Signal processing method and device, equipment and storage medium

[0047] Various exemplary embodiments, features, and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the drawings represent the same or similar elements. Although various aspects of the embodiments are shown in the drawings, unless otherwise indicated, the drawings are not required.

[0048] The word "exemplary" exemplified here is "used as examples, embodiments or illustrative". Any embodiment as described herein as "exemplary" does not have to be construed as better or better than other embodiments.

[0049] In addition, in order to better illustrate the present application, numerous specific details are given in the specific embodiments below. Those skilled in the art will appreciate that there is no specific details, the present application can also be implemented. In some examples, the methods, components, and circuits are not described in detail in the methods well known to those skilled in the art, so that the main purpose of the application is high.

[0050] In order to solve the technical problem of the processing efficiency of the radar frame signal in the prior art, the processing time is long, and the present invention proposes a signal processing method, apparatus, device, and storage medium, and the present invention is specifically implemented in the following technical solutions.

[0051] A signal processing method provided in the present specification is applied to the vehicle radar. The vehicle radar can be a vehicle millimeter wave radar, and the vehicle millimeter wave radar can transmit and receive signals, and the detection target is detected, and the results of the detection object, The detection object can be other vehicles, pedestrians, buildings in the environment where the vehicles are located.

[0052] Combine Figure 1 to 3 As shown, the signal processing method provided by the embodiment of the present specification includes:

[0053] Step S101: Receive the target sub-signal, the target sub-signal is any of a plurality of sub-signals included in the radar subframe signal.

[0054] In the present specification, the radar can sequentially receive a plurality of sub-signals, starting from the first sub-signal received after the radar, the first number of sub-signals can form a radar frame signal; can determine the first as the actual demand A specific value of a number; the first number of the present specification may be 128.

[0055] Step S102: Sampling the target sub-signal to obtain a plurality of sampled signals.

[0056]In the present specification, the radar is provided with a second number of sampling channels, each sampling channel can sample the target sub-signal; the second number of sampling channels are sampled to sample the target sub-signal, respectively, to obtain the second number The sampling signal, the second number of sampling signals correspond to the second number of sampling channels; the second number of specific values ​​can be determined according to the actual needs.

[0057] Step S103: Parallel The plurality of sampling signals are processed from the dimension of FFI, obtain multiple distance dimension signals.

[0058] In the present specification, a sample signal can correspond to a distance dimension signal. Multiple sampling signals can be performed based on field programmable gate array devices (FPGA, FIELDPROGRAMMABLEARRAY), and multiple sampling signals are processed in parallel. Distance Fourier leaf processing can be a fast Fourier handling of the sampling signal and obtain processing from the relevant information, and the processing result can reflect the detection object and the radar between distances between distance.

[0059] Step S104: The plurality of distance dimension signals perform signal fusion processing to obtain the target distance dimension signal.

[0060] In the present specification, the signal fusion processing may be a plurality of distance dimension signals to be combined, and the obtained target distance dimension signal has a large bit wide. A full distance dimension signal corresponding to the target sub-signal can be simultaneously fused, and the target distance dimension signal is obtained by one signal fusion processing.

[0061] Combine Figure 4 As shown in the prior art, only serial processing is serially used when signal processing is performed using a microcontrol unit (MCU, MicroControllerUnit). Only one sample signal can be performed, and other sampling signals need to be cached. After the sampling signal processing is completed, a sampled signal is taken from the cache to process the distance. Multiple sampling signals need to be multiple times, serial, distance to dimensional, low processing efficiency, long time. For example, when multiple sampling signals include 0_adc_data, 1_adc_data, 2_adc_data, and 3_adc_data, a total of serial is required to perform 4 times distance dimensional Fourier leaf processing, which needs to cache 1_adc_data, 2_adc_data when the 0_adc_data is processed, and the 0_adc_data is processed. 3_adc_data.

[0062] Combine Figure 5 and Image 6 As shown in the present specification, the plurality of sampling signals can be performed based on the pool of the field programmable gate array device (FPGA, FieldProgramMable GateArray), and simultaneously, the plurality of sampling signals can be treated simultaneously. Processing result (multiple distance dimension signals); for example, when multiple sampling signals include 0_adc_data, 1_adc_data, 2_adc_data, and 3_adc_data, the distance Vi Fu Lili is processed in parallel, and does not need to cache 1_adc_data, 2_adc_data and 3_adc_data. The use of parallel distance Vi Fu Yibi, which is time consuming, and has high processing efficiency.

[0063] Further, since the plurality of distance dimension signals are simultaneously obtained, the plurality of distance dimensional signals can be fused to process, and the primary signal fusion processing can obtain a target distance dimension signal; for example, in multiple sampling signals include 0_adc_data When 1_adc_data, 2_adc_data, and 3_adc_data, the distance Weili leaf processing is performed in parallel to obtain four distance dimension signals (0_fft_data, 1_fft_data, 2_fft_data, and 3_fft_data), and signal fusion processing for 4 distance dimensional signals to obtain target distance dimension signals. . It can be seen that the treatment method of the present specification is less time consuming, and the processing efficiency is high.

[0064] In a possible implementation, step S103 includes: calling distance dimensional computing resources, parallel to multiple sampling signals to DVFE leaf processing, to obtain multiple distance dimension signals;

[0065] After step S103, the method also includes: Release the distance dimension calculation resource.

[0066] In the present specification embodiment, the calculation resource can be one or more of the following: the central processor resources, memory resources, hard disk resources, network resources required for program runtime. The distance dimensional computing resource can be the calculation resource required to perform distance DVF.

[0067] During the processes of the sampling signal of the target sub-signal, it is necessary to occupy a distance dimensional computing resource; after the end of the DVF, the distance dimension calculation resources can be released; the release dimensional computing resources can be supplied Sampling signal of one sub-signal. For example, after receiving a sub-signal, the sampling signal of the sub-signal is processed, and the distance dimension calculation resource is released after the end of the Division, and then the next sub-signal is then started. The distance dimensional computing resources in the present specification can be reused, improve the utilization of calculating resources, which can reduce the demand for the total computing resource, saving the manufacturing cost of radar.

[0068] In one possible implementation, step S104 further includes step S105: Storing the target distance dimension signal.

[0069] In the prior art, since the distance dimension of multiple sampling signals is serial processing, multiple distance dimension signals can only be sequentially sequentially sequentially; each time a distance dimension is required to first Cache to the random access memory (RAM, Randomaccess Memory); after the distance dimensional signal cache of the last sampling signal of the target sub-signal, multiple distance dimension signals to double rate synchronous dynamic random memory in sequence (DDR, DOUBLE DATA RATE SDRAM), shows that there is a slower storage speed of prior art.

[0070] Combine Image 6 As shown in the present specification, since the distance dimensional Fourier leaf processing of the plurality of sampled signals is a parallel process, a plurality of distance dimension signals can be obtained, and a plurality of distance dimension signals can be fused to process, obtain Large-bit wide data (target distance dimension signal), a single package storage can store the target distance dimension signal to DDR; there is no need to cache to RAM and store it to DDR, no need to store multiple storage. It can be seen that in the present specification embodiment, only the target distance dimension signal is required to be stored once, and all of the distance dimension signals of the target sub-signal can be stored, the storage speed is fast, and the storage efficiency is high. The above manner adds storage efficiency while reducing the occupation of the cache space, and the processing of the later Doppler weigh will also accelerate the gain.

[0071] In the present specification, the plurality of sub-signals of the radar frame signal may be performed in the above steps (step S101 to S105), respectively, until the target distance dimension signal corresponding to the last sub-signal of the radar subframe signal is stored in the DDR. In the present specification embodiment, the radar subframe signal can include 128 sub-signals; the first sub-signal can represent the first sub-signal received by the radar subframe signal, and the second sub-signal can represent the radar frame signal. The second sub-signal received by the radar; the first sub-signal corresponds to the first target distance dimension signal, and the second sub-signal corresponds to the second target distance dimension signal, and so on. In DDR, 128 target distance dimension signals may be sequentially stored in the radar reception order of the corresponding sub-signal.

[0072] In a possible implementation, the method also includes:

[0073] In step S106: When the plurality of target distance dimension signals corresponding to the radar subframe signal, multiple channel signals are obtained according to the plurality of target distances corresponding to the radar subframe signal;

[0074] Step S107: Parallel Doppler Fuier leaf processing for multiple channel signals to obtain multiple Doppler signals.

[0075] In the present specification, the plurality of target distance dimension signals corresponding to the radar subframe signal are ended, and all of the distance dimension signals corresponding to the radar subframe signal have been obtained. In the present specification, a sub-signal is sampled by a multi-channel sampling channel to obtain a plurality of sampled signals, and obtains a plurality of distance dimension signals according to the plurality of sampled signals, thereby obtaining a target distance dimension signal; one target distance dimension signal corresponds to more Sample channel, one target distance dimension signal corresponds to a sub-signal. In step S106, a plurality of channel signals corresponding to a plurality of sample channels may be obtained according to a plurality of sub-signals, and a channel signal may correspond to a sampling channel; in turn, in step S107, multiple Doppler Fuier leaf treatment is performed in parallel. For example, in the radar including 4 sampling channels, a radar subframe signal includes 128 sub-signals, a target distance dimension signal can be obtained for a sub-signal, and one target distance dimension signal includes four distance dimension signals, according to The target distance dimension signal corresponding to 128 sub-signals can obtain four channel signals. Doppler Fuier leaf processing is a fast Fourier handle of Doppleri (speed dimension) for multiple channel signals.

[0076] In the present specification, multi-Doppler signals can be obtained simultaneously based on the FPGA, and multi-channel signals can be performed in parallel. The treatment method of the present specification is less time consuming, and the processing efficiency is high.

[0077] Combine Figure 7 As shown in one possible implementation, step S106 includes:

[0078] In step S1061: When the plurality of target distance dimension signal storage corresponding to the radar subframe signal, the plurality of target distance dimension signals are performed: read the target distance dimension signal, and disassemble the target distance dimension signal. A dismantling signal is obtained;

[0079] In step S1062: When the multiple target distance dimension signal is completed, a plurality of dismantling signals corresponding to a plurality of target distance dimension signals are obtained.

[0080] In the present specification, in the case where a plurality of target distance dimension signal storage corresponding to the radar frame signal, multiple target distance dimension signals are sequentially read and disassembled. For example, the first target distance dimension signal can be read, and the first target distance dimension signal is disassembled to obtain a disassembly signal corresponding to the first target distance dimension signal; then read the second target distance dimension signal, The doctrine distance dimension signal is disassembled to obtain a disassembly signal corresponding to the second target distance dimension signal; and then read the third target distance dimension signal, the third target distance dimension signal is disassembled to obtain a third target distance. Dimensional signal corresponding to the disassembly signal; in this type, the dismantling signal corresponding to the last target distance dimension signal is obtained; thereby obtaining all the disassemble signals corresponding to the radar frame signal, and then according to all disassembly signals Get multiple channel signals.

[0081] In the present specification, serial writing corresponds to serial reading from the DDR cache space. In the present specification, the multi-channel reading can be realized due to the distance dimension signal corresponding to the multi-sampling signal. When the target distance dimension signal is read, since the target distance dimension signal is incorporated herein by one of the all distance dimension signals corresponding to one sub-signal, one sub-signal corresponding to the corresponding distance dimension signal, read High efficiency, improved signal processing efficiency.

[0082] In a possible implementation, step S1062 includes a plurality of disassemble signals corresponding to a plurality of target distance dimension signals to obtain a plurality of channel signals.

[0083]In the present specification, the sorting process may be signal recombination, merge according to sub-signal numbers and sampling channel numbers; sub-signal numbers can be sequential numbers when they are received by radar; the sample channel number can be a number of radar pairs of sampling channels The plurality of dismantling signals corresponding to the same sampling channel number may be sorted in sequential sequential sequential sequence and together to obtain a channel signal corresponding to the sample channel number. After obtaining all the dismantling signals corresponding to the radar frame signal, the entire disassembly signal is sorted, and a plurality of channel signals can be obtained; or it can be sorted after obtaining a sub-signal corresponding to the disassembly signal. After the dismantling signal corresponding to the sub-signal is after the dismantling signal corresponds to the previous sub-signal; whether it is obtained, a plurality of channel signals are obtained after obtaining all the disassemble signals corresponding to the radar subframe signal. The present specification embodiment can simultaneously obtain multiple channel signals to improve signal processing efficiency.

[0084] In a possible implementation, step S107, the method further includes:

[0085] Step S108: Parallel to perform mode processing for multiple Doppler signals to obtain multiple sample results signals;

[0086] Step S109: The plurality of sample results signals are added to obtain a non-coherent accumulation signal.

[0087] In the present specification, a Doppler signal comprises a solid portion and an imaginary portion, and the mode processing is the sum of the sum of the peer squares of the Doppler signal. The accumulation method is divided into coherence and non-coherent accumulation; the coherent accumulation utilizes the phase relationship between the receiving pulses, the amplitude of the signal is superposed, and the advantage of this method is to add all radar echo energy directly; non-coherence accumulation After taking the signal envelope, the information of the complex signal is lost at this time, only the modulus is retained, and there is no strict phase relationship.

[0088] In the present specification, multiple channel signals perform Doppler Fuier leaf processing in parallel, which can be obtained simultaneously, which can be directly, and simultaneously perform moderation processing and addition processing, without the need for Doppler. The signal can be cached, which can obtain a non-coherent accumulation signal faster, and the processing efficiency is high.

[0089] In the prior art, the Radar frame signal is processed based on the MCU and other processors. When processing the data stream, it is a serial way.

[0090] The signal processing method provided in this specification embodiment can be performed based on FPGA, and the maximum advantage of FPGA is the flexibility of the algorithm adaptation. The resource particle size of FPGA is lower, and it is not like MCU to limit a FFT (Fast Fourier Transform) module. The resources contained in FPGA can constitute a variety of FFT modules, and the specific use of flexibility changes. . In the FPGA, the advantages of the device itself can be utilized, and the parallel processing / high clock operation is used to use, seeking a balance between resources and speeds. The embodiment of this specification changes the idea of ​​serial calculation in the MCU era, maximizing the advantages of parallel signal processing using FPGA, opening multiplex, using poolized resources to generate multiple FFT modules, perform parallel calculations.

[0091] In the present specification, the distance dimension is placed within a subframe intact reception, and the sub-signal is received to sample the sub-signal and the distance TiFier leaf processing, and receives a radar frame signal. After completion, the distance dimension of the radar frame signal can be completed synchronously.

[0092] In the present specification, the distance from the maximum 1024bit supported by DDR4 AXI-4 (Advanced ExtensInetFace 4) bus can be used, 32bit * 32 = 1024bit, up to 1024bit, 32 antenna The results of Vi Fu Lili sent to DDR storage, and after the entire distance Taxi Fourier handle, Doppler Weili Lee treatment was performed. When Dopplevi Fourier operation, the AXI-4 bus reads supports the maximum bit wide to 1024bit, the number is 256, and the 128 units of 1024bit bit wide can be selected for the corresponding 32. 128 Chirp data of the road antenna Doppler Fu Ri Ye Ye.

[0093] In the present specification, according to the bus characteristics of AXI-4, the data unit of the plurality of antennas is combined into a large bit width 1024bit data unit in combination with the data characteristics of the radar signal processing. The DDR buffer is written to the DDR buffer. According to the bus characteristics of AXI-4, the maximum utilization of DDR interface efficiency is achieved by multiple data units of the radar signal processing in combination with the data characteristics of the radar signal processing, and the output multi-channel antenna result is calculated in parallel.

[0094] The single subframe processing cycle is shortened, and the amount of point cloud data in unit time is large. The output target real time and reliability are improved, which is conducive to automatic driving faster execution related operations.

[0095] Combine Figure 8 As shown, the present specification embodiment also provides a signal processing apparatus, which is applied to the vehicle radar, and the device includes:

[0096] The receiving module 10 is configured to receive the target sub-signal, and the target sub-signal is any of a plurality of sub-signals included in the radar subframe signal;

[0097] Sampling module 20 for sampling the target sub-signal to obtain a plurality of sampled signals;

[0098] The distance dimensional module 30 is used to process multiple sampling signals in parallel to obtain a plurality of distance dimension signals;

[0099] The signal fusion module 40 is configured to fuse the plurality of distance dimension signals to obtain a target distance dimension signal.

[0100] In the present specification, the plurality of processing results can be obtained simultaneously based on the pool of the field programmable gate array device (FPGA, Field ProgrammableGateArray), in parallel, and simultaneously. (Multiple distance dimension signals). The use of parallel distance Vi Fu Yibi, which is time consuming, and has high processing efficiency. Further, since the plurality of distance dimension signals are simultaneously obtained, the plurality of distance dimensional signals can be fused simultaneously, and the primary signal fusion processing can obtain a target distance dimension signal; it can be seen that the processing method of the present specification embodiment is time consuming. High processing efficiency.

[0101] In one possible implementation, the distance dimensional module 30 is used to call the distance dimensional computing resource, and the plurality of sampling signals are processed by the plurality of sampled signals to obtain multiple distance dimension signals;

[0102] The apparatus also includes a release module for releasing the distance dimension computing resources after obtaining a plurality of distance dimensional signals.

[0103] In a possible implementation, the device further includes a storage module for storing the target distance dimension signal.

[0104] In a possible implementation, the device also includes:

[0105] The first determination module is configured to obtain a plurality of channel signals according to a plurality of target distance dimension signals corresponding to the radar subframe signal at the end of the plurality of target distance dimension signals corresponding to the radar subframe signal;

[0106] Dopplerwei module, used in parallel to perform Doppler Fuier leaf processing for multiple channel signals to obtain multiple Doppler signals.

[0107] In a possible implementation, the first determination module includes:

[0108] The first determination unit is configured to perform a plurality of target distance dimension signals, respectively, read the target distance dimension signal, respectively, and perform the target distance dimension signal, respectively: read the target distance dimension signal. Disassembly treatment to obtain a disassembly signal;

[0109] The second determining unit is configured to obtain a plurality of channel signals corresponding to the plurality of target distance dimension signals, depending on the plurality of target distance dimension signals.

[0110] In a possible implementation, the second determining unit is used to sequencing the plurality of disassembled signals corresponding to the plurality of target distance dimension signals to obtain a plurality of channel signals.

[0111] In a possible implementation, the device also includes:

[0112] The mode unit is used to perform moderation processing on multiple Doppler signals parallel to obtain multiple sample results signals;

[0113] The addition unit is used to add a plurality of sample result signals to obtain a non-coherent accumulation signal.

[0114] It should be noted that the apparatus provided by the above embodiment is illustrated by the division of the respective functional modules, in the actual application, can be done by different function modules as needed, and will be completed by different functional modules as needed. The internal structure of the device is divided into different functional modules to complete all or part of the above described above. Further, the apparatus and method embodiment provided by the above embodiment belongs to the same concept, and the specific implementation process is detailed in the method embodiment, and details are not described herein again.

[0115] Further, the present specification embodiment also provides a signal processing device, which is applied to the vehicle radar, including:

[0116] processor;

[0117] Memory for storing processors executable instructions;

[0118] Where the processor is configured to execute:

[0119] The target sub-signal is received, and the target sub-signal is any of a plurality of sub-signals included in the radar subframe signal;

[0120] Sampling the target sub-signal to obtain multiple sampling signals;

[0121] Parallel multi-sample signals are processed from dimensional Fourier leaf, resulting in multiple distance dimension signals;

[0122] The plurality of distance dimension signals perform signal fusion processing to obtain a target distance dimension signal.

[0123] Further, the present specification embodiment also provides a non-volatile computer readable storage medium, which stores a computer program instruction, and the computer program command is executed by the processor. The signal processing method is implemented.

[0124] The computer program product can include a computer readable storage medium that uploads a computer readable program command for enabling the processor to implement the various aspects of the present application.

[0125] The computer readable storage medium can be a tangible device that can hold and store instructions used by the instruction execution device. Computer readable storage media can be, for example, - but is not limited to, - electric storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the above. More specific examples of computer readable storage media (non-exhaustive lists) include: Portable Computer Disc, Hard Disk, Random Access Memory (RAM), read-only memory (ROM), removable programmable read-only memory (EPROM Or flash memory), static random access memory (SRAM), portable compression disk read only memory (CD-ROM), digital multi-function disk (DVD), memory stick, floppy disk, mechanical encoding device, such as the memory stored The convex structure in the hole or groove, and any suitable combination of the above. The computer readable storage medium used herein is not interpreted as an instantaneous signal itself, such as radio waves or other free propagation electromagnetic waves, electromagnetic waves propagated by waveguide or other transport medium (e.g., through the optical pulse of the fiber optic cable), or pass the wire The transmitted electrical signal.

[0126] The computer readable program instruction described herein can be downloaded from the computer readable storage medium to each computing / processing device, or over the network, such as the Internet, a local area network, a wide area network, and / or wireless network to an external computer or an external storage device. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives a computer readable program command from the network, and forwards the computer readable program command for stored in computer readable storage media in each computing / processing device. .

[0127] Computer program instructions for performing this application can be assembled instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcodes, firmware instructions, status setting data, or in one or more programming languages Any combination of source code or target code, programming language includes object-oriented programming languages, such as SmallTalk, C ++, etc., and conventional process programming languages ​​- such as "C" language or similar programming languages. Computer readable program instructions can be performed on the user's computer, partially executed on the user's computer, execute as a separate package, partially performed on the remote computer on the remote computer, or on the remote computer or server implement. In the case involving remote computers, remote computers can connect to user computers by any kind of network, including a local area network (LAN) or WAN (WAN), or can be connected to external computers (eg, using Internet service providers through the Internet via the Internet connect). In some embodiments, personalized electronic circuitry, such as a programmable logic circuit, field programmable gate array (FPGA), or programmable logic array (PLA), by using a state information of the computer readable program instruction. Perform a computer readable program instruction to implement the various aspects of the present application.

[0128] The various aspects of the present application will be described herein with reference to the flowcharts and / or block diagrams of the method, apparatus (system) and computer program products, in accordance with the present application embodiments. It should be understood that each block of the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by a computer readable program instruction.

[0129] These computer readable program instructions can provide a processor to a general purpose computer, a dedicated computer, or other programmable data processing device, thereby producing a machine such that these instructions are performed at the processor through the computer or other programmable data processing device. , A device that implements a function / action specified in the flowchart and / or block diagram in one or more box diagrams. These computer readable program instructions can also be stored in a computer readable storage medium, such instructions, such instructions, to operate in a particular manner in a particular manner, and stored instructions including A manufacturing product includes instructions that implement the functions / operations specified in the flowchart and / or block diagram.

[0130] You can also load computer readable program instructions to your computer, other programmable data processing devices, or other devices, allowing a series of steps to generate a computer implementation on a computer, other programmable data processing device, or other device to generate a computer implemented process. This makes the functions / operations specified in the flowchart and / or block diagram in the computers, other programmable data processing devices, or other devices implement flowcharts and / or one or more boxes in the block diagram.

[0131] The flowcharts and block diagrams in the drawings are shown in the system, methods, and possible architectural architecture, functions, and operations of the system, method, and computer program products in accordance with various embodiments of the present application. In this regard, each of the flowcharts or block diagrams can represent a portion of a module, block, or instruction, part of the module, block, or instruction contains one or more executable of implementing a predetermined logic function. instruction. In some implementations, the functions labeled in the box can occur in the order than those labeled in the drawings. For example, two consecutive blocks can actually be performed in parallel, and they can sometimes be performed in reverse order, which is determined according to the functions involved. It is also to be noted that each block in block diagram and / or flowchart, and a combination of block diagrams and / or box in block diagram, can be implemented with dedicated hardware-based systems that perform specified functions or motions. Or can be implemented with a combination of dedicated hardware to computer instructions.

[0132] Embodiments of the present application have been described above, and the above description is exemplary, non-exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes in the art will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The selection of the terms disclosed herein is intended to be the principles, practical applications, or techniques for techniques in the market, or other one of ordinary skill in the art will appreciate the embodiments disclosed herein.