A dual-stage channelized radar signal processing method, device and medium

Abstract

The present invention discloses a dual-stage channelized radar signal processing method, device, and medium, belonging to the field of radar signal processing. The method comprises: collecting the radio frequency signal of the radar antenna and sampling the intermediate frequency signal at a set sampling rate to obtain a digital signal; performing frequency domain coarse division processing on the digital signal, performing first-stage coarse division signalization based on wide-band rapid division to generate a time domain signal; outputting a high-resolution output signal under the dynamic impact impulse response of the filter; performing long and short pulse differential detection on the output signal of the filter, performing inverse Fourier transform on the detected pulse signal, reconstructing the time domain waveform, and outputting a pure target echo signal. The terminal device includes a processor, a transceiver, and a memory. The computer storage medium is used to store computer programs. The present invention effectively solves the problems of the conflict between resolution and real-time performance, high resource consumption, and insufficient pulse detection sensitivity of traditional single-stage channelization.

Description

Technical Field

[0001] The present invention relates to the field of radar signal processing, and in particular to a dual-stage channelized radar signal processing method, device, and medium. Background Art

[0002] Radar uses radio waves to detect targets and determine their spatial position. Therefore, radar is also called "radio positioning." Radar is an electronic device that uses electromagnetic waves to detect targets. Radar transmits electromagnetic waves to the target and receives the echoes, thereby obtaining information such as the distance from the target to the point of emission, the rate of change of distance (radial velocity), direction, and altitude.

[0003] In the existing technology, most radar signal processing methods use a single-stage channelized radar signal processing method. However, the traditional single-stage channelized radar signal processing method has the following limitations:

[0004] Conflict between resolution and processing speed: Single-stage channelization requires high resolution within a wide frequency band, which leads to a sharp increase in computational complexity and makes it difficult to meet real-time requirements.

[0005] High resource consumption: When RF is directly sampled, direct processing of high sampling rate signals will cause field programmable gate array (FPGA) resource overflow.

[0006] Insufficient pulse detection sensitivity: The detection sensitivity cannot be taken into account for both long and short pulse signals. The accumulation time of long pulses is insufficient, and the noise accumulation of short pulses is serious. Summary of the Invention

[0007] In response to the above-mentioned deficiencies in the prior art, the present invention provides a two-stage channelized radar signal processing method, device, and medium, which significantly improves radar signal processing efficiency and detection performance through two-stage channelized processing technology.

[0008] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0009] A dual-stage channelized radar signal processing method is provided, comprising:

[0010] Step S1: Collect the radio frequency signal of the radar antenna and sample the intermediate frequency signal at a set sampling rate to obtain a digital signal;

[0011] Step S2: performing frequency domain coarse division processing on the digital signal, performing first-level coarse signalization based on wide-band rapid division, and generating a time domain signal;

[0012] Step S3: performing high-resolution sub-band analysis on the output time domain signal, performing sub-band focusing processing using the second-level subdivision channelization, and outputting a high-resolution output signal under the dynamic impulse response of the filter;

[0013] Step S4: Perform differential detection of long and short pulses on the output signal of the filter, perform inverse Fourier transform on the detected pulse signal, reconstruct the time domain waveform, and output a pure target echo signal.

[0014] Furthermore, step S1 is specifically as follows:

[0015] Collect the RF signal from the radar antenna and use it at a sampling rate Sampling the intermediate frequency signal to obtain a digital signal , is the number of sampling points, is the IF bandwidth, n is the discrete time point of sampling, is the sampling period, It is the sampling process of RF signal.

[0016] Furthermore, step S2 includes:

[0017] Step S21: Convert the digital signal conduct The fast Fourier transform of the point is used to convert the total bandwidth Divided into parallel sub-channels, the bandwidth of each sub-channel is ;

[0018] ;

[0019] in, is the index of the subchannel, e is a natural constant, j is the imaginary unit, , is the frequency domain component of the subchannel;

[0020] Step S22: Perform the following operations on the signal output by each sub-channel: times the resolution of the data frequency down to , generating a time domain signal , m for The time index after decimation.

[0021] Furthermore, step S3 includes:

[0022] Step S31: Output the time domain signal of each sub-channel Proceed to the second level The Fourier transform of the point is used to convert the bandwidth of the subchannel Further divided into sub-channels, the bandwidth of the sub-channels , increasing the resolution of data frequency to ;

[0023] ;

[0024] in, is the index of the second-level subdivision channel, , the joint index formed by the index of the second-level subdivided channel and the index of the first-level channel Uniquely identifies the precise frequency interval where the signal is located, Sub-channel based on joint index The output signal;

[0025] Step S32: Dynamically adjust the filter bandwidth according to the signal characteristics of different sub-channels , is the bandwidth adjustment factor, , then the filter impulse response is:

[0026] ;

[0027] in, is the impulse response of the filter, l is the tap position of the filter, L is the order of the filter, is the filter coefficient, k is the channel index; by the filter coefficient Optimization of the channel size to suppress interference between adjacent sub-channels while reducing computational complexity;

[0028] Step S33: Calculate the output signal of the filter according to the filter impulse response ;

[0029] .

[0030] Furthermore, the filter coefficients The optimization method is:

[0031] The output signal Enter k After a non-uniform filter, the output signal , and output the signal as expected Constructing the minimum mean square error objective function ;

[0032] ;

[0033] in, is the error signal, is the expectation of the error signal;

[0034] For the objective function Find the filter coefficients The partial derivative of

[0035] ;

[0036] The instantaneous approximation is used instead of the expectation of the error signal to update the filter coefficients;

[0037] ;

[0038] in, For the n The filter coefficients corresponding to the discrete time points of the samples are is the updated filter coefficient, is the step size factor, .

[0039] Furthermore, the strategy to reduce computational complexity is:

[0040] Set every M The filter coefficients are updated once at each sampling discrete time point;

[0041] ;

[0042] in, i is the number of the sampling discrete time point, For the i Sub-sampling the error signal at discrete time points, The signal is sampled at discrete time points Yes, the sampled value.

[0043] Furthermore, the method for differential detection of long and short pulses is:

[0044] Step S41: Setting the pulse width of the short pulse , the output signal of the filter Perform short pulse detection and calculate output signal In the sliding window length N Mean difference during short pulse detection ;

[0045] ;

[0046] in, N is the sliding window length for short pulse detection;

[0047] Step S42: If the mean difference , then the short pulse detection is triggered and the pulse signal detected by the short pulse is output. is the threshold value for short pulse detection, otherwise, execute step S43;

[0048] Step S43: Setting the pulse width of the long pulse , the sliding window length for long pulse detection is , calculate the output signal The sliding window length Mean difference of long pulse detection under ;

[0049] ;

[0050] Step S42: If the mean difference , then the short pulse detection is triggered and the pulse signal detected by the long pulse is output. is the threshold value for long pulse detection, otherwise, return to step S41 and reset the pulse width of the short pulse .

[0051] A terminal device is provided, which includes a processor, a transceiver, and a memory. The memory is used to store a computer program. The processor is used to call and run the computer program from the memory and control the transceiver to perform a receiving or transmitting action, so that the terminal device executes the above-mentioned two-stage channelized radar signal processing method.

[0052] A computer storage medium is provided for storing a computer program, wherein the computer program includes instructions for executing the dual-stage channelized radar signal processing method as described above.

[0053] The present invention significantly improves signal processing efficiency and detection performance through a dual-stage channel structure that combines coarse and fine segmentation. The first-stage channel utilizes fast Fourier transforms for wideband coarse segmentation, while the second-stage utilizes subband Fourier transforms for high resolution. Combined with non-uniform filter bank optimization and dual-threshold pulse detection, this effectively addresses the issues of traditional single-stage channelization, such as the conflict between resolution and real-time performance, high resource consumption, and insufficient pulse detection sensitivity. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 Flowchart of a two-stage channelized radar signal processing method. DETAILED DESCRIPTION

[0055] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0056] like Figure 1 As shown, a dual-stage channelized radar signal processing method includes:

[0057] Step S1: Collect the RF signal of the radar antenna and sample the intermediate frequency signal at a set sampling rate to obtain a digital signal. Sampling the intermediate frequency signal to obtain a digital signal , is the number of sampling points, is the IF bandwidth, n is the discrete time point of sampling, is the sampling period, It is the sampling process of RF signal.

[0058] Step S2: Convert the digital signal Perform frequency domain coarse division processing, perform first-level coarse signaling based on broadband rapid division, and generate time domain signals. Step S2 specifically includes the following steps:

[0059] Step S21: Convert the digital signal conduct The fast Fourier transform of the point is used to convert the total bandwidth Divided into parallel sub-channels, the bandwidth of each sub-channel is ;

[0060] ;

[0061] in, is the index of the subchannel, e is a natural constant, j is the imaginary unit, , is the frequency domain component of the subchannel;

[0062] Step S22: Perform the following operations on the signal output by each sub-channel: times the resolution of the data frequency down to , generating a time domain signal , m for The time index after decimation.

[0063] Step S3: Output the time domain signal Perform high-resolution sub-band analysis, use the second-stage subdivision channelization to perform sub-band focusing processing, and output a high-resolution output signal; in two-stage channelization, the first-stage prototype filter usually has a wider transition band, and the second stage uses a high-resolution prototype filter to improve frequency selectivity.

[0064] The specific steps include:

[0065] Step S31: Output the time domain signal of each sub-channel Proceed to the second level The Fourier transform of the point is used to convert the bandwidth of the subchannel Further divided into sub-channels, the bandwidth of the sub-channels , increasing the resolution of data frequency to ;

[0066] ;

[0067] in, is the index of the second-level subdivision channel, , the joint index formed by the index of the second-level subdivided channel and the index of the first-level channel Uniquely identifies the precise frequency interval where the signal is located, Sub-channel based on joint index The output signal;

[0068] Step S32: Dynamically adjust the filter bandwidth according to the signal characteristics of different sub-channels , is the bandwidth adjustment factor, , then the filter impulse response is:

[0069] ;

[0070] in, is the impulse response of the filter, l is the tap position of the filter, L is the order of the filter, is the filter coefficient, k is the channel index; by the filter coefficient Optimization of the channel size to suppress interference between adjacent sub-channels while reducing computational complexity;

[0071] Step S33: Calculate the output signal of the filter according to the filter impulse response ;

[0072] .

[0073] Filter coefficients The optimization method is:

[0074] The output signal Enter k After a non-uniform filter, the output signal , and output the signal as expected Constructing the minimum mean square error objective function ;

[0075] ;

[0076] in, is the error signal, is the expectation of the error signal; ideally, the expected output signal It is a pure signal containing only the target signal and without adjacent channel interference, by minimizing the objective function , which can make the filter output as close to the desired signal as possible, thereby suppressing adjacent channel interference.

[0077] For the objective function Find the filter coefficients The partial derivative of

[0078] ;

[0079] The instantaneous approximation is used instead of the expectation of the error signal to update the filter coefficients;

[0080] ;

[0081] in, For the n The filter coefficients corresponding to the discrete time points of the samples are is the updated filter coefficient, is the step size factor, The step size factor is used to control the amplitude of each filter coefficient update. If the step size is too large, the optimization process may not converge. If the step size is too small, the convergence speed will be slow. , by iteratively adjusting the filter coefficients, the output signal The objective function is satisfied, and the expectation of the error signal reaches the error expectation threshold;

[0082] The strategy to reduce computational complexity is:

[0083] Set every M The filter coefficients are updated once at each sampling discrete time point;

[0084] ;

[0085] in, i is the number of the sampling discrete time point, For the i Sub-sampling the error signal at discrete time points, The signal is sampled at discrete time points Yes, the sampled value.

[0086] When updating filter coefficients, operations are performed based on polyphase branches. Compared to directly operating on the original filter coefficients, the number of multiplication and addition operations is significantly reduced, effectively lowering computational complexity. Combined with this computational complexity reduction strategy, this approach effectively suppresses adjacent channel interference, improving radar signal processing accuracy while also reducing the amount of computation required to meet real-time processing requirements. This approach plays a crucial role in dual-stage channelized radar systems.

[0087] Radar echo signals are often affected by time-varying interference (such as moving interference sources and multipath). Dynamic updating of filter coefficients allows the filter to adaptively adjust the stopband position at different times, continuously suppressing changing adjacent channel interference. For example, if the interference frequency drifts over time, the filter coefficients automatically adjust to ensure that the filter notch position follows the interference frequency.

[0088] The introduction of discrete sampling time points enables online filter updates, eliminating the need for offline coefficient redesign. Within the radar pulse repetition period (PRI), the filter coefficients can be iteratively optimized with each pulse echo, meeting real-time signal processing requirements.

[0089] Step S4: Perform differential detection of long and short pulses on the output signal of the filter, perform inverse Fourier transform on the detected pulse signal, reconstruct the time domain waveform, and output a pure target echo signal.

[0090] The method for differential detection of long and short pulses is:

[0091] Step S41: Setting the pulse width of the short pulse , the output signal of the filter Perform short pulse detection and calculate output signal In the sliding window length N Mean difference during short pulse detection ;

[0092] ;

[0093] in, N is the sliding window length for short pulse detection;

[0094] Step S42: If the mean difference , then the short pulse detection is triggered and the pulse signal detected by the short pulse is output. is the threshold value for short pulse detection, otherwise, execute step S43;

[0095] Step S43: Setting the pulse width of the long pulse , the sliding window length for long pulse detection is , calculate the output signal The sliding window length Mean difference of long pulse detection under ;

[0096] ;

[0097] Step S42: If the mean difference , then the short pulse detection is triggered and the pulse signal detected by the long pulse is output. is the threshold value for long pulse detection, otherwise, return to step S41 and reset the pulse width of the short pulse .

[0098] The processing object of dual threshold pulse detection is the time domain pulse characteristics in the noisy echo signal, and the effective pulses are screened through two levels of thresholds. In the differential detection of long and short pulses, This is a dynamically adjusted coarse detection threshold whose value is inversely proportional to the pulse width and directly proportional to the noise mean. The noise mean is the average power of the Gaussian white noise at the receiver front end, estimated through statistical properties and coupled to the pulse bandwidth. This dual-threshold design achieves a high detection probability for long pulses and a low false alarm probability for short pulses.

[0099] A terminal device includes a processor, a transceiver, and a memory. The memory is used to store a computer program. The processor is used to call and run the computer program from the memory and control the transceiver to perform a receiving or transmitting action, so that the terminal device executes the above-mentioned dual-stage channelized radar signal processing method.

[0100] A computer storage medium is used to store a computer program, wherein the computer program includes instructions for executing the above-mentioned dual-stage channelized radar signal processing method.

[0101] This invention significantly improves signal processing efficiency and detection performance through a dual-stage channel structure that combines coarse and fine segmentation. The first-stage channel utilizes fast Fourier transforms for wideband coarse segmentation, while the second-stage utilizes subband Fourier transforms for high resolution. Combined with non-uniform filter bank optimization and dual-threshold pulse detection, this approach effectively addresses the issues of traditional single-stage channelization, such as the conflict between resolution and real-time performance, high resource consumption, and insufficient pulse detection sensitivity.

Claims

1. A two-stage channelized radar signal processing method, characterized in that: include: Step S1: Collect the radio frequency signal of the radar antenna and sample the intermediate frequency signal at a set sampling rate to obtain a digital signal; Step S2: performing frequency domain coarse division processing on the digital signal, performing first-level coarse signalization based on wide-band rapid division, and generating a time domain signal; Step S3: performing high-resolution sub-band analysis on the output time domain signal, performing sub-band focusing processing using the second-level subdivision channelization, and outputting a high-resolution output signal under the dynamic impulse response of the filter; Step S4: performing differential detection of long and short pulses on the output signal of the filter, performing inverse Fourier transform on the detected pulse signal, reconstructing the time domain waveform, and outputting a pure target echo signal; The step S2 comprises: Step S21: Convert the digital signal conduct N 1-point fast Fourier transform, the total bandwidth B 0 is divided into K 1 parallel sub-channel, the bandwidth of each sub-channel is ; ； in, k 1 is the index of the first-level channel, e is a natural constant, j is the imaginary unit, , For the k Frequency domain component of 1 subchannel; Step S22: Perform the following operations on the signal output by each sub-channel: times the resolution of the data frequency fs down to , generating a time domain signal , m for D Time index after 1x decimation; The step S3 comprises: Step S31: Output the time domain signal of each sub-channel Proceed to the second level N The Fourier transform of the 2-point subchannel is Further divided into K 2 sub-channels, the bandwidth of the sub-channels , increasing the resolution of data frequency to ; ； in, k 2 is the index of the second-level subdivision channel, , the joint index formed by the index of the second-level subdivided channel and the index of the first-level channel Uniquely identifies the precise frequency interval where the signal is located, Sub-channel based on joint index The output signal; Step S32: Dynamically adjust the filter bandwidth according to the signal characteristics of different sub-channels , is the bandwidth adjustment factor, , then the filter impulse response is: ； in, is the impulse response of the filter, l is the tap position of the filter, L is the order of the filter, is the filter coefficient, k is the channel index; by the filter coefficient Optimization of the channel size to suppress interference between adjacent sub-channels while reducing computational complexity; Step S33: Calculate the output signal of the filter according to the filter impulse response ; 。 2. The dual-stage channelized radar signal processing method according to claim 1, wherein: The step S1 is specifically as follows: Collect the RF signal from the radar antenna and use it at a sampling rate Sampling the intermediate frequency signal to obtain a digital signal , is the number of sampling points, is the IF bandwidth, n is the discrete time point of sampling, is the sampling period, It is the sampling process of RF signal.

3. The dual-stage channelized radar signal processing method according to claim 1, wherein: The filter coefficients The optimization method is: The output signal Enter k After a non-uniform filter, the output signal , and output the signal as expected Constructing the minimum mean square error objective function ; ； in, is the error signal, is the expectation of the error signal; For the objective function Find the filter coefficients The partial derivative of ； The instantaneous approximation is used instead of the expectation of the error signal to update the filter coefficients; ； in, For the n The filter coefficients corresponding to the discrete time points of the samples are is the updated filter coefficient, is the step size factor, .

4. The dual-stage channelized radar signal processing method according to claim 3, wherein: The strategy to reduce computational complexity is: Set every M The filter coefficients are updated once at each sampling discrete time point; ； in, i is the number of the sampling discrete time point, For the i Sub-sampling the error signal at discrete time points, The signal is sampled at discrete time points The sampling value of .

5. The dual-stage channelized radar signal processing method according to claim 4, characterized in that: The method for differential detection of long and short pulses is: Step S41: Setting the pulse width of the short pulse , the output signal of the filter Perform short pulse detection and calculate output signal In the sliding window length N Mean difference during short pulse detection ; ； in, N is the sliding window length for short pulse detection; Step S42: If the mean difference , then the short pulse detection is triggered and the pulse signal detected by the short pulse is output. is the threshold value for short pulse detection, otherwise, execute step S43; Step S43: Setting the pulse width of the long pulse , the sliding window length for long pulse detection is , calculate the output signal The sliding window length Mean difference of long pulse detection under ; ； Step S42: If the mean difference , then the long pulse detection is triggered and the pulse signal detected by the long pulse is output. is the threshold value for long pulse detection, otherwise, return to step S41 and reset the pulse width of the short pulse .

6. A terminal device, characterized in that: The method comprises a processor, a transceiver, and a memory, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, control the transceiver to perform a receiving or sending action, so that the terminal device performs the dual-stage channelized radar signal processing method according to any one of claims 1 to 5.

7. A computer storage medium, characterized in that Used to store a computer program, wherein the computer program includes instructions for executing the dual-stage channelized radar signal processing method according to any one of claims 1 to 5.

Images