Range Doppler radar angle sensing method and device

Abstract

The present invention provides a distance Doppler radar angle sensing method and device, the method comprising the following steps: receiving a first sensing signal and a second sensing signal, performing a one-dimensional 1D fast Fourier transform FFT and then performing a two-dimensional 2D FFT on each of the multiple signals to calculate at least one first 2D fast Fourier transform image and at least one second 2D fast Fourier transform image; according to a given Doppler index, selecting a first feature column from the at least one first 2D fast Fourier transform image and selecting a second feature column from the at least one second 2D fast Fourier transform image to perform a three-dimensional 3D FFT calculation to obtain a distance Doppler radar angle. The present invention can reduce the computational burden generated by a gesture recognition function.

Description

Technical Field

[0001] A radar angle sensing method and device, in particular, a range Doppler angle sensing method and device. Background Art

[0002] A Doppler radar gesture recognition system is a conventional gesture recognition system. The Doppler radar gesture recognition system can sense the user's hand movements, shapes, and postures and generate a range Doppler radar image (RDI). The Doppler radar gesture recognition system usually uses a one-dimensional (1D) Fast Fourier Transform (FFT) to sense a distance of an object, a two-dimensional (2D) Fast Fourier Transform to sense a speed of the object, and a three-dimensional (3D) Fast Fourier Transform to sense a radar angle of the object.

[0003] The Doppler radar gesture recognition system can be used in combination with a smart device such as a smart phone or a smart panel, and the Doppler radar gesture recognition system can use a processor of the smart device to recognize multiple gesture sensing signals and to distinguish a user action. The smart device can perform multiple functions according to the distinguished user action, for example, the smart device can sense a user's action and unlock the smart device when the action is consistent with a known user action.

[0004] However, 3D fast Fourier transform is quite complex in calculation, and the above-mentioned gesture recognition function is only one of the functions of the smart device. When the gesture recognition function is calculated, it will form a huge system burden on the smart device, and the general functions of the smart device will be affected in operation.

[0005] Therefore, the Doppler radar gesture recognition system needs to improve its working efficiency so that more space can be used. Summary of the invention

[0006] In view of the above problems, the present invention provides a range Doppler angle sensing method and a range Doppler angle sensing device. The present invention can reduce the computational burden of a gesture recognition function on a system.

[0007] The range Doppler radar angle sensing method comprises the following steps:

[0008] receiving a first sensing signal and a second sensing signal;

[0009] Performing a one-dimensional (1D) Fast Fourier Transform (FFT) on the first sensing signal to calculate a plurality of first 1D Fast Fourier Transform maps (FFT maps);

[0010] Performing a two-dimensional (2D) FFT on the plurality of first 1D fast Fourier transform maps to calculate at least one first 2D fast Fourier transform map;

[0011] performing a 1D FFT on the second sensing signal to calculate a plurality of second 1D fast Fourier transform maps;

[0012] Performing a 2D FFT on the plurality of second 1D fast Fourier transform maps to calculate at least one second 2D fast Fourier transform map;

[0013] Selecting a first feature column from the at least one first 2D fast Fourier transform image and a second feature column from the at least one second 2D fast Fourier transform image according to a given Doppler index;

[0014] A three-dimensional (3D) FFT is performed on the first feature column of the at least one first 2D FFT map and the second feature column of the at least one second 2D FFT map to obtain a range Doppler radar angle.

[0015] In addition, the range Doppler radar angle sensing device comprises a transmitting unit, a first sensing unit, a second sensing unit and a processing unit. The transmitting unit sends a check signal. The first sensing unit senses a first sensing signal, and the second sensing unit senses a second sensing signal. The processing unit is electrically connected to the transmitting unit, the first sensing unit and the second sensing unit.

[0016] The processing unit receives the first sensing signal from the first sensing unit and the second sensing signal from the second sensing unit, performs a one-dimensional (1D) fast Fourier transform (FFT) on the first sensing signal to calculate a plurality of first 1D fast Fourier transform maps (FFT maps), and performs a two-dimensional (2D) FFT on the plurality of first 1D fast Fourier transform maps to calculate at least one first 2D fast Fourier transform map.

[0017] Next, the processing unit performs a 1D FFT on the second sensing signal to calculate a plurality of second 1D fast Fourier transform maps, and performs a 2D FFT on the plurality of second 1D fast Fourier transform maps to calculate at least one second 2D fast Fourier transform map.

[0018] The processing unit further selects a first feature column from the at least one first 2D fast Fourier transform image and a second feature column from the at least one second 2D fast Fourier transform image according to a given Doppler index, and performs a three-dimensional (3D) FFT on the first feature column of the at least one first 2D fast Fourier transform image and the second feature column of the at least one second 2D fast Fourier transform image to obtain a range Doppler radar angle.

[0019] The present invention only performs a 3D FFT on the selected first feature column and the selected second feature column, and does not perform a 3D FFT on all the values ​​in the first 2D fast Fourier transform image and the second 2D fast Fourier transform image. Therefore, the present invention can reduce the computational burden of the gesture recognition function on a system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 FIG. 4 is a flow chart of a range Doppler angle sensing method of the present invention.

[0021] Figure 2 FIG. 4 is a system block diagram of a range Doppler angle sensing device according to the present invention.

[0022] Figure 3 FIG. 4 is another flow chart of the range Doppler radar angle sensing method of the present invention.

[0023] Figure 4 The present invention is a schematic diagram of performing a three-dimensional (3D) Fast Fourier Transform (FFT) on the first feature column of the at least one first 2D Fast Fourier Transform image and the second feature column of the at least one second 2D Fast Fourier Transform image. DETAILED DESCRIPTION

[0024] See also Figure 1 and Figure 2 As shown, the present invention is a range Doppler angle sensing method and a range Doppler angle sensing device.

[0025] The range Doppler radar angle sensing method comprises the following steps:

[0026] Step S101: receiving a first sensing signal and a second sensing signal;

[0027] Step S102: performing a one-dimensional (1D) Fast Fourier Transform (FFT) on the first sensing signal to calculate a plurality of first 1D Fast Fourier Transform maps (FFT maps);

[0028] Step S103: performing a two-dimensional (2D) FFT on the plurality of first 1D fast Fourier transform images to calculate at least one first 2D fast Fourier transform image;

[0029] Step S104: performing a 1D FFT on the second sensing signal to calculate a plurality of second 1D fast Fourier transform images;

[0030] Step S105: performing a 2D FFT on the plurality of second 1D fast Fourier transform images to calculate at least one second 2D fast Fourier transform image;

[0031] Step S106: selecting a first feature column from the at least one first 2D fast Fourier transform image and selecting a second feature column from the at least one second 2D fast Fourier transform image according to a given Doppler index;

[0032] Step S107: performing a three-dimensional (3D) FFT on the first feature column of the at least one first 2D FFT image and the second feature column of the at least one second 2D FFT image to obtain a range Doppler radar angle.

[0033] In addition, the range Doppler radar angle sensing device comprises a transmitting unit 10, a first sensing unit 21, a second sensing unit 22 and a processing unit 30, wherein the processing unit 30 is electrically connected to the transmitting unit 10, the first sensing unit 21 and the second sensing unit 22. The transmitting unit 10 sends a check signal. The first sensing unit 21 senses a first sensing signal, and the second sensing unit 22 senses a second sensing signal.

[0034] The processing unit 30 receives the first sensing signal from the first sensing unit 21 and the second sensing signal from the second sensing unit 22, and then performs a one-dimensional (1D) fast Fourier transform (FFT) on the first sensing signal to calculate a plurality of first 1D fast Fourier transform maps (FFT maps), and performs a two-dimensional (2D) FFT on the plurality of first 1D fast Fourier transform maps to calculate at least one first 2D fast Fourier transform map.

[0035] The processing unit 30 also performs a 1D FFT on the second sensing signal to calculate a plurality of second 1D fast Fourier transform maps, and performs a 2D FFT on the plurality of second 1D fast Fourier transform maps to calculate at least one second 2D fast Fourier transform map.

[0036] The processing unit 30 further selects a first feature column from the at least one first 2D fast Fourier transform image and a second feature column from the at least one second 2D fast Fourier transform image according to a given Doppler index, and performs a three-dimensional (3D) FFT on the first feature column of the at least one first 2D fast Fourier transform image and the second feature column of the at least one second 2D fast Fourier transform image to obtain a range Doppler radar angle.

[0037] Since the present invention only performs a 3D FFT on the selected first feature column and the selected second feature column, a 3D FFT will not be performed on all the values ​​in the first 2D fast Fourier transform image and the second 2D fast Fourier transform image. Therefore, the present invention can reduce the computational burden of the gesture recognition function on a system.

[0038] See also Figure 3 As shown, in detail, the range Doppler radar angle sensing method further includes the following steps in step S107:

[0039] Step S301: Make the first characteristic column of the at least one first 2D fast Fourier transform image a first column in a transformation image;

[0040] Step S302: Make the second characteristic column of the at least one second 2D fast Fourier transform image become a second column in the transformation image;

[0041] Step S303: Fill 0 into the remaining columns in the conversion map;

[0042] Step S304: performing a 3D FFT on the conversion graph to calculate a radar angle fast Fourier transform graph;

[0043] Step S305: Using a coordinate rotation digital computer (CORDIC) to perform fast Fourier transformation of the radar angle to obtain the range Doppler radar angle.

[0044] See also Figure 4As shown, for example, assume that the at least one first 2D fast Fourier transform graph and the at least one second 2D fast Fourier transform graph are each a 32×32 graph, that is, the at least one first 2D fast Fourier transform graph and the at least one second 2D fast Fourier transform graph each contain 32 columns and 32 rows, and each contain 32×32 values.

[0045] When the Doppler index is 0, the first column of the at least one first 2D fast Fourier transform image and the first column of the at least one second 2D fast Fourier transform image will be selected, that is, the first column of the at least one first 2D fast Fourier transform image is the first characteristic column, and the first column of the at least one second 2D fast Fourier transform image is the second characteristic column. Then, the selected columns will be respectively set to the conversion map, such as the first characteristic column is set to the conversion map to become the first column in the conversion map, and the second characteristic column is set to the conversion map to become the second column in the conversion map. Because the conversion map is also a 32×32 map, the remaining columns in the conversion map, that is, the columns from the 3rd column to the 32nd column in the conversion map, will be filled with the number 0. In other words, a total of 32×30 digital 0s will correspond to 32 rows and 30 columns of digital 0s to fill in the columns from the 3rd column to the 32nd column in the conversion map.

[0046] Each row in the transformation map is further processed by 3D FFT to calculate the radar angle fast Fourier transform map, and a coordinate rotation digital computer (CORDIC) is used on the radar angle fast Fourier transform map to obtain the range Doppler radar angle.

[0047] In one embodiment of the present invention, the processing unit 30 of the range Doppler radar angle sensing device executes the range Doppler radar angle sensing method. In this embodiment, the given Doppler index is 0, so that the first feature column of the at least one first 2D fast Fourier transform graph becomes a first column in a transformation graph, and the second feature column of the at least one second 2D fast Fourier transform graph becomes a second column in the transformation graph. In addition, in this embodiment, the first sensing signal and the second sensing signal come from different sensing chains, and the first sensing signal and the second sensing signal are a frequency modulated continuous wave (FMCW) signal.

[0048] In this embodiment, the transmitting unit 10 is a FMCW transmitter, and the first sensing unit 21 and the second sensing unit 22 are each a FMCW receiver. In other words, in this embodiment, the first sensing unit 21 is a first sensing chain, and the second sensing unit 22 is a second sensing chain different from the first sensing chain. In this embodiment, the processing unit 30 is a central processing unit (CPU) of a smart phone.

[0049] Because the processing unit 30 performs a 3D FFT on only one transformed map, and the transformed map only contains the first feature column of the at least one first 2D fast Fourier transform map, the second feature column of the at least one second 2D fast Fourier transform map, and the remaining columns filled with the number 0, the transformation map is calculated. Figure 3 The FFT of D will be very simple. Therefore, the present invention can reduce the computational burden of the gesture recognition function on the system of a smart device, so that the general functions of the smart device will not be affected in operation.

[0050] Although this specification mentions many of the foreseen features and advantages of the present invention, as well as the detailed structures and functions of the present invention, the information disclosed in this specification is only an example. In other words, changes in details such as changing the shape, changing the size, or restructuring the structure are allowed within the scope of the present invention, and the scope of the present invention is the scope of patent protection contained in the text of the claims of the present invention.

Claims

1. A range Doppler radar angle sensing method, characterized in that: The following steps are involved: receiving a first sensing signal and a second sensing signal; Performing a one-dimensional 1D fast Fourier transform FFT on the first sensing signal to calculate a plurality of first 1D fast Fourier transform graphs; Performing a two-dimensional 2D FFT on the plurality of first 1D fast Fourier transform maps to calculate at least one first 2D fast Fourier transform map; performing a 1D FFT on the second sensing signal to calculate a plurality of second 1D fast Fourier transform maps; Performing a 2D FFT on the plurality of second 1D fast Fourier transform maps to calculate at least one second 2D fast Fourier transform map; Selecting a first feature column from the at least one first 2D fast Fourier transform image and a second feature column from the at least one second 2D fast Fourier transform image according to a given Doppler index; Performing a three-dimensional 3D FFT on the first feature column of the at least one first 2D fast Fourier transform image and the second feature column of the at least one second 2D fast Fourier transform image to obtain a range Doppler radar angle; The step of performing a three-dimensional 3D FFT on the first feature column of the at least one first 2D fast Fourier transform image and the second feature column of the at least one second 2D fast Fourier transform image to obtain a range Doppler radar angle further includes the following steps: Making the first characteristic column of the at least one first 2D fast Fourier transform map a first column in a transform map; Making the second characteristic column of the at least one second 2D fast Fourier transform map a second column in the transform map; Fill the remaining columns in the conversion chart with 0; Performing a 3D FFT on the transformation map to calculate a radar angle fast Fourier transform map; The radar angle Fast Fourier Transform is processed using a coordinate rotation digital computer to obtain the range Doppler radar angle.

2. The range Doppler radar angle sensing method according to claim 1, characterized in that: The given Doppler index is 0.

3. The range Doppler radar angle sensing method according to claim 1, characterized in that: The first feature column is a first column of the at least one first 2D fast Fourier transform graph; and The second characteristic column is a first column of the at least one second 2D fast Fourier transform map.

4. The range Doppler radar angle sensing method according to claim 1, characterized in that: The first sensing signal and the second sensing signal come from different sensing chains.

5. The range Doppler radar angle sensing method according to claim 1, characterized in that: The first sensing signal and the second sensing signal are frequency modulated continuous wave signals.

6. A range Doppler radar angle sensing device, characterized in that: include: a transmitting unit, sending out a check signal; a first sensing unit, sensing a first sensing signal; a second sensing unit, sensing a second sensing signal; a processing unit electrically connected to the transmitting unit, the first sensing unit and the second sensing unit; The processing unit receives the first sensing signal from the first sensing unit and the second sensing signal from the second sensing unit, and then performs a one-dimensional 1D fast Fourier transform FFT on the first sensing signal to calculate a plurality of first 1D fast Fourier transform graphs, and performs a two-dimensional 2D FFT on the plurality of first 1D fast Fourier transform graphs to calculate at least one first 2D fast Fourier transform graph; The processing unit also performs a 1D FFT on the second sensing signal to calculate a plurality of second 1D fast Fourier transform maps, and performs a 2D FFT on the plurality of second 1D fast Fourier transform maps to calculate at least one second 2D fast Fourier transform map; The processing unit further selects a first feature column from the at least one first 2D fast Fourier transform image and a second feature column from the at least one second 2D fast Fourier transform image according to a given Doppler index, and performs a three-dimensional 3D FFT on the first feature column of the at least one first 2D fast Fourier transform image and the second feature column of the at least one second 2D fast Fourier transform image to obtain a range Doppler radar angle; When the processing unit performs a three-dimensional 3D FFT on the first feature column and the second feature column to obtain the range Doppler radar angle, the processing unit makes the first feature column a first column in a conversion diagram and makes the second feature column a second column in the conversion diagram, then fills 0 into the remaining columns in the conversion diagram, performs a 3D FFT on the conversion diagram to calculate a radar angle fast Fourier transform diagram, and uses a coordinate rotation digital computer on the radar angle fast Fourier transform diagram to obtain the range Doppler radar angle.

7. The range Doppler radar angle sensing device according to claim 6, characterized in that: The given Doppler index is 0.

8. The range Doppler radar angle sensing device as claimed in claim 6, characterized in that: The first feature column is a first column of the at least one first 2D fast Fourier transform graph; and The second characteristic column is a first column of the at least one second 2D fast Fourier transform map.

9. The range Doppler radar angle sensing device according to claim 6, characterized in that: The first sensing unit is a first sensing chain, and the second sensing unit is a second sensing chain different from the first sensing chain.

10. The range Doppler radar angle sensing device according to claim 6, characterized in that: The transmitting unit is a frequency modulated continuous wave (FMCW) transmitter; and The first sensing unit and the second sensing unit are each a FMCW receiver.

11. The range Doppler radar angle sensing device according to claim 6, characterized in that: The processing unit is a central processing unit of a smart phone.

Images