Fall detection system based on BLE backscattering technology and implementation method thereof

Abstract

The invention discloses a fall detection system based on a BLE backscattering technology, and relates to the technical field of wireless communication. Comprising an acceleration sensing module, a baseband signal generation module, a backscattering modulation module, an energy collection and power supply management module and a terminal information processing system. The acceleration sensing module is used for acquiring three-axis acceleration data; after the baseband signal generation module encapsulates data into a broadcast protocol data unit conforming to the BLE standard, smoothing processing is carried out on the signal through GFSK, and then symbols are mapped into subcarriers with different frequencies through MFSK; the backscattering modulation module switches an impedance unit according to the received subcarriers with different frequencies, and modulates a baseband signal to an environment radio frequency carrier; and the terminal processing system receives and analyzes the BLE broadcast signal and executes a fall detection algorithm. The invention further discloses an implementation method of the fall detection system based on the BLE backscattering technology. The fall detection system based on the BLE backscattering technology has the advantages of being low in power consumption, high in compatibility and small in size.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a fall detection system based on BLE backscattering technology and its implementation method. Background Technology

[0002] Falls are a common and highly dangerous accident among the elderly, often occurring suddenly with serious consequences and difficult to detect in certain environments. Therefore, real-time detection and alarm systems for falls among the elderly have significant practical value. Currently, fall detection systems on the market mainly fall into three categories: user-triggered alarm systems, visual perception-based fall detection systems, and wearable sensor-based fall detection systems. The first type relies on the user actively issuing an alarm signal after a fall, and cannot effectively respond when the user is unconscious or unable to move. The second type is typically limited to fixed indoor monitoring scenarios, lacking adaptability to outdoor environments and thus limiting its application scope. While the third type offers better applicability and real-time performance, its long-term use is significantly limited due to factors such as device size, battery life, and wearing comfort. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a fall detection system and its implementation method based on BLE (Bluetooth Low Energy) backscattering technology. The present invention solves the battery life problem of wearable devices and is suitable for long-term monitoring of fall problems.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0005] A fall detection system based on BLE backscattering technology proposed according to the present invention includes: an acceleration sensing module, a baseband signal generation module, a backscattering modulation module, an energy harvesting and power supply management module, and a terminal information processing system; wherein,

[0006] An acceleration sensing module is used to collect triaxial acceleration data and input it to a baseband signal generation module.

[0007] The baseband signal generation module is responsible for encapsulating triaxial acceleration data into BLE data packet payloads, adding device name identifiers and power consumption information identifiers, and then performing cyclic redundancy check and whitening processing on the triaxial acceleration data after adding device name identifiers and power consumption information identifiers, assembling it into broadcast packet protocol data units that conform to the BLE standard, and sending the baseband signal to the backscatter modulation module in the form of a modulated signal.

[0008] The backscatter modulation module is used to modulate the baseband signal onto the ambient radio frequency carrier to generate a reflected signal;

[0009] The energy harvesting and power management module is connected to the acceleration sensing module, the baseband signal generation module, and the backscatter modulation module. It is used to harvest energy from the ambient radio frequency carrier and power the acceleration sensing module, the baseband signal generation module, and the backscatter modulation module.

[0010] The terminal information processing system is used to receive and parse the reflected signal emitted by the backscatter modulation module, extract acceleration data from it, and execute a fall detection algorithm based on the acceleration data.

[0011] As a further optimization of the fall detection system based on BLE backscattering technology described in this invention, the backscattering modulation module adopts an orthogonal IQ modulator structure. The backscattering modulation module is used to receive the baseband signal from the baseband signal generation module, and switches the impedance unit in the backscattering modulation module according to the received baseband signal to modulate the baseband signal onto the ambient radio frequency carrier to generate a reflected signal.

[0012] As a further optimization of the fall detection system based on BLE backscattering technology described in this invention, the backscattering modulation module adopts an IQ modulator structure, including a power divider, a 90-degree phase shifter, two RF switches controlled by baseband signals, and a combiner. The input of the power divider is used to receive the ambient RF carrier signal, and the output is divided into two paths. One path is directly output to one RF switch, and the other path is output to another RF switch via the 90-degree phase shifter. The outputs of the two RF switches are respectively connected to the input of the combiner, and the output of the combiner is connected to the antenna port. This is used to combine the modulated reflected signal into a single RF output signal and radiate it into space, so as to realize backscattering signal modulation based on IQ quadrature modulation.

[0013] As a further optimization of the fall detection system based on BLE backscattering technology described in this invention, the acceleration sensing module uses an MPU6050 sensor, which communicates with the baseband signal generation module via the I2C protocol. Only the accelerometer function of the MPU6050 sensor is enabled, and the gyroscope function is disabled.

[0014] The baseband signal generation module, after assembling and generating broadcast packet protocol data units, uses Gaussian Frequency Shift Keying (GFSK) and Multi-level Frequency Shift Keying (MFSK) to finally map the symbols in the broadcast packet protocol data units onto subcarriers of corresponding frequencies, thus forming a baseband signal.

[0015] The terminal information processing system is a smartphone or a dedicated wireless receiving device. Its built-in application is configured to: scan and receive reflected signals emitted by the backscatter modulation module, perform cyclic redundancy check (CRC), parse the triaxial acceleration data in the broadcast packet protocol data unit, and determine whether a fall event has occurred by analyzing the acceleration vector and amplitude characteristics.

[0016] As a further optimization of the fall detection system based on BLE backscattering technology described in this invention, the baseband signal generation module is implemented by a low-power field-programmable gate array (FPGA). The internal program of the baseband signal generation module includes a top-level function, an MPU6050 sensor driver module, an uplink control module, and a baseband modulation module. The MPU6050 sensor driver module controls the sensor to collect triaxial acceleration data and transmits the triaxial acceleration data to the uplink control module. The uplink control module performs clock division, CRC check, and whitening operations, and transmits the processed triaxial acceleration data to the baseband modulation module. The baseband modulation module performs GFSK and MFSK operations.

[0017] As a further optimization of the fall detection system based on BLE backscattering technology described in this invention, the MPU6050 sensor drive module communicates with the MPU6050 sensor via the I2C protocol, collects triaxial acceleration data, and outputs the triaxial acceleration data to the uplink control module in 16-bit format.

[0018] As a further optimization of the fall detection system based on BLE backscattering technology described in this invention, the uplink control module has a built-in finite state machine (FSM) for performing clock division, CRC check and whitening operations, generating broadcast packet protocol data units that conform to the specified format and transmitting them to the baseband modulation module.

[0019] As a further optimization of the fall detection system based on BLE backscattering technology described in this invention, the baseband modulation module includes a clock divider submodule, a data shifter submodule, a Gaussian filter submodule, and an FSK modulation submodule. The clock divider submodule provides the working clock for the entire baseband modulation module. The data shifter submodule transmits the broadcast packet protocol data units transmitted by the uplink module serially to the Gaussian filter submodule. The Gaussian filter submodule performs a GFSK operation on the received data and then transmits it to the FSK modulation submodule. The FSK modulation submodule performs an MFSK operation on the received data and outputs the baseband signal.

[0020] As a further optimization of the fall detection system based on BLE backscattering technology described in this invention, the FSK modulation submodule includes a numerically controlled oscillator (NCO). Utilizing the principle of phase accumulation, it obtains four subcarriers of different frequencies required to perform the MFSK operation based on the set base frequency and frequency hopping size.

[0021] The above-mentioned implementation method of a fall detection system based on BLE backscattering technology includes:

[0022] Step 1: After the FPGA is powered on, the energy harvesting and power management module provides the initial operating voltage to each module;

[0023] Step 2: The FPGA starts the acceleration sensing module, collects acceleration data of the X, Y, and Z axes, and transmits it to the baseband signal processing module;

[0024] Step 3: The baseband signal processing module encapsulates the acceleration data into broadcast packet protocol data units conforming to the BLE 5.0 specification. The encapsulation process includes: generating a broadcast packet protocol data unit header according to the BLE 5.0 specification, including the protocol data unit PDU type, transmission address, and length field; embedding the triaxial acceleration data into multiple broadcast packet protocol data units, each broadcast packet protocol data unit including a length field, a data type field, and a data field, wherein the data type field is identified as a custom data type, and the data field includes the sensor identifier and acceleration value, resulting in the framed broadcast packet protocol data unit.

[0025] Step 4: Perform CRC check on the framed broadcast packet protocol data unit, generate a 24-bit cyclic redundancy check code and append it to the end of the broadcast packet protocol data unit; perform whitening processing on the complete broadcast packet protocol data unit after appending CRC, and output the processed broadcast packet protocol data unit.

[0026] Step 5: Perform oversampling and edge detection on the broadcast packet protocol data unit signal, and output a stepped transition symbol according to the signal transition stage;

[0027] Step 6: Based on the stepped transition symbol mapping output by the Gaussian filter submodule, the corresponding frequency offset is mapped and input to the NCO module; the NCO module refers to the input frequency offset, uses the phase accumulator to accumulate the phase increment in real time, extracts the high bits of the accumulator to generate in-phase and quadrature square wave signals, and finally combines them to generate the baseband signal used to control the antenna impedance.

[0028] Step 7: The backscatter modulation module adjusts the impedance unit settings in the backscatter modulation module according to the baseband signals of different frequencies generated by the NCO module, thereby changing the reflection coefficient of the antenna, modulating the baseband signal onto the ambient radio frequency carrier signal and performing frequency shifting, generating a reflected signal that conforms to the BLE broadcast protocol and radiating it outward.

[0029] Step 8: The terminal information processing system receives the reflected signal, demodulates, decodes, and performs CRC verification on the reflected signal, extracts the triaxial acceleration data from the reflected signal, and uses a fall detection algorithm to process and analyze the triaxial acceleration data. If it is determined to be a fall event, a local alarm or remote notification is triggered.

[0030] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0031] (1) Backscatter communication technology is adopted to avoid the high power consumption active transmission chain in the traditional mode; combined with environmental radio frequency energy harvesting, the problem of wearable device battery life is solved, which is suitable for long-term monitoring of fall problems;

[0032] (2) The reflected signal follows the standard BLE broadcast protocol format and can communicate directly with commercial receivers such as smartphones, reducing deployment costs and the barrier to entry.

[0033] (3) Single-sideband modulation effectively suppresses unwanted sidebands through symmetrical impedance design and precise control. Stopband rejection is better than 20dB, significantly improving the signal-to-noise ratio and link budget, and enhancing communication reliability in complex environments;

[0034] (4) The overall system is miniaturized and easy to integrate into various wearable devices, which improves the user's comfort. Attached Figure Description

[0035] Figure 1 This is a block diagram of the overall structure of the system of the present invention;

[0036] Figure 2 Here is a block diagram of the baseband signal generation module;

[0037] Figure 3 This is a picture of the actual test object;

[0038] Figure 4 Signal diagram of the logic analyzer on the simulation platform;

[0039] Figure 5 This is a graph showing the observation results of the physical test spectrum analyzer;

[0040] Figure 6 This is a diagram showing the observation results from a smartphone during a physical test. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] To overcome the shortcomings of the prior art, this invention introduces BLE-based backscattering technology to achieve an ultra-low power, miniaturized wearable fall detection system. This technology modulates and transmits the sensing information in a standard BLE broadcast packet format by adjusting the impedance to change the transmission form of the carrier signal. The system design needs to address the following key technical issues: first, achieving stable carrier reflection and baseband signal modulation transmission under extremely low power consumption; second, achieving highly reliable sensor data acquisition, baseband signal processing, and backend data analysis and judgment.

[0043] like Figure 1As shown, this invention designs and implements a fall detection system based on BLE backscattering technology. The system consists of an acceleration sensing module, a baseband signal generation module, a backscattering modulation module, an energy harvesting and power supply management module, and a terminal information processing system.

[0044] The acceleration sensing module uses an MPU6050 sensor and connects to the MPU6050_CTRL driver module inside the FPGA via an I2C bus. This driver module is based on an FSM design, generating I2C timings through clock division and using a counter to set a sampling period of approximately 10ms. After power-on reset, the FSM sequentially writes configuration parameters: sets power management to wake up the device, sets the configuration register (CONFIG1) to 0x06 to enable 5Hz low-pass filtering to remove noise, and sets the acceleration configuration register (ACC_CONFIG) to 0x01 to set the ±2g range. Once in working mode, the FSM cyclically reads sensor data from address ranges 0x3B to 0x48, concatenates the high and low eight bits of the X, Y, and Z axis acceleration data, and finally outputs a 16-bit wide signal (ACC_X / Y / Z) to the baseband signal generation module, achieving high-precision motion data acquisition.

[0045] like Figure 2 As shown, the baseband signal generation module is implemented based on a low-power FPGA, and its internal logic is mainly divided into three core parts:

[0046] MPU6050 Sensor Drive Module: This module connects to the MPU6050 sensor via the I2C communication protocol and is configured as the master device in the connection to read sensor data. It then converts the read raw X, Y, and Z axis acceleration data into 16-bit binary two's complement format and transmits it to the uplink control module.

[0047] Uplink Control Module: This module implements the cyclic redundancy check (CRC) and whitening process for data packets through an FSM (Functional Signal Processor), ultimately generating payload data conforming to the specified format. Internally, it instantiates three sub-modules: a clock divider, a CRC24 checker, and a data whitening unit. A four-state FSM manages the transitions between the IDLE (idle), CRC (checksum), WHITEN (whitening), and START (transmit) states. The system input 50MHz clock is divided to generate a 10Hz trigger signal, initiating a transmission cycle every 100ms. When the FSM enters the CRC state, the module sends the input PDU data to the serial CRC24 checker, based on a polynomial... A 24-bit checksum is calculated. Then, in the WHITEN state, the PDU data and CRC checksum are concatenated and fed into the whitening module. A 7-bit LFSR with a preset initial value corresponding to the broadcast channel is XORed to eliminate DC bias. Finally, in the START state, the module concatenates the preamble, access address, and whitened payload to form a complete link-layer data packet, which is then output to the baseband modulation module, achieving BLE-compliant data link layer processing.

[0048] Baseband Modulation Module: As the core of the physical layer, this module is responsible for processing the assembled BLE broadcast protocol data units through GFSK and MFSK, ultimately mapping the symbols to subcarriers of corresponding frequencies. This module integrates four sub-modules: the clock_divider module divides the 50MHz system clock to 1MHz to drive data shifting; the shifter module receives complete data packets and serializes them into a bitstream according to the packet length; the gauss_filter module performs GFSK Gaussian filtering preprocessing on the baseband bitstream; and the fsk_modulator module completes the final frequency modulation.

[0049] In the specific modulation process, the shifter module outputs data bit by bit to the gauss_filter module driven by a 1MHz clock. The gauss_filter module detects signal transition edges by oversampling (50 times) and inserts transition symbols (2'b01 or 2'b10) at the moment of transition to smooth the frequency switching. The processed symbol stream enters the fsk_modulator module, where the NCO (numerically controlled oscillator) selects the corresponding frequency hopping increment according to the input symbol, superimposes it with the base frequency, and drives the phase accumulator. The in-phase and quadrature signals generated by the NCO are combined via single-sideband (SSB) logic to finally output two quadrature control signals, which are used to drive the I / Q impedance switches of the backscattering front end to achieve low-power FSK single-sideband modulation transmission.

[0050] In terms of programming, the backscatter modulation module can choose to use either double-sideband (DSB) or SSB modulation. Using an IQ modulator structure under SSB modulation can solve the energy waste and additional spectrum occupation problems of traditional double-sideband backscattering. The module includes a power divider, a 90-degree phase shifter, two RF switches controlled by the digital baseband modulation signals, and a combiner. The power divider splits the RF signal received by the antenna into two paths. The 90-degree phase shifter shifts one of the signals by 90 degrees, forming two carriers: I (in-phase) and Q (quadrature). Each carrier is connected to an RF switch, and the switch states are controlled by the I and Q baseband control signals output from the FPGA. The combiner combines the two modulated signals and reflects them through the antenna. By precisely controlling the phase relationship of the I and Q control signals, one sideband can be canceled, thus achieving single-sideband modulation with a stopband rejection ratio better than 20dB.

[0051] The energy harvesting and power management module supplies power according to the power requirements of the low-power backscatter communication system.

[0052] The terminal information processing system is a smartphone or a dedicated wireless receiving device. Its built-in application is configured to: scan and receive BLE broadcast signals, verify CRC, parse triaxial acceleration data in a custom data format, and determine whether a fall event has occurred by analyzing acceleration vector and amplitude characteristics.

[0053] This invention also discloses a method for implementing a fall detection system based on BLE backscattering technology, specifically including the following steps:

[0054] Step 1: System Initialization and Power Management

[0055] After the system powers on and resets, the energy harvesting module stabilizes the voltage to the operating threshold, and the reset signal rst_n is released. The system then enters the initialization state, and all logic modules within the FPGA reset to their default states, waiting for the clock to stabilize.

[0056] Step 2: Sensor Configuration and Motion Data Acquisition

[0057] The MPU6050_CTRL module within the FPGA starts the I2C interface state machine, first executing the configuration sequence: writing 0x00 to the sensor power management register to wake up the device; writing 0x06 to the configuration register to enable the 5Hz low-pass filter; and writing 0x01 to the acceleration configuration register to set the ±2g range. After configuration, the driver module uses an internal counter to generate a sampling rate of approximately 100Hz and periodically reads the acceleration register. The driver module extracts the high and low eight bits of data from the X, Y, and Z axes, concatenates them into a 16-bit two's complement signal, and latches it for subsequent module calls.

[0058] Step 3: Encapsulation of Broadcast Protocol Data Units

[0059] The uplink control module uses a frequency divider to generate a 10Hz transmit trigger signal. Upon detecting the trigger, the state machine enters the encapsulation process: first, it constructs the BLE broadcast packet header, including the PDU type and sender address; then, it fills the Payload field with the latched triaxial acceleration data. The code pre-sets the access address to 0x8E89BED6 and the preamble to 0xAA to ensure compliance with the BLE physical layer access specification.

[0060] Step 4: CRC Check and Data Whitening

[0061] Perform CRC-24 calculation on the framed data units, based on polynomials. A 24-bit cyclic redundancy check code is generated and appended to the end of the data packet. After the data packet enters the whitening module, a pseudo-random sequence is generated based on the channel index using a 7-bit linear feedback shift register. This sequence is then XORed with the input data to eliminate DC bias caused by long strings of 0s or 1s.

[0062] Step 5: Baseband serialization and Gaussian filter preprocessing

[0063] The encapsulated complete data packet is input in parallel to the baseband modulation module. First, serialization is performed; the shifter module, driven by a 1MHz clock, converts the data packet into a single-bit baseband stream. Next, the gauss_filter module uses a 50MHz fast clock to oversample the 1MHz baseband stream by 50x. The module detects the rising and falling edges of the signal, outputting specific 2-bit wide transition symbols (such as 2'b01 or 2'b10) during the transition phase and a hold symbol during the hold phase. This digitized stepped waveform simulates the smoothing effect of a Gaussian filter, effectively suppressing spectral sidelobes.

[0064] Step 6: FSK Modulation and NCO Waveform Generation

[0065] The `fsk_modulator` module receives the filtered 2-bit symbol stream, performs frequency mapping, and selects the corresponding frequency offset from the preset frequency hopping table `hop_freq` based on the input symbol (00, 01, 10, 11). The offset is then added to the center carrier fundamental frequency and input to the `nco_gen` module. The NCO uses a phase accumulator to accumulate the frequency control word in real time and generates two orthogonal square wave signals by truncating the high-order bits of the accumulator.

[0066] Step 7, Backscatter Control

[0067] To improve spectral efficiency, the modulation module combines the I / Q signals generated by the NCO into a single-sideband control signal, fswitch. This signal is directly output to the RF front end to control the on / off state of the I and Q RF switches in the antenna impedance matching network. Through this phase-precisely controlled impedance switching, the incident RF carrier is modulated and frequency-shifted, generating a BLE reflected signal carrying acceleration data.

[0068] Step 8: Signal Spatial Transmission

[0069] The modulated backscattered signal is radiated into space through the antenna. The signal energy is concentrated in the target BLE broadcast channel and captured by surrounding standard BLE receiving equipment.

[0070] Step 9: Terminal reception and fall judgment

[0071] After receiving the broadcast packet, the smart terminal demodulates, dewhitens, and performs CRC verification. After extracting the X, Y, and Z axis acceleration data, the terminal algorithm calculates the resultant acceleration vector (SVM). If the data characteristics match the time-domain threshold sequence of "weightlessness (<0.5g) — impact (>2.5g) — stillness", the system determines that a fall event has occurred and triggers relevant alarm prompts.

[0072] This invention employs a backscatter communication mechanism. Unlike traditional active radios that require high-power oscillators and phase-locked loops (PLLs) to locally generate high-frequency carriers, this system uses "reflection" rather than "generation" for communication. In common backscatter communication, the core mechanism is based on impedance matching and reflection modulation. According to transmission line theory, part of the electromagnetic energy received by the antenna is absorbed by the load, and the other part is re-radiated into space as a scattered wave. The matching state between the antenna and its terminal load is determined by the complex reflection coefficient. The description, its mathematical expression is:

[0073]

[0074] in, This indicates the variable load impedance connected to the antenna terminal. This represents the inherent impedance of the antenna. It is its conjugate complex number. In this invention, the control signal (i.e., the fswitch signal) generated by the baseband modulation module does not directly drive the antenna transmit power, but rather controls the RF switch in the RF front-end impedance matching network.

[0075] When the control signal changes, the load impedance Rapid switching between different states causes impedance changes that affect the reflection coefficient. Over time Changes occur, forming a time-varying function. Assume that there exists a frequency in space. incident carrier signal The signal reflected back into space by the antenna It can be represented as:

[0076]

[0077] Therefore, baseband frequency control via FPGA By observing the variation pattern, the amplitude (ASK) or phase (PSK) of the reflected radio frequency signal can be modulated without the need for a radio frequency power amplifier (PA), thereby greatly reducing system power consumption.

[0078] This invention, based on the backscatter communication mechanism, employs Broadcast Link (BLE) as the signal transmission method. Related signals, such as those transmitted by the RF front-end, strictly adhere to the physical layer standard of the BLE protocol. BLE operates in the 2.4 GHz ISM band (2400-2483.5 MHz). It divides the band into 40 RF channels, each with a bandwidth of 2 MHz. Channels 37 (2402 MHz), 38 (2426 MHz), and 39 (2480 MHz) are defined as broadcast channels, primarily used for device discovery and connection establishment; the remaining 37 channels are data channels. The uplink_controller and fsk_modulator modules in this invention are designed specifically for the BLE broadcast packet specification to ensure that devices can be quickly scanned and discovered.

[0079] Specifically, a standard BLE broadcast channel PDU structure includes the following four core parts:

[0080] Preamble: Used for receiver automatic gain control and symbol synchronization. For a 1 Mbps PHY, the preamble is an 8-bit alternating sequence, which is fixed at 0xAA in this invention.

[0081] Access address: This address is fixed at 0x8E89BED6 for all broadcast packets.

[0082] Protocol Data Unit: Contains a PDU header and payload. The PDU header indicates the data packet type, while the payload carries the MPU6050 acceleration data acquired by this invention.

[0083] CRC Check: BLE uses a 24-bit CRC check to ensure data integrity. Its generator polynomial is:

[0084]

[0085] In wireless communication, prolonged consecutive "0"s or "1"s can lead to a significant DC component in the signal spectrum, affecting receiver demodulation performance. Therefore, after generating BLE broadcast packets, whitening processing is required. The whitening algorithm uses a 7-bit linear feedback shift register (LFSR), and its generator polynomial is: .

[0086] The initial value of the LFSR depends on the current radio frequency channel index. The whitening module of this invention implements this algorithm by performing an XOR operation between the data stream and the pseudo-random sequence generated by the LFSR, thereby making the distribution probability of "0" and "1" in the output data stream tend to be uniform, thus optimizing the spectral characteristics and eliminating DC bias.

[0087] The modulation scheme used in BLE backscattering systems is GFSK. This modulation scheme can be regarded as an improved form of FSK. Its core lies in preprocessing the digital baseband signal with Gaussian low-pass filter (GLPF) before frequency modulation.

[0088] Let the binary data sequence to be sent be... The values ​​are {+1,−1} (corresponding to bits '1' and '0' respectively). The sequence is first passed through a Gaussian low-pass filter, and its impulse response... for:

[0089]

[0090] in, For the symbol period, With the filter's 3-dB bandwidth satisfy Filtered baseband signal It can be represented as:

[0091]

[0092] This filtering process smooths out steep transitions in the baseband signal and effectively limits the spectral width of the modulated signal.

[0093] Filtered signal Used to control the instantaneous frequency of the carrier wave. The complex form of a GFSK modulated signal can be expressed as:

[0094]

[0095] in, For signal amplitude, The center frequency of the carrier. For peak frequency offset, This is the initial phase.

[0096] instantaneous frequency for:

[0097]

[0098] For the BLE standard, modulation index The value must be between 0.45 and 0.55 (usually 0.5) to ensure that the receiver can demodulate reliably.

[0099] To verify the functional integrity, BLE protocol compatibility, and low-power characteristics of the low-power fall detection system based on BLE backscattering technology proposed in this invention, a relevant experimental platform was built, and simulation and physical verification tests were conducted. (See attached...) Figure 3 As shown, the experimental platform includes a portable signal transmitter, an FPGA-based backscattering tag prototype, a spectrum analyzer, a power supply, a smartphone, and several antennas. The portable signal transmitter transmits a single-tone carrier signal at a frequency of 2412MHz, the spectrum analyzer observes the frequency band and waveform of the reflected signal, and the smartphone receives and analyzes the reflected signal to read acceleration information.

[0100] First, the invention was simulated and tested on the development board using the Signal Tap Logic Analyzer tool provided by the Quartus experimental platform. In the on-board simulation environment, without carrier or reflected signals, the packet information of the uplink control module in the baseband signal generation module of the invention, including triaxial acceleration data, was read using a logic analyzer. The accuracy of the data read from the MPU6050 accelerometer was verified by calculating the acceleration data under specific attitude conditions.

[0101] As attached Figure 4 As shown, with the device stationary and vertically downward, the data received by the baseband signal generation module is as follows:

[0102]

[0103] exist Under the specified range setting, the triaxial acceleration data obtained from the test can be calculated as follows:

[0104]

[0105] From this data analysis, we can conclude that... shaft and The acceleration of the shaft is approximately zero. The acceleration of the shaft is approximately: The readings correspond to the device's orientation. This test demonstrates that the present invention can correctly read acceleration sensor information and use it to generate BLE broadcast packets.

[0106] In the appendix Figure 3 The backscatter communication function described in this invention was tested on the experimental platform shown. A portable signal transmitter was used to transmit a single-tone signal at a frequency of 2412 MHz as the carrier signal for backscatter communication. The spectrum analyzer results are attached. Figure 5 As shown, the center frequency is 2412MHz, and the frequency indicated by marker 1 is 2426MHz. This frequency corresponds to the BLE 38 broadcast signal. A signal peak with an intensity of -63.99dBm can be observed at this frequency, which is the reflected signal received by the spectrum analyzer. The observation results from the smartphone are attached. Figure 6 As shown, the BLE Modulation Assistant APP on a smartphone can receive broadcast packets reflected from a backscattering tag prototype named "SEU". The 1826 field in the broadcast packet content indicates the content type of the broadcast packet, and the triaxial acceleration at this time can be read.

[0107]

[0108] exist Under the specified range setting, the triaxial acceleration data obtained from the test can be calculated as follows:

[0109]

[0110] At this point, the sum of accelerations is

[0111]

[0112] The sum of the calculation results is accelerated to approximately The results are consistent with the effects of gravitational acceleration under static conditions, and the triaxial acceleration matches the device's attitude. This test demonstrates that the present invention can accurately read acceleration sensing information and transmit it to a smartphone terminal in the form of BLE backscattering.

[0113] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A fall detection system based on BLE backscattering technology, characterized in that, include: The system includes an acceleration sensing module, a baseband signal generation module, a backscatter modulation module, an energy harvesting and power management module, and a terminal information processing system; among which, An acceleration sensing module is used to collect triaxial acceleration data and input it to a baseband signal generation module. The baseband signal generation module is responsible for encapsulating triaxial acceleration data into BLE data packet payloads, adding device name identifiers and power consumption information identifiers, and then performing cyclic redundancy check and whitening processing on the triaxial acceleration data after adding device name identifiers and power consumption information identifiers, assembling it into broadcast packet protocol data units that conform to the BLE standard, and sending the baseband signal to the backscatter modulation module in the form of a modulated signal. The backscattering modulation module is used to modulate the baseband signal onto the ambient radio frequency carrier to generate a reflected signal; The energy harvesting and power management module is connected to the acceleration sensing module, the baseband signal generation module, and the backscatter modulation module. It is used to harvest energy from the ambient radio frequency carrier and power the acceleration sensing module, the baseband signal generation module, and the backscatter modulation module. The terminal information processing system is used to receive and parse the reflected signal emitted by the backscatter modulation module, extract acceleration data from it, and execute a fall detection algorithm based on the acceleration data.

2. The fall detection system based on BLE backscattering technology according to claim 1, characterized in that, The backscatter modulation module adopts an orthogonal IQ modulator structure. The backscatter modulation module is used to receive the baseband signal from the baseband signal generation module. According to the received baseband signal, the impedance unit in the backscatter modulation module is switched to modulate the baseband signal onto the ambient radio frequency carrier to generate a reflected signal.

3. A fall detection system based on BLE backscattering technology according to claim 1, characterized in that, The backscatter modulation module adopts an IQ modulator structure, including a power divider, a 90-degree phase shifter, two RF switches controlled by baseband signals, and a combiner. The input of the power divider is used to receive the ambient RF carrier signal, and the output is divided into two paths. One path is directly output to one RF switch, and the other path is output to another RF switch via the 90-degree phase shifter. The outputs of the two RF switches are respectively connected to the input of the combiner, and the output of the combiner is connected to the antenna port. It is used to combine the modulated reflected signal into a single RF output signal and radiate it into space to realize backscatter signal modulation based on IQ quadrature modulation.

4. A fall detection system based on BLE backscattering technology according to claim 1, characterized in that, The acceleration sensing module uses the MPU6050 sensor and communicates with the baseband signal generation module via the I2C protocol. Only the MPU6050 sensor accelerometer function is enabled, and the gyroscope function is disabled. The baseband signal generation module, after assembling and generating broadcast packet protocol data units, uses Gaussian Frequency Shift Keying (GFSK) and Multi-level Frequency Shift Keying (MFSK) to finally map the symbols in the broadcast packet protocol data units onto subcarriers of corresponding frequencies, thus forming a baseband signal. The terminal information processing system is a smartphone or a dedicated wireless receiving device. Its built-in application is configured to: scan and receive reflected signals emitted by the backscatter modulation module, perform cyclic redundancy check (CRC), parse the triaxial acceleration data in the broadcast packet protocol data unit, and determine whether a fall event has occurred by analyzing the acceleration vector and amplitude characteristics.

5. A fall detection system based on BLE backscattering technology according to claim 1, characterized in that, The baseband signal generation module is implemented using a low-power field-programmable gate array (FPGA). The internal program of the baseband signal generation module includes a top-level function, an MPU6050 sensor driver module, an uplink control module, and a baseband modulation module. The MPU6050 sensor driver module controls the sensor to acquire triaxial acceleration data and transmits the triaxial acceleration data to the uplink control module. The uplink control module performs clock division, CRC check, and whitening operations, and transmits the processed triaxial acceleration data to the baseband modulation module. The baseband modulation module performs GFSK and MFSK operations.

6. A fall detection system based on BLE backscattering technology according to claim 5, characterized in that, The MPU6050 sensor driver module communicates with the MPU6050 sensor via the I2C protocol, acquires triaxial acceleration data, and outputs the triaxial acceleration data to the uplink control module in 16-bit format.

7. A fall detection system based on BLE backscattering technology according to claim 5, characterized in that, The uplink control module has a built-in finite state machine (FSM) for performing clock division, CRC check and whitening operations, generating broadcast packet protocol data units that conform to the specified format and transmitting them to the baseband modulation module.

8. A fall detection system based on BLE backscattering technology according to claim 5, characterized in that, The baseband modulation module includes a clock divider submodule, a data shifter submodule, a Gaussian filter submodule, and an FSK modulation submodule. The clock divider submodule provides the operating clock for the entire baseband modulation module. The data shifter submodule transmits the broadcast packet protocol data units transmitted by the uplink module serially to the Gaussian filter submodule. The Gaussian filter submodule performs a GFSK operation on the received data and then transmits it to the FSK modulation submodule. The FSK modulation submodule performs an MFSK operation on the received data and outputs the baseband signal.

9. A fall detection system based on BLE backscattering technology according to claim 8, characterized in that, The FSK modulation submodule includes a numerically controlled oscillator (NCO) that uses the principle of phase accumulation to obtain four different frequency subcarriers required to perform MFSK operations based on the set base frequency and frequency hopping size.

10. A method for implementing a fall detection system based on BLE backscattering technology as described in claim 1, characterized in that, include: Step 1: After the FPGA is powered on, the energy harvesting and power management module provides the initial operating voltage to each module; Step 2: The FPGA starts the acceleration sensing module, collects acceleration data of the X, Y, and Z axes, and transmits it to the baseband signal processing module; Step 3: The baseband signal processing module encapsulates the acceleration data into broadcast packet protocol data units conforming to the BLE 5.0 specification. The encapsulation process includes: generating a broadcast packet protocol data unit header according to the BLE 5.0 specification, including the protocol data unit PDU type, transmission address, and length field; embedding the triaxial acceleration data into multiple broadcast packet protocol data units, each broadcast packet protocol data unit including a length field, a data type field, and a data field, wherein the data type field is identified as a custom data type, and the data field includes the sensor identifier and acceleration value, resulting in the framed broadcast packet protocol data unit. Step 4: Perform CRC check on the framed broadcast packet protocol data unit, generate a 24-bit cyclic redundancy check code and append it to the end of the broadcast packet protocol data unit; perform whitening processing on the complete broadcast packet protocol data unit after appending CRC, and output the processed broadcast packet protocol data unit. Step 5: Perform oversampling and edge detection on the broadcast packet protocol data unit signal, and output a stepped transition symbol according to the signal transition stage; Step 6: Based on the stepped transition symbol mapping output by the Gaussian filter submodule, the corresponding frequency offset is mapped and input to the NCO module; the NCO module refers to the input frequency offset, uses the phase accumulator to accumulate the phase increment in real time, extracts the high bits of the accumulator to generate in-phase and quadrature square wave signals, and finally combines them to generate the baseband signal used to control the antenna impedance. Step 7: The backscatter modulation module adjusts the impedance unit settings in the backscatter modulation module according to the baseband signals of different frequencies generated by the NCO module, thereby changing the reflection coefficient of the antenna, modulating the baseband signal onto the ambient radio frequency carrier signal and performing frequency shifting, generating a reflected signal that conforms to the BLE broadcast protocol and radiating it outward. Step 8: The terminal information processing system receives the reflected signal, demodulates, decodes, and performs CRC verification on the reflected signal, extracts the triaxial acceleration data from the reflected signal, and uses a fall detection algorithm to process and analyze the triaxial acceleration data. If it is determined to be a fall event, a local alarm or remote notification is triggered.

Images