes exhibit unavoidable asynchronicity and random timing jitter.

[0005] Existing software-based Bluetooth synchronization methods cannot eliminate this timing discrepancy at the hardware level, resulting in data streams from pressure, inertial, and electromyography sensors not being strictly aligned on the timeline. This timing mismatch of multi-source data introduces serious feature extraction errors and recognition error rates when performing analyses requiring high-precision time correlation, such as gait event segmentation and gait phase recognition, directly affecting the reliability and clinical usability of monitoring results.

[0006] Therefore, how to achieve high-precision synchronous sampling at the hardware level of multimodal biosensors to ensure the effectiveness of data fusion has become an urgent technical challenge. Summary of the Invention

[0007] The purpose of this invention is to provide a multimodal biosensor monitoring sock system to solve the problem of multi-source data timing mismatch caused by the use of independent communication units for each sensor module in the prior art, thereby ensuring the effectiveness of high-precision synchronous sampling and data fusion of multimodal biosignals such as pressure, inertia and electromyography at the hardware level.

[0008] This invention provides a multimodal biosensor monitoring sock system, which includes a flexible sock body, a multimodal sensor array integrated on the flexible sock body, a main control synchronization module, a data processing and communication module, and a flexible power supply module for powering the above modules.

[0009] The flexible sock body adopts an elastic knitted structure, and sensor integration areas are pre-positioned at specific anatomical locations on the sole, instep, and posterior calf. The multimodal sensor array consists of distributed pressure sensor units, inertial measurement unit clusters, and surface electromyography sensor units, each embedded in its corresponding sensor integration area.

[0010] The main control synchronization module is the core timing control center of the system, which includes a high-precision crystal oscillator, a global synchronization signal generator, and a multi-channel synchronous sampling controller. The high-precision crystal oscillator generates the system's unique reference clock signal.

[0011] The global synchronization signal generator periodically generates hardware synchronization pulse signals with fixed pulse widths based on this reference clock signal.

[0012] The multi-channel synchronous sampling controller receives the hardware synchronization pulse signal and sends a unified sampling start command to the analog-to-digital converters of all sensor channels accordingly.

[0013] Furthermore, the distributed pressure sensor unit consists of multiple flexible piezoresistive sensing nodes arranged in an array in the main weight-bearing area of ​​the sole of the foot.

[0014] Each flexible piezoresistive sensing node is connected to a multiplexer via an analog signal line, and the output of the multiplexer is connected to a shared high-precision analog-to-digital converter.

[0015] The sampling trigger pin of the shared high-precision analog-to-digital converter is directly controlled by the multi-channel synchronous sampling controller of the main control synchronization module.

[0016] When the synchronous sampling start command is issued, the multiplexer, driven by the control logic, sequentially switches the analog signals of each pressure sensing node to a shared analog-to-digital converter for digitization in a preset scanning order, thereby ensuring that the sampling time of all pressure channels is strictly aligned with the rising edge of the same hardware synchronization pulse.

[0017] In one embodiment of the present invention, the inertial measurement unit cluster includes a combination of a triaxial accelerometer and a triaxial gyroscope respectively fixed on the dorsum of the foot, the outer side of the ankle joint, and the back of the calf.

[0018] Each inertial measurement unit integrates a dedicated analog-to-digital converter. The synchronization input pins of all inertial measurement units are connected in parallel to the hardware synchronization pulse signal line output by the global synchronization signal generator of the main control synchronization module.

[0019] When the hardware synchronization pulse signal arrives, the analog-to-digital converters inside each inertial measurement unit are simultaneously triggered, and synchronous sampling of acceleration and angular velocity signals begins.

[0020] Furthermore, the surface electromyography sensor unit adopts a dry electrode design, which includes a pair of differential detection electrodes and a reference electrode, and is attached to the belly of the gastrocnemius muscle in the calf.

[0021] The surface electromyography (EMG) sensor unit internally includes a preamplifier, a bandpass filter, and an independent analog-to-digital converter (ADC). The external trigger pin of this independent ADC is also connected to the hardware synchronization pulse signal line output by the global synchronization signal generator.

[0022] Therefore, the sampling time of electromyographic signals is precisely aligned with the sampling time of pressure signals and inertial signals by the same hardware synchronization pulse.

[0023] The data processing and communication module includes a microprocessor and a Bluetooth Low Energy transceiver. The microprocessor reads synchronized digital data from the multi-channel synchronous sampling controller of the main control synchronization module and the analog-to-digital converters of each sensor unit via a high-speed serial peripheral interface bus.

[0024] The microprocessor runs data preprocessing and packaging firmware, which performs timestamp marking, data packaging and caching operations on the read synchronous data stream.

[0025] The timestamp is generated based on the reference clock counter of the master control synchronization module, giving each set of synchronously sampled multimodal data a unified absolute time identifier.

[0026] The encapsulated data packet is then sent to the external terminal device as a single data stream via a Bluetooth Low Energy transceiver.

[0027] The flexible power module is a rechargeable flexible lithium polymer battery. It is connected to all power-consuming modules in the system through flexible circuits and provides multiple voltage levels after voltage regulation to meet the power supply requirements of different sensors and chips.

[0028] In a preferred embodiment, the multi-channel synchronous sampling controller of the main control synchronization module also integrates a programmable sampling rate configuration register.

[0029] Users or external terminal devices can write instructions to this register via Bluetooth communication to dynamically adjust the frequency of the global synchronization pulse signal, thereby steplessly adjusting the synchronization sampling rate of the entire multimodal sensor array within the range of 50 Hz to 1000 Hz to adapt to the data rate requirements of different monitoring scenarios.

[0030] Furthermore, the microprocessor of the data processing and communication module also executes an online data quality verification algorithm. This algorithm analyzes the synchronous data stream of each sensor channel in real time, checking whether the data is continuous and whether the values ​​are within a preset reasonable physiological range. When an anomaly is detected in a certain channel's data, the microprocessor embeds an error flag bit in the corresponding data packet and attempts to diagnose the fault source through the built-in sensor self-test program. The diagnostic results are then reported to the external terminal device.

[0031] In a more preferred embodiment of the present invention, the system further includes an ambient temperature and humidity compensation unit. This unit is integrated into the non-weight-bearing area of ​​the arch of the flexible sock and includes a temperature and humidity sensor.

[0032] The analog-to-digital converter trigger pin of the temperature and humidity sensor is also connected to the global synchronization pulse signal line, so that its sampling is synchronized with the environmental data.

[0033] During the data preprocessing stage, the microprocessor performs real-time compensation and correction on the readings of the piezoresistive pressure sensor based on the synchronously collected ambient temperature and humidity data, in order to eliminate the influence of environmental factors on the accuracy of pressure measurement.

[0034] Furthermore, the sensor integration area of ​​the flexible sock body adopts a multi-layer composite structure.

[0035] The innermost layer is a skin-friendly conductive fabric layer, which is used to maintain stable contact with the skin and serve as electrodes for part of the sensor; The middle layer is a flexible printed circuit layer, on which sensor connection traces and power / signal buses are etched. The outermost layer is a protective encapsulation layer, which is coated or laminated with silicone or thermoplastic polyurethane materials to ensure the system's washing reliability and mechanical durability.

[0036] The data fusion of the multimodal sensor array is completed in an external terminal device or cloud server.

[0037] The external processing unit receives a synchronous multimodal data stream with a unified timestamp sent by the monitoring system.

[0038] Based on this, a high-precision gait event detection algorithm can be executed. This algorithm comprehensively utilizes the synchronized plantar pressure center trajectory, ankle joint angular velocity changes, and calf electromyography burst mode. Through multi-feature joint decision-making, it accurately identifies key event points in the gait cycle, such as heel strike, full foot flat, and toe lift-off.

[0039] Furthermore, gait phase segmentation and parameter calculation can be performed to generate a comprehensive evaluation report including stride length, stride speed, gait symmetry, joint range of motion, and muscle synergistic activation patterns.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention introduces a hardware-level master control synchronization module. A single high-precision crystal oscillator generates a reference clock and produces a unified hardware synchronization pulse signal to directly trigger the analog-to-digital conversion process of all sensor channels. This eliminates the asynchronous sampling timing and random jitter caused by crystal oscillator differences and channel competition among multiple independent wireless communication modules. This invention ensures sub-millisecond high-precision time alignment of multimodal biosignals such as pressure, inertia, and electromyography at the data acquisition source. This lays a reliable hardware foundation for subsequent multi-source data fusion and advanced gait analysis that require strict time correlation, improving the temporal consistency and clinical usability of monitoring data.

[0041] 2. This invention employs a combination of a shared analog-to-digital converter (ADC) and a multiplexer to process a distributed pressure sensor array. This not only achieves synchronous sampling of all pressure nodes but also significantly reduces the number of ADCs required by the system, lowering overall power consumption and hardware costs. Simultaneously, the inertial measurement unit and the electromyography (EMG) sensor unit are connected in parallel to the same synchronization pulse signal line for hardware triggering. This results in a simple and efficient system architecture, ensuring the reliability and stability of the synchronization mechanism and avoiding the additional delays and uncertainties caused by complex software protocol synchronization.

[0042] 3. This invention designs an auxiliary function system including online data quality verification, synchronous compensation for environmental temperature and humidity, and programmable sampling rate adjustment. Data quality verification ensures the integrity and validity of the data; synchronous environmental compensation improves the accuracy and robustness of pressure measurement in varying usage scenarios; and the programmable sampling rate gives the system the flexibility to adapt to different applications, from daily activity monitoring to high-speed motion analysis. These features together constitute a high-performance, highly reliable, and user-configurable complete monitoring system solution, transcending the existing technology level that only focuses on data acquisition. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the overall technical architecture of the multimodal biosensor monitoring sock system proposed in this invention; Figure 2 This is a schematic diagram of the core principle framework of the main control synchronization module in this invention to realize hardware synchronous sampling of multi-modal signals; Figure 3 This is a flowchart illustrating the synchronous data acquisition and preprocessing logic of the multimodal sensor array (pressure, inertia, electromyography) in this invention. Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow between the data processing and communication module and the external terminal device in this invention; Figure 5 This is a flowchart of the external processing workflow for gait analysis and evaluation based on synchronous multimodal data in this invention. Detailed Implementation

[0044] This embodiment details the specific implementation architecture and workflow of the multimodal biosensor monitoring sock system. Please refer to the appendix. Figures 1 to 5 The system physically takes the form of a flexible sock that can be worn on the user's feet and calves, and integrates a complete electronic system for biosignal acquisition, synchronization, processing and wireless transmission.

[0045] The core objective of the system is to ensure that multimodal biosignals from sensors based on different physical principles are aligned with high precision at the sampling time in the sub-millisecond level through unified timing control at the hardware level, thereby providing a raw data foundation with extremely high temporal consistency for subsequent precise gait analysis and biomechanical research.

[0046] The flexible sock body, as the physical carrier and wearing interface of the entire system, directly affects the signal quality of the sensors, the user's wearing comfort, and the overall durability of the system through its design and manufacturing process. Please refer to the appendix. Figure 1 The flexible socks are made of a highly elastic and breathable knitted structure, usually made of a blend of spandex or nylon, to ensure a good fit and movement for the feet and calves.

[0047] Specially designed sensor integration areas are pre-positioned at specific anatomically critical locations on the sock.

[0048] These areas are mainly distributed on the sole, dorsum, and back of the lower leg. The sole area is further subdivided into the forefoot, arch, and heel areas, which are the main weight-bearing and pressure distribution areas of the foot during the gait cycle.

[0049] The dorsum of the foot is primarily used to fix inertial measurement units that monitor the spatial motion of the foot. The gastrocnemius muscle belly on the posterior side of the lower leg is the optimal location for acquiring surface electromyography (EMG) signals.

[0050] To achieve stable sensor integration and reliable electrical connection, the sensor integration area of ​​the flexible sock body adopts a multi-layer composite structure. The innermost layer is a skin-friendly conductive fabric layer, which is in direct contact with the user's skin.

[0051] Its material is silver-plated nylon or carbon fiber blended fabric, which has good conductivity, flexibility and biocompatibility.

[0052] For the surface electromyography sensor unit, the conductive fabric layer itself constitutes part of its differential detection electrode and reference electrode, ensuring a stable contact impedance with the skin.

[0053] For other sensor areas, this layer primarily serves as the underlying fixation and electromagnetic shielding layer. The middle layer is a flexible printed circuit layer, which forms the electrical framework of the system.

[0054] This layer uses polyimide or polyester film as a substrate, and a precise copper trace pattern is formed on it through photolithography and etching processes.

[0055] These traces include analog signal lines connecting various sensor nodes, synchronization pulse signal lines connecting the main control synchronization module, data buses connecting the data processing and communication modules, power buses connecting the flexible power supply modules, and serial peripheral interface buses and general-purpose input / output control lines for inter-module communication.

[0056] All stitching uses a serpentine or curved design to withstand the repeated stretching and bending stresses generated during wearing and exercise without breaking.

[0057] The outermost layer is a protective encapsulation layer, which completely seals the internal electronic components and flexible circuits to resist the corrosion of sweat, moisture, and dust, and to ensure that the system is washable.

[0058] This layer is typically formed by coating with liquid silicone, laminating with a thermoplastic polyurethane film, or a combination of both.

[0059] The packaging process must ensure that the pressure sensing area of ​​the pressure sensing node, the mounting plane of the inertial measurement unit, and the contact point of the electromyography electrode have precise openings or use specific sound- and pressure-permeable materials to ensure the normal functioning of the sensor.

[0060] The encapsulated system module has excellent flexibility, allowing it to bend and twist along with the sock and maintain its function under normal machine washing conditions.

[0061] The multimodal sensor array is the sensing front end of the system, responsible for converting the biomechanical and physiological activities of the foot and lower leg into electrical signals.

[0062] The array consists of three functional sub-units: Distributed pressure sensor unit, inertial measurement unit cluster, and surface electromyography sensor unit.

[0063] Please refer to the attached document. Figure 1 With appendix Figure 2 These three sub-units are physically distributed in different locations on the flexible sock body, but they are all controlled by the same main control synchronization module in terms of electricity and timing.

[0064] The distributed pressure sensor unit is responsible for capturing the dynamic pressure distribution during the contact between the sole of the foot and the ground.

[0065] The unit consists of multiple flexible piezoresistive sensing nodes arranged in a two-dimensional array in the main weight-bearing areas of the sole of the foot. A typical configuration is to arrange 8 nodes in the forefoot area, 4 nodes in the arch area, and 6 nodes in the heel area, for a total of 18 nodes.

[0066] Each flexible piezoresistive sensing node consists of two layers of flexible electrodes and a piezoresistive composite material sandwiched in the middle.

[0067] When pressure is applied to the sole of the foot, the resistance of the pressure-sensitive material changes linearly or nonlinearly in proportion to the pressure.

[0068] Each sensing node is led out through two independent analog signal lines, one of which is connected to a constant pressure excitation source, and the other outputs a voltage signal that reflects pressure changes.

[0069] In order to acquire signals from all pressure nodes in an economical and synchronous manner, this system adopts an architecture that combines a shared analog-to-digital converter and a multiplexer.

[0070] The analog output signal lines of all pressure sensing nodes are connected to different input channels of a high-speed analog multiplexer.

[0071] This multiplexer is typically a 16-to-1 or 32-to-1 model, and its channel selection is controlled by a set of digital address lines.

[0072] The single analog output of the multiplexer is connected to a high-precision, high-sampling-rate analog-to-digital converter, such as a 24-bit precision trigonometric integrator analog-to-digital converter.

[0073] The sampling trigger pin of the shared analog-to-digital converter, which is the start signal input terminal for initiating an analog-to-digital conversion, is directly connected to the dedicated synchronous output pin of the multi-channel synchronous sampling controller of the main control synchronization module.

[0074] The inertial measurement unit cluster is responsible for measuring the kinematic parameters of the foot and lower leg in three-dimensional space.

[0075] The cluster consists of three independent inertial measurement units, which are fixed at the center of the dorsum of the foot, the lateral side of the ankle joint, and the posterior side of the calf below the gastrocnemius muscle near the Achilles tendon.

[0076] Each inertial measurement unit is an integrated microelectromechanical system chip that encapsulates a three-axis accelerometer and a three-axis gyroscope.

[0077] A triaxial accelerometer measures linear acceleration along the three orthogonal axes of the chip, including gravitational acceleration and motion acceleration. A triaxial gyroscope measures angular velocity around the three orthogonal axes of the chip.

[0078] Each inertial measurement unit chip integrates a dedicated analog-to-digital converter to convert analog acceleration and angular velocity signals into digital values.

[0079] These internal analog-to-digital converters typically output data through the chip's serial peripheral interface or integrated circuit bus interface.

[0080] To achieve synchronous sampling, each inertial measurement unit chip is equipped with an external synchronization input pin.

[0081] In this system, the external synchronization input pins of all three inertial measurement units are electrically connected in parallel to a single hardware synchronization pulse signal line output by the global synchronization signal generator of the main control synchronization module.

[0082] The surface electromyography sensor unit is responsible for collecting electrophysiological activity signals of the calf muscles during exercise.

[0083] The unit employs a dry electrode design to enhance wearability and comfort, and includes a pair of differential detection electrodes and a reference electrode.

[0084] The electrode material is medical stainless steel or gold-plated, and it forms an electrical connection with the skin through conductive gel or the aforementioned skin-friendly conductive fabric layer.

[0085] The electrodes are attached at the most prominent points of the medial and lateral heads of the gastrocnemius muscle, according to international standards, while the reference electrode is placed on the nearby bone surface where there is no muscle activity.

[0086] The surface electromyography sensor unit contains an analog signal conditioning circuit, which is typically integrated into a dedicated biopotential amplifier chip.

[0087] The circuit includes: A preamplifier with high input impedance and high common-mode rejection ratio is used to amplify microvolt-level electromyographic signals; Bandpass filters, typically with a passband frequency range of 20 Hz to 500 Hz, are used to filter out ECG artifacts, motion artifacts, and high-frequency noise. And programmable gain amplifiers.

[0088] The conditioned analog electromyographic signals are then sent to a separate, high-resolution analog-to-digital converter for digitization.

[0089] Similar to the inertial measurement unit, this independent analog-to-digital converter also has an external trigger pin, which is connected to the same hardware synchronization pulse signal line output by the global synchronization signal generator of the main control synchronization module.

[0090] The main control synchronization module is the core and timing control center for achieving hardware-level synchronization in the entire system. Please refer to the appendix. Figure 2 Physically, this module is typically a tiny printed circuit board that integrates specific functional integrated circuits and interconnects with other parts of the system via flexible connectors.

[0091] Its core function is to generate a unique and highly stable timing reference for the entire system and generate a unified trigger signal to command all sensor channels to start a data sampling simultaneously via hardware interrupt.

[0092] The core of the main control synchronization module is a high-precision temperature-compensated crystal oscillator.

[0093] The crystal oscillator generates a highly stable square wave clock signal, such as 16 MHz or 32 MHz. This clock signal serves as the sole time reference for all digital logic and timing operations in the entire system.

[0094] Its frequency stability is typically better than ±10ppm, meaning the error is extremely small every 1 / 1000000 seconds. This ensures the consistency of the time base during long-term sampling from the source and avoids the problem of gradual misalignment of the data time axis caused by the drift differences of multiple independent clock sources.

[0095] A global synchronization signal generator is a functional unit based on digital logic circuits or the internal timer peripheral of a microcontroller.

[0096] It takes the output clock of the aforementioned high-precision crystal oscillator as input, performs programmable frequency division and counting operations, and periodically generates hardware synchronization pulse signals with fixed pulse widths.

[0097] The frequency of this pulse signal is the global sampling rate of the entire multimodal sensor array.

[0098] The rising or falling edge of the pulse is defined as the precise alignment moment for all sensor samples.

[0099] For example, when the system is set to a sampling rate of 200 Hz, the global synchronization signal generator generates a positive pulse with a width of 10 microseconds every 5 milliseconds.

[0100] The pulse signal is output to various parts of the system through a dedicated low-impedance trace.

[0101] The multi-channel synchronous sampling controller is the actuator of the main control synchronization module.

[0102] It receives hardware synchronization pulse signals from a global synchronization signal generator and generates a series of precisely synchronized control commands accordingly.

[0103] One of its core tasks is to send sampling start commands to the shared analog-to-digital converter of the distributed pressure sensor unit.

[0104] When the rising edge of the synchronization pulse arrives, the multi-channel synchronous sampling controller immediately generates a start pulse that meets the timing requirements of the analog-to-digital converter on its output line connected to the trigger pin of the pressure analog-to-digital converter, commanding it to start an analog-to-digital conversion.

[0105] Meanwhile, the multi-channel synchronous sampling controller is also responsible for controlling the multiplexing logic of pressure signal acquisition.

[0106] After the pressure analog-to-digital converter starts the conversion, the controller outputs the channel address of the next pressure node to be sampled through its digital address lines, driving the multiplexer to switch to the corresponding sensing node.

[0107] When the next synchronization pulse arrives, the analog-to-digital converter is triggered again to convert the new node signal that has been switched to the output.

[0108] This process is repeated, with all pressure nodes being scanned and converted sequentially within the sampling period. However, since each conversion is triggered by the same global synchronization pulse, the effective sampling times of all pressure nodes are strictly aligned with the pulse edge, even though their digital readings are read out sequentially.

[0109] For inertial measurement unit clusters and surface electromyography (EMG) sensor units, the synchronization mechanism is more direct. The hardware synchronization pulse signal line output by the global synchronization signal generator of the main control synchronization module is directly connected to the external trigger pins of all inertial measurement units and EMG analog-to-digital converters in a star or bus topology.

[0110] When the synchronization pulse arrives, the analog-to-digital converters inside these units are simultaneously hardware-triggered and immediately begin sampling and converting the current acceleration, angular velocity, and electromyographic voltage signals.

[0111] This hardware triggering method eliminates the time uncertainty caused by any software instruction delay, achieving true simultaneous sampling.

[0112] The data processing and communication module serves as the system's intelligent hub and external interface. Please refer to the appendix. Figure 3 With appendix Figure 4 The core of this module is a low-power microprocessor, such as a 32-bit microcontroller based on the ARM Cortex-M series.

[0113] The microprocessor communicates with the main control synchronization module and the analog-to-digital converters of each sensor unit via a high-speed serial peripheral interface bus to read the synchronously sampled digital data.

[0114] The data reading process is strictly time-sequential. After each global synchronization pulse triggers all sensors to sample, the microprocessor waits for a preset time interval, slightly longer than the conversion time of all analog-to-digital converters, to ensure that the data is ready. Subsequently, the microprocessor initiates a series of serial peripheral interface transactions.

[0115] It first reads the 32-bit system timestamp counter value from the main control synchronization module. This counter is driven by a high-precision crystal oscillator and latches the current count value when each synchronization pulse arrives, assigning an absolute time tag to this synchronization sampling event.

[0116] Next, the microprocessor reads the digital pressure value of the currently active node from the shared analog-to-digital converter of the pressure sensor unit.

[0117] Then, it accesses the three inertial measurement units sequentially through the serial peripheral interface and reads the register data of the six-axis acceleration and angular velocity inside each unit.

[0118] Finally, it reads the digital value of the electromyographic voltage from the independent analog-to-digital converter of the surface electromyography sensor unit.

[0119] This series of read operations is completed sequentially within the time window immediately following the sampling trigger, ensuring that data is acquired in a timely manner and avoiding loss.

[0120] The microprocessor runs dedicated data preprocessing and packaging firmware. The firmware first associates all the raw data read with the corresponding timestamps, combining them into a complete multimodal data record.

[0121] The data structure of this record contains the following fields: a 32-bit timestamp, 18 16-bit pressure values, 3 sets of 6 16-bit inertial data each, and 1 16-bit electromyography value.

[0122] Subsequently, the firmware performs a data packaging operation, caching multiple such records in the internal static random access memory, and encapsulating them according to the frame format of the Bluetooth Low Energy protocol, adding information such as packet header, packet serial number, and device identifier.

[0123] The encapsulated data packet is sent to the transmit buffer of the low-power Bluetooth transceiver integrated into the microprocessor or externally.

[0124] The Bluetooth Low Energy transceiver is responsible for establishing a wireless connection with external terminal devices and sending out the encapsulated data packets in the form of a single, continuous data stream.

[0125] The communication uses a reliable connection mode with an acknowledgment mechanism to ensure reliable data transmission over the wireless channel.

[0126] The microprocessor also manages Bluetooth connection parameters, such as connection intervals, to strike a balance between data throughput and system power consumption.

[0127] The flexible power module provides power to the entire system. It uses a thin, flexible lithium polymer battery with a capacity designed according to the system's continuous operating time requirements, such as 100 mAh.

[0128] The battery output is regulated and distributed through a high-efficiency DC-DC converter and multiple low-dropout linear regulators to produce various voltage levels required by the system, such as 3.3 volts for digital logic and microprocessors, 2.5 volts for analog circuits, and 1.8 volts for some sensor cores.

[0129] All power traces are designed with sufficient width on a flexible printed circuit layer and supplemented with bypass capacitors to reduce impedance and noise.

[0130] This system also includes a series of enhancement units to improve its performance and reliability.

[0131] The ambient temperature and humidity compensation unit is integrated into the non-weight-bearing area of ​​the arch of the flexible sock and includes a digital temperature and humidity sensor.

[0132] The sensor's data readout clock or trigger pin is also connected to the global synchronization pulse signal line, so that its environmental parameter sampling time is synchronized with other biological signals.

[0133] During the preprocessing stage, the microprocessor inputs the synchronously collected temperature and relative humidity data into a preset compensation model to perform real-time correction on the readings of each pressure sensor node.

[0134] The compensation model is based on the properties of piezoresistive materials, and its core is to correct the zero-point drift and sensitivity changes of the sensor caused by changes in ambient temperature and humidity.

[0135] This model can be expressed as a representation of the original pressure reading. Correction: ; It depends on the temperature and humidity The changing zero-point offset function, It depends on the temperature and humidity The sensitivity coefficient function of change.

[0136] These two functions were obtained through prior calibration experiments and are stored in the microprocessor's non-volatile memory in the form of lookup tables or polynomial coefficients.

[0137] Microprocessors according to current and Value, calculated in real time and This is applied to each pressure reading, resulting in a corrected pressure value that is unaffected by environmental interference. .

[0138] The online data quality verification algorithm is executed in real time by the microprocessor. This algorithm performs multiple checks on the synchronized data stream for each sensor channel.

[0139] First, a continuity check is performed, which determines whether data packets are lost by verifying the monotonically increasing nature of the timestamps and the expected interval.

[0140] Next, a range check is performed, comparing the readings of each sensor with the preset reasonable physiological range. For example, the plantar pressure should be between 0 and 1000 kPa, the acceleration should be within ±16g, and the electromyographic signal should be within ±2 mV.

[0141] Perform a rate of change check again to identify any unreasonable abrupt changes between adjacent sampling points, which may indicate poor sensor contact or impact.

[0142] When any check fails, the microprocessor sets a specific error flag in the corresponding data packet and records the error type and channel number.

[0143] At the same time, it can initiate built-in sensor self-test programs, such as sending test excitations to pressure sensors, reading the internal diagnostic registers of the inertial measurement unit, or measuring the contact impedance of electromyographic electrodes.

[0144] The self-test results, along with error flags, are reported to external terminal devices to provide users with clear fault diagnosis information.

[0145] The programmable sampling rate configuration function is implemented through registers within the master control synchronization module.

[0146] Users can send configuration commands to the microprocessor of the data processing and communication module via Bluetooth connection through an application on an external terminal device.

[0147] After the microprocessor parses the instructions, it writes the new frequency division coefficient value to the configuration register of the multi-channel synchronous sampling controller in the main control synchronization module through the serial peripheral interface bus.

[0148] This value directly changes the timer division ratio of the global synchronization signal generator, thereby dynamically adjusting the frequency of the hardware synchronization pulse signal. The sampling rate can be steplessly adjusted from 50 Hz to 1000 Hz.

[0149] Low sampling rates are suitable for long-term monitoring of daily activities to save power; high sampling rates are suitable for detailed analysis of high-speed movements such as running and jumping.

[0150] After the sampling rate is changed, the sampling times of all sensors in the system are automatically re-aligned according to the new frequency without any hardware modifications.

[0151] The system collects synchronous multimodal data streams with unified high-precision timestamps and sends them to external terminal devices, such as smartphones, tablets, or dedicated data receivers, via Bluetooth.

[0152] Please refer to the attached document. Figure 5 Subsequent advanced data processing and analysis are mainly completed on external terminal devices or further uploaded to cloud servers.

[0153] After receiving the data stream, the external processing unit first parses and reassembles it to recover the time-aligned pressure array data, inertial data, and electromyographic data.

[0154] Based on this synchronized data, a high-precision gait event detection algorithm can be executed. This algorithm comprehensively utilizes the spatiotemporal correlation of multimodal information.

[0155] For example, the algorithm simultaneously monitors the pressure rise edge of the pressure sensor in the heel region, the zero-crossing point of the angular velocity signal of the inertial measurement unit around the sagittal axis at the ankle joint from positive to negative, and the bursting start point of the electromyographic signal of the calf gastrocnemius muscle.

[0156] By setting up joint decision logic, when these three events occur successively within a very short time window, the system can determine the moment of occurrence of the "heel touching the ground" event with great accuracy, which is far greater than the method using only a single signal source.

[0157] Similarly, the "full foot flat" event can be determined by the movement of the center of pressure trajectory of the foot to the mid-foot and the stable pressure distribution; the "toe off the ground" event can be determined by the disappearance of forefoot pressure, the positive peak value of ankle joint angular velocity, and the weakening of gastrocnemius muscle electromyography signal.

[0158] Based on the accurate identification of key gait event points, complete gait cycle segmentation and parameter calculation can be performed.

[0159] A gait cycle is defined as the time interval between one heel strike and the next heel strike on the same side. Within each cycle, a series of quantitative parameters can be calculated: Time parameters such as gait cycle duration, standing phase duration, swing phase duration, and double support phase duration; Spatial parameters such as step size and step speed estimated based on inertial data integration; Symmetry parameters include the ratio of left and right step cycle durations and the ratio of left and right foot pressure peak values. Joint range of motion parameters, such as the range of motion of the ankle joint in the sagittal, coronal, and transverse planes calculated based on data from the lower leg and foot inertial measurement units; Muscle activity parameters include the root mean square value of electromyography (EMG) signals, integrated EMG values, muscle activation sequence, and duration.

[0160] All these parameters are aggregated to generate a comprehensive gait and biomechanical assessment report, which can be used for clinical rehabilitation assessment, motor performance analysis, or scientific research data collection.

[0161] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0162] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multimodal biosensor monitoring sock system, characterized in that, include: The flexible sock body uses an elastic knitted structure and has sensor integration areas pre-positioned in specific anatomical locations on the sole, instep, and back of the calf. A multimodal sensor array, integrated into the sensor integration area of ​​the flexible sock, includes a distributed pressure sensor unit, an inertial measurement unit cluster, and a surface electromyography sensor unit. The main control synchronization module, as the timing control center of the system, includes a high-precision crystal oscillator, a global synchronization signal generator, and a multi-channel synchronous sampling controller. The high-precision crystal oscillator is used to generate the system's unique reference clock signal; The global synchronization signal generator periodically generates hardware synchronization pulse signals based on the reference clock signal; The multi-channel synchronous sampling controller is used to receive the hardware synchronization pulse signal and send a unified sampling start command to the analog-to-digital converters of all sensor channels; The data processing and communication module includes a microprocessor and a low-power Bluetooth transceiver; The microprocessor is used to read synchronized digital data from the main control synchronization module and each analog-to-digital converter of the multimodal sensor array via a high-speed serial peripheral interface bus, and to perform timestamp marking, data packaging and caching operations on the read synchronized data stream to generate encapsulated data packets. The low-power Bluetooth transceiver is used to send the encapsulated data packet to an external terminal device in the form of a single data stream. The flexible power module is a rechargeable flexible lithium polymer battery that connects to all power-consuming modules in the system via flexible circuitry and provides multiple voltage levels with voltage regulation.

2. The multimodal biosensor monitoring sock system according to claim 1, characterized in that, The distributed pressure sensor unit consists of multiple flexible piezoresistive sensing nodes arranged in an array in the main weight-bearing area of ​​the sole of the foot. Each flexible piezoresistive sensing node is connected to a multiplexer via an analog signal line, and the output of the multiplexer is connected to a shared high-precision analog-to-digital converter. The sampling trigger pin of the shared high-precision analog-to-digital converter is directly controlled by the multi-channel synchronous sampling controller of the main control synchronization module; The inertial measurement unit cluster includes a combination of a triaxial accelerometer and a triaxial gyroscope, which are respectively fixed to the dorsum of the foot, the outer side of the ankle joint, and the back of the lower leg. Each inertial measurement unit integrates a dedicated analog-to-digital converter; the synchronization input pins of all inertial measurement units are connected in parallel to the hardware synchronization pulse signal line output by the global synchronization signal generator of the main control synchronization module. The surface electromyography sensor unit adopts a dry electrode design, including a pair of differential detection electrodes and a reference electrode, and is attached to the belly of the gastrocnemius muscle in the calf. The surface electromyography sensor unit internally includes a preamplifier, a bandpass filter, and an independent analog-to-digital converter; the external trigger pin of the independent analog-to-digital converter is connected to the hardware synchronization pulse signal line output by the global synchronization signal generator.

3. The multimodal biosensor monitoring sock system according to claim 2, characterized in that, The multi-channel synchronous sampling controller of the main control synchronization module also integrates a programmable sampling rate configuration register. Users or external terminal devices write instructions to the programmable sampling rate configuration register via Bluetooth communication to dynamically adjust the frequency of the hardware synchronization pulse signal output by the global synchronization signal generator, thereby steplessly adjusting the synchronization sampling rate of the entire multimodal sensor array within a specified Hertz range.

4. The multimodal biosensor monitoring sock system according to claim 3, characterized in that, The microprocessor of the data processing and communication module also executes an online data quality verification algorithm; The online data quality verification algorithm analyzes the synchronous data stream of each sensor channel in real time to check whether the data continuity and values ​​are within the preset reasonable physiological range. When an anomaly is detected, the microprocessor embeds an error flag in the corresponding data packet and diagnoses the fault source through the built-in sensor self-test program, then reports the diagnosis results to the external terminal device.

5. The multimodal biosensor monitoring sock system according to claim 4, characterized in that, The system also includes an environmental temperature and humidity compensation unit; The environmental temperature and humidity compensation unit is integrated into the non-weight-bearing area of ​​the arch of the flexible sock and includes a temperature and humidity sensor; the analog-to-digital converter trigger pin of the temperature and humidity sensor is connected to the hardware synchronization pulse signal line output by the global synchronization signal generator. During the data preprocessing stage, the microprocessor performs real-time compensation and correction on the readings of the distributed pressure sensor unit based on the synchronously collected ambient temperature and humidity data.

6. The multimodal biosensor monitoring sock system according to claim 5, characterized in that, The real-time compensation and correction process is as follows: Acquire the raw readings of the pressure sensor, and synchronously collect temperature and humidity data; based on the preset compensation model, calculate the zero-point offset function and sensitivity coefficient function as temperature and humidity change; Subtract the value of the zero-point offset function from the original reading, and then divide by the value of the sensitivity coefficient function to obtain the corrected pressure value.

7. The multimodal biosensor monitoring sock system according to claim 6, characterized in that, The sensor integration area of ​​the flexible sock body adopts a multi-layer composite structure; The multilayer composite structure includes an innermost skin-friendly conductive fabric layer, a middle flexible printed circuit layer, and an outermost encapsulation and protective layer. The skin-friendly conductive fabric layer is used to maintain stable contact with the skin and serves as an electrode for part of the sensor. Sensor connection traces and power and signal buses are etched on the flexible printed circuit layer. The encapsulation protective layer is coated or laminated with silicone or thermoplastic polyurethane material.

8. The multimodal biosensor monitoring sock system according to claim 7, characterized in that, The timestamp annotation operation performed by the microprocessor of the data processing and communication module is generated based on the reference clock counter of the main control synchronization module, which assigns a unified absolute time identifier to each group of synchronously sampled multimodal data.

9. The multimodal biosensor monitoring sock system according to claim 8, characterized in that, Driven by the control logic of the multi-channel synchronous sampling controller, the multiplexer sequentially switches the analog signals of each pressure sensing node to the shared high-precision analog-to-digital converter for digitization in a preset scanning order, ensuring that the sampling time of all pressure channels is strictly aligned with the rising edge of the same hardware synchronization pulse signal.

10. The multimodal biosensor monitoring sock system according to claim 9, characterized in that, The external terminal device or cloud server receives the synchronous multimodal data stream with a unified timestamp sent by the monitoring system, executes the gait event detection algorithm and gait parameter calculation, and generates a comprehensive evaluation report.

Images