Radar processing system and radar processing method

Abstract

We provide a radar processing system capable of high-speed processing. [Solution] The radar processing system comprises a plurality of processing units that operate in parallel to detect a target based on a group of radar data received by the radar, and a control unit that controls the plurality of processing units. The radar data group is divided into a plurality of data groups corresponding to each of the plurality of processing units. The control unit instructs each of the plurality of processing units to start the operation of detecting a target from the corresponding data group, and when the operation of detecting a target is completed by one of the plurality of processing units, it instructs the other processing unit to start the operation of detecting a target from the other data groups that have not yet been completed.

Description

Technical Field

[0001] The present disclosure relates to a radar processing system and a radar processing method for detecting an object.

Background Art

[0002] Generally, radar processing systems are used for various purposes. For example, in recent years, radar processing systems have been mounted on vehicles for the purpose of collision prevention and auto cruise. Specifically, after transmitting a transmission wave, a reflected wave from an object is received, and the distance to the object and the like are calculated (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In this regard, a radar processing system that detects an object and enables accurate estimation of the position and velocity of the object is required to process data at high speed in order to process a large amount of radar data. In this regard, a radar processing system may include a plurality of processing devices and execute distributed processing.

[0005] The present disclosure has been made to solve the above problems, and provides a radar processing system and a radar processing method capable of high-speed processing.

[0006] Other problems and novel features will become apparent from the description of this specification and the accompanying drawings.

Means for Solving the Problems

[0007] The radar processing system of this disclosure comprises a plurality of processing units that operate in parallel to detect a target based on a set of radar data received by a radar, and a control unit that controls the plurality of processing units. The radar data set is divided into a plurality of data sets, each corresponding to one of the plurality of processing units. The control unit instructs each of the plurality of processing units to start the operation of detecting a target from its corresponding data set, and when the operation of detecting a target is completed by one of the plurality of processing units, it instructs the other processing unit to start the operation of detecting a target from the other data sets that have not yet been completed.

[0008] The radar processing method of this disclosure comprises the step of operating each of a plurality of processing units in parallel to detect a target based on a set of radar data received by a radar. The set of radar data is divided into a plurality of data sets, each corresponding to one of the plurality of processing units. The step of operating each of the plurality of processing units includes the step of instructing each of the plurality of processing units to start the operation of detecting a target from its corresponding data set, and, if the operation of detecting a target is completed by one of the plurality of processing units, the step of instructing that processing unit to start the operation of detecting a target from the other data sets that have not yet been completed. [Effects of the Invention]

[0009] The radar processing system and radar processing method of this disclosure are capable of high-speed processing. [Brief explanation of the drawing]

[0010] [Figure 1] This figure illustrates a radar processing system 1 according to Embodiment 1 of the present disclosure. [Figure 2] This figure illustrates the processing flow for object detection in a signal processing device 30 according to Embodiment 1 of this disclosure. [Figure 3] This figure illustrates the method for calculating distance Bin according to Embodiment 1 of this disclosure. [Figure 4]This figure illustrates a data cube according to Embodiment 1 of this disclosure. [Figure 5] This figure illustrates the functional configuration of a signal processing device 30 according to Embodiment 1 of this disclosure. [Figure 6] This figure illustrates a map of the 2D FFT results according to Embodiment 1 of this disclosure. [Figure 7] This diagram illustrates the peak detection process according to the comparative example. [Figure 8] This diagram illustrates a specific example of peak detection processing according to the comparative example. [Figure 9] This figure illustrates a peak detection process according to Embodiment 1 of this disclosure. [Figure 10] This diagram illustrates the flow of the peak detection process according to Embodiment 1 of this disclosure. [Figure 11] This figure illustrates the functional configuration of the signal processing device 30# according to Embodiment 2 of this disclosure. [Figure 12] This figure illustrates a peak detection process according to Embodiment 2 of the present disclosure. [Figure 13] Another diagram illustrating a peak detection process according to Embodiment 2 of this disclosure. [Figure 14] This diagram illustrates the flow of the peak detection process according to Embodiment 2 of this disclosure. [Figure 15] This figure illustrates a subroutine for calculating the required processing location according to Embodiment 2 of this disclosure. [Figure 16] This figure illustrates the calculation method for the number of predicted targets Nexp1 and Nexp2 calculated by the required processing position calculation unit 142 according to Embodiment 2 of this disclosure. [Figure 17] This figure illustrates a signal processing device 30#A according to a modified example of Embodiment 2 of the present disclosure. [Modes for carrying out the invention]

[0011] The embodiments will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and the description thereof will not be repeated.

[0012] (Embodiment 1) FIG. 1 is a diagram for explaining a radar processing system 1 of the present disclosure according to Embodiment 1 of the present disclosure. Referring to FIG. 1, the radar processing system 1 is mounted on a vehicle and can be used to detect targets existing around the vehicle, such as other vehicles, signs, guardrails, etc. Note that the radar processing system 1 may be used for various applications other than in-vehicle radar devices (for example, monitoring of an aircraft in flight or a ship in navigation, etc.).

[0013] As an example, a radar processing system for detecting an object will be described with respect to a radar processing system of the FCM (Fast Chirp Modulation) method. The FCM method is a method for obtaining the distance to a target based on the frequency of a beat signal generated from a transmission signal and a reception signal received after being reflected by an object, and obtaining the relative speed to the target based on the phase change of the beat signal. The distance and the relative speed can be obtained separately, and more accurate detection of an object is possible.

[0014] The radar processing system 1 includes a transmission antenna 10, a transmitter 12, a signal generation circuit 14, a reception antenna 20, a receiver 22, an A / D conversion circuit 24, and a signal processing device 30. The signal generation circuit 14 generates a transmission signal (chirp) whose frequency changes in a sawtooth waveform, modulates it with the transmitter 12, and transmits it via the transmission antenna 10. In this example, signals of a plurality of channels (as an example, ch0 - ch12) are received via the reception antenna 20.

[0015] The radar processing system 1 receives, as a reception signal, a reflected wave from an object via the reception antenna 20. The receiver 22 mixes the reception signal with a part of the transmission signal, and generates a beat signal by taking the absolute value of the difference between the transmission signal and the reception signal.

[0016] The signal processing device 30 acquires radar data according to the beat signal input via the A / D conversion circuit 24, analyzes the data, and performs a process to detect objects.

[0017] Figure 2 is a diagram illustrating the processing flow for object detection in a signal processing device 30 according to Embodiment 1 of the present disclosure. Referring to Figure 2, the processing includes distance FFT processing 32, Doppler FFT processing 34, merging processing 36, peak detection processing 38, direction determination processing 40, and aggregation processing 42.

[0018] First, a first Fast Fourier Transform (FFT) process, also known as distance FFT processing 32, is performed. Next, a second Fast Fourier Transform (FFT) process, also known as Doppler FFT processing 34, is performed.

[0019] Here, we will briefly explain how distance and relative velocity are calculated in the FCM (Forward Measurement Compression) method. Figure 3 illustrates the method for calculating distance Bin according to Embodiment 1 of this disclosure. Referring to Figure 3, the FCM method uses one waveform of the transmitted wave, whose frequency changes in a sawtooth pattern, as one chirp, and receives the reflected wave from the object as the received signal. A beat signal is obtained by taking the difference between the received signal and the transmitted signal, and the distance to the object and the relative velocity are calculated by performing a two-dimensional FFT (Fast Fourier Transform) on this beat signal. Specifically, the time delay of the received signal increases as the distance to the transmitted signal increases, so the frequency of the beat signal is proportional to the distance. Therefore, by processing the beat signal with FFT, a peak appears at a frequency position corresponding to the distance to the object.

[0020] Furthermore, since FFT extracts received level and phase information for each frequency point (sometimes referred to as a range bin) set at predetermined frequency intervals, a peak appears in the range bin corresponding to the frequency of the object's distance. Therefore, by detecting the peak frequency, it is possible to calculate the distance to the object.

[0021] Next, we will explain how to calculate relative velocity. In the FCM method, when a relative velocity exists between the radar processing system and an object, the Doppler frequency is detected by utilizing the phase change that occurs between the beat signals according to the Doppler frequency, and the relative velocity is calculated. That is, if the relative velocity is 0, there is no Doppler component in the received signal, so the phase of the received signal for each chirp is all the same. On the other hand, when there is a relative velocity with the object, a phase change occurs between the received signals for each chirp according to the Doppler frequency. Since the peak information obtained by processing the beat signal with FFT contains this phase information, if the peak information of the same object obtained from each beat signal is arranged in time series and a second FFT is performed, the Doppler frequency can be determined from the phase information, and a peak will appear at that frequency position. This peak frequency corresponds to the relative velocity. In this way, it is possible to calculate distance and relative velocity by performing a two-dimensional FFT on the beat signal.

[0022] Figure 4 illustrates a data cube according to Embodiment 1 of this disclosure. Referring to Figure 4, the state of the map of 2D FFT results (one dimension corresponding to distance and the other dimension corresponding to velocity) obtained by the Doppler FFT process 34 is shown. Then, a join process 36 is performed to combine these for each channel. In this example, a case is shown in which a data cube is generated by joining the 2D FFT results in the channel direction.

[0023] The peak detection process 38 detects the velocity and distance of an object (target) as well as peak power information by identifying peaks within the generated data cube, for example, using the CFAR (Constant False Alarm Rate) algorithm.

[0024] The direction determination process 40 performs a process to estimate the angle of the detected object (target). Specifically, it performs a process to estimate the angle of the detected object (target) based on the velocity and distance of the target (detected position information) and peak power information detected in the peak detection process, as well as the Doppler FFT results of each channel. Note that the method is not limited to this method, and the angle may be estimated based on other methods. For example, ambient power (peak interpolation) may be used to fine-tune Bin, or CFAR threshold information or the power difference between the target and that threshold (noise margin) may be used.

[0025] Aggregation process 42 is a process that aggregates and outputs the distance, velocity, and angle of the identified object.

[0026] Figure 5 is a diagram illustrating the functional configuration of a signal processing device 30 according to Embodiment 1 of the present disclosure. Referring to Figure 5, the signal processing device 30 includes an overall processing device 100 and a memory 200. The overall processing device 100 performs various processes of the signal processing device 30 in cooperation with the memory 200.

[0027] This example primarily describes the process of detecting an object (target), that is, estimating the object's distance, velocity, and angle. Specifically, it describes the case where the peak detection process 38 and the direction determination process 40 are performed.

[0028] The overall processing unit 100 includes a first processing unit 110, a second processing unit 120, and a control unit 130. The first processing unit 110, the second processing unit 120, and the control unit 130 may each be different hardware devices, or they may be implemented functionally differently by software programs.

[0029] The first processing unit 110 and the second processing unit 120 operate in parallel to detect a target based on the radar data received by the radar. The control unit 130 controls the first processing unit 110 and the second processing unit 120.

[0030] The above-described merging process 36 generates a data cube, and the process for detecting a target is executed. Specifically, the control unit 130 divides the radar data group of the data cube into multiple data groups. For example, the radar data group is stored in the memory 200. The control unit 130 divides the radar data group into two, the first processing unit 110 processes the first divided data group, and the second processing unit 120 processes the second divided data group.

[0031] In this example, the control unit 130 instructs each of the multiple processing units (first processing unit 110 and second processing unit 120) to start the operation of detecting an object (target) from the corresponding data group. When the operation of detecting a target is completed in one of the multiple processing units, the control unit 130 instructs that processing unit to start the operation of detecting a target from the other data groups that have not yet been completed.

[0032] Specifically, the control unit 130 instructs the first processing unit 110 to start the operation of detecting a target from the first data group, and when the target detection operation is completed, to start the operation of detecting a target from the incomplete second data group. The control unit 130 also instructs the second processing unit 120 to start the operation of detecting a target from the second data group, and when the target detection operation is completed, to start the operation of detecting a target from the incomplete first data group.

[0033] The control unit 130 includes a position setting unit 132, a detection operation instruction unit 134, an operation determination unit 136, and a reverse operation instruction unit 138.

[0034] The position setting unit 132 sets the start and end positions for the detection operation of the first and second data groups.

[0035] The detection operation instruction unit 134 instructs the first and second processing units 110 and 120 to perform a detection operation.

[0036] The operation determination unit 136 determines the completion of the detection operations of the first and second processing units 110 and 120.

[0037] The reverse operation instruction unit 138, based on the determination result of the operation determination unit 136's determination of completion of the detection operation, instructs one of the first and second processing units 110 and 120 to execute the detection operation of the corresponding data group of the other.

[0038] Figure 6 illustrates a map of a two-dimensional FFT result according to Embodiment 1 of this disclosure. Referring to Figure 6, as an example for brief explanation, a map of a two-dimensional FFT result for a channel of a data cube is shown. One dimension corresponds to distance, and the other dimension corresponds to velocity. In this example, data sets from distance Bin0 to distance Bin1023 are shown according to distance.

[0039] Figure 7 illustrates the target detection process according to the comparative example. Referring to Figure 7, in the comparative example, the position setting unit 132 sets, for example, distance Bin0 as the starting position and distance Bin511 as the ending position for the first processing unit 110. The position setting unit 132 also sets, for example, distance Bin512 as the starting position and distance Bin1023 as the ending position for the second processing unit 120. In the comparative example, the first processing unit 110 performs a peak detection process 38 and a direction determination process 40 as operations to detect targets from distance Bin0 to distance Bin511 with respect to the assigned data group.

[0040] The second processing unit 120 performs a peak detection process 38 and a direction determination process 40 as operations to detect targets from distance Bin 512 to distance Bin 1023 with respect to the assigned data set.

[0041] Figure 8 illustrates a specific example of peak detection processing according to the comparative example. Referring to Figure 8, it is shown that, as an example, the number of targets detected by the first processing unit 110 is 1500, and the number of targets detected by the second processing unit 120 is 500. The processing time increases as the number of detected targets in the data set increases.

[0042] Therefore, in this example, the processing time of the first processing unit 110 is long and the processing time of the second processing unit 120 is short, resulting in a large amount of idle time for the second processing unit 120, which required improvement in terms of high-speed processing of data sets.

[0043] Figure 9 is a diagram illustrating the target detection process according to Embodiment 1 of the present disclosure. Referring to Figure 9(A), the position setting unit 132 sets, for example, distance Bin0 as the starting position and distance Bin511 as the ending position for the first processing unit 110. The position setting unit 132 sets, for example, distance Bin512 as the starting position and distance Bin1023 as the ending position for the second processing unit 120.

[0044] In this embodiment, the first processing unit 110 performs a peak detection process 38 and a direction determination process 40 as operations to detect targets from distance Bin0 to distance Bin511 with respect to the assigned data group. The second processing unit 120 performs a peak detection process 38 and a direction determination process 40 as operations to detect targets from distance Bin512 to distance Bin1023 with respect to the assigned data group.

[0045] In this example, the second processing unit 120 is shown to have reached distance Bin 1023 from distance Bin 512, and the number of detected targets is 750. The first processing unit 110 is shown to have started from distance Bin 0, not yet reached distance Bin 511, and the number of detected targets is 750.

[0046] Referring to Figure 9(B), the second processing unit 120 has reached its end position at distance Bin 1023, and therefore executes the process of detecting targets in reverse order from the end position of the first processing unit 110 at distance Bin 511. The operation is then completed at the position where the processes collide. In this example, the first processing unit 110 detects 1000 targets, and the second processing unit 120 detects 750 targets plus an additional 250 targets.

[0047] Therefore, referring to Figure 9(C), the target detection processes of the first processing unit 110 and the second processing unit 120 are completed simultaneously. That is, the processing times of the first processing unit 110 and the second processing unit 120 are the same, there is no idle time, and it is possible to process the data set at high speed. In the comparative example, idle time was spent waiting, but in this example, instead of waiting, the system supports the processing of other processing units, so it is possible to perform data processing efficiently. Note that in this example, for the sake of simplicity, the case in which the same number of targets are detected is described, but the number of targets is not necessarily the same.

[0048] Figure 10 is a diagram illustrating the flow of the target detection process according to Embodiment 1 of the present disclosure. Referring to Figure 10, the control unit 130 performs an initialization process (step S2). Specifically, the position setting unit 132 sets distance Bin0 as the starting position and distance Bin511 as the ending position for the first processing unit 110. The position setting unit 132 also sets distance Bin512 as the starting position and distance Bin1023 as the ending position for the second processing unit 120. Then, the position setting unit 132 sets the processing direction for both the first processing unit 110 and the second processing unit 120 to forward. Next, the control unit 130 performs a peak detection process at the processing position (step S4). Specifically, the detection operation instruction unit 134 instructs the first processing unit 110 and the second processing unit 120 to perform a peak detection process based on the information set. As a result, the first processing unit 110 and the second processing unit 120 perform a peak detection process at the specified processing position.

[0049] Next, the control unit 130 performs direction determination processing for the processing position (step S5). Specifically, the detection operation instruction unit 134 instructs the first processing unit 110 and the second processing unit 120 to perform direction determination processing. As a result, the first processing unit 110 and the second processing unit 120 perform direction determination processing at the specified processing position. Specifically, they perform processing to estimate the angle based on the velocity and distance of the target detected in the peak detection processing (detected position information), the peak power information, and the Doppler FFT results for each channel.

[0050] Next, the control unit 130 determines whether the processing position is the end position or not (step S6). The operation determination unit 136 determines whether the processing position of the first processing unit 110 has reached the distance Bin 511, which is the end position. The operation determination unit 136 determines whether the processing position of the second processing unit 120 has reached the distance Bin 1023, which is the end position.

[0051] In step S6, if the control unit 130 determines that the processing position is the end position (YES in step S6), it determines whether the processing of the other processing unit has finished (step S8). If the operation determination unit 136 determines that the processing position of the first processing unit 110 is the end position, it determines whether the processing of the second processing unit 120 has finished. Also, if the operation determination unit 136 determines that the processing position of the second processing unit 120 is the end position, it determines whether the processing of the first processing unit 110 has finished.

[0052] In step S8, if the control unit 130 determines that the processing of other processing units has finished (YES in step S8), it terminates its processing (end). In this case, the processing of the first processing unit 110 and the second processing unit 120 is considered to have finished almost simultaneously, and the target detection process is terminated.

[0053] On the other hand, in step S8, if the control unit 130 determines that the processing of the other processing unit has not finished (NO in step S8), it sets the start of processing to the end position of the other processing unit and the end of processing to the start position of the other processing unit. The operation determination unit 136 instructs the reverse operation instruction unit 138, which in turn instructs the first processing unit 110 to set distance Bin 1023, which is the end position of the second processing unit 120, as the start position and distance Bin 512, which is the start position of the second processing unit 120, as the end position. Alternatively, the reverse operation instruction unit 138 instructs the second processing unit 120 to set distance Bin 511, which is the end position of the first processing unit 110, as the start position and distance Bin 0, which is the start position of the first processing unit, as the end position.

[0054] Next, the control unit 130 sets the processing direction in reverse order (step S12). The reverse order operation instruction unit 138 sets the processing direction in reverse order for the first processing unit 110 or the second processing unit 120. Then, the control unit 130 proceeds to "A". That is, it returns to step S4 and executes the peak detection process 38 and the direction determination process 40.

[0055] On the other hand, in step S6, if the control unit 130 determines that the processing position is not the end position (NO in step S6), it determines whether the processing direction is forward or not (step S14). If the operation determination unit 136 determines that the processing position of the first processing unit 110 or the second processing unit 120 is not the end position, it instructs the detection operation instruction unit 134, and the detection operation instruction unit 134 determines whether the set processing direction of the first processing unit 110 or the second processing unit 120 is forward or not.

[0056] In step S14, if the control unit 130 determines that the processing direction is forward, it adds to the processing position (step S18). The detection operation instruction unit 134 adds to the processing position if the processing direction of the set first processing unit 110 or second processing unit 120 is forward. Then, the process proceeds to step S20.

[0057] On the other hand, in step S14, if the control unit 130 determines that the processing is not in the forward direction, that is, in the reverse forward direction (NO in step S14), it subtracts the processing position (step S16). The detection operation instruction unit 134 subtracts the processing position if the processing direction of the set first processing unit 110 or second processing unit 120 is in the reverse forward direction. Then, the process proceeds to step S20.

[0058] In step S20, the control unit 130 determines whether the processing position is the same as that of another processing unit (step S20). The detection operation instruction unit 134 determines whether the processing positions of the first processing unit 110 and the second processing unit 120 are the same.

[0059] In step S20, if the control unit 130 determines that the processing position is the same as that of another processing unit (YES in step S20), it terminates the process (end). If the detection operation instruction unit 134 determines that the processing positions of the first processing unit 110 and the second processing unit 120 are the same, it determines that the peak detection process 38 and the direction determination process 40 are duplicated and terminates the process.

[0060] On the other hand, in step S20, if the control unit 130 determines that the processing position is not the same as that of another processing unit (NO in step S20), it returns to step S4 and executes the peak detection process 38 and the direction determination process 40. If the detection operation instruction unit 134 determines that the processing positions of the first processing unit 110 and the second processing unit 120 are not the same, the peak detection process 38 and the direction determination process 40 will not overlap, so it instructs the first processing unit 110 or the second processing unit 120 to execute the peak detection process 38 and the direction determination process 40 for the added or subtracted processing positions.

[0061] As a result of this process, the radar processing system completes the target detection processing of the first processing unit 110 and the second processing unit 120 almost simultaneously, as explained in Figure 9.

[0062] In this example, it is possible to make the number of objects (targets) detected approximately the same. That is, the total processing time of the first processing unit 110 and the second processing unit 120 will be approximately the same, and it will be possible to process the data set at high speed.

[0063] In the above description, we have mainly explained the case where the peak detection process 38 and the direction determination process 40 are performed when detecting an object (target). However, this method is not limited to this, and the same methods can be applied to cases where an object (target) is detected by applying only one of the processes.

[0064] (Embodiment 2) In the above embodiment 1, a method for detecting objects (targets) in all the data of the generated data cube was described.

[0065] On the other hand, if there is a limit to the number of objects (targets) that can be detected, it is possible to perform the detection process within a limited range rather than performing the detection process on all data, thereby speeding up the process.

[0066] Embodiment 2 of this disclosure describes a method in which the peak detection process 38 and the direction determination process 40 are performed within a limited range while monitoring the number of detected objects (targets).

[0067] Figure 11 is a diagram illustrating the functional configuration of signal processing device 30# according to Embodiment 2 of this disclosure. Referring to Figure 11, signal processing device 30# differs from the signal processing device 30 in Figure 5 in that the overall processing device 100 is replaced by overall processing device 100#. Other aspects are the same as those described in Figure 5, so a detailed explanation will not be repeated.

[0068] The overall processing unit 100# differs from the overall processing unit 100 in that the control unit 130 has been replaced by the control unit 130#. The control unit 130# differs from the control unit 130 in that it further includes a counting unit 140 and a required processing position calculation unit 142. Other points are the same as those explained in Figure 5, so a detailed explanation will not be repeated.

[0069] The counting unit 140 counts the number of objects detected by the first processing unit 110 and the second processing unit 120.

[0070] The required processing position calculation unit 142 calculates the required processing position based on the counting result of the counting unit 140.

[0071] Figure 12 is a diagram illustrating the target detection process according to Embodiment 2 of the present disclosure. Referring to Figure 12(A), the position setting unit 132 sets, for example, distance Bin0 as the starting position and distance Bin511 as the ending position for the first processing unit 110. The position setting unit 132 sets, for example, distance Bin512 as the starting position and distance Bin1023 as the ending position for the second processing unit 120.

[0072] In Embodiment 2, as described above, the first processing unit 110 performs a peak detection process as an operation to detect a target from distance Bin0 with respect to the assigned data group. The second processing unit 120 performs a peak detection process 38 and a direction determination process 40 as an operation to detect a target from distance Bin512 with respect to the assigned data group.

[0073] In Embodiment 2, the number of detected objects (targets) is counted, and the required processing location is calculated based on the count result.

[0074] For example, in this example, the first processing unit 110 is shown starting from distance Bin0 and detecting a target number of 300. Similarly, the second processing unit 120 is shown starting from distance Bin512 and detecting a target number of 300.

[0075] For example, if the maximum number of objects (targets) detected in the detection process of the signal processing device 30# is 2000, the total number of targets predicted from the number of targets detected by the first processing unit 110 and the second processing unit 120 is estimated.

[0076] If the predicted total number of targets is greater than the maximum number of detections, the required processing position is calculated. Specifically, the first processing unit 110 prioritizes data sets with shorter processing distances over data sets with longer processing distances. Then, by adjusting the end position processed by the second processing unit 120, the maximum number of detected objects (targets) is kept within 2000. In this regard, the required processing position calculation unit 142 calculates the required processing position based on the count results so that the maximum number of detected objects (targets) is kept within 2000. The required processing position calculation unit 142 then sets the calculated required processing position as the end position of the second processing unit 120.

[0077] Referring to Figure 12(B), the second processing unit 120, after reaching the calculated required processing position, executes the process of detecting targets in reverse order from the end position of the first processing unit 110, which is distance Bin 511. The operation is then completed at the position where the processing collides. In this example, the first processing unit 110 detects 1000 targets, and the second processing unit 120 detects 500 targets plus another 500 targets.

[0078] Therefore, referring to Figure 12(C), the target detection processing of the first processing unit 110 and the second processing unit 120 is completed simultaneously. That is, while adjusting within the limited range of a maximum number of detected objects of 2000, the processing times of the first processing unit 110 and the second processing unit 120 become the same, making it possible to process the data set at high speed. In this case, since the target detection processing is not performed beyond the maximum number of detected objects, processing can be performed efficiently.

[0079] Figure 13 is another diagram illustrating the target detection process according to Embodiment 2 of the present disclosure. Referring to Figure 13(A), the position setting unit 132 sets, for example, distance Bin0 as the starting position and distance Bin511 as the ending position for the first processing unit 110. The position setting unit 132 sets, for example, distance Bin512 as the starting position and distance Bin1023 as the ending position for the second processing unit 120.

[0080] In Embodiment 2, similar to the above, the first processing unit 110 executes a peak detection process 38 and a direction determination process 40 as operations to detect a target from distance Bin0 with respect to the assigned data group. The second processing unit 120 executes a peak detection process 38 and a direction determination process 40 as operations to detect a target from distance Bin512 with respect to the assigned data group.

[0081] In Embodiment 2, the number of detected objects (targets) is counted, and the required processing position is calculated based on the count result. The required processing position calculation unit 142 calculates the required processing position if the predicted total number of targets is greater than the maximum number of detected targets. The required processing position calculation unit 142 sets the calculated required processing position as the end position of the second processing unit 120. For example, in this example, the first processing unit 110 starts from distance Bin 0, reaches distance Bin 511, and detects a target count of 500. Similarly, the second processing unit 120 starts from distance Bin 512 and detects a target count of 500.

[0082] Referring to Figure 13(B), the first processing unit 110, after reaching the end position, executes the process of detecting targets in reverse order from the required processing position, which is the end position of the second processing unit 120. Then, it completes its operation at the position where the processes collide. In this example, the second processing unit 120 detects a target count of 1000, and the first processing unit 110 detects a target count of 500 plus an additional 500 targets.

[0083] Therefore, referring to Figure 13(C), the target detection processing of the first processing unit 110 and the second processing unit 120 is completed simultaneously. That is, while adjusting within the limited range of a maximum number of detected objects of 2000, the processing times of the first processing unit 110 and the second processing unit 120 become the same, making it possible to process the data set at high speed. In this case, since peak detection processing is not performed beyond the maximum number of detected objects, processing can be performed efficiently.

[0084] Figure 14 is a diagram illustrating the flow of the target detection process according to Embodiment 2 of this disclosure. Referring to Figure 14, the difference from the flow described in Figure 10 is that steps S22 and S24 have been added. The other processes are the same as those described in Figure 10, so a detailed explanation will not be repeated.

[0085] Specifically, in step S4, the control unit 130# performs the processing position peak detection process 38 and the direction determination process 40, and then performs the required processing position calculation process (step S22). Details of the required processing position calculation process will be described later.

[0086] Figure 15 illustrates a subroutine for calculating the required processing location according to Embodiment 2 of this disclosure. Referring to Figure 15, the control unit 130# calculates the predicted number of targets Nexp1 and Nexp2 to be detected by each processing unit (step S30). Specifically, the required processing location calculation unit 142 obtains the number of objects (targets) detected by the first processing unit 110 and the second processing unit 120 from the counting unit 140 and calculates the predicted number of targets Nexp1 and Nexp2 based on a predetermined method.

[0087] Figure 16 illustrates the calculation method for the number of predicted targets Nexp1 and Nexp2 calculated by the required processing position calculation unit 142 according to Embodiment 2 of this disclosure. Referring to Figure 16, in this example, the required processing position calculation unit 142 calculates prediction lines for targets detected by the first processing unit 110 and the second processing unit 120, respectively, and estimates the number of targets Nexp1 and Nexp2 that will be detected if the first processing unit 110 and the second processing unit 120 perform detection operations up to the end position based on the calculated prediction lines. For example, the slopes R1 (=N1 / (P1-Pmin1)) and R2 (=N2 / (P2-Pmin2)) of the prediction lines are calculated from the number of targets N1 and N2 at processing positions P1 and P2.

[0088] The predicted number of targets Nexp1 can be calculated using the slope R1*(Pmax1-Pmin1). The predicted number of targets Nexp2 can be calculated using the slope R2*(Pmax2-Pmin2). Here, Pmin1 and Pmin2 are the starting positions of the first processing unit 110 and the second processing unit 120, and Pmax1 and Pmax2 are the ending positions of the first processing unit 110 and the second processing unit 120. In this example, as an example, Pmin1 is Bin0 and Pmin2 is Bin512. Pmax1 is Bin511 and Pmax2 is Bin1023.

[0089] Referring again to Figure 15, the control unit 130# then determines whether the calculated total number of predicted targets Nexp1 + Nexp2 exceeds the maximum number of detections Nmax (step S32). The required processing position calculation unit 142 compares the predicted estimated total of Nexp1 + Nexp2 with the maximum number of detections Nmax and determines whether the condition Nexp1 + Nexp2 > Nmax is met.

[0090] If the control unit 130# determines that the total number of predicted targets Nexp1 + Nexp2 exceeds the maximum number of detections Nmax (YES in step S32), it calculates the number of target detections Nt for the lower-priority data group.

[0091] Specifically, the required processing position calculation unit 142 calculates the surplus number of targets ΔN (=Nexp1+Nexp2-Nmax) by subtracting the maximum number of detections Nmax from the total number of predicted targets Nexp1+Nexp2. Next, the required processing position calculation unit 142 calculates the number of detection target targets Nt (=Nexp2-ΔN) for the second processing unit 120 with lower priority according to the surplus number of targets ΔN.

[0092] Next, the control unit 130# calculates the required processing position Pt from the number of detected target targets Nt (step S36). The required processing position calculation unit 142 calculates the required processing position Pt (=Nt / R2) corresponding to the number of detected target targets Nt based on the target prediction line.

[0093] Then, terminate the process (press Return). On the other hand, if the control unit 130# determines that the total number of predicted targets Nexp1 + Nexp2 does not exceed the maximum number of detections Nmax (NO in step S32), it proceeds to process "B". In other words, since it is not necessary to calculate the required processing position, this process is skipped and the process proceeds to step S6.

[0094] In this example, we describe a method for calculating the number of target predictions using a linear prediction line. However, it is also possible to perform more accurate predictions using autoregressive models (AR models) or moving average models (MA models). Furthermore, if the processing position is small, the accuracy of the predicted value may be low, so it may be advisable to apply weighting based on the number of target predictions or the processing position.

[0095] Referring again to Figure 14, the control unit 130# resets the calculated end position to the required processing position (step S24). Specifically, the required processing position calculation unit 142 sets the required processing position Pt to the end position of the second processing unit 120.

[0096] Next, the control unit 130# determines whether the processing position is the end position or not (step S6). The operation determination unit 136 determines whether the first processing unit 110 has reached the distance Bin 511 where the processing position is the end position. The operation determination unit 136 determines whether the second processing unit 120 has reached the required processing position Pt where the processing position is the end position.

[0097] In step S6, if the control unit 130 determines that the processing position is the end position (YES in step S6), it determines whether the processing of the other processing unit has finished (step S8). If the operation determination unit 136 determines that the processing position of the first processing unit 110 is the end position, it determines whether the processing of the second processing unit 120 has finished. Also, if the operation determination unit 136 determines that the processing position of the second processing unit 120 is the required processing position Pt, it determines whether the processing of the first processing unit 110 has finished.

[0098] In step S8, if the control unit 130# determines that the processing of other processing units has finished (YES in step S8), it terminates its processing (end). In this case, the processing of the first processing unit 110 and the second processing unit 120 is considered to have finished almost simultaneously, and the target detection process is terminated.

[0099] On the other hand, in step S8, if the control unit 130# determines that the processing of the other processing unit has not finished (NO in step S8), it sets the start of processing to the end position of the other processing unit and the end of processing to the start position of the other processing unit. The operation determination unit 136 instructs the reverse operation instruction unit 138, which sets the required processing position Pt, which is the end position of the second processing unit 120, as the start position for the first processing unit 110, and sets the distance Bin512, which is the start position of the second processing unit 120, as the end position. Alternatively, the reverse operation instruction unit 138 sets the distance Bin511, which is the end position of the first processing unit 110, as the start position for the second processing unit 120, and sets the distance Bin0, which is the start position of the first processing unit, as the end position.

[0100] Next, the control unit 130 sets the processing direction in reverse order (step S12). The reverse order operation instruction unit 138 sets the processing direction in reverse order for the first processing unit 110 or the second processing unit 120. Then, the control unit 130 proceeds to "A". That is, it returns to step S4 and executes the peak detection process 38 and the direction determination process 40.

[0101] On the other hand, in step S6, if the control unit 130 determines that the processing position is not the end position (NO in step S6), it determines whether the processing direction is forward or not (step S14). The subsequent processing is the same as that described in Figure 10, so a detailed explanation will not be repeated.

[0102] As a result of this process, the radar processing system completes the target detection processing of the first processing unit 110 and the second processing unit 120 almost simultaneously, as explained in Figures 12 and 13.

[0103] In this example, it is possible to make the number of objects (targets) detected approximately the same. That is, the total processing time of the first processing unit 110 and the second processing unit 120 will be approximately the same, and it will be possible to process the data set at high speed.

[0104] In the above description, we have mainly explained the case in which the peak detection process 38 and the direction determination process 40 are performed when detecting an object (target). However, this method is not limited to this, and the same methods can be applied to cases in which an object (target) is detected by applying only one of the processes.

[0105] Figure 17 illustrates a modified signal processing device 30#A according to Embodiment 2. Referring to Figure 17, the signal processing device 30#A differs from the signal processing device 30 in Figure 11 in that the overall processing device 100# is replaced with the overall processing device 100#A. Other points are the same as those described in Figure 11, so a detailed explanation will not be repeated.

[0106] The overall processing unit 100#A differs from the overall processing unit 100# in that it adds a third processing unit 125 in addition to the first processing unit 110 and the second processing unit 120. Other aspects are the same as those explained in Figure 11, so a detailed explanation will not be repeated.

[0107] In other words, this method is also applicable to cases where multiple processing units are added in addition to the first processing unit 110 and the second processing unit 120 for distributed processing.

[0108] Specifically, regarding the 2D FFT result map explained in Figure 6, in the above embodiment, the case where it is divided into two parts corresponding to the first and second processing units 110 and 120 was described. However, in this example, for example, it would be possible to divide it into three parts according to the number of processing units. Other points are the same as described above.

[0109] The radar processing system can process data even faster by incorporating multiple processing units and operating them in parallel.

[0110] In this embodiment, a method of dividing the data set according to the distance Bin has been described, but the method of division is not limited to this, and for example, a method of division according to speed may be adopted.

[0111] Although the present disclosure has been described in detail based on embodiments, it goes without saying that the present disclosure is not limited to these embodiments and can be modified in various ways without departing from its essence. [Explanation of Symbols]

[0112] 1 Radar processing system, 10 Transmitting antenna, 12 Transmitter, 14 Signal generation circuit, 20 Receiving antenna, 22 Receiver, 24 A / D conversion circuit, 30, 30#, 30#A Signal processing device, 100, 100# Overall processing device, 110 First processing unit, 120 Second processing unit, 125 Third processing unit, 130 Control unit, 132 Position setting unit, 134 Detection operation instruction unit, 136 Operation determination unit, 138 Reverse operation instruction unit, 140 Counting unit, 142 Required processing position calculation unit, 200 Memory.

Claims

1. Multiple processing units that operate in parallel to detect a target based on a set of radar data received by the radar, The system includes a control unit that controls the plurality of processing units, The radar data set is divided into multiple data sets corresponding to each of the multiple processing units, A radar processing system in which the control unit instructs each of the plurality of processing units to start the operation of detecting the target from a corresponding data group, and when the operation of detecting the target is completed by one of the plurality of processing units, the control unit instructs that one processing unit to start the operation of detecting the target from the other data group that has not yet been completed.

2. The plurality of processing units include a first and a second processing unit, The radar data set is divided into a first data set and a second data set. The control unit, The first processing unit is instructed to start the operation of detecting the target from the first data group, and when the operation of detecting the target is completed, to start the operation of detecting the target from the incomplete second data group. The radar processing system according to claim 1, wherein the second processing unit is instructed to start an operation to detect the target from the second data group, and when the operation to detect the target is completed, to start an operation to detect the target from the incomplete first data group.

3. The control unit, A position setting unit that sets the start and end positions of the detection operation of the first and second data groups, A detection operation instruction unit that instructs the first and second processing units to perform a detection operation, An operation determination unit that determines the completion of the detection operations of the first and second processing units, The radar processing system according to claim 2, further comprising a reverse operation instruction unit that, based on the determination result of the operation determination unit's determination of completion of the detection operation, instructs one of the first and second processing units to perform the detection operation of the corresponding data group of the other unit.

4. The position setting unit sets first and second start positions indicating the beginning of the data in the first and second data groups, and first and second end positions indicating the end of the data. The detection operation instruction unit instructs the first and second processing units to execute the detection operation from the first and second starting positions. The radar processing system according to claim 3, wherein the reverse-sequence operation instruction unit instructs the first and second processing units to execute the detection operations in reverse order from the end position of the other data group, based on the determination result of the operation determination unit's determination of the completion of one of the detection operations of the first and second processing units.

5. The control unit, A counting unit for counting the number of targets detected by the first and second processing units, The system includes a required processing position calculation unit that calculates the required processing position based on the counting result of the counting unit, The radar processing system according to claim 3, wherein the position setting unit resets the end position of the detection operation of one of the first and second data groups based on the calculated required processing position.

6. The aforementioned required processing position calculation unit is: Based on the counting result of the counting unit, the predicted count numbers of the first and second processing units are estimated. The radar processing system according to claim 5, which calculates the required processing location based on the estimation result.

7. A step of operating multiple processing units in parallel to detect a target based on a set of radar data received by the radar, The radar data set is divided into multiple data sets corresponding to each of the multiple processing units, The aforementioned step of performing the operation is: The steps include instructing each of the plurality of processing units to start the operation of detecting the target from the corresponding data set, A radar processing method comprising the step of instructing one of the plurality of processing units to start an operation to detect the target from other data sets that have not yet been completed, once the operation to detect the target has been completed by one of the processing units.

Images