Radar data compression method and system, electronic equipment and storage medium

Abstract

The invention discloses a radar data compression method and system, electronic equipment and a storage medium, the storage amount of radar data is increased under the condition of a certain memory, a data compression module and a data decompression module are arranged in a storage unit, distance dimension FFT (Fast Fourier Transform) and data compression are processed in parallel, the decompression process is parallel to speed dimension FFT, and the data compression efficiency is improved. The time consumed by the system in the compression and decompression processes is saved; in addition, the requirement of the system for the chirp period is lowered, configuration of waveform parameters is more flexible, and a more complex data compression algorithm can be selected to achieve a better compression effect. The total time overhead of the system is smaller, and the refresh rate of the system is improved; a data compression module and a data decompression module are placed in a storage unit, and the data compression module and the data decompression module can be reused by other processing flows.

Description

Technical Field

[0001] The present invention relates to the field of millimeter wave radar, and in particular to a radar data compression method, system, electronic device and storage medium based on near-memory computing. Background Art

[0002] Millimeter-wave radar can provide information such as the speed, angle, and distance of a target. Millimeter-waves have low atmospheric attenuation, better penetration into smoke and dust, and are less affected by weather. These characteristics determine that millimeter-wave radar has the ability to work around the clock and in all weather conditions, so it is widely used in the fields of intelligent driving and smart transportation.

[0003] With the development of intelligent driving and intelligent transportation, millimeter wave radar has higher requirements, requiring higher resolution, longer detection distance and target height information. In order to achieve higher angular resolution, more antennas are needed. In order to achieve a longer detection distance, more chirps may need to be added or the number of sampling points of each chirp (a signal whose frequency changes continuously over time within a specific frequency range, generally referring to a "linear frequency modulation signal") may need to be increased, which will cause the ADC (Analog-to-Digital Converter) data of each radar frame to become larger and larger.

[0004] The huge amount of data brings certain challenges to data transmission, processing and storage. Because the dynamic range in the radar measurement process is limited, the dynamic range between people and cars is about 30dB. Assuming that the data is stored with a bit width of 16 bits, the dynamic range that can be represented is 96dB, and there is a lot of redundancy, so data compression can be used to reduce the amount of data while maintaining the key information of the data.

[0005] Because each chirp of ADC data is processed directly, there is no need for storage, and the result of two-dimensional FFT (Fast Fourier Transform, FFT) can be selected to detect and measure the angle after each distance unit is completed, which is not necessary to store. However, no matter what signal processing architecture is selected, the result of the distance dimension FFT must be stored. Therefore, data compression in radar signal processing is generally performed after the distance FFT. The traditional radar data compression method is to complete the compression in the calculation unit and then transfer the data storage to the memory. This means that each frame of ADC data must complete FFT and data compression before processing the next frame of data. This inevitably has requirements for a single chirp period. If the chirp period is too short, the distance dimension FFT and data compression cannot be completed within a single frame period.

[0006] Therefore, there is an urgent need to improve and optimize the radar data compression method to save the system time spent in the compression and decompression process. Summary of the invention

[0007] The purpose of the present invention is to propose a radar data compression method based on near-memory computing, so that the data compression and decompression process is parallel with the distance dimension FFT and the speed dimension FFT, saving the time spent by the system in the compression and decompression process.

[0008] In order to achieve the above object, the present invention provides a radar data compression method, comprising:

[0009] Perform a first calculation process on the ADC data of the nth chirp;

[0010] The first calculation result of the nth chirp is compressed, and the first calculation process of the n+1th chirp is performed at the same time;

[0011] The second calculation process is performed on the ADC data of the nth chirp, and the data is decompressed on the first calculation result of the n+1th chirp; wherein n is a natural number greater than 0.

[0012] The ADC data refers to the data converted by an analog-to-digital converter (ADC).

[0013] Optionally, the first calculation process is distance dimension FFT, and the second calculation process is speed dimension FFT.

[0014] Optionally, the data compression and data decompression processes are performed in a storage unit.

[0015] Optionally, after the data compression, the step of packaging and storing the data after the data compression is completed is also included.

[0016] Optionally, the storage process includes transposed storage.

[0017] Optionally, the transposition operation, the data compression of the nth chirp, and the first calculation process of the n+1 chirps are performed in parallel.

[0018] The present invention also provides a radar data compression system, comprising: a radio frequency front end, a signal processing unit, a storage unit and a data processing unit;

[0019] The storage unit includes a data compression module and a data decompression module;

[0020] The RF front end is used to input the received ADC data into the signal processing unit for a first calculation, and the first calculation result is input into the data compression module for compression; while compressing, the signal processing unit performs a first calculation of the next chirp;

[0021] The data decompression module is used to decompress the compressed data and output it to the signal processing unit for a second calculation. While the signal processing unit is performing the second calculation, the data decompression module decompresses the compressed data of the next chirp;

[0022] The point cloud obtained after calculation and processing is input into the data processing unit.

[0023] Optionally, the data compression module includes a compression parameter configuration module, an encoder and a data packaging module;

[0024] The compression parameter configuration module is used to receive the input parameters and configure them to the encoder and the data packaging module;

[0025] The encoder is used to select an encoding method according to the parameters of the compression parameter configuration module to encode the input data;

[0026] The data packaging module is used to post-process the encoded code elements to determine whether the amount of encoded data meets the configured compression requirement, thereby selecting a data storage method.

[0027] Optionally, the parameters include at least one of enabling and disabling a data compression function, an encoding method, a configuration of a compression amount, and a data storage method.

[0028] The present invention also provides an electronic device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to perform the steps of the method described above.

[0029] The present invention also provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are executed.

[0030] Compared with the prior art, the scheme of the present invention provides a radar data compression method and system. By placing the data compression and decompression modules in a storage unit, the data compression and decompression process is paralleled with the distance dimension and speed dimension FFT, thereby saving the time spent by the system in the compression and decompression process; in addition, the system provided by the present invention has lower requirements on the chirp cycle, the configuration of waveform parameters is more flexible, and a more complex data compression algorithm can be selected to achieve a better compression effect; the overall time overhead of the system is smaller, which is conducive to improving the refresh rate of the system; the data compression module and the data decompression module are placed in the storage unit, and the data compression module and the data decompression module can be reused by other processing flows. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a flow chart of a radar data compression method according to an embodiment of the present invention;

[0032] Figure 2 is a block diagram of a radar data compression system in an embodiment of the present invention;

[0033] Figure 3 4 is a block diagram of a storage unit in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The present invention will be described in more detail below in conjunction with the accompanying drawings, wherein preferred embodiments of the present invention are shown, and it should be understood that those skilled in the art may modify the present invention described herein and still achieve the beneficial effects of the present invention. Therefore, the following description should be understood as being widely known to those skilled in the art and not as a limitation of the present invention.

[0035] The present invention is described in more detail in the following paragraphs by way of example with reference to the accompanying drawings. The advantages and features of the present invention will become more apparent from the following description. It should be noted that the accompanying drawings are in very simplified form and are not in exact proportions, and are only used to facilitate and clearly assist in illustrating the purpose of the embodiments of the present invention.

[0036] Embodiment 1

[0037] The signal processing unit mainly performs distance dimension FFT, velocity dimension FFT, CFAR, angle dimension FFT and other processing steps to obtain point cloud. With the development of radar, the number of channels is increasing, resulting in a huge amount of data to be processed. The transmission, processing and storage of data bring certain challenges. In order to solve this problem, a radar data compression method is proposed based on the redundancy of the dynamic range of radar data. Radar data compression can store more data in the same memory, and more data can improve radar performance. For example, increasing the number of sampling points of each chirp under a certain maximum detection distance can improve the radar's distance resolution; increasing the number of chirps under a certain speed measurement range can improve the speed resolution; designing more channels can improve the radar's angle resolution. Data compression allows us to use a smaller memory space to get better radar performance.

[0038] The existing radar data compression scheme is generally to perform data compression after the ADC data has completed the distance dimension FFT, and then transfer the data to the storage unit after the data compression is completed, and then take it out from the storage unit when performing the speed dimension FFT, and then perform decompression and subsequent processing in the calculation. Because the ADC data of each channel is immediately subjected to the distance dimension FFT, there is no storage requirement for the ADC, and the data after the two-dimensional FFT does not have to be stored (the back-end processing process can select a range gate to detect the distance, speed, angle and other information of the target), so the data compression is selected after the distance dimension FFT. The current data compression and decompression are both implemented in the calculation unit, so that the distance FFT of the next chirp must be performed after the data compression of the current chirp is completed during the distance FFT. The distance dimension FFT and data compression must be completed within a chirp cycle, which makes the setting of the chirp cycle not too small. The calculation unit needs to be decompressed before the speed dimension FFT. From the whole process, data compression, decompression and signal processing are not parallel. Overall, the system needs to spend some extra time to compress and decompress the data, which is not conducive to improving the refresh rate of the radar.

[0039] In an embodiment of the present invention, a radar data compression method based on near-memory computing is proposed. Data compression and decompression are performed in a storage unit, and data compression and decompression can be performed in parallel with the signal processing process. When compressing the data of this data block, the computing unit can perform the distance FFT of the next data block. Similarly, when performing the speed dimension FFT of this distance unit, the data of the next distance unit can be decompressed, thereby saving the system's time overhead for realizing data compression and decompression. The saved time overhead can improve the refresh rate of the system. On the other hand, a more complex data compression algorithm can be selected to achieve a better compression effect.

[0040] For details, please refer to Figure 1 The present invention provides a radar data compression method, comprising the following steps:

[0041] S1, performing a first calculation process on the ADC data of the nth chirp;

[0042] S2, compressing the first calculation result of the nth chirp, and performing the first calculation process of the n+1th chirp at the same time;

[0043] S3. Perform a second calculation process on the ADC data of the nth chirp, and decompress the first calculation result of the (n+1)th chirp; wherein n is a natural number greater than 0.

[0044] Furthermore, the first calculation process is distance dimension FFT, and the second calculation process is speed dimension FFT.

[0045] Furthermore, the data compression and data decompression processes are performed in the storage unit.

[0046] Specifically, in steps S1-S2, the radar receiver receives the ADC data of a chirp and performs FFT calculation, and then inputs the distance FFT result of this chirp into the data compression module of the storage unit. While compressing the data of the previous chirp, the radar system can perform the distance dimension FFT of the next chirp in parallel, thereby realizing parallel processing of data compression and distance FFT.

[0047] Because the distance dimension FFT and data compression are processed in parallel, different compression algorithms can be selected according to the duration of the distance dimension FFT. If the distance dimension FFT takes a long time, a more complex compression algorithm can be selected to achieve better compression effect.

[0048] Furthermore, after the data is compressed, the method further includes the steps of packaging and storing the data after the data is compressed.

[0049] Specifically, after data compression is completed, the data needs to be packaged and stored. There are two ways to store the packaged data: normal storage or transposed storage.

[0050] Among them, the advantage of transposed storage is that the transposition operation when performing velocity dimension FFT is advanced to the storage process. In this way, transposition and data compression are carried out in parallel with distance dimension FFT, saving the data reading time of subsequent velocity dimension FFT.

[0051] That is, through the steps of transposing the storage, the transposition operation, the data compression of the nth chirp, and the distance dimension FFT process of the n+1 chirps are performed in parallel.

[0052] Furthermore, when storing the distance dimension FFT results, the first step is to configure the data compression parameters, enable the data compression function, select the encoding method, configure the data compression amount, and configure the data storage format. Then, when the data is transmitted to the storage unit, the encoder will encode and compress it according to the configured parameters. Finally, the encoded code elements will be transmitted to the data packaging module. The data packaging module will determine whether the encoded data size meets the preset compression amount. If it meets the preset compression amount, the data will be directly packaged and stored. If it does not meet the preset compression amount requirements, further processing is required, and lossy compression is performed to meet the preset compression amount requirements.

[0053] Furthermore, in step S3, the first calculation result data needs to be decompressed before performing the velocity dimension FFT. The decompression process and the velocity dimension FFT are also parallel, that is, the data of the next distance unit is decompressed while the velocity dimension FFT of this distance unit is performed.

[0054] Furthermore, when performing velocity dimension FFT, the velocity dimension FFT is performed after being decompressed by the data decompression module inside the storage unit. After the velocity dimension FFT is completed, the result of the velocity dimension FFT is directly input into the CFAR detection module for CFAR detection. After the detected points are processed by angle dimension FFT, the point cloud is obtained for the subsequent data processing part.

[0055] It should be noted that, in this embodiment, when n=1, the process of decompressing the data of the first calculation result of the first chirp is not parallel; but a single step, and the decompression is performed before the second calculation process of the first chirp.

[0056] Similarly, the data compression of the first calculation result of the last chirp is also single-line, so this is already the last chirp and does not need to be parallel with the first calculation process of the next chirp.

[0057] In summary, the method provided in this embodiment performs FFT calculation of the next chirp in the computing unit while compressing the chirp data. At this time, only the distance dimension FFT needs to be completed within the chirp cycle. Placing the decompression in the storage unit can also make the decompression process and the speed dimension FFT parallel, saving the time spent by the system in the compression and decompression process; data compression increases the storage capacity of radar data (the memory size is constant), and helps the radar detection performance; this method saves the time consumed by the system for data compression and decompression, helps the system to improve the refresh rate and shorten the chirp cycle, and also provides more possibilities for the selection of data compression algorithms.

[0058] Embodiment 2

[0059] This embodiment provides a radar data compression system. Figure 2 - Figure 3 , including: a radio frequency front end, a signal processing unit, a storage unit and a data processing unit; the storage unit includes a data compression module and a data decompression module; the radio frequency front end is used to input the received ADC data into the signal processing unit for a first calculation, and the first calculation result is input into the data compression module for compression; while compressing, the signal processing unit performs a first calculation of the next chirp; the data decompression module is used to decompress the compressed data and output it to the signal processing unit for a second calculation, while the signal processing unit performs the second calculation, the data decompression module decompresses the compressed data of the next chirp; the point cloud obtained after calculation and processing is input into the data processing unit.

[0060] It should be noted that the decompression method of the decompression module is related to the encoding method and data truncation rule used, and each encoding method corresponds to a decompression method.

[0061] Furthermore, the data compression module includes a compression parameter configuration module, an encoder and a data packaging module.

[0062] The compression parameter configuration module is used to receive the input parameters and configure them to the encoder and the data packaging module.

[0063] The encoder is used to select an encoding method according to the parameters of the compression parameter configuration module to encode the input data.

[0064] In other embodiments, the encoder may adaptively select an encoding method according to the characteristics of the input data to achieve the best data compression effect. The basis for selecting the adaptive encoding method is that the encoding method with the smallest total bit width after encoding is the optimal encoding method.

[0065] The data packaging module is used to post-process the encoded codewords, determine whether the amount of encoded data meets the configured compression requirement, and select a data storage method; if it does, the compressed data is stored according to the configured data storage method; if it does not meet the compression requirement, it is further lossily compressed and then stored according to the configured storage method.

[0066] Furthermore, the parameters received by the compression parameter configuration module include at least one of enabling and disabling of a data compression function, an encoding method, configuration of a compression amount, and a data storage method.

[0067] The compression amount is defined as the ratio of the compressed data amount to the uncompressed data amount.

[0068] It should be noted that the data compression and decompression module described above is not dedicated to range-dimensional FFT data compression and decompression, but can be reused in any step of signal processing and data processing that requires data compression.

[0069] In this embodiment, data compression and decompression are placed in the storage unit, data compression and distance dimension FFT can be processed in parallel, the requirements for the chirp period are reduced, the chirp period can be set smaller, the configuration of waveform parameters is more flexible, and a more complex data compression algorithm can be selected to achieve better compression effect.

[0070] Compared with the traditional architecture, placing data compression and decompression in the storage unit reduces the overall system time overhead and helps improve the system refresh rate.

[0071] By placing data compression and decompression in the storage unit, data compression and decompression can be reused by other processing flows.

[0072] Embodiment 3

[0073] An embodiment of the present invention further provides an electronic device, which may be a server, a computer or other device.

[0074] The electronic device may include: a processor, a storage medium and a bus, the storage medium stores machine-readable instructions executable by the processor, when the electronic device is running, the processor and the storage medium communicate via the bus, the processor executes the machine-readable instructions to perform the steps described in the above embodiments. The specific implementation method and technical effect are similar and will not be repeated here.

[0075] For ease of explanation, only one processor is described in the above electronic device. However, it should be noted that in some embodiments, the electronic device in the present invention may also include multiple processors, so the steps performed by one processor described in the present invention may also be performed jointly or individually by multiple processors. For example, if the processor of the electronic device executes step A and step B, it should be understood that step A and step B may also be performed jointly by two different processors or individually in one processor. For example, the first processor executes step A, the second processor executes step B, or the first processor and the second processor execute steps A and B together, etc.

[0076] In some embodiments, the processor may include one or more processing cores (e.g., a single-core processor (S) or a multi-core processor (S)). By way of example only, the processor may include a central processing unit (CPU), an application specific integrated circuit (ASIC), an application specific instruction set processor (ASIP), a graphics processing unit (GPU), a physical processing unit (PPU), a digital signal processor (DSP), a field programmable gate array (FPGA), a programmable logic device (PLD), a controller, a microcontroller unit, a reduced instruction set computer (RISC) or a microprocessor, or any combination thereof.

[0077] Accordingly, other embodiments of the present application may also provide a computer-readable storage medium in which computer executable instructions are stored, and when the computer executable instructions are executed by the processor, the various method embodiments of the present application are implemented. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be a computer-readable instruction, a data structure, a module of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0078] In summary, the radar data compression method, system electronic device and storage medium provided by the present invention place the data compression and decompression modules in the storage unit so that the data compression and decompression process is parallel with the distance dimension and speed dimension FFT, thereby saving the time spent by the system in the compression and decompression process; in addition, the system provided by the present invention has lower requirements on the chirp cycle, the configuration of waveform parameters is more flexible, and a more complex data compression algorithm can be selected to achieve better compression effect; the overall time overhead of the system is smaller, which is conducive to improving the refresh rate of the system; the data compression module and the data decompression module are placed in the storage unit, and the data compression module and the data decompression module can be reused by other processing flows.

[0079] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A radar data compression method, characterized in that: include: Perform a first calculation process on the ADC data of the nth chirp; The first calculation result of the nth chirp is compressed, and the first calculation process of the n+1th chirp is performed at the same time; The second calculation process is performed on the ADC data of the nth chirp, and the data is decompressed on the first calculation result of the n+1th chirp; wherein n is a natural number greater than 0.

2. The radar data compression method according to claim 1, characterized in that: The first calculation process is distance dimension FFT, and the second calculation process is speed dimension FFT.

3. The radar data compression method according to claim 1, characterized in that: The data compression and data decompression processes are performed in the storage unit.

4. The radar data compression method according to claim 1, characterized in that: The data compression also includes the following steps: packaging and storing the data after the data compression is completed.

5. The radar data compression method according to claim 4, characterized in that: The storage process includes transposition storage.

6. The radar data compression method according to claim 5, characterized in that: The transposition operation, data compression of the nth chirp, and the first calculation process of the n+1 chirp are performed in parallel.

7. A radar data compression system, characterized in that: include: RF front end, signal processing unit, storage unit and data processing unit; The storage unit includes a data compression module and a data decompression module; The RF front end is used to input the received ADC data into the signal processing unit for a first calculation, and the first calculation result is input into the data compression module for compression; while compressing, the signal processing unit performs a first calculation of the next chirp; The data decompression module is used to decompress the compressed data and output it to the signal processing unit for a second calculation. While the signal processing unit is performing the second calculation, the data decompression module decompresses the compressed data of the next chirp; The point cloud obtained after calculation and processing is input into the data processing unit.

8. The radar data compression system according to claim 7, characterized in that: The data compression module includes a compression parameter configuration module, an encoder and a data packaging module; The compression parameter configuration module is used to receive the input parameters and configure them to the encoder and the data packaging module; The encoder is used to select an encoding method according to the parameters of the compression parameter configuration module to encode the input data; The data packaging module is used to post-process the encoded code elements to determine whether the amount of encoded data meets the configured compression requirement, thereby selecting a data storage method.

9. The radar data compression system according to claim 8, characterized in that: The parameters include at least one of enabling and disabling a data compression function, an encoding method, a configuration of a compression amount, and a data storage method.

10. An electronic device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of the method according to any one of claims 1 to 6.

11. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are executed.

Images