TW202240198A - Doppler signal processing device and doppler signal processing method - Google Patents

Abstract

The Doppler signal processing device includes a frequency analyzer, an interference suppressor, an interference baseline estimator and a synthesizer. The frequency analyzer can generate a frequency domain signal according to a digital signal. The interference suppressor can perform a suppression operation to generate an interference suppressed frequency domain signal. The interference baseline estimator can generate or update the frequency domain interference estimation signal according to the frequency domain signal. The synthesizer can generate an interference suppressed time domain digital signal according to the interference suppressed frequency domain signal, where the interference suppressed time domain digital signal is related to the digital signal with the interference energy suppressed.

Description

Doppler signal processing device and Doppler signal processing method

The invention relates to a Doppler signal processing device, in particular to a Doppler signal processing device for suppressing background interference.

Background interference often causes problems when using radar devices to detect objects. For example, if a radar device is used to detect vital signs or human bodies, background interference is often significant, resulting in incorrect detection results. In order to reduce the influence of background interference, a signal energy threshold device or a frequency energy detection method can be used to judge the detection result. However, none of these methods can provide high-precision detection results.

The embodiment provides a Doppler signal processing device for suppressing a background interference according to a received wireless signal, the Doppler signal processing device includes a frequency analyzer, an interference suppressor, an interference baseline estimator and a synthesizer. The frequency analyzer is used to generate a frequency domain signal according to a digital signal, wherein the digital signal corresponds to the received wireless signal, and the digital signal contains interference energy generated by the interference of the background. The interference suppressor is used to perform a suppression operation according to the frequency-domain signal and a frequency-domain interference estimation signal to generate an interference-suppressed frequency-domain signal, wherein in the interference-suppressed frequency-domain signal, the interference energy in the frequency domain is suppressed by the The frequency-domain interference estimation signal is suppressed. The interference baseline estimator is used for generating or updating the frequency domain interference estimation signal according to the frequency domain signal, wherein the frequency domain interference estimation signal corresponds to an energy distribution of the interference in the frequency domain. The synthesizer is used to generate an interference-suppressed time-domain digital signal according to the interference-suppressed frequency-domain signal, wherein the interference-suppressed time-domain digital signal is related to the digital signal whose interference energy has been suppressed.

Another embodiment provides a Doppler signal processing method for suppressing background interference based on a received wireless signal, the Doppler signal processing method includes generating a frequency domain signal based on a digital signal, wherein the digital signal corresponds to The received wireless signal, and the digital signal contains interference energy generated by the background interference; performing a suppression operation according to the frequency domain signal and a frequency domain interference estimation signal to generate an interference suppressed frequency domain signal, wherein in the interference In suppressing the frequency domain signal, the interference energy is suppressed in the frequency domain; generating or updating the frequency domain interference estimation signal according to the frequency domain signal, wherein the frequency domain interference estimation signal corresponds to the interference in one of the frequency domains energy distribution; and generating an interference-suppressed time-domain digital signal based on the interference-suppressed frequency-domain signal, wherein the interference-suppressed time-domain digital signal is related to the digital signal whose interference energy has been suppressed.

Exemplary embodiments will be described in detail below with reference to the accompanying drawings for easy implementation by those skilled in the art. The inventive concept can be embodied in various forms without being limited to the exemplary embodiments set forth herein. For clarity, descriptions of well-known components have been omitted, and like reference numbers refer to like elements.

FIG. 1 is a schematic diagram of a signal detection device 10 for detecting an object Obj in an embodiment. The signal detection device 10 can be a radar. The signal detection device 10 may include antennas ANtx and ANrx, amplifiers Atx and Arx, an oscillator OSC, a receiving unit 1066 , and a Doppler signal processing device 100 . The Doppler signal processing device 100 can suppress interference, as described below. The signal detection device 10 may further include digital-to-analog converters 15 and 16 , and a digital serial interference unit 17 . The coupling relationship of these components can be shown in FIG. 1 .

The oscillator OSC can provide a carrier signal fc, and the carrier signal fc has a substantially fixed frequency. The amplifying device Atx (such as a power amplifier) can output a wireless signal d (such as a radio frequency transmission signal), which is transmitted by the antenna ANtx. The wireless signal d can be generated according to the carrier signal fc. The generator for generating the wireless signal d can be selectively arranged between the oscillator OSC and the amplifying device Atx according to the carrier signal fc. If the object Obj (such as a human body) is within the detection range of the signal detection device 10, the object Obj can reflect the wireless signal d to generate a received wireless signal d', such as a radio frequency received signal. The antenna ANrx can receive the reflected received wireless signal d', the amplifying device Arx (such as a low noise amplifier) can amplify the received wireless signal d', and the received wireless signal d' can be input to the receiving unit 1066.

The receiving unit 1066 can generate at least one digital signal according to the carrier signal fc and the received wireless signal d'. For example, the receiving unit 1066 may include mixers 1071 and 1072 , a phase adjustment unit 1073 , and analog-to-digital converters (ADCs) 1074 and 1075 . The mixer 1071 can mix the carrier signal fc and the received wireless signal d', and convert the result into a digital signal S1 by the analog-to-digital converter 1074. The phase adjusting unit 1073 can rotate the phase of the carrier signal fc by a specific phase (for example, 90 degrees) to generate an adjusted carrier signal fc'. The mixer 1072 can mix the adjusted carrier signal fc' and the received wireless signal d', and convert the result into a digital signal S2 by the analog-to-digital converter 1075. The at least one digital signal output by the receiving unit 1066 may include the digital signals S1 and/or S2 output by the analog-to-digital converters 1074 and/or 1075 .

For example, the digital signals S1 and S2 can be time-domain signals, which are respectively related to an in-phase channel and a quadrature channel of In-phase/quadrature (I/Q) modulation. The Doppler signal processing device 100 can process the digital signals S1 and S2 to suppress the interference part of the digital signals S1 and S2, so as to generate digital signals S1' and S2'.

The digital signals S1 ′ and S2 ′ can be time-domain signals, which are respectively processed by the digital-to-analog converters 15 and 16 to generate the analog signals S1 a and S2 a. For example, the analog signals S1a and S2a may relate to the in-phase channel and the quadrature channel of quadrature/in-phase (I/Q) modulation, respectively. The digital serial interference unit 17 can process the digital signals S1 ′ and S2 ′ to generate data for detecting objects. The data generated by the digital serial interference unit 17 can be transmitted to the outside through the serial peripheral interface (SPI) or the inter-integrated circuit (I ² C) interface, such as a device selected by the user for analysis. For example, a microcontroller (MCU) can be used to analyze the data generated by the serial jamming unit 17 to detect the object Obj. Because the interference part of the time-domain signal has been suppressed, compared with the digital signals S1 and S2, the digital signals S1' and S2' can be "cleaner" (that is, contain less interference energy in the time domain), and the user It allows relatively simple hardware/software to process data related to the detected object Obj. The structure of the receiving unit 1066 in FIG. 1 is just an example, not intended to limit the structure of the circuit.

FIG. 2 is a schematic diagram of the Doppler signal processing device 100 in FIG. 1 . The Doppler signal processing device 100 can suppress background interference according to the received wireless signal d'. The suppressed interference may be stationary interference (stationary interference), for example, may be quasi-stationary interference (quasi-stationary interference) whose interference frequency is substantially constant or does not change much. The Doppler signal processing device 100 may include a frequency analyzer 110 , an interference suppressor 120 , an interference baseline estimator 130 and a synthesizer 140 . In FIG. 2, i can be an index of time, n can be an index of a time domain block, and k can be an index of frequency, that is, a frequency bin.

The frequency analyzer 110 can generate a frequency-domain signal _Xn (k) according to the digital signal x(i). The digital signal x(i) in Figure 2 can correspond to the digital signal S1 and/or S2 in Figure 1, and is the original data of the in-phase channel and quadrature channel related to quadrature/in-phase (I/Q) modulation . The digital signal x(i) may correspond to the received wireless signal d' and contains interference energy (to be suppressed) caused by interference in the background.

The interference suppressor 120 can perform a suppression operation according to the frequency-domain signal X _n (k) and the frequency-domain interference estimation signal U _n (k) to generate the interference-suppressed frequency-domain signal Y _n (k). In the interference-suppressed frequency-domain signal Y _n (k), the interference energy in the frequency domain can be suppressed by using the frequency-domain interference estimation signal U _n (k).

The interference baseline estimator 130 can generate or update the frequency-domain interference estimation signal U _n (k) according to the frequency-domain signal X _n (k). The frequency-domain interference estimation signal U _n (k) may correspond to the energy distribution of the interference in the frequency domain (eg, quasi-stationary interference energy).

The synthesizer 140 can generate an interference-suppressed time-domain digital signal y(i) according to the interference-suppressed frequency-domain signal Y _n (k). The interference-suppressed time-domain digital signal y(i) can be related to the digital signal x(i) whose interference energy has been suppressed. In FIG. 1 and FIG. 2, the interference-suppressed time-domain digital signal y(i) in FIG. 2 may correspond to the digital signals S1' and S2' in FIG. 1.

From FIG. 2 to FIG. 6, the frequency-domain signal X _n (k) may include a frequency-domain signal vector, and the frequency-domain interference estimation signal U _n (k) may include a frequency-domain interference estimation signal vector.

FIG. 3 is a schematic diagram of the Doppler signal processing device 100 shown in FIG. 1 and FIG. 2 in an embodiment. As shown in FIG. 3 , frequency analyzer 110 may perform STFT 1102 to generate frequency-domain signal _Xn (k) using digital signal x(i). The frequency analyzer 110 can further perform a buffering operation 1101 to collect the digital signal x(i) at different time points as the input of the STFT 1102 .

As shown in FIG. 3 , the synthesizer 140 can use the interference-suppressed frequency-domain signal Y _n (k) to perform an inverse fast Fourier transform (Inverse FFT) 1401 to generate a time-domain signal corresponding to the interference-suppressed time-domain digital signal y(i) y _n (m). The synthesizer 140 can also use the time-domain signal y _n (m) to perform an overlap-and-accumulate operation 1402 to generate an interference-suppressed time-domain digital signal y(i).

FIG. 4 is a schematic diagram of signal processing related to the short-time Fourier transform (STFT) in FIG. 3 . In FIG. 4, M, K, L, and O may be indexes related to discrete-time signals. A signal denoted X may be related to the frequency domain, a signal denoted x may be related to the time domain, and m and n may be indexes of time domain blocks.

In FIGS. 3 and 4, for example, buffering operation 1101 may include collecting K samples of digital signal x(i) to generate signal xn( _m ). The frequency domain signal X _n (k) can be expressed as {X _n (k): k = 0~(K-1)}, and the interference suppressed frequency domain signal Y _n (k) can be expressed as {Y _n (k): k = 0~(K-1)}.

In Figure 3, the interference suppression frequency domain signal Y _n (k) can be expressed as Y _n (k) = A _n (k) × e ^{j ∠ Xn(k)} , or as Y _n (k) = B _n ( k) × e ^{j ∠Xn (k)} , as described later.

According to an embodiment, the interference-suppressed frequency-domain signal Y _n (k) can be expressed as Y _n (k) = A _n (k) × e ^{j ∠ Xn(k)} according to the spectral subtraction method, where A _n (k) = MAX {|X _n (k)| - α•U _n (k),η }. The parameter η can be the minimum value of A _n (k), and the parameter α can be adjusted to control the signal-to-noise ratio (SNR) to achieve a target SNR.

According to an embodiment, the interference-suppressed frequency-domain signal Y _n (k) can be expressed as Y _n (k) = B _n (k) × e ^{j ∠ Xn(k)} according to the spectrum weighting method, where B _n (k) = (|X _n (k)| / (1+ β·U _n (k)). The parameter β can be adjusted to control the signal-to-noise ratio (SNR) to achieve the target SNR.

In FIG. 3 , STFT 1102 can analyze and convert time-domain data (eg, signal xn( _m )) to generate frequency-domain data (eg, frequency-domain signal _Xn (k)). The interference baseline estimator 130 can estimate the interference baseline relative to a fixed frequency range. For example, the interference may be (but not limited to) interference from fans, fluorescent lamps, and air conditioner adapters. Interference can be suppressed to improve the accuracy of detecting the object Obj. In Fig. 3, the interference part of the interference-suppressed frequency-domain signal Y _n (k) has been reduced, and an inverse fast Fourier transform 1401 can be performed to process the interference-suppressed frequency-domain signal Y _n (k) in the frequency domain to generate the time-domain The time-domain signal y _n (m) of .

The STFT 1102 in FIG. 3 can be as shown in FIG. 4 . Figure 4 is only an example, not limiting the scope of the embodiment. Figure 4 illustrates the operation of the six blocks B0 to B5. The horizontal axis of Fig. 4 may correspond to a sample index in the time domain, which is used in digital signal processing (DSP). Block B0 may correspond to n=1, block B1 may correspond to n=2, and so on. Regarding block BO, there may be K samples, of which L samples do not overlap, but O samples overlap with some samples of block B1.

…eq-1; X _n(k) = FFT {x _n(m)} =

_n(m)•exp (-j

)   …eq-2; y _n(m) = IFFT{Y _n(k)}=

…eq-3; y(i) =

…eq-4; According to an embodiment, the aforementioned signals x _n (m), X _n (k), y _n (m) and y (i) can be expressed and obtained according to the following equations eq-1 to eq-3: x _n (m) =

…eq-1; X _n (k) = FFT {x _n (m)} =

_n (m)·exp (-j

) …eq-2; y _n (m) = IFFT{Y _n (k)}=

...eq-3; y(i) =

...eq-4;

Wherein M and K can be positive integers, M ≤ K, and M can be optionally equal to K. Equation eq-1 may be related to the buffering operation 1101 of FIG. 3 , where the pane signal w(m) may be a known pane signal, as shown in FIG. 4 . Equation eq-2 can be related to the short-time Fourier transform shown in Fig. 3 . Equation eq-3 can be related to the inverse fast Fourier transform 1401 shown in FIG. 3 . Equation eq-4 may be related to the overlap-accumulate operation 1402 shown in FIG. 3 . Equations eq-1 to eq-4 can describe the operation principle of the Doppler signal processing device 100 .

Fig. 5 is a schematic diagram of the Doppler signal processing device 100 in Fig. 1 and Fig. 2 in the embodiment. As shown in FIG. 5 , the frequency analyzer 110 may include a filter bank (filter bank) 110A for filtering the digital signal x(i) to generate a frequency domain signal X _n (k). As shown in FIG. 5, the synthesizer 140 may include a filter bank 140A for filtering the interference-suppressed frequency-domain signal _Yn (k) to generate the interference-suppressed time-domain digital signal y(i). According to an embodiment, the filter bank 110A may be an analysis filter bank, and the filter bank 140A may be a synthesis filter bank. For example, the filter bank 110A may be decimated, and the filter bank 140A may be interpolated accordingly; or, the filter bank 110A may be non-decimated, and the filter bank 140A may be interpolated accordingly; 140A may be correspondingly non-interpolated.

In Figure 5, the interference suppression frequency domain signal Y _n (k) can be expressed as Y _n (k) = A _n (k)·e ^{j ∠ Xn(k)} , or as Y _n (k) = B _n ( k)•e ^{j ∠Xn (k)} , as described below.

According to an embodiment, the interference-suppressed frequency-domain signal Y _n (k) can be expressed as Y _n (k) = A _n (k)•e ^{j ∠Xn (k)} according to the spectral subtraction method, where A _n (k) = MAX {|X _n (k)| - α•U _n (k),η}. The parameter η can be the minimum value of A _n (k), and the parameter α can be adjusted to control the signal-to-noise ratio (SNR) to achieve a target SNR.

According to an embodiment, the interference-suppressed frequency-domain signal Y _n (k) can be expressed as Y _n (k) = B _n (k)·e ^{j ∠ Xn(k)} according to the spectrum weighting method, where B _n (k) = (|X _n (k)| / (1+β·U _n (k)). The parameter β can be adjusted to control the signal-to-noise ratio (SNR) to achieve the target SNR. In this case, zero can be used Zero-forcing method. For example, this case can be related to the correlation filter of the Wiener (Wiener) solution. In this case, the signal strength can be attenuated while maintaining the phase.

FIG. 6 is a schematic diagram of the interference baseline estimator 130 described in FIG. 2 , FIG. 3 and FIG. 5 . As shown in FIG. 6 , the interference baseline estimator 130 may include an absolute value data unit 610 , a filter unit 620 , a subtractor 630 , a control unit 640 , an adder 650 and a delay unit 660 .

The absolute value data unit 610 can generate absolute value data |X _n (k)| of the frequency domain signal X _n (k). The filtering unit 620 can generate a filtering signal (R _n (k) or E _n (k)) according to the absolute value data |X _n (k)|. The subtractor 630 can generate difference data Δ _n (k) between the filtered signal (R _n (k) or E _n (k)) and the frequency-domain interference estimation signal U _n (k). The control unit 640 can process the difference data Δ _n (k) to generate the adjustment data Q _n (k). The adder 650 can add the adjustment data Q _n (k) and the frequency-domain interference estimation signal U _n (k) to generate an updated frequency-domain interference estimation signal U _n+1 (k). The delay unit 660 can use the updated frequency-domain interference estimation signal U _n+1 (k) to replace the frequency-domain interference estimation signal U _n (k). In other words, the frequency-domain interference estimation signal U _n (k) and the updated frequency-domain interference estimation signal U _n+1 (k) can be respectively related to a frequency-domain signal block index and the next frequency-domain signal block on the time line index, wherein the difference between the two frequency-domain signal block indices can be related to the delay unit 660 .

According to an embodiment, the absolute value data | _Xn (k)| contains absolute values, the difference data _Δn (k) contains difference values, the adjustment data _Qn (k) contains adjustment values, and the filtered signal _Rn (k) or _En (k) contains the filtered data vector.

Regarding the filtering unit 620 , the filtering unit 620 may at least include a filter 6201 and optionally further include an envelope detector 6202 . The filter 6201 can perform a filtering operation at block index n on the timeline, and/or a filtering operation at frequency bin k.

If the filtering unit 620 includes a filter 6201 and does not include an envelope detector 6202, the filter 6201 can filter the absolute value data |X _n (k)| to generate a filtered signal R _n (k), and the filtered signal R _n ( k) may be input to the subtractor 630 .

In another case, if the filtering unit 620 includes a filter 6201 and an envelope detector 6202, the filter 6201 can filter the absolute value data |X _n (k)| to generate a filtered signal R _n (k) as a preliminary filter signal, wherein the preliminary filtered signal may include a preliminary filtered signal vector. The envelope detector 6202 can obtain the envelope information of the preliminary filtered signal R _n (k) to generate the filtered signal E _n (k). The envelope detector 6202 can make the result smoother on the timeline, and can reduce unexpected jitter caused by burst noise, because the system is mainly used to deal with fixed noise.

FIG. 7 is a schematic diagram of a finite state machine (FSM) of the control unit 640 in FIG. 6 . The control unit 640 can be used for tracking control. The tracking control finite state machine of the k-th frequency grid can be shown in FIG. 7 . A finite state machine may include states 710-730.

The state 710 can be an idle state, corresponding to the situation that the Doppler signal processing device 100 is idle. When the Doppler signal processing device 100 starts up, it can send out the startup signal S _{system_startup} . When receiving the information of the object Obj, the initial data U _initial (k) of the frequency-domain estimation signal U _n (k) can be obtained, and then enter the state 715 . The control unit 640 can obtain the initial data U _initial (k) according to the difference data Δ _n (k) and the filter signal R _n (k) or E _n (k) for a period of time. According to an embodiment, the initial data U _initial (k) may include an initial value.

For example, if the filtered signal _is the filtered signal _En ( _k ) generated by the envelope detector ₆₂₀₂ in FIG. (k), E ₂ (k)...E _Nacq-1 (k) elapsed time, that is, the time when the index n of the time domain block changes from 0 to Nacq-1. The state 715 may be an information acquisition state, where U _initial may be an average of a plurality of amplitude values (ie, signal strength). For example, U _initial (k) may be AVG{E ₀ (k), E ₁ (k), E ₂ (k) . . . E _Nacq-1 (k)}, and AVG{} may be an average function.

After state 715 , the finite state machine may enter state 720 . State 720 may be a slow tracking state. The control unit 640 can enter a slow tracking state (state 720 ) to update the frequency-domain interference estimation signal U _n (k) at a first speed. In this case, the update of the signal can be slower and more original parts of the signal can be preserved.

After state 720, the finite state machine can enter state 725 according to the signal S _{to_fast} . State 725 may be a fast track state. The control unit 640 can enter the fast tracking state (state 725 ) to update the frequency-domain interference estimation signal U _n (k) at a second speed, wherein the second speed is higher than the first speed. In this case, the update of the signal can be faster and less original parts of the signal can be preserved.

After state 720, the finite state machine can enter state 725 according to the signal S _{~to_fast} . As shown in FIG. 7, the finite state machine can alternately enter state 720 or state 725 as desired.

After the state 720, the finite state machine can enter the state 730 according to the signal S _{~(to_slow|to_fast)} . The state 730 may be a freezing state, which is used to stop updating the frequency-domain interference estimation signal _Un (k). The finite state machine can enter the state 720 from the state 730 according to the signal S _{to_slow} .

FIG. 8 is a schematic diagram of the range of the filtered signal generated by the filtering unit 620 in FIG. 6 . The range of FIG. 8 can be related to the state shown in FIG. 7 . As mentioned above, the filtered signal can be the filtered signal R _n (k) or E _n (k) in FIG. 6 , and the filtered signal E _n (k) is taken as an example here. When the intensity of the filtered signal E _n (k) is between the first threshold value and the second threshold value higher than the first threshold value, the control unit 640 can enter the slow tracking state (for example, the finite state machine in FIG. 7 status 720). For example, as shown in FIG. 8, the first threshold may be -λ U _n (k), and the second threshold may be γ U _n (k), where λ and γ may be given Positive value. When the intensity of the filtered signal _{En(k) is lower than the first threshold value -λ•U n (} k), the control unit 640 enters a fast-tracking state (eg, the state 725 of the finite state machine in FIG. 7 ). When the intensity of the filtered signal _{En(k) is higher than the second threshold γ•U n (} k), the control unit 640 may enter into a freezing state (eg, the state 730 of the finite state machine in FIG. 7 ). In other words, the above-mentioned signal S _{to_slow} may be related to the condition of E _n (k) < γ • U _n (k) and E _n (k) ≥ -λ • U _n (k). The above-mentioned signal S _{to_fast} may be related to another situation, that is, the situation of E _n (k) < -λ·U _n (k). The above signal S _{~(to_slow|to_fast)} can be related to another situation, that is, the situation of E _n (k) ≥ γ•U _n (k).

…eq-5; Regarding the adjustment data Q _n (k) in Fig. 6, the adjustment data Q _n (k) can be expressed as the following equation eq-5: Q _n (k) =

...eq-5;

Wherein α _fast and α _slow are positive values between 0 and 1, and can be selected to generate the adjustment data Q _n (k). For example, the "fast track" state, "slow track" state, and "freeze" state described in equation eq-5 may correspond to state 725, state 720, and state 730 of FIG. 7, respectively.

Figures 1 to 7 can describe the structure and state of the device provided by the embodiment. In the Doppler signal processing device 100 in FIG. 1, the blocks in FIG. 2, FIG. 3, FIG. 5 and FIG. and implemented, or can be implemented with software functions embedded in a microprocessor (MCU) or a digital signal processor (DSP).

FIG. 9 is a flowchart of a Doppler signal processing method 900 for suppressing background interference according to the received wireless signal d' in FIG. 1 . The Doppler signal processing method 900 can be implemented by the above-mentioned signal detection device 10 and the Doppler signal processing device 100 , and can include the following steps.

Step 910: Generate a frequency-domain signal X _n (k) according to the digital signal x(i), wherein the digital signal x(i) can correspond to the received wireless signal d', and the digital signal x(i) includes interference caused by background interference energy;

Step 920: Perform a suppression operation according to the frequency-domain signal X _n (k) and the frequency-domain interference estimation signal U _n (k) to generate an interference-suppressed frequency-domain signal Y _n (k), wherein the interference-suppressed frequency-domain signal Y _n ( In k), the interference energy can be suppressed in the frequency domain;

Step 930: Generate or update a frequency-domain interference estimation signal U _n (k) according to the frequency-domain signal X _n (k), wherein the frequency-domain interference estimation signal U _n (k) corresponds to the energy distribution of the interference in the frequency domain; and

Step 940: Generate an interference-suppressed time-domain digital signal y(i) according to the interference-suppressed frequency-domain signal Y _n (k), wherein the interference-suppressed time-domain digital signal y(i) is related to the digital signal x( i).

Steps 910, 920, 930, and 940 may be associated with the frequency analyzer 110, the interference suppressor 120, the interference baseline estimator 130, and the synthesizer 140 shown in FIG. 2, FIG. 3, and FIG. 5, respectively. Relevant details can be as above, so it will not be repeated.

In summary, the signal detection device 10, the Doppler signal processing device 100 and the Doppler signal processing method 900 provided by the embodiment can effectively update the detection result by suppressing the interference, and the fixed interference in the time-domain signal ( For example, quasi-stationary disturbances in the background) can be greatly reduced and, therefore, precise tracking control can be achieved. In addition, the interference suppression calculation can reduce the interference energy corresponding to a specific frequency band in the background in the time domain signal, so the detection accuracy can be improved. The above descriptions are only preferred embodiments of the present invention, and all equivalent changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope of the present invention.

10: signal detection device 100: Doppler signal processing device 1066: receiving unit 1071, 1072: mixer 1073: phase adjustment unit 1074, 1075: analog to digital converter 110: frequency analyzer 1101: buffer operation 1102: short Time-distance Fourier Transform 110A, 140A: Filter Bank 120: Interference Suppressor 130: Interference Baseline Estimator 140: Synthesizer 1401: Inverse Fast Fourier Transform 1402: Overlap-Accumulate Operation 15, 16: Digital-to-Analog Converter 17: Digital Serial interference unit 610: absolute value data unit 620: filter unit 6201: filter 6202: envelope detector 630: subtractor 640: control unit 650: adder 660: delay unit 710, 715, 720, 725, 730: state 900: Doppler signal processing Methods 910, 920, 930, 940: Steps ANtx, ANrx: Antenna Atx, Arx: Amplifying devices B0, B1, B2, B3, B4, B5: Block d: Wireless signal d': Received wireless signal fc: Carrier signal fc': Adjusted carrier signal Obj: Object OSC: Oscillator Q _n (k): Adjustment data R _n (k), E _n (k): Filtered signals S1, S2, x(i), S1', S2': Digital signals S1a, S2a: Analog signal S _{system_startup} : start signal U _n (k): frequency domain interference estimation signal U _n+1 (k): update frequency domain interference estimation signal w(m): pane signal X _n (k): frequency domain Signal x _n (m), S _{to_fast} , S _{~to_fast} , S _{to_slow} , S _{~(to_slow|to_fast)} : Signal x(i): Digital signal y(i): Interference suppression time-domain digital signal Y _n (k): Interference suppression frequency domain signal y _n (m): time domain signal |X _n (k)|: absolute value data Δ _n (k): difference data

Fig. 1 is a schematic diagram of a signal detection device for detecting objects in an embodiment. Fig. 2 is a schematic diagram of the Doppler signal processing device in Fig. 1. Fig. 3 is a schematic diagram of the Doppler signal processing device in Fig. 1 and Fig. 2 in an embodiment. FIG. 4 is a schematic diagram of signal processing related to the short-time Fourier transform in FIG. 3 . Fig. 5 is a schematic diagram of the Doppler signal processing device in Fig. 1 and Fig. 2 in an embodiment. FIG. 6 is a schematic diagram of the interference baseline estimator described in FIG. 2 , FIG. 3 and FIG. 5 . FIG. 7 is a schematic diagram of the finite state machine of the control unit in FIG. 6 . FIG. 8 is a schematic diagram of the scope of the filtered signal generated by the filtering unit in FIG. 6 . FIG. 9 is a flowchart of a Doppler signal processing method for suppressing interference in an embodiment.

100: Doppler signal processing device

110:Frequency Analyzer

120: Interference suppressor

130: Interference baseline estimator

140: Synthesizer

U _n (k): frequency domain interference estimation signal

X _n (k): frequency domain signal

x(i): digital signal

y(i): Interference suppressed time-domain digital signal

Y _n (k): Interference suppression frequency domain signal

Claims (20)

A Doppler signal processing device for suppressing interference in a background according to a received wireless signal, the Doppler signal processing device comprising: a frequency analyzer for generating a frequency-domain signal based on a digital signal corresponding to the received wireless signal, and the digital signal includes interference energy generated by the interference of the background; An interference suppressor for performing a suppression operation according to the frequency domain signal and a frequency domain interference estimation signal to generate an interference suppressed frequency domain signal, wherein in the interference suppressed frequency domain signal, the interference energy is in the frequency domain system suppressed by the frequency-domain interference estimate signal; an interference baseline estimator for generating or updating the frequency-domain interference estimation signal according to the frequency-domain signal, wherein the frequency-domain interference estimation signal corresponds to an energy distribution of the interference in the frequency domain; and A synthesizer is used for generating an interference-suppressed time-domain digital signal according to the interference-suppressed frequency-domain signal, wherein the interference-suppressed time-domain digital signal is related to the digital signal whose interference energy has been suppressed. The Doppler signal processing device as claimed in claim 1, wherein the frequency-domain signal includes a frequency-domain signal vector, and the frequency-domain interference estimation signal includes a frequency-domain interference estimation signal vector. The Doppler signal processing device as claimed in claim 1, wherein the frequency analyzer is used to perform a short-time Fourier transform to generate the frequency domain signal using the digital signal. The Doppler signal processing device as claimed in claim 3, wherein the frequency analyzer is further configured to perform a buffering operation to collect the digital signals at different time points as an input of the short-time-distance Fourier transform. The Doppler signal processing device as claimed in claim 3, wherein the synthesizer is used to perform an inverse fast Fourier transform using the interference-suppressed frequency-domain signal to generate a time-domain signal corresponding to the interference-suppressed time-domain digital signal. The Doppler signal processing device as claimed in claim 5, wherein the synthesizer is further used to perform an overlap-and-accumulate operation using the time-domain signal to generate the interference-suppressed time-domain digital signal. The Doppler signal processing device as claimed in claim 1, wherein the frequency analyzer includes a filter bank for filtering the digital signal to generate the frequency domain signal. The Doppler signal processing device as claimed in claim 7, wherein the synthesizer includes a filter bank for filtering the interference-suppressed frequency-domain signal to generate the interference-suppressed time-domain digital signal. The Doppler signal processing device as described in Claim 1, wherein the interference baseline estimator includes: an absolute value data unit for generating an absolute value data of the frequency domain signal; a filter unit, used to generate a filter signal according to the absolute value data; a subtractor for generating difference data between the filtered signal and the frequency-domain interference estimation signal; a control unit for processing the difference data to generate an adjustment data; an adder for adding the adjustment data and the frequency domain interference estimation signal to generate an updated frequency domain interference estimation signal; and A delay unit is used for replacing the frequency domain interference estimation signal with the updated frequency domain interference estimation signal. The Doppler signal processing device as described in claim 9, wherein the absolute value data includes an absolute value, the difference data includes a difference value, the adjustment data includes an adjustment value, and the filtered signal includes a filter data vector . The Doppler signal processing device as described in Claim 9, wherein the filtering unit includes: A filter is used for filtering the absolute value data to generate the filtered signal. The Doppler signal processing device as described in Claim 9, wherein the filtering unit includes: a filter for filtering the absolute value data to generate a preliminary filtered signal; and An envelope detector is used to obtain envelope information of the preliminary filtered signal to generate the filtered signal. The Doppler signal processing device as claimed in claim 12, wherein the preliminary filtered signal comprises a preliminary filtered signal vector. The Doppler signal processing device as described in claim 12, wherein the control unit is additionally used for: Obtain an initial data according to the difference data and the filtered signal of a period, enter a slow tracking state to use a first speed to update the frequency domain interference estimation signal, and enter a fast tracking state to use a speed higher than the first A second one of the speeds updates the frequency-domain interference estimation signal and enters a freeze state to stop updating the frequency-domain interference estimation signal. The Doppler signal processing device as claimed in claim 14, wherein the initial data includes an initial value. The Doppler signal processing device as described in claim 14, wherein: when the strength of the filtered signal is between a first threshold and a second threshold higher than the first threshold, the control unit enters the slow tracking state; when the strength of the filtered signal is lower than the first threshold, the control unit enters the fast tracking state; and When the intensity of the filtered signal is higher than the second threshold, the control unit enters the freezing state. A Doppler signal processing method for suppressing a background interference according to a received wireless signal, the Doppler signal processing method comprising: generating a frequency domain signal based on a digital signal, wherein the digital signal corresponds to the received wireless signal, and the digital signal includes interference energy generated by the background interference; performing a suppression operation based on the frequency-domain signal and a frequency-domain interference estimation signal to generate an interference-suppressed frequency-domain signal, wherein the interference energy is suppressed in the frequency domain in the interference-suppressed frequency-domain signal; generating or updating the frequency domain interference estimation signal according to the frequency domain signal, wherein the frequency domain interference estimation signal corresponds to an energy distribution of the interference in the frequency domain; and An interference-suppressed time-domain digital signal is generated according to the interference-suppressed frequency-domain signal, wherein the interference-suppressed time-domain digital signal is related to the digital signal whose interference energy has been suppressed. The Doppler signal processing method as described in Claim 17, wherein: generating the frequency domain signal from the digital signal by performing a short-time Fourier transform on the digital signal to generate the frequency domain signal; and The method further includes performing an inverse fast Fourier transform using the interference-suppressed frequency-domain signal to generate a time-domain signal related to the interference-suppressed time-domain digital signal. The Doppler signal processing method as described in Claim 18, further comprising: An overlap-and-accumulate operation is performed using the time-domain signal to generate the interference-suppressed time-domain digital signal. The Doppler signal processing method as described in Claim 17, wherein: generating the frequency domain signal from the digital signal filters the digital signal to generate the frequency domain signal; and Generating the interference-suppressed time-domain digital signal based on the interference-suppressed frequency-domain signal is filtering the interference-suppressed frequency-domain signal to generate the interference-suppressed time-domain digital signal.

Images