FMCW radar signal processing device and FMCW radar signal processing program

By defining dimensions for Fourier transforms and targeting detected data only, the method reduces signal processing and memory needs, addressing the challenges of conventional FMCW radar devices.

JP7620488B2Active Publication Date: 2025-01-23JAPAN RADIO CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2021073651
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-04-23
Publication Date
2025-01-23
Estimated Expiration
2041-04-23

AI Technical Summary

Technical Problem

Conventional FMCW radar signal processing devices face challenges in reducing the amount of signal processing and data memory requirements when implemented in actual devices.

Method used

The proposed method involves defining specific dimensions for Fourier transforms and target detection, performing transforms in selected directions, and extracting data only for detected targets, thereby reducing the number of transforms and memory requirements.

Benefits of technology

This approach significantly reduces the number of Fourier transforms, target object detections, and data memory requirements, making it feasible to implement the FMCW radar signal processing device in practical applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007620488000001
    Figure 0007620488000001
  • Figure 0007620488000002
    Figure 0007620488000002
  • Figure 0007620488000003
    Figure 0007620488000003
Patent Text Reader

Abstract

To reduce a signal processing amount and a data memory amount when implementing an FMCW radar signal processing device in an actual machine.SOLUTION: In a first stage, the present disclosure executes a Fourier transform in a first dimensional and a second dimensional direction and then detect a target in a reverse spatial direction to the first dimensional direction and a reverse spatial direction to the second dimensional direction. Meanwhile, in a second stage, the present disclosure only extracts data in which a target is detected, while not extracting data in which a target is not detected, when executing the Fourier transform and then detects a target in a reverse spatial direction to a third dimensional direction.SELECTED DRAWING: Figure 6
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a frequency modulation-continuous wave (FMCW) radar signal processing technique that uses an FMCW radar. [Background technology]

[0002] An FMCW radar signal processing technique using an FMCW radar is disclosed in Patent Document 1 and Non-Patent Document 1, and is applied to high frequency bands such as millimeter wave bands and antenna arrangements such as MIMO (Multiple-Input and Multiple-Output). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-315447 A [Non-patent literature]

[0004] [Non-Patent Document 1] Millimeter Wave Radar Basics 1 [Distance, Speed, and Angle Detection], [online], Marubun Corporation, [Retrieved April 7, 2021],<URL:https: / / www.marubun.co.jp / service / technicalsquare / a7ijkd000000dtyd.html> Summary of the Invention [Problem to be solved by the invention]

[0005] The configuration of a conventional FMCW radar signal processing device is shown in Fig. 1. The procedure of a conventional FMCW radar signal processing program is shown in Fig. 2. The FMCW radar signal processing device R1 includes a transmitting antenna 1, receiving antennas 2-1 to 2-a, mixers 3-1 to 3-a, an oscillator 4, A / D (Analog / Digital) conversion units 5-1 to 5-a, distance analysis units 6-1 to 6-a, a data storage unit 7, speed analysis units 8-1 to 8-a, a data storage unit 9, an angle analysis unit 10, a data storage unit 11, and a target detection unit 12.

[0006] The mixers 3-1 to 3-a, the oscillator 4, and the A / D converters 5-1 to 5-a calculate beat signals between the transmission signals and the reception signals sampled for a number of samples Ns, for a number of chirp numbers Nc and a number of antennas Na (top left column of FIG. 2).

[0007] When the distance analysis units 6-1 to 6-a perform a Fourier transform on the beat signals calculated by the mixers 3-1 to 3-a in the time direction, they extract beat signals including all coordinate values ​​in the chirp number direction and the antenna arrangement direction (upper left column and upper right column in FIG. 2). The number of Fourier transforms in the distance analysis units 6-1 to 6-a is Nc×Na.

[0008] The data storage unit 7 stores the Fourier transform results of the distance analysis units 6-1 to 6-a (top right column in FIG. 2). The memory capacity required by the data storage unit 7 is Br×Nc×Na×D (Br is the total number of bins in the distance bin direction, and D is the amount of data according to the data type).

[0009] When the velocity analysis units 8-1 to 8-a perform a Fourier transform on the phase values ​​of the Fourier transform results of the distance analysis units 6-1 to 6-a in the chirp number direction, they extract phase values ​​including all bin values ​​in the distance bin direction and all coordinate values ​​in the antenna array direction (upper right column and lower left column in FIG. 2). The number of Fourier transforms in the velocity analysis units 8-1 to 8-a is Br×Na.

[0010] The data storage unit 9 stores the Fourier transform results from the velocity analysis units 8-1 to 8-a (lower left column in FIG. 2). The memory capacity required by the data storage unit 9 is Br×Bv×Na×D (Bv is the total number of bins in the velocity bin direction, and D is the amount of data according to the data type).

[0011] When the angle analysis unit 10 performs a Fourier transform on the phase values ​​of the Fourier transform results of the velocity analysis units 8-1 to 8-a in the antenna array direction, it extracts phase values ​​including all bin values ​​in the distance bin direction and the velocity bin direction (lower left column and lower right column in FIG. 2). The number of Fourier transforms in the angle analysis unit 10 is Br×Bv.

[0012] The data storage unit 11 stores the Fourier transform results from the angle analysis unit 10 (lower right column in FIG. 2). The memory capacity required by the data storage unit 11 is Br×Bv×Ba×D (Ba is the total number of bins in the angle bin direction, and D is the amount of data according to the data type).

[0013] After performing threshold detection, the target detection unit 12 detects the target distance, target speed, and target angle based on the target bin values ​​in the distance bin direction, speed bin direction, and angle bin direction (lower right column in FIG. 2). The number of target detections by the target detection unit 12 is Br×Bv×Ba.

[0014] The total number of Fourier transforms performed by the FMCW radar signal processing device R1 is Nc×Na+Br×Na+Br×Bv. The total amount of Fourier transform calculations performed by the FMCW radar signal processing device R1 is Nc×Na×Br×log2Br+Br×Na×Bv×log2Bv+Br×Bv×Ba×log2Ba. The total number of target detections by the FMCW radar signal processing device R1 is Br×Bv×Ba. The minimum total memory capacity required by the FMCW radar signal processing device R1 is Br×Bv×Ba×D (memory is overwritten).

[0015] For example, assume that Ns=128, Nc=128, Na=16, Br=128, Bv=256, and Ba=32. The total number of Fourier transforms performed by the FMCW radar signal processing device R1 is 36,864. The total number of Fourier transform calculations performed by the FMCW radar signal processing device R1 is 11,272,192. The total number of target detections performed by the FMCW radar signal processing device R1 is 1,048,576. The total minimum required memory amount for the FMCW radar signal processing device R1 is 1,048,576×D. When performing zero interpolation in the Fourier transform or when virtually expanding the array antenna, the data amount D according to the data type increases, and therefore the total minimum required memory amount for the FMCW radar signal processing device R1 also increases.

[0016] As described above, in the conventional technology, it was difficult to reduce the amount of signal processing and the amount of data memory required when implementing the FMCW radar signal processing device R1 in an actual aircraft.

[0017] Therefore, in order to solve the above problem, an object of the present disclosure is to reduce the amount of signal processing and the amount of data memory required when implementing an FMCW radar signal processing device in an actual device. [Means for solving the problem]

[0018] In order to solve the above problem, any one of the time passage direction, the chirp number direction, and the antenna array direction is defined as the first dimension direction, the second dimension direction, and the third dimension direction, and any one of the distance bin direction, the velocity bin direction, and the angle bin direction is defined as the reciprocal space direction of the first dimension direction, the reciprocal space direction of the second dimension direction, and the reciprocal space direction of the third dimension direction.

[0019] In order to solve the above problem, in the first step, a Fourier transform is performed in the first and second dimensional directions, and a target is detected in the reciprocal space direction of the first and second dimensional directions. Then, in the second step, when a Fourier transform is performed in the third dimensional direction, data in which a target is not detected is not extracted, and only data in which a target is detected is extracted, and a target is detected in the reciprocal space direction of the third dimensional direction.

[0020] Specifically, the present disclosure relates to an FMCW radar signal processing device that uses an FMCW radar, the device including: a beat signal calculation unit that calculates beat signals between a transmission signal and a reception signal sampled by a plurality of sample numbers, the number of beat signals corresponding to a plurality of chirp numbers and the number of antennas; a first Fourier transform unit that extracts beat signals including all coordinate values ​​in a second dimensional direction and a third dimensional direction among the time elapse direction, the chirp number direction and the antenna array direction when Fourier transforming the beat signal calculated by the beat signal calculation unit in a first dimensional direction among the time elapse direction, the chirp number direction and the antenna array direction; a second Fourier transform unit that extracts phase values ​​including all coordinate values ​​in a reciprocal space direction of the first dimensional direction and a specific coordinate value in the third dimensional direction when Fourier transforming a phase value of a Fourier transform result in the first Fourier transform unit in the second dimensional direction; a first target detection unit that detects any two of a target distance, a target velocity, and a target angle based on a target coordinate value in a reciprocal space direction of the first dimensional direction and a reciprocal space direction of the second dimensional direction; a third Fourier transform unit that extracts phase values ​​including target coordinate values ​​in a reciprocal space direction of the first dimensional direction and all coordinate values ​​in the third dimensional direction when Fourier transforming a phase value of a Fourier transform result in the first Fourier transform unit in the second dimensional direction; a fourth Fourier transform unit that extracts phase values ​​including target coordinate values ​​in a reciprocal space direction of the second dimensional direction when Fourier transforming a phase value of a Fourier transform result in the third Fourier transform unit in the third dimensional direction; and a second target detection unit that detects a target distance, a target velocity, and a target angle based on target coordinate values ​​in the first dimensional direction, the second dimensional direction, and the reciprocal space direction of the third dimensional direction.

[0021] The present disclosure also provides an FMCW radar signal processing program for using an FMCW radar, the program including: a beat signal calculation step of calculating beat signals between a transmission signal and a reception signal sampled by a plurality of sample numbers, the number of beat signals corresponding to a plurality of chirp numbers and the number of antennas; a first Fourier transform step of extracting beat signals including all coordinate values ​​in a second dimensional direction and a third dimensional direction among the time elapse direction, the chirp number direction and the antenna array direction when Fourier transforming the beat signal calculated in the beat signal calculation step in a first dimensional direction among the time elapse direction, the chirp number direction and the antenna array direction; a second Fourier transform step of extracting phase values ​​including all coordinate values ​​in a reciprocal space direction of the first dimensional direction and a specific coordinate value in the third dimensional direction when Fourier transforming a phase value of a Fourier transform result in the first Fourier transform step in the second dimensional direction; a first target detection step of detecting any two of a target distance, a target speed, and a target angle based on a target coordinate value in a reciprocal space direction of a second dimensional direction; a third Fourier transform step of extracting phase values ​​including target coordinate values ​​in a reciprocal space direction of the first dimensional direction and all coordinate values ​​in the third dimensional direction when Fourier transforming a phase value of a Fourier transform result in the first Fourier transform step in the second dimensional direction; a fourth Fourier transform step of extracting phase values ​​including target coordinate values ​​in a reciprocal space direction of the second dimensional direction when Fourier transforming a phase value of a Fourier transform result in the third Fourier transform step in the third dimensional direction; and a second target detection step of detecting a target distance, a target speed, and a target angle based on target coordinate values ​​in the reciprocal space directions of the first dimensional direction, the second dimensional direction, and the third dimensional direction.

[0022] According to these configurations, when an FMCW radar signal processing device is implemented in an actual aircraft, it is possible to reduce the number of Fourier transforms, the number of target detections, and the amount of data memory required.

[0023] The present disclosure also relates to an FMCW radar signal processing device, characterized in that the number of Fourier transforms performed in the entire FMCW radar signal processing device is N2×N3+B1+T1×N3+T2 (N2 and N3 are the total number of coordinate values ​​in the second dimensional direction and the third dimensional direction, respectively, B1 is the total number of coordinate values ​​in the reciprocal space direction of the first dimensional direction, and T1 and T2 are the number of target coordinate values ​​in the reciprocal space direction of the first dimensional direction and the second dimensional direction, respectively).

[0024] According to this configuration, when the FMCW radar signal processing device is implemented in an actual device, the number of Fourier transforms can be reduced to N2×N3+B1+T1×N3+T2.

[0025] The present disclosure also provides an FMCW radar signal processing device, characterized in that the total number of target detections in the FMCW radar signal processing device is B1×B2+B3×T3 (B1, B2, and B3 are the total numbers of coordinate values ​​in the reciprocal spatial directions of the first dimension, the second dimension, and the third dimension, respectively, and T3 is the total number of detections in the second target detection unit).

[0026] According to this configuration, when the FMCW radar signal processing device is implemented in an actual aircraft, the number of target detections can be reduced to B1×B2+B3×T3.

[0027] The present disclosure also provides an FMCW radar signal processing device, characterized in that the overall minimum required memory amount of the FMCW radar signal processing device is B1×N2×N3×D (B1 is the total number of coordinate values ​​in the reciprocal space direction of the first dimension, N2 and N3 are the total numbers of coordinate values ​​in the second dimension and the third dimension, respectively, and D is the data amount according to the data type of each Fourier transform result in the first Fourier transform section).

[0028] According to this configuration, when the FMCW radar signal processing device is implemented in an actual aircraft, the minimum required data memory size can be reduced to B1×N2×N3×D. Effect of the Invention

[0029] In this way, the present disclosure makes it possible to reduce the amount of signal processing and the amount of data memory required when implementing an FMCW radar signal processing device in an actual device. [Brief description of the drawings]

[0030] [Figure 1] FIG. 1 is a diagram showing the configuration of a conventional FMCW radar signal processing device. [Diagram 2] FIG. 1 is a diagram showing the procedure of a conventional FMCW radar signal processing program. [Diagram 3] 1 is a diagram illustrating a configuration of an FMCW radar signal processing device according to a first embodiment. [Figure 4] FIG. 4 is a diagram showing a procedure of an FMCW radar signal processing program according to the first embodiment. [Diagram 5] FIG. 11 is a diagram illustrating a configuration of an FMCW radar signal processing device according to a second embodiment. [Figure 6] FIG. 11 is a diagram showing a procedure of an FMCW radar signal processing program according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0031] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present disclosure will be described with reference to the accompanying drawings. The embodiments described below are examples of the present disclosure, and the present disclosure is not limited to the following embodiments.

[0032] (FMCW radar signal processing device of the first embodiment) In the first stage, a Fourier transform is performed in the time direction and the chirp number direction, and then targets are detected in the distance bin direction and the velocity bin direction. In the second stage, when performing a Fourier transform in the antenna array direction, data in which targets are not detected is not extracted, and only data in which targets are detected is extracted, and then targets are detected in the angle bin direction. Each stage will be described in detail below.

[0033] The configuration of the FMCW radar signal processing device of the first embodiment is shown in Fig. 3. The procedure of the FMCW radar signal processing program of the first embodiment is shown in Fig. 4. The FMCW radar signal processing device R2 includes a transmitting antenna 21, receiving antennas 22-1 to 22-a, mixers 23-1 to 23-a, an oscillator 24, A / D conversion units 25-1 to 25-a, distance analysis units 26-1 to 26-a, a data storage unit 27, a speed analysis unit 28, a target detection unit 29, a list storage unit 30, a speed analysis unit 31, an angle analysis unit 32, a target detection unit 33, and a list storage unit 34.

[0034] The mixers 23-1 to 23-a, the oscillator 24, and the A / D converters 25-1 to 25-a calculate beat signals between the transmission signals and the reception signals sampled for a number of samples Ns, for a number of chirp numbers Nc and a number of antennas Na (top left column of FIG. 4).

[0035] When distance analysis units 26-1 to 26-a perform Fourier transform on beat signals calculated by mixers 23-1 to 23-a etc. in the time direction, they extract beat signals including all coordinate values ​​in the chirp number direction and the antenna arrangement direction (upper left column and upper center column in FIG. 4). The number of Fourier transforms in distance analysis units 26-1 to 26-a is Nc×Na.

[0036] The data storage unit 27 stores the Fourier transform results of the distance analysis units 26-1 to 26-a (see the upper center column of FIG. 4). The memory capacity required by the data storage unit 27 is Br×Nc×Na×D (Br is the total number of bins in the distance bin direction, and D is the amount of data according to the data type).

[0037] When the velocity analysis unit 28 performs a Fourier transform on the phase values ​​of the Fourier transform results of the distance analysis units 26-1 to 26-a in the chirp number direction, it extracts phase values ​​including all bin values ​​in the distance bin direction and a specific coordinate value (one coordinate value) in the antenna array direction (upper middle and upper right columns in FIG. 4). The number of Fourier transforms in the velocity analysis unit 28 is Br×1=Br.

[0038] After performing threshold detection, the target detection unit 29 detects the target distance and target speed based on the target bin values ​​in the distance bin direction and the speed bin direction (upper right column in FIG. 4). The number of times the target is detected by the target detection unit 29 is Br×Bv (Bv is the total number of bins in the speed bin direction).

[0039] The list storage unit 30 stores the target detection results and Fourier transform intensity from the target detection unit 29 (top right column in FIG. 4). The memory capacity required by the list storage unit 30 is Tr×D or Tv×D (Tr (1 in FIG. 4) is the total number of detections in the distance bin direction, Tv (1 in FIG. 4) is the total number of detections in the speed bin direction, and D is the amount of data according to the data type. Since multiple targets may exist at the same distance / speed bin value, Tr and Tv may be different).

[0040] When the velocity analysis unit 31 performs a Fourier transform on the phase values ​​of the Fourier transform results of the distance analysis units 26-1 to 26-a in the chirp number direction, it extracts the target bin value in the distance bin direction (one bin value in FIG. 4) and the phase values ​​including all coordinate values ​​in the antenna array direction (lower left column and lower center column in FIG. 4). The number of Fourier transforms in the velocity analysis unit 31 is Tr×Na.

[0041] Here, the velocity analysis unit 28 performs a Fourier transform on only the phase value including a specific coordinate value (one coordinate value) in the antenna arrangement direction. Therefore, the velocity analysis unit 31 performs a Fourier transform on the phase value including all coordinate values ​​in the antenna arrangement direction. Then, the processing of the angle analysis unit 32 becomes possible.

[0042] When the angle analysis unit 32 performs a Fourier transform on the phase values ​​of the Fourier transform results from the speed analysis unit 31 in the antenna array direction, it extracts phase values ​​including the target bin value in the speed bin direction (one bin value in FIG. 4) (lower middle and lower right columns in FIG. 4). The number of Fourier transforms performed by the angle analysis unit 32 is Tv (Tv (1 in FIG. 4) is the total number of detections in the speed bin direction).

[0043] After performing threshold detection, the target detection unit 33 detects the target distance, target speed, and target angle based on the target bin values ​​in the distance bin direction, speed bin direction, and angle bin direction (lower right column in FIG. 4). The number of target detections by the target detection unit 33 is Ba×Ta (Ba is the total number of bins in the angle bin direction, and Ta (1 in FIG. 4) is the total number of detections by the target detection unit 33).

[0044] The list storage unit 34 stores the target detection results and Fourier transform intensities from the target detection unit 33 (lower right column in FIG. 4). The memory capacity required by the list storage unit 34 is Ta×D (D is the amount of data according to the data type. Since multiple targets may exist at the same distance / velocity / angle bin value, Ta may be different from Tr and Tv).

[0045] The total number of Fourier transforms performed by the FMCW radar signal processing device R2 is Nc×Na+Br+Tr×Na+Tv. The total amount of Fourier transform calculations performed by the FMCW radar signal processing device R2 is Nc×Na×Br×log2Br+Br×Bv×log2Bv+Tr×Na×Bv×log2Bv+Tv×Ba×log2Ba. The total number of target detections performed by the FMCW radar signal processing device R2 is Br×Bv+Ba×Ta. The total minimum memory capacity required by the FMCW radar signal processing device R2 is Br×Nc×Na×D (after the speed analysis unit 28, necessary data is read from the data storage unit 27 and discarded without storing intermediate calculation results, so it is sufficient to secure the memory capacity of the data storage unit 27).

[0046] For example, assume that Ns = 128, Nc = 128, Na = 16, Br = 128, Bv = 256, and Ba = 32. However, in the worst case, the total number of detections in the target detection unit 33 is the maximum number of detections, 512, and the number of target bin values ​​in the range bin direction is the total number of bin values, 128.

[0047] The total number of Fourier transforms performed by the FMCW radar signal processing device R2 is 4,736. The total number of Fourier transform calculations performed by the FMCW radar signal processing device R2 is 6,373,376. The total number of target detections performed by the FMCW radar signal processing device R2 is 49,152. The total minimum required memory amount for the FMCW radar signal processing device R2 is 262,144×D. Even if the data amount D increases according to the data type when performing zero interpolation in the Fourier transform or when virtually expanding the array antenna, the total minimum required memory amount for the FMCW radar signal processing device R2 does not increase significantly.

[0048] In this way, in the first embodiment, when the FMCW radar signal processing device R2 is implemented in an actual device, it becomes easy to reduce the amount of signal processing and the amount of data memory required.

[0049] (FMCW radar signal processing device of the second embodiment) In the second embodiment, the first embodiment is expanded. Any of the time passage direction, the chirp number direction, and the antenna array direction is defined as the first dimension direction, the second dimension direction, and the third dimension direction. Any of the distance bin direction, the velocity bin direction, and the angle bin direction is defined as the reciprocal space direction of the first dimension direction, the reciprocal space direction of the second dimension direction, and the reciprocal space direction of the third dimension direction.

[0050] In the first stage, a Fourier transform is performed in the first and second dimensional directions, and then a target is detected in the reciprocal space direction of the first and second dimensional directions. In the second stage, when a Fourier transform is performed in the third dimensional direction, data in which a target is not detected is not extracted, and only data in which a target is detected is extracted, and then a target is detected in the reciprocal space direction of the third dimensional direction. Each stage will be specifically described below.

[0051] The configuration of the FMCW radar signal processing device of the second embodiment is shown in Fig. 5. The procedure of the FMCW radar signal processing program of the second embodiment is shown in Fig. 6. The FMCW radar signal processing device R3 includes a transmitting antenna 41, receiving antennas 42-1 to 42-a, mixers 43-1 to 43-a, an oscillator 44, A / D conversion units 45-1 to 45-a, a first analysis unit 46, a data storage unit 47, a second analysis unit 48, a target detection unit 49, a list storage unit 50, a second analysis unit 51, a third analysis unit 52, a target detection unit 53, and a list storage unit 54.

[0052] The mixers 43-1 to 43-a, the oscillator 44, and the A / D converters 45-1 to 45-a calculate beat signals between the transmission signals sampled by a plurality of samples and the reception signals by a plurality of chirp numbers and a plurality of antenna numbers (upper left column in FIG. 6). Any of the plurality of sample numbers, the plurality of chirp numbers, and the plurality of antenna numbers is defined as N1, N2, and N3.

[0053] When the first analyzer 46 performs a Fourier transform on the beat signals calculated by the mixers 43-1 to 43-a in the first dimensional direction, the first analyzer 46 extracts beat signals including all coordinate values ​​in the second and third dimensional directions (upper left and upper center columns in FIG. 6). The number of Fourier transforms performed by the first analyzer 46 is N2×N3.

[0054] The data storage unit 47 stores the Fourier transform results from the first analysis unit 46 (see the upper center column of FIG. 6). The memory capacity required by the data storage unit 47 is B1×N2×N3×D (B1 is the number of all coordinate values ​​in the reciprocal space direction of the first dimension, and D is the amount of data according to the data type).

[0055] When the second analysis unit 48 performs a Fourier transform on the phase values ​​of the Fourier transform result in the first analysis unit 46 in the second dimensional direction, it extracts phase values ​​including all coordinate values ​​in the reciprocal space direction of the first dimensional direction and a specific coordinate value (one coordinate value) in the third dimensional direction (upper middle and upper right columns in FIG. 6). The number of Fourier transforms in the second analysis unit 48 is B1×1=B1.

[0056] After performing threshold detection, the target detection unit 49 detects any two of the target distance, target speed, and target angle based on the target coordinate values ​​in the reciprocal space directions of the first and second dimensions (upper right column in FIG. 6). The number of times the target is detected by the target detection unit 49 is B1×B2 (B2 is the total number of coordinate values ​​in the reciprocal space direction of the second dimension).

[0057] The list storage unit 50 stores the target detection results and Fourier transform intensity from the target detection unit 49 (top right column in FIG. 6). The memory capacity required by the list storage unit 50 is T1×D or T2×D (T1 (1 in FIG. 6) is the total number of detections in the reciprocal space direction of the first dimension, T2 (1 in FIG. 6) is the total number of detections in the reciprocal space direction of the second dimension, and D is the amount of data according to the data type. Since multiple targets may exist at the same coordinate value in the reciprocal space direction of the first / second dimension, T1 and T2 may be different).

[0058] When the second analysis unit 51 performs a Fourier transform on the phase values ​​of the Fourier transform result in the first analysis unit 46 in the second dimensional direction, it extracts the target coordinate value in the reciprocal space direction of the first dimensional direction (one coordinate value in FIG. 6) and phase values ​​including all coordinate values ​​in the third dimensional direction (lower left column and lower center column in FIG. 6). The number of Fourier transforms in the second analysis unit 51 is T1×N3.

[0059] Here, the second analysis unit 48 performs a Fourier transform on only the phase values ​​including a specific coordinate value (one coordinate value) in the third dimensional direction. Therefore, the second analysis unit 51 performs a Fourier transform on the phase values ​​including all coordinate values ​​in the third dimensional direction. Then, the processing of the third analysis unit 52 becomes possible.

[0060] When the third analysis unit 52 performs a Fourier transform on the phase values ​​of the Fourier transform results in the second analysis unit 51 in the third dimensional direction, it extracts phase values ​​including the target coordinate values ​​in the reciprocal space direction in the second dimensional direction (one coordinate value in FIG. 6) (lower middle and lower right columns in FIG. 6). The number of Fourier transforms in the third analysis unit 52 is T2 (T2 (1 in FIG. 6) is the total number of detections in the reciprocal space direction in the second dimensional direction).

[0061] After performing threshold detection, the target detection unit 53 detects the target distance, target speed, and target angle based on the target coordinate values ​​in the first, second, and third dimensional reciprocal space directions (lower right column in FIG. 6). The number of target detections by the target detection unit 53 is B3×T3 (B3 is the total number of coordinate values ​​in the third dimensional reciprocal space direction, and T3 (1 in FIG. 6) is the total number of detections by the target detection unit 53).

[0062] The list storage unit 54 stores the target detection results and Fourier transform intensities from the target detection unit 53 (lower right column in FIG. 6). The memory capacity required by the list storage unit 54 is T3×D (D is the amount of data according to the data type. Since multiple targets may exist at the same coordinate value in the reciprocal space direction of the first / second / third dimensional direction, T3 may be different from T1 and T2).

[0063] The total number of Fourier transforms performed by the FMCW radar signal processing device R3 is N2×N3+B1+T1×N3+T2. The total number of Fourier transform calculations performed by the FMCW radar signal processing device R3 is N2×N3×B1×log2B1+B1×B2×log2B2+T1×N3×B2×log2B2+T2×B3×log2B3. The total number of target detections performed by the FMCW radar signal processing device R3 is B1×B2+B3×T3. The total minimum memory capacity required by the FMCW radar signal processing device R3 is B1×N2×N3×D (after the second analysis unit 48, necessary data is read from the data storage unit 47 and intermediate calculation results are discarded without being stored, so it is sufficient to secure the memory capacity of the data storage unit 47).

[0064] In this way, also in the second embodiment, it becomes easy to reduce the amount of signal processing and the amount of data memory when the FMCW radar signal processing device R3 is implemented in an actual device. [Industrial Applicability]

[0065] The FMCW radar signal processing device and the FMCW radar signal processing program disclosed herein can be applied to high frequency bands such as millimeter wave bands and antenna arrangements such as MIMO. [Explanation of symbols]

[0066] R1: FMCW radar signal processing device 1: Transmitting antenna 2-1, 2-a: Receiving antenna 3-1, 3-a: Mixer 4: Oscillator 5-1, 5-a: A / D conversion section 6-1, 6-a: Distance analysis section 7: Data storage section 8-1, 8-a: Speed ​​analysis section 9: Data storage section 10: Angle analysis section 11: Data storage unit 12: Target detection unit R2: FMCW radar signal processing device 21: Transmitting antenna 22-1, 22-a: Receiving antenna 23-1, 23-a: Mixer 24: Oscillator 25-1, 25-a: A / D conversion section 26-1, 26-a: Distance analysis section 27: Data storage unit 28: Speed ​​analysis section 29: Target detection unit 30: List storage section 31: Speed ​​analysis section 32: Angle analysis section 33: Target detection unit 34: List storage section R3: FMCW radar signal processing device 41: Transmitting antenna 42-1, 42-a: Receiving antenna 43-1, 43-a: Mixer 44: Oscillator 45-1, 45-a: A / D conversion section 46: 1st analysis section 47: Data storage unit 48:Second Analysis Department 49: Target detection unit 50: List storage section 51:Second analysis section 52:Third Analysis Department 53: Target detection unit 54: List storage section

Claims

1. A frequency modulation-continuous wave (FMCW) radar signal processing device using an FMCW radar, a beat signal calculation unit that calculates beat signals between a transmission signal and a reception signal sampled by a plurality of samples, the number of the beat signals corresponding to a plurality of chirp numbers and the number of the antennas; a first Fourier transform unit that extracts beat signals including all coordinate values ​​in a second and third dimensional directions among the time elapse direction, the chirp number direction, and the antenna array direction when performing a Fourier transform on the beat signal calculated by the beat signal calculation unit in a first dimensional direction among the time elapse direction, the chirp number direction, and the antenna array direction; a second Fourier transform unit that extracts phase values ​​including all coordinate values ​​in a reciprocal space direction of the first dimensional direction and specific coordinate values ​​in the third dimensional direction when performing a Fourier transform on phase values ​​of the Fourier transform result in the first Fourier transform unit in the second dimensional direction; a first target detection unit that detects any two of a target distance, a target velocity, and a target angle based on target coordinate values ​​in a reciprocal space direction of the first dimension direction and the second dimension direction; a third Fourier transform unit that extracts phase values ​​including a target coordinate value in a reciprocal space direction of the first dimensional direction and all coordinate values ​​in the third dimensional direction when performing a Fourier transform on a phase value of a Fourier transform result in the first Fourier transform unit in the second dimensional direction; a fourth Fourier transform unit that extracts a phase value including a target coordinate value in a reciprocal space direction of the second dimensional direction when performing a Fourier transform on a phase value of a Fourier transform result in the third Fourier transform unit in the third dimensional direction; a second target detection unit that detects a target distance, a target velocity, and a target angle based on target coordinate values ​​in reciprocal spatial directions of the first dimension direction, the second dimension direction, and the third dimension direction; 1. An FMCW radar signal processing device comprising:

2. The number of Fourier transforms performed in the entire FMCW radar signal processing device is N2×N3+B1+T1×N3+T2 (N2 and N3 are the total number of coordinate values ​​in the second dimensional direction and the third dimensional direction, respectively, B1 is the total number of coordinate values ​​in the reciprocal space direction of the first dimensional direction, and T1 and T2 are the number of target coordinate values ​​in the reciprocal space direction of the first dimensional direction and the second dimensional direction, respectively).

2. An FMCW radar signal processing device according to claim 1 .

3. The total number of target detections in the FMCW radar signal processing device is B1×B2+B3×T3 (B1, B2, and B3 are the total number of coordinate values ​​in the reciprocal space directions of the first dimension, the second dimension, and the third dimension, respectively, and T3 is the total number of detections in the second target detection unit).

3. An FMCW radar signal processing device according to claim 1 or 2.

4. The minimum memory amount required for the entire FMCW radar signal processing device is B1×N2×N3×D (B1 is the total number of coordinate values ​​in the reciprocal space direction of the first dimension, N2 and N3 are the total numbers of coordinate values ​​in the second dimension and the third dimension, respectively, and D is the data amount according to the data type of each Fourier transform result in the first Fourier transform unit).

4. An FMCW radar signal processing device according to claim 1, wherein the FMCW radar signal processing device comprises:

5. A frequency modulation-continuous wave (FMCW) radar signal processing program using an FMCW radar, a beat signal calculation step of calculating beat signals between a transmission signal and a reception signal sampled by a plurality of samples, the beat signals being calculated for a plurality of chirp numbers and a plurality of antennas; a first Fourier transform step of extracting beat signals including all coordinate values ​​in a second and third dimensional directions among the time elapse direction, the chirp number direction, and the antenna array direction, when performing a Fourier transform on the beat signal calculated in the beat signal calculation step in a first dimensional direction among the time elapse direction, the chirp number direction, and the antenna array direction; a second Fourier transform step of extracting phase values ​​including all coordinate values ​​in a reciprocal space direction of the first dimensional direction and specific coordinate values ​​in the third dimensional direction when performing a Fourier transform on the phase values ​​of the Fourier transform result in the first Fourier transform step in the second dimensional direction; a first target detection step of detecting any two of a target distance, a target velocity, and a target angle based on target coordinate values ​​in a reciprocal space direction of the first dimension direction and the second dimension direction; a third Fourier transform step of extracting phase values ​​including target coordinate values ​​in a reciprocal space direction of the first dimensional direction and all coordinate values ​​in the third dimensional direction when performing a Fourier transform on the phase values ​​of the Fourier transform result in the first Fourier transform step in the second dimensional direction; a fourth Fourier transform step of extracting a phase value including a target coordinate value in a reciprocal space direction of the second dimensional direction when performing a Fourier transform on the phase value of the Fourier transform result in the third Fourier transform step in the third dimensional direction; a second target detection step of detecting a target distance, a target velocity, and a target angle based on target coordinate values ​​in reciprocal spatial directions of the first dimension direction, the second dimension direction, and the third dimension direction; An FMCW radar signal processing program for causing a computer to execute the above steps in sequence.

Citation Information

Patent Citations

  • Antenna switching method for scanning type fmcw radar and scanning type fmcw radar

    JP2003315447A

  • Speed detector

    JP2017207368A

  • Distributed radar signal processing in radar systems

    JP2017521669A

  • Radar system with optimized storage of intermediate data

    JP2017524898A

  • Evaluation device and method for evaluating at least one radar sensor

    US20200408879A1