Millimeter wave radar baseband module and system for multi-target vital signs monitoring

Abstract

The present invention discloses a millimeter wave radar baseband module and system for multi-target vital signs monitoring, wherein the radar baseband module comprises a time domain processing submodule, a frequency domain processing submodule, and an FFT submodule, wherein the time domain processing submodule is used to filter and preprocess the digital intermediate frequency signal of the echo signal reflected by the target to obtain a time domain sampling point sequence, and the FFT submodule is used to perform FFT transformation on the time domain sampling point sequence output by the time domain processing submodule to obtain a one-dimensional frequency domain signal, and the frequency domain processing submodule is used to extract the phase characteristics of multiple targets from the one-dimensional frequency domain signal, and the FFT submodule is also used to perform FFT transformation on the phase characteristics of multiple targets to obtain a two-dimensional frequency domain signal to extract the vital signs information of each target; wherein the frequency domain processing submodule comprises a distance extraction unit, a phase extraction unit, multiple parallel dewrapping units, a parallel-to-serial conversion unit, and a null filter. The present invention has high integration, low power consumption, low cost, and low latency.

Description

Technical Field

[0001] The present invention relates to the field of millimeter wave radars, and in particular to a millimeter wave radar baseband module and system for multi-target vital signs monitoring. Background Art

[0002] Heart rate and respiratory rate are important vital signs of the human body and can well reflect the health of the human body. Early detection of abnormal heart rate and respiratory rate can effectively prevent the occurrence of diseases. Compared with other monitoring methods, millimeter wave radar has the characteristics of all-day, all-weather, low-power operation, is not affected by temperature, light and weather, and can well protect user privacy. This technology transmits electromagnetic waves and receives the echo reflected by the monitored object. After a series of signal processing, it obtains the distance information of the monitored target and the corresponding vital signs information. With the development of science and technology, people's demand for non-contact detection is increasing, and vital signs detection based on frequency modulated continuous wave (FMCW) radar has been widely used.

[0003] Traditional vital sign monitoring radar systems mainly process intermediate frequency signals by writing software programs in digital signal processing (DSP) or microcontroller unit (MCU) to achieve vital sign detection. However, using DSP and MCU as processing devices often results in resource redundancy, poor real-time performance, and high power consumption. Summary of the invention

[0004] In view of the problems existing in the prior art, the purpose of the present invention is to provide a millimeter-wave radar baseband module and system for multi-target vital signs monitoring with high integration, low power consumption, low cost and low latency.

[0005] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0006] A millimeter wave radar baseband module for multi-target vital signs monitoring, comprising a time domain processing submodule, a frequency domain processing submodule, an FFT submodule and a vital signs extraction submodule, wherein the time domain processing submodule is used to filter and preprocess a digital intermediate frequency signal of an echo signal reflected from a target to obtain a time domain sampling point sequence, the FFT submodule is used to perform an FFT transformation on the time domain sampling point sequence output by the time domain processing submodule to obtain a one-dimensional frequency domain signal, the frequency domain processing submodule is used to extract phase features of multiple targets from the one-dimensional frequency domain signal, and the FFT submodule is also used to perform an FFT transformation on the phase features of multiple targets to obtain a two-dimensional frequency domain signal for extracting vital signs information of each target;

[0007] Among them, the frequency domain processing submodule includes a distance extraction unit, a phase extraction unit, multiple parallel dewrapping units, a parallel-to-serial conversion unit and a null filter. The distance extraction unit is used to separate the frequency domain signal of each target from the one-dimensional frequency domain signal. The phase extraction unit is used to extract phase information from the frequency domain signal of each target. Each of the dewrapping units is used to dewrapped the phase information belonging to the same target, and different dewrapping units correspond to different targets. The parallel-to-serial conversion unit is used to convert the dewrapped phase information from parallel to serial. The null filter is used to filter and obtain the phase characteristics of each target.

[0008] Furthermore, the module also includes an APB interface, and the time domain processing submodule, frequency domain processing submodule, FFT submodule and vital sign extraction submodule are all connected to the APB interface.

[0009] Furthermore, the time domain processing submodule specifically includes a filter, a FIFO unit, a Chirp extraction unit and a data preprocessing unit connected in sequence, the filter is used to filter the digital intermediate frequency signal of the echo signal reflected by the target, the FIFO unit is used to cache the filtered echo signal according to a first-in-first-out mechanism, the Chirp extraction unit is used to perform frame extraction on the signal output by the FIFO unit, and the data preprocessing unit is used to sample a number of sampling points from the extracted frames, and number the frames and the sampling points in the frames in sequence to generate a time domain sampling point sequence.

[0010] Furthermore, the data preprocessing unit is used to circularly number the frames according to 0, 1..., N-1, and the sampling points in the frames are circularly numbered according to 0, 1..., M-1. If the sampling points are out of range, a truncation operation is performed, and if the sampling points are insufficient, a zero-padding operation is performed. N is the number of two-dimensional frequency domain signal points, and M is the number of one-dimensional frequency domain signal points.

[0011] Furthermore, the data preprocessing unit has a configurable sampling point delay. When the sampling point delay is configured as k, the numbering will start from the kth valid sampling point.

[0012] Furthermore, the filter includes a CIC filter and a FIR filter connected in sequence.

[0013] Furthermore, the distance extraction unit specifically includes:

[0014] A memory, used to assign a storage address to each frequency point in the one-dimensional frequency domain signal and store it;

[0015] A comparator is used to compare the frequency points in all storage addresses outside the protection interval Q in the memory, find the frequency point with the largest modulus value and its storage address and output them to the judgment calculator until the number of found modulus values ​​reaches the target number;

[0016] The judgment calculator is used to judge whether the storage address of the frequency point with the largest modulus value is within the current protection interval Q; if it is, the frequency point is discarded and the execution comparator is returned; otherwise, the IQ data of the frequency point is extracted as the frequency domain signal of the identified target, and the protection interval Q is updated to: Returns the execution comparator, where the initial value of Q is the empty set and ADDR max is the storage address of the frequency point with the largest modulus value, WIN is the preset unilateral protection window width, and M is the number of frequency points in the one-dimensional frequency domain signal.

[0017] Furthermore, the phase extraction unit is specifically configured to perform phase extraction using a CORDIC method.

[0018] Furthermore, the unwinding unit specifically includes:

[0019] A memory, used to select the phase information belonging to the target corresponding to the current dewrapping unit from the phase information sent by the phase extraction unit, and allocate storage addresses in sequence from front to back according to the time of each phase information for storage;

[0020] The controller is used to determine whether the number of phase information stored in the memory is greater than or equal to N, if not, then wait, if yes, then obtain the phase information in the latest N consecutive storage addresses, N is the number of frames of the two-dimensional frequency domain signal;

[0021] The unwrapping operator is used to unwrap the phase information in the latest N consecutive storage addresses in the order of storage addresses from front to back according to the following method: obtain the phase information at the current storage address ADDR i The previous storage address ADDR i-1 The phase information D stored in i-1 and the phase information D in the current storage address i Subtract, when the difference is greater than π, the unwrapped phase D i '=D i -2π is output to the parallel-serial conversion unit, and the previous storage address ADDR is i-1 The phase information D stored in i-1 Updated to D i , when the difference is less than -π, the unwrapped phase D i '=D i +2π is output to the parallel-to-serial conversion unit, and the previous storage address ADDR is i-1 The phase information D stored ini-1 Updated to D i , when the difference is between -π and π, the unwrapped phase D i '=D i Output to the parallel-to-serial conversion unit and store the previous address ADDR i-1 The phase information D stored in i-1 Updated to D i .

[0022] Furthermore, the FFT submodule includes a data selection unit, an FFT calculation unit, a data cache unit and a data control unit, and the data control unit is respectively connected to the data selection unit, the FFT calculation unit and the data cache unit. The data control unit is used to set the transformation parameters of the FFT calculation unit to one-dimensional FFT transformation parameters when the data received by the data selection unit is a time domain sampling point sequence, and control the data of the data cache unit to be output as a one-dimensional frequency domain signal to the frequency domain processing submodule. When the data received by the data selection unit is the phase characteristics of multiple targets, the transformation parameters of the FFT calculation unit are set to two-dimensional FFT transformation parameters, and control the data of the data cache unit to be output as a two-dimensional frequency domain signal. The FFT calculation unit is used to perform FFT calculation under the control of the data control unit, and the data cache unit is used to cache and output the data output by the FFT calculation unit under the control of the data control unit.

[0023] A multi-target vital sign monitoring system based on a millimeter wave radar baseband module, comprising a time domain processing submodule, a frequency domain processing submodule, an FFT submodule and a vital sign extraction submodule, wherein the time domain processing submodule is used to filter and preprocess a digital intermediate frequency signal of an echo signal reflected from a target to obtain a time domain sampling point sequence, the FFT submodule is used to perform an FFT transformation on the time domain sampling point sequence output by the time domain processing submodule to obtain a one-dimensional frequency domain signal, the frequency domain processing submodule is used to extract phase features of multiple targets from the one-dimensional frequency domain signal, the FFT submodule is also used to perform an FFT transformation on the phase features of multiple targets to obtain a two-dimensional frequency domain signal, and the vital sign extraction submodule is used to extract vital sign information of each target from the two-dimensional frequency domain signal;

[0024] Among them, the frequency domain processing submodule includes a distance extraction unit, a phase extraction unit, multiple parallel dewrapping units, a parallel-to-serial conversion unit and a null filter. The distance extraction unit is used to separate the frequency domain signal of each target from the one-dimensional frequency domain signal. The phase extraction unit is used to extract phase information from the frequency domain signal of each target. Each of the dewrapping units is used to dewrapped the phase information belonging to the same target, and different dewrapping units correspond to different targets. The parallel-to-serial conversion unit is used to convert the dewrapped phase information from parallel to serial. The null filter is used to filter and obtain the phase characteristics of each target.

[0025] Compared with the prior art, the present invention has the following beneficial effects: the present invention provides a millimeter-wave radar baseband module for multi-target vital signs monitoring and a system based on the module. The millimeter-wave radar baseband module integrates multiple hardware modules to realize the processing flow of vital signs monitoring. Compared with the prior art of programming in a CPU or FPGA to realize the processing flow, the present invention has lower power consumption, lower cost and lower delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a schematic structural diagram of a millimeter wave radar baseband module for multi-target vital signs monitoring provided by an embodiment of the present invention;

[0027] Figure 2 It is the output timing diagram of the data preprocessing unit;

[0028] Figure 3 is a schematic diagram of the processing flow of the distance extraction unit;

[0029] Figure 4 is a schematic diagram of data processing of the unwinding unit;

[0030] Figure 5 It is the state transition diagram of unwinding unit strings and converting them. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0032] The embodiment of the present invention provides a millimeter wave radar baseband module for multi-target vital signs monitoring. The module is implemented by hardware and can be integrated into a chip. It can replace the original DSP / FPGA solution at a lower cost and lower power consumption to achieve real-time monitoring of multi-target respiratory rate and heart rate. The millimeter wave radar baseband module described in the present invention is suitable for FMCW radar. The echo signal of the FMCW radar is converted into a digital intermediate frequency signal through mixing, filtering, sampling, etc. and enters the millimeter wave radar baseband module. At the same time, the chirp synchronization signal CHIRP_SYNC is also sent to the chip.

[0033] In this embodiment, the requirements need to be clarified first, including: ADC sampling rate; chirp parameters, including single chirp cycle, slope, bandwidth and other parameters; system refresh rate; target distance resolution and maximum detection distance. According to the above requirements, the core register configuration of this embodiment is determined, and the corresponding register is written through the APB interface during the system initialization phase. The APB interface unit is mainly used for external interaction, responsible for configuring the parameters of each module in the baseband and checking the working status of each module.

[0034] 1. Determine the decimation multiple of the CIC filter according to the ADC sampling rate and chirp period, and determine the gain, FIR filter order, coefficient and other parameters of the CIC filter.

[0035] 2. Determine the chirp extraction rate based on the chirp period and refresh rate.

[0036] 3. Determine the number of one-dimensional FFT and two-dimensional FFT points, and set the FFT window function.

[0037] 4. Determine the single-side protection window width setting in the distance extraction module.

[0038] Figure 1 It is a structural schematic diagram of the millimeter wave radar baseband module provided by the present invention, including a time domain processing submodule, a frequency domain processing submodule, and an FFT submodule. The time domain processing submodule is used to filter and preprocess the digital intermediate frequency signal of the echo signal reflected by the target to obtain a time domain sampling point sequence. The FFT submodule is used to perform FFT transformation on the time domain sampling point sequence output by the time domain processing submodule to obtain a one-dimensional frequency domain signal. The frequency domain processing submodule is used to extract the phase characteristics of multiple targets from the one-dimensional frequency domain signal. The FFT submodule is also used to perform FFT transformation on the phase characteristics of multiple targets to obtain a two-dimensional frequency domain signal for extracting the vital sign information of each target.

[0039] In this embodiment, the analog signal output by the RF chip of the radar is converted into digital form by the ADC, and the ADC outputs 12-bit quantized data, that is, a digital intermediate frequency signal, with a rate of ≤50Mbps.

[0040] The time domain processing submodule specifically includes a filter, a FIFO unit, a Chirp extraction unit and a data preprocessing unit which are connected in sequence.

[0041] The filter includes a CIC filter and an FIR filter connected in sequence, which are used to filter the digital intermediate frequency signal. For different scenarios, the bandwidth of the intermediate frequency signal input to the ADC is different, the highest may be 25MHz, and the lowest may be at the kHz level. In order to improve the resolution and reduce the amount of data processing, it is necessary to use the CIC filter to downsample the ADC data. The downsampling rate is 0 to 1023 (1 to 1024), and supports a sampling rate of 50kHz to 50MHz; after the downsampling process, there is signal aliasing, so it is necessary to use the FIR filter for anti-aliasing filtering. The FIR filter supports configurable coefficients and configurable orders from 1 to 99.

[0042] The FIFO unit is used to cache the filtered echo signal according to the first-in-first-out mechanism, and switches the data from the ADC sampling clock domain (50MHz) to the baseband processing clock domain (200MHz) through the FIFO.

[0043] The chirp extraction unit is used to extract frames from the signal output by the FIFO unit. The chirp rate directly determines the system refresh rate. Generally, the refresh rate required for vital sign monitoring is 20Hz. When the chirp rate is too fast, the amount of data is too large, and the subsequent data processing capability is too high. Therefore, the chirp can be extracted according to the preset extraction rate, and the redundant chirp can be discarded to adjust the system refresh rate.

[0044] The data preprocessing unit is used to sample a number of sampling points from the extracted frames, and to number the frames and the sampling points in the frames in sequence to generate a time domain sampling point sequence. Specifically, when CHIRP_SYNC is high, it indicates that it is in the rising stage of the chirp and is a valid data segment. At this time, each chirp (frame) and the sampling points in the chirp are numbered separately. The frames are numbered in a circular manner of 0, 1..., N-1, and the sampling points in the frames are numbered in a circular manner of 0, 1..., M-1. Each sampling point will have two numbers, chirp_num and pulse_num, to facilitate subsequent processing. If the sampling point exceeds the range, a truncation operation is performed. If the sampling points are insufficient, a zero-filling operation is performed. At the same time, the module has a configurable sampling point delay. When the sampling point delay is configured to k, the numbering will start from the kth valid sampling point. The timing diagram is shown as follows. Figure 2 As shown, the data is also expanded from 12 bits of ADC to 16 bits. N is the number of two-dimensional frequency domain signal points, M is the number of one-dimensional frequency domain signal points, and is also the number of sampling points in the frame.

[0045] The FFT submodule includes a data selection unit, an FFT calculation unit, a data cache unit and a data control unit, and the data control unit is connected to the data selection unit, the FFT calculation unit and the data cache unit respectively. The data control unit is used to set the transformation parameters of the FFT calculation unit to one-dimensional FFT transformation parameters when the data received by the data selection unit is a time domain sampling point sequence, and control the data of the data cache unit to be output as a one-dimensional frequency domain signal to the frequency domain processing submodule. When the data received by the data selection unit is a phase feature of multiple targets, the transformation parameters of the FFT calculation unit are set to two-dimensional FFT transformation parameters, and control the data of the data cache unit to be output as a two-dimensional frequency domain signal. The FFT calculation unit is used to perform FFT calculation under the control of the data control unit, and can perform FFT transformation of the sampling point sequence output by the time domain processing submodule into a one-dimensional frequency domain signal (1D-FFT data) including M FFT frequency points, and perform FFT transformation of the phase feature output by the frequency domain processing submodule into a two-dimensional frequency domain signal (2D-FFT data) including N FFT points. The data cache unit is used to cache and output the data output by the FFT calculation unit under the control of the data control unit. In FFT calculation, in order to reduce the impact of data truncation on FFT spectrum leakage, the data is windowed. The length of the window function is consistent with the pulse length, and the FFT processing data width is 16 bits. In order to improve the FFT processing speed, the pulse data is stored in ping-pong mode. A pseudo dual-port RAM is used to support read operations on one port and write operations on the other port. One-dimensional FFT (distance dimension) calculation is performed, and the number of points is configurable, ranging from 2^2 to 2^10.

[0046] The frequency domain processing submodule includes a distance extraction unit, a phase extraction unit, a plurality of parallel dewrapping units, a parallel-to-serial conversion unit and a null filter.

[0047] The distance extraction unit is used to separate the frequency domain signal of each target from the one-dimensional frequency domain signal. Specifically, it includes a memory, a comparator and a judgment calculator. Figure 3 As shown, the memory may be an SRAM, which is used to assign a storage address to each frequency point in the one-dimensional frequency domain signal and store it; the comparator is used to compare the frequency points in all storage addresses outside the protection interval Q in the memory, find the frequency point with the largest modulus value and its storage address and output it to the judgment calculator until the number of found modulus values ​​reaches the target number; the judgment calculator is used to judge whether the storage address of the frequency point with the largest modulus value is located in the current protection interval Q; if it is, the frequency point is discarded and the execution comparator is returned; otherwise, the IQ data of the frequency point is extracted as the frequency domain signal of the identified target, and the protection interval Q is updated to: Returns the execution comparator, where the initial value of Q is the empty set and ADDR maxis the storage address of the frequency point with the largest modulus value, WIN is the preset unilateral protection window width, and setting a unilateral protection window for each target can reduce the influence of the signal side lobe on the target distance extraction, and M is the number of frequency points in the one-dimensional frequency domain signal. Among them, the extraction of IQ data is an existing technology and will not be repeated here.

[0048] For example, if the maximum number of monitoring targets is 3, first store the data of the positive frequency part of the one-dimensional frequency domain signal into an SRAM with a depth of 512. If M is less than 1024 and the positive frequency is less than 512, then only Then, the address is traversed in sequence to find the frequency point with the largest modulus value, and the address ADDR1 and its IQ data are output, and a protection interval is generated at the same time. Update the current protection interval Q = Q ∪ Q ′. Traverse again, if the address is in the current protection interval Q, discard the frequency point, otherwise keep it, output the address ADDR2 and its IQ data, and generate the protection interval Update the current protection interval Q = Q ∪ Q ′, repeat the above operation to get the ADDR and IQ data of the three targets, and convert the ADDR into the actual distance according to the frequency resolution. The processing flow chart is as follows: Figure 3 shown.

[0049] The phase extraction unit is used to extract the phase information from the frequency domain signal (IQ data) of each target. The CORDIC method is used for phase extraction. The output range is [-π, π] and the quantization is 16 bits.

[0050] The unwrapping unit can be set according to the number of targets. For example, three unwrapping units can be set for three targets. Different unwrapping units correspond to different targets, and each unwrapping unit unwraps the phase information belonging to the same target. The unwrapping unit specifically includes a memory, a controller, and an unwrapping operator. The memory can be an SRAM, which is used to select the phase information belonging to the target corresponding to the current unwrapping unit from the phase information sent by the phase extraction unit, and allocate storage addresses in sequence from front to back according to the time of each phase information for storage. The controller is used to determine whether the number of phase information stored in the memory is greater than or equal to N. If not, it waits. If so, it obtains the phase information in the most recent N consecutive storage addresses. The unwrapping operator is used to unwrap the phase information in the most recent N consecutive storage addresses in the order of the storage addresses from front to back according to the following method: Get the phase information located at the current storage address ADDR i The previous storage address ADDR i-1 The phase information D stored in i-1 and the phase information D in the current storage address i Subtract, when the difference is greater than π, the unwrapped phase D i '=Di -2π is output to the parallel-serial conversion unit, and the previous storage address ADDR is i-1 The phase information D stored in i-1 Updated to D i , when the difference is less than -π, the unwrapped phase D i '=D i +2π is output to the parallel-to-serial conversion unit, and the previous storage address ADDR is i-1 The phase information D stored in i-1 Updated to D i , when the difference is between -π and π, the unwrapped phase D i '=D i Output to the parallel-to-serial conversion unit and store the previous address ADDR i-1 The phase information D stored in i-1 Updated to D i .like Figure 4 As shown in the figure, the process of the de-wrapping unit is as follows: the target phase value output by the phase extraction module is written into the SRAM in sequence according to the addresses 0, 1, 2...N-1. When the N address data is filled, the 0 address data D0 is output normally first, the 1 address data D1 is compared with the 0 address data D0 to determine whether to compensate ±2π, and the processed 1 address data D1' is output, and the 1 address original data is written into the 0 address, and so on, and the N point data is output serially. After that, each incoming data will be stored in the N address, and the previous process will be repeated to output the N point data. Among them, the de-wrapping operator delays the previous stage data by one beat and then subtracts it from the current address data and compares it with π / -π through the comparator to control the data path of the data selector. When the difference is greater than π, the current value -2π is output. If the difference is less than -π, the current value +2π is output, otherwise the current value is output. When the number of two-dimensional FFT points is set to 1024, the maximum output data width of the de-wrapping module can reach 16+log1024=26bit.

[0051] The parallel-to-serial conversion unit is used to convert the unwrapped phase information from parallel to serial. The three unwrapping units perform unwrapping operations in sequence according to the specified priority. The state transition diagram is shown in Figure 5As shown. In the initial state IDLE, the three unwinding units are in a waiting state. When new data arrives, they enter the WORK1 state, the No. 1 unwinding unit starts working, and the other two unwinding units are in a waiting state; when the No. 1 unwinding unit completes the operation, it enters the WORK2 state, at which time the No. 2 unwinding unit starts working, and the other two unwinding units are in a waiting state; when the No. 2 unwinding unit completes the operation, it enters the WORK3 state, at which time the No. 3 unwinding unit starts working, and the other two unwinding units are in a waiting state; when the No. 3 unwinding unit completes the operation, it enters the IDLE state, at which time the three unwinding units are in a waiting state; if new data arrives when the three unwinding units are working, they will enter the ERROR state, the data will be discarded, and the ERROR flag will be output at the same time, and return to the previous state.

[0052] The null filter is used for filtering to obtain the phase characteristics of each target. The null filter uses a 3rd order IIR filter with a cutoff frequency of 0.02Hz and a transfer function of Among them, z1 and z2 are zero points, and μ is the pole distance parameter. Due to the ±2π compensation of the unwrapped data, the data dynamic range will increase sharply in extreme cases, causing the data bit width to increase to 26 bits. The use of the null filter can not only filter out the zero-frequency component in the unwrapped signal, but also reduce the dynamic range of the data. Therefore, the low 16-bit data after the null filter is intercepted as the output data.

[0053] After the null filter, the data width is reduced to 16 bits, so it can share an FFT calculation unit with the one-dimensional FFT. The data source is selected through the FFT control module, and the data is windowed. The length of the window function is consistent with the FFT data length. To support three targets, three one-dimensional vector FFT calculations are required. After the calculation is completed, a two-dimensional frequency domain signal (2D-FFT) is obtained and the signal is written to the specified storage space. The two-dimensional frequency domain signal can be used to extract the vital signs information of each target.

[0054] The embodiment of the present invention also provides a multi-target vital signs monitoring system based on the above-mentioned millimeter wave radar baseband module, which includes, in addition to the above-mentioned millimeter wave radar baseband module, a vital signs extraction submodule, which is used to extract the vital signs information of each target from the two-dimensional frequency domain signal. For example, continuing with the above example, the vital signs extraction submodule can read the two-dimensional frequency domain data of three targets, intercept the corresponding respiratory rate and heart rate intervals respectively according to the frequency resolution, find the respective extreme values ​​in these two intervals, perform frequency conversion according to the coordinates of the extreme value points, and obtain the respiratory rate and heart rate of the three monitored targets.

[0055] It is worth noting that in the embodiments of the above modules, the various units and modules included are only divided according to functional logic, but are not limited to the above divisions, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0056] It should be understood that the above embodiments and descriptions only describe the principles, main features and advantages of the present invention. Without departing from the spirit and scope of the present invention, the present invention may be subject to various changes and improvements, and these changes and improvements all fall within the scope of protection of the present invention.

Claims

1. A millimeter wave radar baseband module for multi-target vital signs monitoring, characterized in that: It includes a time domain processing submodule, a frequency domain processing submodule and an FFT submodule. The time domain processing submodule is used to filter and preprocess the digital intermediate frequency signal of the echo signal reflected by the target to obtain a time domain sampling point sequence. The FFT submodule is used to perform FFT transformation on the time domain sampling point sequence output by the time domain processing submodule to obtain a one-dimensional frequency domain signal. The frequency domain processing submodule is used to extract the phase characteristics of multiple targets from the one-dimensional frequency domain signal. The FFT submodule is also used to perform FFT transformation on the phase characteristics of multiple targets to obtain a two-dimensional frequency domain signal for extracting the vital sign information of each target. Among them, the frequency domain processing submodule includes a distance extraction unit, a phase extraction unit, a plurality of parallel dewrapping units, a parallel-to-serial conversion unit and a null filter. The distance extraction unit is used to separate the frequency domain signal of each target from the one-dimensional frequency domain signal. The phase extraction unit is used to extract the phase information from the frequency domain signal of each target. Each of the dewrapping units is used to dewrapped the phase information belonging to the same target. Different dewrapping units correspond to different targets. The parallel-to-serial conversion unit is used to convert the dewrapped phase information from parallel to serial. The null filter is used to filter and obtain the phase characteristics of each target. The distance extraction unit specifically includes: A memory, used to assign a storage address to each frequency point in the one-dimensional frequency domain signal and store it; A comparator is used to compare the frequency points in all storage addresses outside the protection interval Q in the memory, find the frequency point with the largest modulus value and its storage address and output them to the judgment calculator until the number of found modulus values ​​reaches the target number; The judgment calculator is used to judge whether the storage address of the frequency point with the largest modulus value is within the current protection interval Q; if it is, the frequency point is discarded and the execution comparator is returned; otherwise, the IQ data of the frequency point is extracted as the frequency domain signal of the identified target, and the protection interval Q is updated to: Returns the execution comparator, where the initial value of Q is the empty set and ADDR max is the storage address of the frequency point with the largest modulus value, WIN is the preset unilateral protection window width, and M is the number of frequency points in the one-dimensional frequency domain signal.

2. The millimeter wave radar baseband module for multi-target vital signs monitoring according to claim 1, characterized in that: It also includes an APB interface, and the time domain processing submodule, frequency domain processing submodule, FFT submodule and vital sign extraction submodule are all connected to the APB interface.

3. The millimeter wave radar baseband module for multi-target vital signs monitoring according to claim 1, characterized in that: The time domain processing submodule specifically includes a filter, a FIFO unit, a Chirp extraction unit and a data preprocessing unit connected in sequence, wherein the filter is used to filter the digital intermediate frequency signal of the echo signal reflected from the target, the FIFO unit is used to cache the filtered echo signal according to a first-in-first-out mechanism, the Chirp extraction unit is used to perform frame extraction on the signal output by the FIFO unit, and the data preprocessing unit is used to sample a number of sampling points from the extracted frames, and number the frames and the sampling points in the frames in sequence to generate a time domain sampling point sequence.

4. The millimeter wave radar baseband module for multi-target vital signs monitoring according to claim 3, characterized in that: The data preprocessing unit is used to circularly number the frames according to 0, 1..., N-1, and the sampling points in the frames according to 0, 1..., M-1. If the sampling points are out of range, a truncation operation is performed, and if the sampling points are insufficient, a zero padding operation is performed. N is the number of two-dimensional frequency domain signal points, and M is the number of one-dimensional frequency domain signal points.

5. The millimeter wave radar baseband module for multi-target vital signs monitoring according to claim 3, characterized in that: The data preprocessing unit has a configurable sampling point delay. When the sampling point delay is configured as k, the numbering will start from the kth valid sampling point.

6. The millimeter wave radar baseband module for multi-target vital signs monitoring according to claim 3, characterized in that: The filter includes a CIC filter and a FIR filter connected in sequence.

7. The millimeter wave radar baseband module for multi-target vital signs monitoring according to claim 1, characterized in that: The unwinding unit specifically comprises: A memory, used to select the phase information belonging to the target corresponding to the current dewrapping unit from the phase information sent by the phase extraction unit, and allocate storage addresses in sequence from front to back according to the time of each phase information for storage; The controller is used to determine whether the number of phase information stored in the memory is greater than or equal to N, if not, then wait, if yes, then obtain the phase information in the latest N consecutive storage addresses, N is the number of frames of the two-dimensional frequency domain signal; The unwrapping operator is used to unwrap the phase information in the latest N consecutive storage addresses in the order of storage addresses from front to back according to the following method: obtain the phase information at the current storage address ADDR i The previous storage address ADDR i-1 The phase information D stored in i-1 and the phase information D in the current storage address i Subtract, when the difference is greater than π, the unwrapped phase D i '=D i -2π is output to the parallel-serial conversion unit, and the previous storage address ADDR is i-1 The phase information D stored in i-1 Updated to D i , when the difference is less than -π, the unwrapped phase D i '=D i +2π is output to the parallel-to-serial conversion unit, and the previous storage address ADDR is i-1 The phase information D stored in i-1 Updated to D i , when the difference is between -π and π, the unwrapped phase D i '=D i Output to the parallel-to-serial conversion unit and store the previous address ADDR i-1 The phase information D stored in i-1 Updated to D i .

8. The millimeter wave radar baseband module for multi-target vital signs monitoring according to claim 1, characterized in that: The FFT submodule includes a data selection unit, an FFT calculation unit, a data cache unit and a data control unit. The data control unit is connected to the data selection unit, the FFT calculation unit and the data cache unit respectively. The data control unit is used to set the transformation parameters of the FFT calculation unit to one-dimensional FFT transformation parameters when the data received by the data selection unit is a time domain sampling point sequence, and control the data of the data cache unit to be output as a one-dimensional frequency domain signal to the frequency domain processing submodule. When the data received by the data selection unit is the phase characteristics of multiple targets, the transformation parameters of the FFT calculation unit are set to two-dimensional FFT transformation parameters, and control the data of the data cache unit to be output as a two-dimensional frequency domain signal. The FFT calculation unit is used to perform FFT calculation under the control of the data control unit, and the data cache unit is used to cache and output the data output by the FFT calculation unit under the control of the data control unit.

9. A multi-target vital signs monitoring system based on a millimeter wave radar baseband module, characterized in that: It includes a time domain processing submodule, a frequency domain processing submodule, an FFT submodule and a vital sign extraction submodule, wherein the time domain processing submodule is used to filter and preprocess the digital intermediate frequency signal of the echo signal reflected by the target to obtain a time domain sampling point sequence, the FFT submodule is used to perform FFT transformation on the time domain sampling point sequence output by the time domain processing submodule to obtain a one-dimensional frequency domain signal, the frequency domain processing submodule is used to extract the phase characteristics of multiple targets from the one-dimensional frequency domain signal, the FFT submodule is also used to perform FFT transformation on the phase characteristics of multiple targets to obtain a two-dimensional frequency domain signal, and the vital sign extraction submodule is used to extract the vital sign information of each target from the two-dimensional frequency domain signal; Among them, the frequency domain processing submodule includes a distance extraction unit, a phase extraction unit, a plurality of parallel dewrapping units, a parallel-to-serial conversion unit and a null filter. The distance extraction unit is used to separate the frequency domain signal of each target from the one-dimensional frequency domain signal. The phase extraction unit is used to extract the phase information from the frequency domain signal of each target. Each of the dewrapping units is used to dewrapped the phase information belonging to the same target. Different dewrapping units correspond to different targets. The parallel-to-serial conversion unit is used to convert the dewrapped phase information from parallel to serial. The null filter is used to filter and obtain the phase characteristics of each target. The distance extraction unit specifically includes: A memory, used to assign a storage address to each frequency point in the one-dimensional frequency domain signal and store it; A comparator is used to compare the frequency points in all storage addresses outside the protection interval Q in the memory, find the frequency point with the largest modulus value and its storage address and output them to the judgment calculator until the number of found modulus values ​​reaches the target number; The judgment calculator is used to judge whether the storage address of the frequency point with the largest modulus value is within the current protection interval Q; if it is, the frequency point is discarded and the execution comparator is returned; otherwise, the IQ data of the frequency point is extracted as the frequency domain signal of the identified target, and the protection interval Q is updated to: Returns the execution comparator, where the initial value of Q is the empty set and ADDR max is the storage address of the frequency point with the largest modulus value, WIN is the preset unilateral protection window width, and M is the number of frequency points in the one-dimensional frequency domain signal.

Images