WO2019080093A1 - 雷达数据处理方法、设备及可移动平台 - Google Patents

雷达数据处理方法、设备及可移动平台

Info

Publication number
WO2019080093A1
WO2019080093A1 PCT/CN2017/108026 CN2017108026W WO2019080093A1 WO 2019080093 A1 WO2019080093 A1 WO 2019080093A1 CN 2017108026 W CN2017108026 W CN 2017108026W WO 2019080093 A1 WO2019080093 A1 WO 2019080093A1
Authority
WO
WIPO (PCT)
Prior art keywords
radar data
packet
frequency point
sorted
frequency
Prior art date
Application number
PCT/CN2017/108026
Other languages
English (en)
French (fr)
Inventor
朱磊
喻庆东
肖巍
Original Assignee
深圳市大疆创新科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to CN201780029118.9A priority Critical patent/CN109313260A/zh
Priority to PCT/CN2017/108026 priority patent/WO2019080093A1/zh
Publication of WO2019080093A1 publication Critical patent/WO2019080093A1/zh
Priority to US16/856,747 priority patent/US20200264269A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/03—Details of HF subsystems specially adapted therefor, e.g. common to transmitter and receiver
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
    • H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
    • H03M7/3068—Precoding preceding compression, e.g. Burrows-Wheeler transformation
    • H03M7/3077—Sorting
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/06—Systems determining position data of a target
    • G01S13/08—Systems for measuring distance only
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/933—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of aircraft or spacecraft
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/003—Transmission of data between radar, sonar or lidar systems and remote stations
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
    • H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
    • H03M7/40—Conversion to or from variable length codes, e.g. Shannon-Fano code, Huffman code, Morse code
    • H03M7/4031—Fixed length to variable length coding
    • H03M7/4037—Prefix coding
    • H03M7/4043—Adaptive prefix coding
    • H03M7/4068—Parameterized codes
    • H03M7/4075—Golomb codes
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
    • H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
    • H03M7/60—General implementation details not specific to a particular type of compression
    • H03M7/6064—Selection of Compressor
    • H03M7/6076—Selection between compressors of the same type
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/881—Radar or analogous systems specially adapted for specific applications for robotics
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles

Definitions

  • the embodiments of the present invention relate to the field of drones, and in particular, to a radar data processing method, device, and a movable platform.
  • a movable platform such as a drone, a movable robot, or the like is usually provided with a radar device, and the radar device can be used to detect a target object around the movable platform and detect the distance of the target object from the movable platform.
  • the radar data detected by the radar device can be processed online, or the radar data detected by the radar device can be stored online. After the drone returns to the ground, the radar data stored by the ground device to the drone is performed. analysis.
  • the radar data needs to be compressed.
  • the prior art lacks a method for efficiently compressing the radar data.
  • Embodiments of the present invention provide a radar data processing method, device, and a movable platform to implement efficient compression of radar data.
  • a first aspect of the embodiments of the present invention provides a radar data processing method, including:
  • each radar data in each group is encoded to obtain encoded radar data.
  • a second aspect of the embodiments of the present invention provides a radar data processing device, including: a processor;
  • the processor is used to:
  • each radar data in each group is encoded to obtain encoded radar data.
  • a third aspect of the embodiments of the present invention provides a mobile platform, including:
  • a power system mounted to the fuselage for providing mobile power
  • a radar device for detecting a target object around the movable platform
  • a radar data processing device provided by the second aspect.
  • the radar data processing method, device and mobile platform provided by this embodiment determine the coding parameters of each packet according to at least one radar data in each group by grouping the radar data to be compressed, and according to each grouping
  • the coding parameters encode the radar data in each group to obtain the encoded radar data, and encode the radar data by means of grouping, which improves the coding efficiency of each group and realizes the efficiency of the radar data. compression.
  • FIG. 1 is a flowchart of a method for processing radar data according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of an encoding storage process according to an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of a decoding process according to an embodiment of the present invention.
  • FIG. 4 is a flowchart of a method for processing radar data according to another embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a ZigZag scan according to an embodiment of the present invention.
  • FIG. 6 is a schematic diagram of a matrix of ZigZag scanning according to an embodiment of the present invention.
  • FIG. 7 is a flowchart of a method for processing radar data according to another embodiment of the present invention.
  • FIG. 8 is a structural diagram of a radar data processing device according to an embodiment of the present invention.
  • FIG. 9 is a structural diagram of a drone according to an embodiment of the present invention.
  • a component when referred to as being "fixed” to another component, it can be directly on the other component or the component can be present. When a component is considered to "connect” another component, it can be directly connected to another component or possibly a central component.
  • FIG. 1 is a flowchart of a method for processing radar data according to an embodiment of the present invention. As shown in FIG. 1, the method in this embodiment may include:
  • Step S101 grouping the radar data to be compressed.
  • mobile platforms such as drones, mobile robots, vehicles, and the like are provided with radar devices that can detect obstacles around the movable platform.
  • the radar device can detect that the obstacles are relatively movable. The position, movement speed, posture, etc. of the platform.
  • the mobile platform performs obstacle avoidance and route planning based on the radar data detected by the radar equipment.
  • the drone when the drone is in flight, the drone can perform online processing or online analysis on the radar data detected by the radar device, and can also store the radar data detected by the radar device online, and wait until When the drone returns to the ground, the radar data stored by the drone is processed or analyzed offline by the ground end equipment.
  • the radar data stored online can also be sent to the ground end device, so that the ground end device processes or analyzes the radar data stored by the UAV online. After that, the processed or analyzed result is returned to the drone.
  • the radar data detected by the radar device may be large, in order to save the storage space of the drone, when the drone stores the radar data online, the radar data needs to be compressed.
  • the radar data to be compressed can be grouped first.
  • the radar device installed on the UAV is a millimeter wave radar device, and may be other types of radar devices, which are not specifically limited in this embodiment.
  • the UAV can compress and store the original radar data measured by the millimeter wave radar equipment. It can also perform de-correlation transformation on the original radar data measured by the millimeter wave radar equipment to obtain the de-correlated transformed radar data. The radar data after the correlation transformation is compressed and stored.
  • the UAV can perform original two-dimensional Fast Fourier Transform (FFT) or Discrete Cosine Transform (DCT) transformation methods on the original radar data measured by the millimeter wave radar device to obtain the original radar data.
  • FFT Fast Fourier Transform
  • DCT Discrete Cosine Transform
  • the frequency point coefficient is then compressed and stored for the frequency point coefficient of the original radar data.
  • the drone can segment the original radar data according to the scale of the two-dimensional FFT.
  • the scale of the two-dimensional FFT is M*N, then no one
  • the machine can block the original radar data according to the size of M*N, and perform two-dimensional FFT in units of original radar data of M*N size.
  • the original radar data is time domain information, and the original radar data passes through two-dimensional.
  • the FFT will be converted into frequency domain information, and the M*N original radar data will be converted into M*N frequency point coefficients after performing two-dimensional FFT.
  • the drone compresses M*N frequency point coefficients in units of M*N frequency point coefficients.
  • the drone uses the M*N frequency point coefficients as the radar data to be compressed, and groups the M*N frequency point coefficients.
  • This embodiment does not limit a specific grouping method. After grouping, each group includes at least one frequency point coefficient.
  • the frequency point coefficients included in each group may be the same or different.
  • Step S102 Determine, according to at least one radar data in each group, an encoding parameter of each packet.
  • the UAV After the UAV groups the M*N frequency point coefficients, it is necessary to encode the frequency point coefficients in each packet. Before encoding, it is necessary to determine the coding parameters of each packet. This embodiment does not limit the specific coding mode.
  • the UAV encodes the frequency point coefficients in each group by using the K-th order Columbus coding method. Before performing the K-th order Columbus coding on the frequency point coefficients in each group, it is necessary to determine the coding parameters of each group. For example K parameters.
  • the drone determines the K parameter corresponding to each group when performing K-th order Columbus encoding for each packet according to each frequency point coefficient in each group.
  • Step S103 Encode each radar data in each group according to coding parameters of each group to obtain coded radar data.
  • the encoded radar data includes: a K parameter corresponding to each packet, and a K-th Columbus-encoded codeword for each frequency point coefficient in each packet.
  • a mobile platform such as a drone, a mobile robot, a car, etc.
  • the recorder 20 includes a radar data encoder 21 and a storage device 22, and a radar data encoder 21
  • the input is radar data
  • the radar data may be the frequency point coefficient after the FFT of the original radar data measured by the millimeter wave radar device described in the above step, or may be the original radar data.
  • the radar data encoder 21 encodes the frequency point coefficients or the original radar data to obtain the encoded radar data, and stores the encoded radar data in the storage device 22.
  • the mobile platform can perform wired communication or wireless communication with the ground offline analysis device, either after the mobile platform stops working or stops moving, or when the mobile platform is working or moving.
  • 30 denotes a parser of the ground offline analysis device, and the parser 30 includes a storage device 31 and a radar data decoder 32.
  • the mobile platform is specifically a drone, and when the drone returns to the ground, it performs wired communication or wireless communication with the ground offline analysis device, so that the drone will store the code stored in the storage device 22 as shown in FIG. 2.
  • the post radar data is sent to the storage device 31 of the ground offline analysis device, and the radar data decoder 32 further obtains the encoded radar data from the storage device 31 and decodes the encoded radar data.
  • the radar data is recovered, so that the ground offline analysis device analyzes or processes the radar data solved by the radar data decoder 32. This embodiment does not limit the analysis or processing process of the radar data by the ground offline analysis device.
  • the coding parameters of each packet are determined according to at least one radar data in each packet, and each radar data in each packet is determined according to the coding parameters of each packet.
  • the coding is performed to obtain the encoded radar data, and the radar data is encoded by grouping, which improves the coding efficiency of each group and realizes efficient compression of the radar data.
  • FIG. 4 is a flowchart of a method for processing radar data according to another embodiment of the present invention. As shown in FIG. 4, on the basis of the embodiment shown in FIG. 1, the method in this embodiment may include:
  • Step S401 Perform decorrelation transformation on the original radar data measured by the radar device to obtain de-correlated transformed radar data.
  • the decorrelation transform may specifically be a two-dimensional FFT, and the radar data after the decorrelation may specifically be a frequency point coefficient after the two-dimensional FFT.
  • the decorrelation transform may also be other than a two-dimensional FFT.
  • a two-dimensional FFT is performed on the original radar data measured by the radar device to obtain a frequency point coefficient of the original radar data.
  • Step S402 Determine, according to the de-correlated transformed radar data, the radar data to be compressed.
  • Determining, according to the de-correlated transformed radar data, the radar data to be compressed includes the following feasible implementation manners:
  • a feasible implementation manner is: using the decorrelated transformed radar data as the radar data to be compressed. For example, after performing two-dimensional FFT on the original radar data measured by the radar device, the frequency point coefficient of the original radar data is used as the radar data to be compressed.
  • Another feasible implementation manner is: sorting the de-correlated transformed radar data to obtain sorted radar data; and using the sorted radar data as the radar data to be compressed. For example, after performing two-dimensional FFT on the original radar data measured by the radar device, the frequency point coefficients of the original radar data are sorted to obtain the sorted frequency point coefficients, and further The sorted frequency point coefficients are used as the radar data to be compressed.
  • the step of sorting the de-correlated transformed radar data to obtain the sorted radar data includes: de-correlated transformed radar data according to the frequency of the de-correlated transformed radar data Sort and get the sorted radar data.
  • the frequency point coefficients may be sorted according to the frequency.
  • the frequency point coefficients may be sorted in order from low frequency to high frequency, or from high frequency to The order of the low frequencies sorts the frequency point coefficients.
  • the step of sorting the de-correlated transformed radar data to obtain the sorted radar data includes: sorting the de-correlated transformed radar data in order from low frequency to high frequency, and obtaining the sorting After the radar data. It is assumed that the energy amplitudes of adjacent frequency points of the radar data are close and correlated. After sorting the frequency point coefficients in order from low frequency to high frequency, it is equivalent to sorting the energy amplitudes of adjacent frequency points of the radar data. Usually, the energy amplitude of the mid-low frequency is larger than the energy amplitude of the high frequency.
  • the step of sorting the de-correlated transformed radar data in order from low frequency to high frequency comprises: performing ZigZag scanning on the de-correlated transformed radar data in order from low frequency to high frequency.
  • ZigZag scanning can be specifically performed on the frequency point coefficients in order from low frequency to high frequency.
  • 50 denotes a 4*4 matrix, which is only illustrative here, and does not limit the size of the matrix.
  • the matrix 50 may be scanned according to the arrow shown in FIG. 5. For example, if the matrix 50 is specifically a matrix as shown in FIG.
  • the scanning sequence obtained after performing ZigZag scanning on the matrix 50 is 1, 5, 3, 9, 7, 3, 9, 5, 4, 7, 3, 6, 6, 4, 1, 3.
  • the elements of the first row and the first column of the matrix 50 are 1, the elements of the first row and the second column are 5, the elements of the first column of the second row are 3, and the first row of the matrix 50 is The element 1 of the first column, the element 5 of the first row and the second column, and the element 3 of the second row and the first column are adjacent to each other.
  • the ZigZag scan of the matrix 50 the element 1 of the first row and the first column
  • the element 5 of the second column of the row and the element 3 of the first column of the second row are also adjacent in the order of arrangement.
  • the ZigZag scan is performed on the frequency point coefficients in order from low frequency to high frequency, so that the adjacent frequency points are adjacent to each other after scanning.
  • Step S403 grouping the radar data to be compressed.
  • the frequency point coefficients with similar frequencies can be divided into one group.
  • the number of frequency point coefficients in each packet may be equal or unequal.
  • Step S404 Quantize each radar data in each group according to a quantization step size corresponding to each group.
  • the frequency point coefficients in each group can be quantized, and each group can correspond to one quantization step.
  • the quantization step size corresponding to each group can be equal or not. equal.
  • the quantization steps of some radar data may also be unequal.
  • Step S405 Determine coding parameters of each packet according to at least one radar data in each packet.
  • Step S405 is similar to the specific principle and implementation manner of step S102, and details are not described herein again.
  • Step S406 Encode each radar data in each packet according to the coding parameters of each packet to obtain encoded radar data.
  • Step S406 is similar to the specific principle and implementation manner of step S103, and details are not described herein again.
  • Step S407 storing the encoded radar data.
  • the de-correlation transform is performed on the original radar data measured by the radar device to obtain the de-correlated transformed radar data, and the de-correlated transformed radar data is compressed, thereby further improving the compression efficiency of the radar data.
  • FIG. 7 is a flowchart of a method for processing radar data according to another embodiment of the present invention. As shown in FIG. 7, the method in this embodiment may include:
  • Step S701 performing two-dimensional FFT on the original radar data measured by the radar device to obtain a frequency point coefficient of the original radar data.
  • Step S702 Determine a frequency point coefficient to be compressed according to a frequency point coefficient of the original radar data.
  • determining the frequency point coefficient to be compressed according to the frequency point coefficient of the original radar data includes: sorting frequency point coefficients of the original radar data to obtain a sorted frequency point coefficient; The following frequency point coefficient is used as the frequency point coefficient to be compressed. For example, sorting the frequency point coefficients according to the frequency, optionally, from low frequency to The order of the high frequency is used to sort the frequency point coefficients, or the frequency point coefficients are sorted in order from high frequency to low frequency.
  • the frequency point coefficients of the original radar data are sorted to obtain the sorted frequency point coefficients, and the frequency point coefficients of the original radar data are sorted in order from low frequency to high frequency, The frequency point coefficient after sorting. It is assumed that the energy amplitudes of adjacent frequency points of the radar data are close and correlated. After sorting the frequency point coefficients in order from low frequency to high frequency, it is equivalent to sorting the energy amplitudes of adjacent frequency points of the radar data.
  • the frequency point coefficients of the original radar data are sorted in order from low frequency to high frequency, including: performing ZigZag scanning on the frequency point coefficients of the original radar data in order from low frequency to high frequency. For example, when the frequency point coefficients are sorted in order from low frequency to high frequency, ZigZag scanning may be performed on the frequency point coefficients in order from low frequency to high frequency, so that the adjacent frequency points are sorted after ZigZag scanning. Also adjacent, the coefficients of the near frequency points are arranged on adjacent scanning orders.
  • Step S703 grouping the sorted frequency point coefficients, each packet including at least one frequency point coefficient.
  • the purpose of grouping the sorted frequency point coefficients is to group frequency points close to the energy amplitude into a group.
  • the number of frequency point coefficients included in each packet is equal; or the number of frequency point coefficients included in each packet is not equal.
  • the grouping method for grouping the sorted frequency point coefficients may be various, and is not specifically limited in this embodiment.
  • Step S704 Quantize each frequency point coefficient in each group according to the quantization step size corresponding to each group.
  • the quantization step size corresponding to each packet is not equal; or the quantization step size corresponding to each packet is equal.
  • the quantization step sizes of the partial frequency point coefficients in the packet are not equal.
  • the quantization step size corresponding to each packet is equal to Q, assuming that the frequency point coefficient before quantization is F, and the frequency point coefficient after quantization is recorded as FQ.
  • the Q value can control the quantization error.
  • Q it is lossless coding (meaning that the decoder can reconstruct the same data as the original data); when Q is not equal to 1, it is lossy coding.
  • the compression ratio of a typical scene is 30% to 50%, saving 70% to 50% of storage space, or the length of recording time.
  • the doubling of the quantization step size increases the average compression ratio by 1 bit; For example, the savings are 1/16 on the original basis.
  • Step S705 Determine, according to each frequency point coefficient in each group, a K parameter corresponding to each group when performing K-th order Columbus encoding for each group. Specifically, the determining, according to each frequency point coefficient in each group, a K parameter corresponding to each group when performing K-th order Columbus encoding for each group, including: determining, according to each frequency point coefficient in the group An estimated value of the K parameter corresponding to each frequency point coefficient is respectively determined; and the K parameter corresponding to the group is determined according to an estimated value of the K parameter corresponding to each frequency point coefficient in the group.
  • the frequency point coefficient before quantization in a certain group is F
  • the frequency point coefficient after quantization is FQ
  • the optimal K parameter corresponding to the frequency point coefficient FQ after quantization is determined according to the following judgment condition, and the optimal K parameter corresponding to the FQ is assumed. Recorded as Ke.
  • a group includes four frequency point coefficients of F1, F2, F3, and F4.
  • the optimal K parameters corresponding to F1, F2, F3, and F4 are respectively determined, and it is assumed that the optimal K parameter corresponding to F1 is the optimal K parameter corresponding to Ke1 and F2, and the optimal K parameter corresponding to Ke2 and F3 is The optimal K parameter corresponding to Ke3 and F4 is Ke4.
  • the K parameters corresponding to the group are calculated according to the optimal K parameters corresponding to F1, F2, F3, and F4, namely, Ke1, Ke2, Ke3, and Ke4.
  • the specific calculation manner is not limited herein.
  • the calculation method of the K parameter corresponding to other groups is the same as this, and will not be repeated here.
  • the estimated values are statistically averaged to obtain the K parameters corresponding to the packets.
  • each frequency point coefficient after quantization in the same group corresponds to an optimal K parameter, that is, Ke, and statistically averages Ke corresponding to each frequency point coefficient after quantization in the same group to obtain the group.
  • K parameters corresponding K parameters.
  • the above packet Includes F1, F2, F3, and F4.
  • the optimal K parameters corresponding to F1, F2, F3, and F4 are Ke1, Ke2, Ke3, and Ke4, respectively.
  • One way to calculate the K parameter corresponding to the group according to Ke1, Ke2, Ke3, and Ke4 is: to Ke1 Ke2, Ke3, and Ke4 perform statistical averaging, and the average value obtained is the K parameter corresponding to the group.
  • the calculation of the K parameters corresponding to other groups is similar to this, and will not be repeated here.
  • the K parameters corresponding to different groups may be the same or different.
  • Step S706 Perform K-th order Columbus coding on each frequency point coefficient in each group according to the K parameter corresponding to each group.
  • the result of K-th order Columbus encoding for FQ includes two code words, one code word is recorded as a, and the other code word is recorded as b.
  • code word a is before code word b, that is, K-order Columbus
  • the coded FQ is ab, the codeword length of codeword a is (Me/2 K )+1, the codeword value of codeword a is 1; the codeword length of codeword b is K, and the codeword of codeword b The value is Me&(2 K -1).
  • the K-code Columbus encoded FQ is 00111.
  • the process of encoding the frequency point coefficients after other quantization is similar to this, and is not enumerated here.
  • the M*N frequency point coefficients are divided into H groups, and the quantization step size corresponding to each group is Q, the K parameter corresponding to the first group is specifically K1, and the K parameter corresponding to the second group is specifically K2.
  • the quantization step size Q corresponding to each packet may be sequentially encoded, for the first The K parameter K1 corresponding to the packet is encoded, and the quantized frequency point coefficients in the first packet are encoded, and the K parameter K2 corresponding to the second packet is encoded, and the quantized frequency points in the second packet are encoded.
  • the encoded radar data includes: a quantization step size corresponding to each packet, a K parameter corresponding to each packet, and a K-th order Columbus-encoded codeword for each frequency point coefficient in each packet.
  • the encoded radar data is The volume storage format may be: the encoding result of Q, K1, K1, the encoding result of K2, K2, the encoding result of KH, KH.
  • the specific storage format of the radar data may be: the encoding result of Q1, K1, K1, the encoding result of Q2, K2, K2, the encoding result of QH, KH, KH.
  • the unmanned aerial vehicle is the transmitting end of the encoded radar data
  • the ground end device is the receiving end
  • the preset grouping method used by the drone to group the sorted frequency point coefficients may be a plurality of preset grouping methods.
  • the preset grouping method adopted by the drone is not a grouping method agreed by the drone and the ground end device, when the drone encodes the radar data, the preset grouping for the drone is also required.
  • the identification information of the method is encoded, that is, the encoded radar data further includes: identification information of a preset grouping method used when grouping the sorted frequency point coefficients.
  • the preset grouping method adopted by the drone can be parsed, so that the ground station device adopts the same grouping method for decoding.
  • the radar data is grouped. If the preset grouping method adopted by the drone is a grouping method agreed by the drone and the ground end device, the identification information of the preset grouping method adopted by the drone is not required to be encoded.
  • Step S707 storing the encoded radar data.
  • the UAV further encodes the encoded radar data after encoding each packet, and realizes compression storage of the radar data.
  • the frequency point coefficient of the original radar data is obtained by performing two-dimensional FFT on the original radar data measured by the radar device, and the frequency points of the energy amplitudes are grouped into a group after the frequency points are sorted and grouped. After grouping, each group shares a K value for displaying expression, further quantizes each frequency point coefficient in each group, and performs K-th order Columbus encoding for each quantized frequency point coefficient in each group, thereby improving the radar data. Compression efficiency.
  • FIG. 8 is a structural diagram of a radar data processing device according to an embodiment of the present invention.
  • the radar data processing device 80 includes: a processor 81.
  • the processor 81 is configured to: group the radar data to be compressed. Determining the encoding parameters of each packet based on at least one radar data in each packet; The coding parameters of the group encode each radar data in each group to obtain the encoded radar data.
  • the coding parameters of each packet are determined according to at least one radar data in each packet, and each radar data in each packet is determined according to the coding parameters of each packet.
  • the coding is performed to obtain the encoded radar data, and the radar data is encoded by grouping, which improves the coding efficiency of each group and realizes efficient compression of the radar data.
  • Embodiments of the present invention provide a radar data processing device.
  • the processor 81 encodes each radar data in each packet according to the coding parameters of each packet, and obtains the encoded radar data. Also used to: store the encoded radar data to a memory.
  • the method further comprises: performing decorrelation transformation on the original radar data measured by the radar device to obtain de-correlated transformed radar data; and performing the decorrelation transformation according to the correlation.
  • the radar data after the determination determines the radar data to be compressed.
  • the determining, by the processor 81, the radar data to be compressed according to the de-correlated transformed radar data specifically, using the de-correlated transformed radar data as the radar data to be compressed.
  • the processor 81 determines, according to the de-correlated transformed radar data, the radar data to be compressed, specifically: sorting the de-correlated transformed radar data to obtain a sorted radar. Data; the sorted radar data is used as the radar data to be compressed.
  • the processor 81 sorts the de-correlated transformed radar data to obtain the sorted radar data, and is specifically used to: perform the decorrelation according to the frequency of the de-correlated transformed radar data.
  • the transformed radar data is sorted to obtain the sorted radar data.
  • the processor 81 sorts the de-correlated transformed radar data to obtain the sorted radar data, where the radar data is used to sequence the de-correlated transformed radar data from low frequency to high frequency. Sort and get the sorted radar data.
  • the processor 81 sorts the de-correlated transformed radar data in order from low frequency to high frequency, specifically, the following is: using the de-correlated transformed radar data from low frequency to high frequency. Perform a ZigZag scan in sequence.
  • the de-correlation transform is performed on the original radar data measured by the radar device to obtain the de-correlated transformed radar data, and the de-correlated transformed radar data is compressed, thereby further improving the compression efficiency of the radar data.
  • Embodiments of the present invention provide a radar data processing device.
  • the processor 81 performs de-correlation transformation on the original radar data measured by the radar device to obtain radar data after decorrelation, and is specifically used for: measuring the radar device.
  • the original radar data is subjected to two-dimensional FFT to obtain a frequency point coefficient of the original radar data.
  • the processor 81 when determining the radar data to be compressed according to the de-correlated transformed radar data, is specifically configured to: determine a frequency point coefficient to be compressed according to a frequency point coefficient of the original radar data.
  • the processor 81 is specifically configured to: sort the frequency point coefficients of the original radar data to obtain the sorted frequency points. a coefficient; the sorted frequency point coefficient is used as the frequency point coefficient to be compressed.
  • the processor 81 sorts the frequency point coefficients of the original radar data to obtain the sorted frequency point coefficients, and is specifically used to: perform frequency point coefficients on the original radar data according to low frequency to high frequency. Sort in order to get the sorted frequency point coefficients.
  • the processor 81 sorts the frequency point coefficients of the original radar data in order from low frequency to high frequency, specifically, the frequency point coefficient of the original radar data is from low frequency to high frequency. Perform a ZigZag scan in sequence.
  • the method is specifically configured to: group the sorted frequency point coefficients, and each packet includes at least one frequency point coefficient.
  • the number of frequency point coefficients included in each packet is equal; or the number of frequency point coefficients included in each packet is not equal.
  • the method when determining, by the processor 81, the coding parameters of each packet according to the at least one radar data in each packet, the method is specifically configured to: determine, according to each frequency point coefficient in each packet, a K-order for each packet.
  • the processor 81 determines, according to each frequency point coefficient in each packet, a K parameter corresponding to each packet when the K-th order Columbus encoding is performed for each packet, specifically, according to each frequency in the packet.
  • the point coefficient determines an estimated value of the K parameter corresponding to each of the frequency point coefficients, and determines the K parameter corresponding to the group according to an estimated value of the K parameter corresponding to each frequency point coefficient in the group.
  • the processor 81 is configured to: determine, according to the estimated value of the K parameter corresponding to each frequency point coefficient in the packet, the K parameter corresponding to the packet, specifically: The estimated values of the corresponding K parameters are statistically averaged to obtain the K parameters corresponding to the packets.
  • the processor 81 when the processor 81 encodes each radar data in each packet according to the coding parameter of each packet, the processor 81 is specifically configured to: according to the K parameter corresponding to each packet, each frequency point in each packet The coefficient is K-order Columbus coded.
  • the encoded radar data includes: a K parameter corresponding to each packet, and a K-th Columbus-encoded codeword for each frequency point coefficient in each packet.
  • the processor 81 before encoding each radar data in each group according to the encoding parameter of each packet, is further configured to: for each frequency in each group according to a quantization step size corresponding to each packet The point coefficients are quantized.
  • the quantization step size corresponding to each packet is not equal; or the quantization step size corresponding to each packet is equal.
  • the quantization step sizes of the partial frequency point coefficients in the packet are not equal.
  • the encoded radar data includes: a quantization step size corresponding to each packet, a K parameter corresponding to each packet, and a K-th order Columbus encoded codeword for each frequency point coefficient in each packet. .
  • the encoded radar data further includes: identifier information of a preset grouping method used when grouping the sorted frequency point coefficients.
  • the frequency point coefficient of the original radar data is obtained by performing two-dimensional FFT on the original radar data measured by the radar device, and the frequency points of the energy amplitudes are grouped into a group after the frequency points are sorted and grouped. After grouping, each group shares a K value for displaying expression, further quantizes each frequency point coefficient in each group, and performs K-th order Columbus encoding for each quantized frequency point coefficient in each group, thereby improving the radar data. Compression efficiency.
  • Embodiments of the present invention provide a mobile platform.
  • the movable platform includes: a fuselage; a power system mounted on the body for providing mobile power; a radar device for detecting a target object around the movable platform; and radar data processing described in the above embodiment device.
  • the movable platform includes at least one of the following: a drone, a movable robot, and a vehicle.
  • FIG. 9 is a structural diagram of a drone according to an embodiment of the present invention.
  • the drone 900 includes: a fuselage, a power system, a radar device 910, a radar data processing device 908, and a flight controller 918.
  • the power system includes at least one of: a motor 907, a propeller 906, and an electronic governor 917, the power system is mounted to the airframe for providing flight power; and the flight controller 918 is communicatively coupled to the power system for Controlling the drone flight.
  • the specific principles and implementation manners of the radar data processing device 908 are similar to the foregoing embodiments, and are not described herein again.
  • the coding parameters of each packet are determined according to at least one radar data in each packet, and each radar data in each packet is determined according to the coding parameters of each packet.
  • the coding is performed to obtain the encoded radar data, and the radar data is encoded by grouping, which improves the coding efficiency of each group and realizes efficient compression of the radar data.
  • the disclosed apparatus and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • Another point, displayed or The mutual coupling or direct coupling or communication connection discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
  • the above-described integrated unit implemented in the form of a software functional unit can be stored in a computer readable storage medium.
  • the above software functional unit is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform the methods of the various embodiments of the present invention. Part of the steps.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, which can store program codes. .

Abstract

一种雷达数据处理方法，包括：将待压缩的雷达数据进行分组；根据每个分组中的至少一个雷达数据，确定每个分组的编码参数；根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据，如此通过分组的方式对雷达数据进行编码，提高了对每个分组的编码效率，实现了对雷达数据的高效压缩。另外还涉及一种实施上述雷达数据处理方法的设备，以及一种搭载该设备的可移动平台。

Description

雷达数据处理方法、设备及可移动平台 技术领域
本发明实施例涉及无人机领域，尤其涉及一种雷达数据处理方法、设备及可移动平台。
背景技术
现有技术中可移动平台例如无人机、可移动机器人等通常设置有雷达设备，雷达设备可用于探测可移动平台周围的目标物体，并探测目标物体距离可移动平台的远近。
无人机在飞行时可以对雷达设备探测的雷达数据进行在线处理，或者对雷达设备探测的雷达数据进行在线存储，等无人机返回地面后，由地面设备对无人机存储的雷达数据进行分析。
当可移动平台在线存储雷达数据时，需要对雷达数据进行压缩，但是，现有技术中缺乏对雷达数据进行高效压缩的方法。
发明内容
本发明实施例提供一种雷达数据处理方法、设备及可移动平台，以实现对雷达数据进行高效压缩。
本发明实施例的第一方面是提供一种雷达数据处理方法，包括：
将待压缩的雷达数据进行分组；
根据每个分组中的至少一个雷达数据，确定每个分组的编码参数；
根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据。
本发明实施例的第二方面是提供一种雷达数据处理设备，包括：处理器；
所述处理器用于：
将待压缩的雷达数据进行分组；
根据每个分组中的至少一个雷达数据，确定每个分组的编码参数；
根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据。
本发明实施例的第三方面是提供一种可移动平台，包括：
机身；
动力系统，安装在所述机身，用于提供移动动力；
雷达设备，用于检测所述可移动平台周围的目标物体；以及
第二方面提供的雷达数据处理设备。
本实施例提供的雷达数据处理方法、设备及可移动平台，通过将待压缩的雷达数据进行分组，根据每个分组中的至少一个雷达数据，确定每个分组的编码参数，并根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据，通过分组的方式对雷达数据进行编码，提高了对每个分组的编码效率，实现了对雷达数据的高效压缩。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为本发明实施例提供的雷达数据处理方法的流程图；
图2为本发明实施例提供的编码存储过程的示意图；
图3为本发明实施例提供的解码过程的示意图；
图4为本发明另一实施例提供的雷达数据处理方法的流程图；
图5为本发明实施例提供的ZigZag扫描的示意图；
图6为本发明实施例提供的ZigZag扫描的矩阵的示意图；
图7为本发明另一实施例提供的雷达数据处理方法的流程图；
图8为本发明实施例提供的雷达数据处理设备的结构图；
图9为本发明实施例提供的无人机的结构图。
附图标记：
20-录制器   21-雷达数据编码器   22-存储设备
30-解析器   31-存储设备    32-雷达数据解码器
50-矩阵        80-雷达数据处理设备
81-处理器      900-无人机
910-雷达设备    908-雷达数据处理设备
918-飞行控制器   907-电机       906-螺旋桨
917-电子调速器
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
需要说明的是，当组件被称为“固定于”另一个组件，它可以直接在另一个组件上或者也可以存在居中的组件。当一个组件被认为是“连接”另一个组件，它可以是直接连接到另一个组件或者可能同时存在居中组件。
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本发明。本文所使用的术语“及/或”包括一个或多个相关的所列项目的任意的和所有的组合。
下面结合附图，对本发明的一些实施方式作详细说明。在不冲突的情况下，下述的实施例及实施例中的特征可以相互组合。
本发明实施例提供一种雷达数据处理方法。图1为本发明实施例提供的雷达数据处理方法的流程图。如图1所示，本实施例中的方法，可以包括：
步骤S101、将待压缩的雷达数据进行分组。
通常情况下，可移动平台例如无人机、可移动机器人、交通工具等设备均设置有雷达设备，该雷达设备可以探测可移动平台周围的障碍物，例如，雷达设备可探测障碍物相对可移动平台的位置、移动速度、姿态等。可移动平台根据雷达设备探测到的雷达数据进行避障、航线规划等。
以无人机为例，当无人机处于飞行状态时，无人机可以对雷达设备探测到的雷达数据进行在线处理或在线分析，也可以对雷达设备探测到的雷达数据进行在线存储，等到无人机返回地面时，由地面端设备对无人机存储的雷达数据进行离线处理或分析。另外，无人机对雷达设备探测到的雷达数据进行在线存储之后，还可以将在线存储的雷达数据发送给地面端设备，以使地面端设备对无人机在线存储的雷达数据进行处理或分析后，将处理或分析后的结果返回给无人机。
由于雷达设备探测到的雷达数据的数据量可能会很大，为了节省无人机的存储空间，当无人机在线存储雷达数据时，需要对雷达数据进行压缩。
当无人机需要对雷达数据进行压缩时，可以先将待压缩的雷达数据进行分组。可选的，无人机上安装的雷达设备为毫米波雷达设备，还可以是其他类型的雷达设备，本实施例不作具体的限定。无人机可以对毫米波雷达设备测出的原始雷达数据进行压缩后存储，也可以先对毫米波雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据，再对去相关变换后的雷达数据进行压缩后存储。
例如，无人机可以对毫米波雷达设备测出的原始雷达数据进行二维快速傅立叶变换(Fast Fourier Transform，简称FFT)或者离散余弦变换(Discrete Cosine Transform，简称DCT)等变换方法得到原始雷达数据的频率点系数，再对原始雷达数据的频率点系数进行压缩后存储。
假设毫米波雷达设备测出的原始雷达数据的数据量较大，无人机可以根据二维FFT的尺度对原始雷达数据进行分块，例如，二维FFT的尺度为M*N，则无人机可按照M*N的大小对原始雷达数据进行分块，并以M*N大小的原始雷达数据为单位进行二维FFT，可以理解，原始雷达数据为时域信息，原始雷达数据经过二维FFT之后将转换成频域信息，M*N个原始雷达数据进行二维FFT之后将转换成M*N个频率点系数。进一步的，无人机以M*N个频率点系数为单位，对M*N个频率点系数进行压缩。具体的，无人机将M*N个频率点系数作为待压缩的雷达数据，并对M*N个频率点系数进行分组。本实施例不限定具体的分组方法，分组之后，每组包括至少一个频率点系数，另外，每组包括的频率点系数可以相同，也可以不同。
步骤S102、根据每个分组中的至少一个雷达数据，确定每个分组的编码参数。
无人机对M*N个频率点系数进行分组之后，需要对每个分组中的频率点系数进行编码，在编码之前，需要确定每个分组的编码参数。本实施例不限定具体的编码方式。可选的，无人机采用K阶哥伦布编码方式对每个分组中的频率点系数进行编码，在对每个分组中的频率点系数进行K阶哥伦布编码之前，需要确定每个分组的编码参数例如K参数。
具体的，无人机根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数。
步骤S103、根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据。
无人机确定出每个分组对应的K参数后，根据每个分组对应的K参数，对每个分组中的各频率点系数进行K阶哥伦布编码。所述编码后的雷达数据包括：每个分组对应的K参数、以及对每个分组中的各频率点系数进行K阶哥伦布编码后的码字。
不是一般性，可移动平台例如无人机、可移动机器人、汽车等设备设置有如图2所示的录制器20，录制器20包括雷达数据编码器21和存储设备22，雷达数据编码器21的输入是雷达数据，该雷达数据可以是上述步骤所述的对毫米波雷达设备测出的原始雷达数据进行FFT后的频率点系数，也可以是原始雷达数据。雷达数据编码器21对频率点系数或原始雷达数据编码后得到编码后的雷达数据，并将编码后的雷达数据存储在存储设备22中。
可移动平台可与地面离线分析设备进行有线通信或无线通信，可以在可移动平台停止作业或停止移动后，也可以在可移动平台正在作业或者正在移动的过程中。如图3所示，30表示地面离线分析设备的解析器，解析器30包括存储设备31和雷达数据解码器32。例如，可移动平台具体为无人机，当无人机返回地面后，与地面离线分析设备进行有线通信或无线通信，以使无人机将如图2所示的存储设备22中存储的编码后的雷达数据发送给地面离线分析设备的存储设备31，雷达数据解码器32进一步从存储设备31中获取编码后的雷达数据，并对编码后的雷达数据进行解码以 恢复出雷达数据，以便地面离线分析设备对雷达数据解码器32解出的雷达数据进行分析或处理，本实施例并不限定地面离线分析设备对雷达数据的分析或处理过程。
本实施例通过将待压缩的雷达数据进行分组，根据每个分组中的至少一个雷达数据，确定每个分组的编码参数，并根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据，通过分组的方式对雷达数据进行编码，提高了对每个分组的编码效率，实现了对雷达数据的高效压缩。
本发明实施例提供一种雷达数据处理方法。图4为本发明另一实施例提供的雷达数据处理方法的流程图。如图4所示，在图1所示实施例的基础上，本实施例中的方法，可以包括：
步骤S401、对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据。
在本实施例中，去相关变换具体可以是二维FFT，去相关变换后的雷达数据具体可以是经过二维FFT后的频率点系数。在其他实施例中，去相关变换还可以是除二维FFT之外的其他方式。
例如，对雷达设备测出的原始雷达数据进行二维FFT，得到原始雷达数据的频率点系数。
步骤S402、根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据。
所述根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据包括如下几种可行的实现方式：
一种可行的实现方式是：将所述去相关变换后的雷达数据作为所述待压缩的雷达数据。例如，对雷达设备测出的原始雷达数据进行二维FFT后，将原始雷达数据的频率点系数作为所述待压缩的雷达数据。
另一种可行的实现方式是：对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据；将所述排序后的雷达数据作为所述待压缩的雷达数据。例如，对雷达设备测出的原始雷达数据进行二维FFT后，对原始雷达数据的频率点系数进行排序，得到排序后的频率点系数，进一步将 排序后的频率点系数作为所述待压缩的雷达数据。
具体的，所述对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据，包括：根据所述去相关变换后的雷达数据的频率，对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据。例如，对原始雷达数据的频率点系数进行排序时，可以根据频率高低对频率点系数进行排序，可选的，按照从低频到高频的顺序对频率点系数进行排序，或者按照从高频到低频的顺序对频率点系数进行排序。
具体的，所述对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据，包括：对所述去相关变换后的雷达数据按照从低频到高频的顺序进行排序，得到排序后的雷达数据。假定雷达数据相邻频点的能量幅度是接近的，是相关的，对频率点系数按照从低频到高频的顺序进行排序后，相当于对雷达数据相邻频点的能量幅度进行了排序。通常情况下，中低频频点的能量幅度比高频频点的能量幅度大。
具体的，所述对所述去相关变换后的雷达数据按照从低频到高频的顺序进行排序，包括：对所述去相关变换后的雷达数据按照从低频到高频的顺序进行ZigZag扫描。例如，对频率点系数按照从低频到高频的顺序进行排序时，具体可以对频率点系数按照从低频到高频的顺序进行ZigZag扫描。如图5所示，50表示4*4的矩阵，此处只是示意性说明，并不限定矩阵的大小。对矩阵50进行ZigZag扫描时具体可以是按照图5所示的箭头对矩阵50进行扫描，例如，矩阵50具体为如图6所示的矩阵，则对矩阵50进行ZigZag扫描后得到的扫描序列为1、5、3、9、7、3、9、5、4、7、3、6、6、4、1、3。如图6所示，矩阵50的第一行第一列的元素是1、第一行第二列的元素是5、第二行第一列的元素是3，在矩阵50中，第一行第一列的元素1、第一行第二列的元素5、以及第二行第一列的元素3相邻，对矩阵50进行ZigZag扫描后，第一行第一列的元素1、第一行第二列的元素5、以及第二行第一列的元素3在排列顺序上也相邻。在本实施例中，对频率点系数按照从低频到高频的顺序进行ZigZag扫描，是为了让相邻的频点在扫描后其排列顺序也相邻。
步骤S403、将待压缩的雷达数据进行分组。
可选的，如果将排序后的频率点系数作为所述待压缩的雷达数据，则 对排序后的频率点系数进行分组时，可以将频率相近的频率点系数分在一个组内。每个分组内的频率点系数的个数可以相等，也可以不等。
步骤S404、根据每个分组对应的量化步长，对每个分组中的各雷达数据进行量化。
由于一个分组内，频率点系数相近，可以对每个分组中的频率点系数进行量化，每个分组可对应一个量化步长，具体的，每个分组对应的量化步长可以相等，也可以不相等。另外，同一个分组内，部分雷达数据的量化步长也可以不相等。
步骤S405、根据每个分组中的至少一个雷达数据，确定每个分组的编码参数。
步骤S405与步骤S102的具体原理和实现方式类似，此处不再赘述。
步骤S406、根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据。
步骤S406与步骤S103的具体原理和实现方式类似，此处不再赘述。
步骤S407、存储所述编码后的雷达数据。
本实施例通过对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据，对去相关变换后的雷达数据进行压缩，可进一步提高对雷达数据的压缩效率。
本发明实施例提供一种雷达数据处理方法。图7为本发明另一实施例提供的雷达数据处理方法的流程图。如图7所示，在上述实施例的基础上，本实施例中的方法，可以包括：
步骤S701、对雷达设备测出的原始雷达数据进行二维FFT，得到所述原始雷达数据的频率点系数。
步骤S702、根据所述原始雷达数据的频率点系数，确定待压缩的频率点系数。
具体的，所述根据所述原始雷达数据的频率点系数，确定待压缩的频率点系数包括：将所述原始雷达数据的频率点系数进行排序，得到排序后的频率点系数；将所述排序后的频率点系数作为所述待压缩的频率点系数。例如，根据频率高低对频率点系数进行排序，可选的，按照从低频到 高频的顺序对频率点系数进行排序，或者按照从高频到低频的顺序对频率点系数进行排序。
具体的，所述将所述原始雷达数据的频率点系数进行排序，得到排序后的频率点系数，包括：对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行排序，得到排序后的频率点系数。假定雷达数据相邻频点的能量幅度是接近的，是相关的，对频率点系数按照从低频到高频的顺序进行排序后，相当于对雷达数据相邻频点的能量幅度进行了排序。
具体的，所述对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行排序，包括：对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行ZigZag扫描。例如，对频率点系数按照从低频到高频的顺序进行排序时，具体可以对频率点系数按照从低频到高频的顺序进行ZigZag扫描，以使相邻的频点经过ZigZag扫描后其排列顺序也相邻，即将相近频点的系数排列在相邻的扫描序上。
步骤S703、对所述排序后的频率点系数进行分组，每个分组包括至少一个频率点系数。
在本实施例中，对所述排序后的频率点系数进行分组的目的是将能量幅度接近的频点分到一组内。每个分组中包括的频率点系数的个数相等；或者每个分组中包括的频率点系数的个数不相等。
对所述排序后的频率点系数进行分组所采用的分组方法可以有多种，本实施例不作具体的限定。
步骤S704、根据每个分组对应的量化步长，对每个分组中的各频率点系数进行量化。
可选的，所述每个分组对应的量化步长不相等；或者所述每个分组对应的量化步长相等。或者，所述分组中部分频率点系数的量化步长不相等。
例如，每个分组对应的量化步长相等均为Q，假设量化之前的频率点系数为F，量化之后频率点系数记为FQ。另外，Q值可以控制量化误差，当Q为1时，为无损编码(意味着解码端可以重建出与原始数据一样的数据)；Q不等于1时，是有损编码。无损条件下，典型场景的压缩率为：30％～50％，节省了70％～50％的存储空间，或相比的录制时间长度。有损条件下，量化步长每增加一倍，平均压缩率提升1个bit；以原始16bit精 度为例，在原有基础上多节省1/16。
步骤S705、根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数。具体的，所述根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数，包括：根据所述分组中的各频率点系数，确定所述各频率点系数分别对应的K参数的估计值；根据所述分组中各频率点系数分别对应的K参数的估计值，确定所述分组对应的所述K参数。
例如，某一分组中量化之前的频率点系数为F，量化之后频率点系数为FQ，根据如下判断条件确定量化之后的频率点系数FQ对应的最佳K参数，假设FQ对应的最佳K参数记为Ke。
If(FQ＝＝0)
Ke＝0；
Else
Ke＝log2(FQ)；
假设一个分组中包括F1、F2、F3、F4四个频率点系数，在确定该分组对应的K参数之前，先确定该分组中F1、F2、F3、F4分别对应的最佳K参数，具体的，根据上述判断条件确定F1、F2、F3、F4分别对应的最佳K参数，假设F1对应的最佳K参数为Ke1、F2对应的最佳K参数为Ke2、F3对应的最佳K参数为Ke3、F4对应的最佳K参数为Ke4。进一步根据F1、F2、F3、F4分别对应的最佳K参数即Ke1、Ke2、Ke3、Ke4，计算出该分组对应的K参数。可选的，根据Ke1、Ke2、Ke3、Ke4计算K参数可以有多种方式，此处并不限定具体的计算方式。其他分组对应的K参数的计算方法同理于此，此处不再一一赘述。
具体的，所述根据所述分组中各频率点系数分别对应的K参数的估计值，确定所述分组对应的所述K参数，包括：对所述分组中各频率点系数分别对应的K参数的估计值进行统计平均，得到所述分组对应的所述K参数。
例如，根据上述条件，同一分组中量化之后的各频率点系数分别对应一个最佳K参数即Ke，对同一分组内量化之后的各频率点系数分别对应的Ke进行统计平均，即可得到该分组对应的K参数。例如，上述分组包 括F1、F2、F3、F4。F1、F2、F3、F4分别对应的最佳K参数依次为Ke1、Ke2、Ke3、Ke4，根据Ke1、Ke2、Ke3、Ke4计算该分组对应的K参数的一种可实现方式是：对Ke1、Ke2、Ke3、Ke4进行统计平均，得到的平均值为该分组对应的K参数。其他分组对应的K参数的计算方式与此类似，此处不再一一赘述。不同分组对应的K参数可能相同，也可能不同。
步骤S706、根据每个分组对应的K参数，对每个分组中的各频率点系数进行K阶哥伦布编码。
以量化之后的频率点系数FQ为例，对FQ进行K阶哥伦布编码的算法具体如下：
Me＝(Sign(FQ)！＝0)？2*Level(FQ)-Sign(FQ):2*Level(FQ)
其中，Sign(FQ)＝(FQ<0)？1:0；Level(FQ)＝(FQ<0)？-FQ:FQ。
对FQ进行K阶哥伦布编码后的结果包括两个码字，一个码字记为a,另一个码字记为b，具体的，码字a在码字b之前，也就是说，K阶哥伦布编码后的FQ为ab，码字a的码字长度为(Me/2K)+1，码字a的码字值为1；码字b的码字长度为K，码字b的码字值为Me&(2K-1)。假设码字a的码字长度为3，码字b的码字长度为2，码字b的码字值为11，则K阶哥伦布编码后的FQ为00111。其他量化之后的频率点系数的编码的过程与此类似，此处不一一列举。
例如，M*N个频率点系数分为H个分组，每个分组对应的量化步长为Q，第一个分组对应的K参数具体为K1，第二个分组对应的K参数具体为K2，一直到第H个分组对应的K参数具体为KH，则对M*N个频率点系数进行K阶哥伦布编码时，具体可以依次对每个分组对应的量化步长Q进行编码，对第一个分组对应的K参数K1进行编码，对第一个分组中各量化后的频率点系数进行编码，对第二个分组对应的K参数K2进行编码，对第二个分组中各量化后的频率点系数进行编码，……，对第H个分组对应的K参数KH进行编码，对第H个分组中各量化后的频率点系数进行编码。如此，所述编码后的雷达数据包括：每个分组对应的量化步长、每个分组对应的K参数、以及对每个分组中的各频率点系数进行K阶哥伦布编码后的码字。
如果每个分组对应的量化步长Q是统一的，则编码后的雷达数据的具 体存储格式可以是：Q、K1、K1的编码结果、K2、K2的编码结果、……KH、KH的编码结果。
如果每个分组对应的量化步长Q不同，例如，第一个分组的量化步长为Q1、第二分组的量化步长为Q2……第H个分组的量化步长为QH，则编码后的雷达数据的具体存储格式可以是：Q1、K1、K1的编码结果、Q2、K2、K2的编码结果、……QH、KH、KH的编码结果。
例如，无人机是编码后的雷达数据的发送端，地面端设备是接收端，无人机对排序后的频率点系数进行分组时采用的预设分组方法可以是多个预设分组方法中的一个，如果无人机采用的预设分组方法不是无人机和地面端设备约定好的分组方法，则当无人机对雷达数据进行编码时，还需要对无人机采用的预设分组方法的标识信息进行编码，也就是说，所述编码后的雷达数据还包括：对所述排序后的频率点系数进行分组时采用的预设分组方法的标识信息。如此，当地面端设备接收到编码后的雷达数据并对编码后的雷达数据进行解码时，可解析出无人机采用的预设分组方法，以便地面端设备采用同样的分组方法对解码后的雷达数据进行分组。如果无人机采用的预设分组方法是无人机和地面端设备约定好的分组方法，则不需要对无人机采用的预设分组方法的标识信息进行编码。
步骤S707、存储所述编码后的雷达数据。
无人机对每个分组编码后进一步存储编码后的雷达数据，实现了对雷达数据的压缩存储。
本实施例通过对雷达设备测出的原始雷达数据进行二维FFT，得到所述原始雷达数据的频率点系数，对各频率点系数排序分组后使得能量幅度接近的频点分到一组内，分组后每组共享一个显示表达的K值，进一步对每个分组中的各频率点系数进行量化，对每个分组中量化后的各频率点系数进行K阶哥伦布编码，提高了对雷达数据的压缩效率。
本发明实施例提供一种雷达数据处理设备。图8为本发明实施例提供的雷达数据处理设备的结构图，如图8所示，雷达数据处理设备80包括：处理器81；其中，处理器81用于：将待压缩的雷达数据进行分组；根据每个分组中的至少一个雷达数据，确定每个分组的编码参数；根据每个分 组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据。
本发明实施例提供的雷达数据处理设备的具体原理和实现方式均与图1所示实施例类似，此处不再赘述。
本实施例通过将待压缩的雷达数据进行分组，根据每个分组中的至少一个雷达数据，确定每个分组的编码参数，并根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据，通过分组的方式对雷达数据进行编码，提高了对每个分组的编码效率，实现了对雷达数据的高效压缩。
本发明实施例提供一种雷达数据处理设备。在图8所示实施例提供的技术方案的基础上，可选的，处理器81根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据之后，还用于：将所述编码后的雷达数据存储到存储器。
可选的，处理器81将待压缩的雷达数据进行分组之前，还用于：对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据；根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据。
可选的，处理器81根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据时，具体用于：将所述去相关变换后的雷达数据作为所述待压缩的雷达数据。
可选的，处理器81根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据时，具体用于：对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据；将所述排序后的雷达数据作为所述待压缩的雷达数据。
可选的，处理器81对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据时，具体用于：根据所述去相关变换后的雷达数据的频率，对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据。
可选的，处理器81对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据时，具体用于：对所述去相关变换后的雷达数据按照从低频到高频的顺序进行排序，得到排序后的雷达数据。
可选的，处理器81对所述去相关变换后的雷达数据按照从低频到高频的顺序进行排序时，具体用于：对所述去相关变换后的雷达数据按照从低频到高频的顺序进行ZigZag扫描。
本发明实施例提供的雷达数据处理设备的具体原理和实现方式均与图4所示实施例类似，此处不再赘述。
本实施例通过对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据，对去相关变换后的雷达数据进行压缩，可进一步提高对雷达数据的压缩效率。
本发明实施例提供一种雷达数据处理设备。在图8所示实施例提供的技术方案的基础上，处理器81对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据时，具体用于：对雷达设备测出的原始雷达数据进行二维FFT，得到所述原始雷达数据的频率点系数。
相应的，处理器81根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据时，具体用于：根据所述原始雷达数据的频率点系数，确定待压缩的频率点系数。
可选的，处理器81根据所述原始雷达数据的频率点系数，确定待压缩的频率点系数时，具体用于：将所述原始雷达数据的频率点系数进行排序，得到排序后的频率点系数；将所述排序后的频率点系数作为所述待压缩的频率点系数。
可选的，处理器81将所述原始雷达数据的频率点系数进行排序，得到排序后的频率点系数时，具体用于：对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行排序，得到排序后的频率点系数。
可选的，处理器81对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行排序时，具体用于：对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行ZigZag扫描。
可选的，处理器81将待压缩的雷达数据进行分组时，具体用于：对所述排序后的频率点系数进行分组，每个分组包括至少一个频率点系数。
可选的，每个分组中包括的频率点系数的个数相等；或者每个分组中包括的频率点系数的个数不相等。
可选的，处理器81根据每个分组中的至少一个雷达数据，确定每个分组的编码参数时，具体用于：根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数。
可选的，处理器81根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数时，具体用于：根据所述分组中的各频率点系数，确定所述各频率点系数分别对应的K参数的估计值；根据所述分组中各频率点系数分别对应的K参数的估计值，确定所述分组对应的所述K参数。
可选的，处理器81根据所述分组中各频率点系数分别对应的K参数的估计值，确定所述分组对应的所述K参数时，具体用于：对所述分组中各频率点系数分别对应的K参数的估计值进行统计平均，得到所述分组对应的所述K参数。
可选的，处理器81根据每个分组的编码参数，对每个分组中的各雷达数据进行编码时，具体用于：根据每个分组对应的K参数，对每个分组中的各频率点系数进行K阶哥伦布编码。
可选的，所述编码后的雷达数据包括：每个分组对应的K参数、以及对每个分组中的各频率点系数进行K阶哥伦布编码后的码字。
可选的，处理器81根据每个分组的编码参数，对每个分组中的各雷达数据进行编码之前，还用于：根据每个分组对应的量化步长，对每个分组中的各频率点系数进行量化。
可选的，所述每个分组对应的量化步长不相等；或者所述每个分组对应的量化步长相等。
可选的，所述分组中部分频率点系数的量化步长不相等。
可选的，所述编码后的雷达数据包括：每个分组对应的量化步长、每个分组对应的K参数、以及对每个分组中的各频率点系数进行K阶哥伦布编码后的码字。
可选的，所述编码后的雷达数据还包括：对所述排序后的频率点系数进行分组时采用的预设分组方法的标识信息。
本发明实施例提供的雷达数据处理设备的具体原理和实现方式均与图7所示实施例类似，此处不再赘述。
本实施例通过对雷达设备测出的原始雷达数据进行二维FFT，得到所述原始雷达数据的频率点系数，对各频率点系数排序分组后使得能量幅度接近的频点分到一组内，分组后每组共享一个显示表达的K值，进一步对每个分组中的各频率点系数进行量化，对每个分组中量化后的各频率点系数进行K阶哥伦布编码，提高了对雷达数据的压缩效率。
本发明实施例提供一种可移动平台。可移动平台包括：机身；动力系统，安装在所述机身，用于提供移动动力；雷达设备，用于检测所述可移动平台周围的目标物体；以及上述实施例所述的雷达数据处理设备。可选的，该可移动平台包括如下至少一种：无人机、可移动机器人、交通工具。
本发明实施例提供一种无人机。图9为本发明实施例提供的无人机的结构图，如图9所示，无人机900包括：机身、动力系统、雷达设备910、雷达数据处理设备908和飞行控制器918，所述动力系统包括如下至少一种：电机907、螺旋桨906和电子调速器917，动力系统安装在所述机身，用于提供飞行动力；飞行控制器918与所述动力系统通讯连接，用于控制所述无人机飞行。
具体的，雷达数据处理设备908的具体原理和实现方式均与上述实施例类似，此处不再赘述。
本实施例通过将待压缩的雷达数据进行分组，根据每个分组中的至少一个雷达数据，确定每个分组的编码参数，并根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据，通过分组的方式对雷达数据进行编码，提高了对每个分组的编码效率，实现了对雷达数据的高效压缩。
在本发明所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或 讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元，可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本发明各个实施例所述方法的部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
本领域技术人员可以清楚地了解到，为描述的方便和简洁，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。上述描述的装置的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
最后应说明的是：以上各实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述各实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的范围。

Claims (50)

  1. 一种雷达数据处理方法，其特征在于，包括：
    将待压缩的雷达数据进行分组；
    根据每个分组中的至少一个雷达数据，确定每个分组的编码参数；
    根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据。
  2. 根据权利要求1所述的方法，其特征在于，所述根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据之后，还包括：
    存储所述编码后的雷达数据。
  3. 根据权利要求1或2所述的方法，其特征在于，所述将待压缩的雷达数据进行分组之前，还包括：
    对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据；
    根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据。
  4. 根据权利要求3所述的方法，其特征在于，所述根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据，包括：
    将所述去相关变换后的雷达数据作为所述待压缩的雷达数据。
  5. 根据权利要求3所述的方法，其特征在于，所述根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据，包括：
    对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据；
    将所述排序后的雷达数据作为所述待压缩的雷达数据。
  6. 根据权利要求5所述的方法，其特征在于，所述对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据，包括：
    根据所述去相关变换后的雷达数据的频率，对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据。
  7. 根据权利要求5或6所述的方法，其特征在于，所述对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据，包括：
    对所述去相关变换后的雷达数据按照从低频到高频的顺序进行排序，得到排序后的雷达数据。
  8. 根据权利要求7所述的方法，其特征在于，所述对所述去相关变换后的雷达数据按照从低频到高频的顺序进行排序，包括：
    对所述去相关变换后的雷达数据按照从低频到高频的顺序进行ZigZag扫描。
  9. 根据权利要求3-8任一项所述的方法，其特征在于，所述对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据，包括：
    对雷达设备测出的原始雷达数据进行二维FFT，得到所述原始雷达数据的频率点系数；
    相应的，所述根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据，包括：
    根据所述原始雷达数据的频率点系数，确定待压缩的频率点系数。
  10. 根据权利要求9所述的方法，其特征在于，所述根据所述原始雷达数据的频率点系数，确定待压缩的频率点系数，包括：
    将所述原始雷达数据的频率点系数进行排序，得到排序后的频率点系数；
    将所述排序后的频率点系数作为所述待压缩的频率点系数。
  11. 根据权利要求10所述的方法，其特征在于，所述将所述原始雷达数据的频率点系数进行排序，得到排序后的频率点系数，包括：
    对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行排序，得到排序后的频率点系数。
  12. 根据权利要求11所述的方法，其特征在于，所述对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行排序，包括：
    对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行ZigZag扫描。
  13. 根据权利要求10-12任一项所述的方法，其特征在于，所述将待压缩的雷达数据进行分组，包括：
    对所述排序后的频率点系数进行分组，每个分组包括至少一个频率点系数。
  14. 根据权利要求13所述的方法，其特征在于，每个分组中包括的 频率点系数的个数相等；或者
    每个分组中包括的频率点系数的个数不相等。
  15. 根据权利要求1所述的方法，其特征在于，所述根据每个分组中的至少一个雷达数据，确定每个分组的编码参数，包括：
    根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数。
  16. 根据权利要求15所述的方法，其特征在于，所述根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数，包括：
    根据所述分组中的各频率点系数，确定所述各频率点系数分别对应的K参数的估计值；
    根据所述分组中各频率点系数分别对应的K参数的估计值，确定所述分组对应的所述K参数。
  17. 根据权利要求16所述的方法，其特征在于，所述根据所述分组中各频率点系数分别对应的K参数的估计值，确定所述分组对应的所述K参数，包括：
    对所述分组中各频率点系数分别对应的K参数的估计值进行统计平均，得到所述分组对应的所述K参数。
  18. 根据权利要求15-17任一项所述的方法，其特征在于，所述根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，包括：
    根据每个分组对应的K参数，对每个分组中的各频率点系数进行K阶哥伦布编码。
  19. 根据权利要求18所述的方法，其特征在于，所述编码后的雷达数据包括：每个分组对应的K参数、以及对每个分组中的各频率点系数进行K阶哥伦布编码后的码字。
  20. 根据权利要求13-19任一项所述的方法，其特征在于，所述根据每个分组的编码参数，对每个分组中的各雷达数据进行编码之前，还包括：
    根据每个分组对应的量化步长，对每个分组中的各频率点系数进行量化。
  21. 根据权利要求20所述的方法，其特征在于，所述每个分组对应 的量化步长不相等；或者
    所述每个分组对应的量化步长相等。
  22. 根据权利要求20所述的方法，其特征在于，所述分组中部分频率点系数的量化步长不相等。
  23. 根据权利要求20-22任一项所述的方法，其特征在于，所述编码后的雷达数据包括：每个分组对应的量化步长、每个分组对应的K参数、以及对每个分组中的各频率点系数进行K阶哥伦布编码后的码字。
  24. 根据权利要求23所述的方法，其特征在于，所述编码后的雷达数据还包括：
    对排序后的频率点系数进行分组时采用的预设分组方法的标识信息。
  25. 一种雷达数据处理设备，其特征在于，处理器；
    所述处理器用于：
    将待压缩的雷达数据进行分组；
    根据每个分组中的至少一个雷达数据，确定每个分组的编码参数；
    根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据。
  26. 根据权利要求25所述的雷达数据处理设备，其特征在于，所述处理器根据每个分组的编码参数，对每个分组中的各雷达数据进行编码，得到编码后的雷达数据之后，还用于：
    将所述编码后的雷达数据存储到存储器。
  27. 根据权利要求25或26所述的雷达数据处理设备，其特征在于，所述处理器将待压缩的雷达数据进行分组之前，还用于：
    对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据；
    根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据。
  28. 根据权利要求27所述的雷达数据处理设备，其特征在于，所述处理器根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据时，具体用于：
    将所述去相关变换后的雷达数据作为所述待压缩的雷达数据。
  29. 根据权利要求27所述的雷达数据处理设备，其特征在于，所述 处理器根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据时，具体用于：
    对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据；
    将所述排序后的雷达数据作为所述待压缩的雷达数据。
  30. 根据权利要求29所述的雷达数据处理设备，其特征在于，所述处理器对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据时，具体用于：
    根据所述去相关变换后的雷达数据的频率，对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据。
  31. 根据权利要求29或30所述的雷达数据处理设备，其特征在于，所述处理器对所述去相关变换后的雷达数据进行排序，得到排序后的雷达数据时，具体用于：
    对所述去相关变换后的雷达数据按照从低频到高频的顺序进行排序，得到排序后的雷达数据。
  32. 根据权利要求31所述的雷达数据处理设备，其特征在于，所述处理器对所述去相关变换后的雷达数据按照从低频到高频的顺序进行排序时，具体用于：
    对所述去相关变换后的雷达数据按照从低频到高频的顺序进行ZigZag扫描。
  33. 根据权利要求27-32任一项所述的雷达数据处理设备，其特征在于，所述处理器对雷达设备测出的原始雷达数据进行去相关变换，得到去相关变换后的雷达数据时，具体用于：
    对雷达设备测出的原始雷达数据进行二维FFT，得到所述原始雷达数据的频率点系数；
    相应的，所述处理器根据所述去相关变换后的雷达数据，确定所述待压缩的雷达数据时，具体用于：
    根据所述原始雷达数据的频率点系数，确定待压缩的频率点系数。
  34. 根据权利要求33所述的雷达数据处理设备，其特征在于，所述处理器根据所述原始雷达数据的频率点系数，确定待压缩的频率点系数时，具体用于：
    将所述原始雷达数据的频率点系数进行排序，得到排序后的频率点系数；
    将所述排序后的频率点系数作为所述待压缩的频率点系数。
  35. 根据权利要求34所述的雷达数据处理设备，其特征在于，所述处理器将所述原始雷达数据的频率点系数进行排序，得到排序后的频率点系数时，具体用于：
    对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行排序，得到排序后的频率点系数。
  36. 根据权利要求35所述的雷达数据处理设备，其特征在于，所述处理器对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行排序时，具体用于：
    对所述原始雷达数据的频率点系数按照从低频到高频的顺序进行ZigZag扫描。
  37. 根据权利要求34-36任一项所述的雷达数据处理设备，其特征在于，所述处理器将待压缩的雷达数据进行分组时，具体用于：
    对所述排序后的频率点系数进行分组，每个分组包括至少一个频率点系数。
  38. 根据权利要求37所述的雷达数据处理设备，其特征在于，每个分组中包括的频率点系数的个数相等；或者
    每个分组中包括的频率点系数的个数不相等。
  39. 根据权利要求25所述的雷达数据处理设备，其特征在于，所述处理器根据每个分组中的至少一个雷达数据，确定每个分组的编码参数时，具体用于：
    根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数。
  40. 根据权利要求39所述的雷达数据处理设备，其特征在于，所述处理器根据每个分组中的各频率点系数，确定对每个分组进行K阶哥伦布编码时每个分组对应的K参数时，具体用于：
    根据所述分组中的各频率点系数，确定所述各频率点系数分别对应的K参数的估计值；
    根据所述分组中各频率点系数分别对应的K参数的估计值，确定所述分组对应的所述K参数。
  41. 根据权利要求40所述的雷达数据处理设备，其特征在于，所述处理器根据所述分组中各频率点系数分别对应的K参数的估计值，确定所述分组对应的所述K参数时，具体用于：
    对所述分组中各频率点系数分别对应的K参数的估计值进行统计平均，得到所述分组对应的所述K参数。
  42. 根据权利要求39-41任一项所述的雷达数据处理设备，其特征在于，所述处理器根据每个分组的编码参数，对每个分组中的各雷达数据进行编码时，具体用于：
    根据每个分组对应的K参数，对每个分组中的各频率点系数进行K阶哥伦布编码。
  43. 根据权利要求42所述的雷达数据处理设备，其特征在于，所述编码后的雷达数据包括：每个分组对应的K参数、以及对每个分组中的各频率点系数进行K阶哥伦布编码后的码字。
  44. 根据权利要求37-43任一项所述的雷达数据处理设备，其特征在于，所述处理器根据每个分组的编码参数，对每个分组中的各雷达数据进行编码之前，还用于：
    根据每个分组对应的量化步长，对每个分组中的各频率点系数进行量化。
  45. 根据权利要求44所述的雷达数据处理设备，其特征在于，所述每个分组对应的量化步长不相等；或者
    所述每个分组对应的量化步长相等。
  46. 根据权利要求44所述的雷达数据处理设备，其特征在于，所述分组中部分频率点系数的量化步长不相等。
  47. 根据权利要求44-46任一项所述的雷达数据处理设备，其特征在于，所述编码后的雷达数据包括：每个分组对应的量化步长、每个分组对应的K参数、以及对每个分组中的各频率点系数进行K阶哥伦布编码后的码字。
  48. 根据权利要求47所述的雷达数据处理设备，其特征在于，所述 编码后的雷达数据还包括：
    对排序后的频率点系数进行分组时采用的预设分组方法的标识信息。
  49. 一种可移动平台，其特征在于，包括：
    机身；
    动力系统，安装在所述机身，用于提供移动动力；
    雷达设备，用于检测所述可移动平台周围的目标物体；以及
    如权利要求25-48任一项所述的雷达数据处理设备。
  50. 根据权利要求49所述的可移动平台，其特征在于，所述可移动平台包括如下至少一种：
    无人机、可移动机器人、交通工具。
PCT/CN2017/108026 2017-10-27 2017-10-27 雷达数据处理方法、设备及可移动平台 WO2019080093A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201780029118.9A CN109313260A (zh) 2017-10-27 2017-10-27 雷达数据处理方法、设备及可移动平台
PCT/CN2017/108026 WO2019080093A1 (zh) 2017-10-27 2017-10-27 雷达数据处理方法、设备及可移动平台
US16/856,747 US20200264269A1 (en) 2017-10-27 2020-04-23 Radar data processing method and device and mobile platform

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/108026 WO2019080093A1 (zh) 2017-10-27 2017-10-27 雷达数据处理方法、设备及可移动平台

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/856,747 Continuation US20200264269A1 (en) 2017-10-27 2020-04-23 Radar data processing method and device and mobile platform

Publications (1)

Publication Number Publication Date
WO2019080093A1 true WO2019080093A1 (zh) 2019-05-02

Family

ID=65205099

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/108026 WO2019080093A1 (zh) 2017-10-27 2017-10-27 雷达数据处理方法、设备及可移动平台

Country Status (3)

Country Link
US (1) US20200264269A1 (zh)
CN (1) CN109313260A (zh)
WO (1) WO2019080093A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH11183623A (ja) * 1997-12-25 1999-07-09 Nec Corp レーダ装置及びレーダ測定方法
EP1136839A1 (en) * 2000-03-24 2001-09-26 Terma Elektronik A/S System and method for transfer of radar information
US20030156054A1 (en) * 2002-02-19 2003-08-21 Murata Manufacturing Co., Ltd. Radar system
CN104237852A (zh) * 2013-06-20 2014-12-24 英飞凌科技股份有限公司 用于处理雷达信号的方法、装置和系统
CN104931977A (zh) * 2015-06-11 2015-09-23 同济大学 一种用于智能车辆的障碍物识别方法

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5473376A (en) * 1994-12-01 1995-12-05 Motorola, Inc. Method and apparatus for adaptive entropy encoding/decoding of quantized transform coefficients in a video compression system
JPH10283343A (ja) * 1997-04-02 1998-10-23 Mitsubishi Electric Corp データ処理方法
JP2007295157A (ja) * 2006-04-24 2007-11-08 Matsushita Electric Ind Co Ltd データ符号化装置、データ符号化方法、データ符号化プログラム及びデータ符号化プログラムを記録した情報記録媒体
ITPI20070012A1 (it) * 2007-02-14 2008-08-15 Univ Pisa Metodo di elaborazione di dati radar multipassaggio per il rilevamento e l'analisi di componenti multiple di retrodiffusori non stazionari
CN101655551B (zh) * 2008-08-20 2012-06-27 中国科学院电子学研究所 定帧长合成孔径雷达原始数据压缩的编解码方法及装置
US10890648B2 (en) * 2014-10-24 2021-01-12 Texas Instruments Incorporated Method and apparatus for generating alignment matrix for camera-radar system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH11183623A (ja) * 1997-12-25 1999-07-09 Nec Corp レーダ装置及びレーダ測定方法
EP1136839A1 (en) * 2000-03-24 2001-09-26 Terma Elektronik A/S System and method for transfer of radar information
US20030156054A1 (en) * 2002-02-19 2003-08-21 Murata Manufacturing Co., Ltd. Radar system
CN104237852A (zh) * 2013-06-20 2014-12-24 英飞凌科技股份有限公司 用于处理雷达信号的方法、装置和系统
CN104931977A (zh) * 2015-06-11 2015-09-23 同济大学 一种用于智能车辆的障碍物识别方法

Also Published As

Publication number Publication date
CN109313260A (zh) 2019-02-05
US20200264269A1 (en) 2020-08-20

Similar Documents

Publication Publication Date Title
US11627339B2 (en) Methods and devices for encoding and reconstructing a point cloud
US20220108494A1 (en) Method, device, and storage medium for point cloud processing and decoding
WO2020014590A1 (en) Generating a compressed representation of a neural network with proficient inference speed and power consumption
JP6923677B2 (ja) ビデオ圧縮におけるエントロピーコーディングのための選択的ミキシング
EP3751304A1 (en) Compressing and decompressing data about radio signals
US20220303537A1 (en) Method and device for encoding/decoding video signal by using optimized conversion based on multiple graph-based model
CN104918046A (zh) 一种局部描述子压缩方法和装置
Baker et al. Compression Of High Spectral Resolution Imagery'
Wiseman Can flight data recorder memory be stored on the cloud?
US8428372B2 (en) Method, apparatus and computer program product for compressing data
Sun et al. Dictionary learning for image coding based on multisample sparse representation
Yu et al. Rate-distortion optimized geometry compression for spinning LiDAR point cloud
Sahitya et al. Real time application of Raspberry Pi in compression of images
CN110928848A (zh) 一种文件碎片分类方法及系统
WO2019080093A1 (zh) 雷达数据处理方法、设备及可移动平台
Polyakov et al. Multi-criteria selection of the radar data compression method
CN116250008A (zh) 点云的编码、解码方法、编码器、解码器以及编解码系统
CN105100810A (zh) 一种成像声纳实时处理系统中的图像压缩解压方法及系统
CN115102934A (zh) 点云数据的解码方法、编码方法、装置、设备及存储介质
Yamazaki et al. Deep Feature Compression using Rate-Distortion Optimization Guided Autoencoder
CN103780904A (zh) 基于边缘的高光谱图像自适应无损预测编码系统及方法
CN116508333A (zh) 基于信道测量的资源高效室内定位
CN116325732A (zh) 点云的解码、编码方法、解码器、编码器和编解码系统
Poomrittigul et al. Reversible color transform for Bayer color filter array images
Pareek et al. An Image Compression Using Multilayer Wavelet Transform with 2DTCWT: A Review

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17929540

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17929540

Country of ref document: EP

Kind code of ref document: A1