Carrier frequency and repetition frequency random agility semi-active pulse radar phase-coherent accumulation method

Abstract

The invention belongs to the technical field of radar signal processing, and particularly relates to a carrier frequency and repetition frequency random agility semi-active pulse radar phase-coherent accumulation method. The method specifically comprises the following steps: carrying out carrier frequency identification on a reference pulse signal received by a semi-active pulse radar receiver, carrying out digital orthogonal down-conversion processing on a residual intermediate frequency component identified in the carrier frequency identification process to obtain a zero intermediate frequency reference pulse signal, and carrying out repetition frequency identification on the zero intermediate frequency reference pulse signal; processing the radio frequency signal within the echo pulse signal receiving time to obtain a zero intermediate frequency echo pulse signal; using the zero intermediate frequency reference pulse signal to perform correlation matching processing on the receiving component of the zero intermediate frequency echo pulse signal to select a target signal; on the basis of a repetition frequency identification result, performing signal phase-coherent accumulation on the target signal by adopting non-uniform Fourier transform; and searching the range gate, the speed and the acceleration.

Description

Technical Field

[0001] The invention belongs to the technical field of radar signal processing, and in particular relates to a coherent accumulation method for a semi-active pulse radar with random agile carrier frequency and repetition frequency. Background Art

[0002] Semi-active pulse radar is a radar that separates the external radiation source and the receiving radar, receives the external radiation source echo pulse signal reflected by the target through the radar echo pulse receiving channel (hereinafter referred to as "echo pulse signal"), receives the direct wave pulse signal of the external radiation source through the radar reference pulse receiving channel (hereinafter referred to as "reference pulse signal"), and detects the target in a coherent processing manner. Compared with active pulse radar, semi-active pulse radar separates the radiation source transmitter and the radar receiver. This feature makes the operation of the semi-active radar receiver carrier independent of high-power transmitters and does not require the use of additional frequency band resources. It has great advantages in anti-electronic interference, anti-stealth detection, etc., and has the advantages of no electromagnetic pollution and low development cost.

[0003] Considering that the signal-to-noise ratio of a single pulse echo signal is low, signal accumulation is required to achieve target detection, and the carrier frequency or repetition frequency remains unchanged within the accumulation pulse segment, but changes between multiple accumulation pulse segments. The traditional coherent accumulation technology based on DFT uses the characteristics of the linear phase change and consistent pulse repetition interval of the radiation source echo signal to achieve coherent accumulation of the echo signal. However, in order to improve the anti-electronic interference, anti-information interception and anti-stealth detection capabilities of the semi-active pulse radar system, it is of great practical significance to agilely change the radiation source carrier frequency (hereinafter referred to as "carrier frequency", CF) and pulse repetition frequency (hereinafter referred to as "repetition frequency", PRF) of the radiation source pulse signal within each pulse repetition period (PRT). In the article "Long-term coherent accumulation algorithm based on non-uniform FFT" published by Tian Chao et al. in the Journal of Electronics and Information Technology, Vol. 36, No. 6, pp. 1374-1380, a non-uniform fast Fourier transform (FFT) algorithm is proposed under the condition of non-uniform phase change caused by time domain sampling, and coherent accumulation is performed on the signal with non-uniform phase change. In the paper "Doppler scaling correction in OFDM" published by Chung HimGeorge Yuen et al. on pages 4713-4717 of the 2013 IEEE International Conference on Communications (ICC), a Fourier transform Doppler frequency estimation method under frequency variation was proposed. However, these traditional accumulation techniques using discrete Fourier transform (DFT) and direct power accumulation are difficult to adapt to the scenario in which the carrier frequency repetition rate changes agilely between each pulse (fast time dimension) in the semi-active radar system.

[0004] On the other hand, the existing pulse agile radars with agile carrier frequency and repetition rate are mostly used in active pulse radars. In the article "Range-Doppler reconstruction for frequency agile and PRF-jittering radar" published by Ying-Hui Quan et al. in IETRadar, Sonar & Navigation, Vol. 12, No. 3, pp. 348-352, based on the theory of compressed sensing, the radar target information reconstruction with agile carrier frequency and repetition rate was realized through the orthogonal matching pursuit algorithm. In the doctoral dissertation "Research on Anti-interference and Coherent Processing Technology of Agile Radar" by Liu Zhixing of Xidian University, a carrier frequency-repetition rate joint agile radar target parameter estimation method based on the MUSIC algorithm was proposed, which equivalently converted the time domain signal processing into the spatial domain signal processing, and completed the joint estimation of the target distance and speed by using the super-resolution signal processing method in array signal processing.

[0005] However, the above technologies are basically studied and designed based on the situation that the carrier frequency and repetition frequency of active radar are known. In semi-active radar, the carrier frequency and repetition frequency are unknown to the radar receiver and echo signal processor, and no related work has specifically considered the coherent accumulation method of the echo signal of the radiation source with unknown and randomly changing carrier frequency and repetition frequency. In addition, the signal coherent accumulation and target parameter measurement method when the target kinematic parameters include acceleration are not systematically given in the above technical solutions. Summary of the invention

[0006] To achieve the above-mentioned purpose, the present invention provides a coherent accumulation method for a semi-active pulse radar with randomly agile carrier frequency and repetition rate. In a semi-active pulse radar system, the method can be applied to the coherent accumulation of echo signals of radiation sources with unknown carrier frequency and repetition rate and randomly agile changes between pulses.

[0007] The technical solution for implementing the present invention is as follows:

[0008] A coherent accumulation method for semi-active pulse radar with random carrier frequency and repetition frequency agility, the specific process is as follows:

[0009] The first step is to perform carrier frequency identification on the reference pulse signal received by the semi-active pulse radar receiver, perform digital orthogonal down-conversion processing on the residual intermediate frequency component identified in the carrier frequency identification process to obtain a zero intermediate frequency reference pulse signal, and perform repetition frequency identification on the zero intermediate frequency reference pulse signal;

[0010] The second step is to down-convert the RF signal within the echo pulse signal receiving time using the RF frequency, and then perform digital orthogonal down-conversion processing using the residual intermediate frequency to obtain a zero intermediate frequency echo pulse signal;

[0011] The third step is to use the zero intermediate frequency reference pulse signal to perform correlation matching processing on the received component of the zero intermediate frequency echo pulse signal;

[0012] The fourth step is to detect the matched signal according to the set detection threshold and select the target signal according to the peak position;

[0013] The fifth step is to use non-uniform Fourier transform to perform signal coherent accumulation on the target signal based on the repetition frequency identification result;

[0014] Step 6: Complete the search for distance gates, velocity, and acceleration.

[0015] Furthermore, the carrier frequency identification in the first step of the present invention is: the radio frequency echo pulse signal received by the radar receiver is detected at a radio frequency Perform down-conversion processing to obtain the residual intermediate frequency The intermediate frequency echo signal; according to the residual intermediate frequency frequency distribution interval The length of the carrier frequency is determined by the identification method, and the residual intermediate frequency f is obtained. I , Indicates the carrier frequency range; according to the intermediate frequency f I Realize carrier frequency f m estimate.

[0016] Furthermore, the re-frequency identification of the zero intermediate frequency reference pulse signal of the present invention is as follows: timing the time interval between two adjacent reference pulse signals to obtain T r The count value of (i) realizes the repetition frequency estimation.

[0017] Furthermore, the specific process of the third step of the present invention is:

[0018] For zero intermediate frequency reference pulse signal Discrete sampling is performed to obtain from Extract the zero intermediate frequency reference signal from Construct finite impulse response (FIR) filters;

[0019] For zero intermediate frequency echo pulse signal Discrete sampling is performed to obtain Input it into the FIR filter to complete the relevant matching process and obtain the matching result c m (k).

[0020] Furthermore, the matching result of the present invention is:

[0021]

[0022] Furthermore, the specific process of the fourth step of the present invention is:

[0023] When the signal c after correlation matching m (k) Reach the detection threshold T h When c m (k) performing peak selection, and determining the peak selection position as the position of the echo pulse signal;

[0024] Let the distance gate number corresponding to the peak selection position be k=k max When cache[k max -n,k max +n] range, complete the extraction of the target signal and peak alignment cache, n is the number of effective sampling points within the echo pulse width, f s is the sampling rate, T p is the pulse width, Indicates rounding up.

[0025] Furthermore, in the fifth step of the present invention, the non-uniform Fourier transform is used to transform the multiple correlated matched signals c m (k) Perform signal coherent accumulation, and the accumulated result is expressed as:

[0026]

[0027] Among them, T s represents the sampling interval, τ(m) represents the time delay of the echo pulse signal, and M is the number of accumulated pulses; φ m is the phase rotation factor.

[0028] Furthermore, the phase rotation factor φ calculated in the non-uniform Fourier transform of the present invention is m Based on the following recursion, we get:

[0029]

[0030] Among them, f m represents the carrier frequency of the mth pulse, v(m) represents the speed of the target when it is irradiated by the pulse signal of the radiation source, T r (m+1) represents the time interval of emitting radiation source pulses, a represents the speed of the target when it is irradiated by the radiation source pulse signal, and c represents the speed of light.

[0031] Furthermore, the speed search interval of the present invention is Δv

[0032]

[0033] Furthermore, the acceleration search interval Δa of the present invention is:

[0034]

[0035] Wherein, Δv is the speed change during the accumulation time.

[0036] Furthermore, the speed estimation interval [v min ,v max ]for:

[0037]

[0038] in, and They represent the range gate numbers corresponding to the Mth and 1st echo pulse signals selected by correlation matching, k p is the number of distance protection units, [a min ,a max ] indicates the range of acceleration.

[0039] Beneficial effects:

[0040] First, the present invention obtains the carrier frequency by analyzing the carrier frequency information of the reference pulse signal received by the reference receiving channel, and obtains the pulse repetition frequency by timing the pulse repetition interval of the reference pulse signal, and then uses the reference pulse signal to perform correlation matching on the echo pulse signal, extracts the correlation matching processing signal of the target echo, and finally uses the non-uniform discrete Fourier transform (NUDFT) processing method proposed by the present invention to realize the coherent accumulation of the signal. This method effectively improves the signal-to-noise ratio of the signal after radar signal accumulation (i.e. before detection), and improves the detection performance of the radar.

[0041] Secondly, the present invention adopts a carrier frequency repetition frequency identification method, so that the processing of radar received signals does not require prior information on the carrier frequency repetition frequency of the signal transmitted by the radiation source, and can independently identify the carrier frequency and repetition frequency characteristics of the signal.

[0042] Third, the non-uniform discrete Fourier (NUDFT) coherent accumulation method of the present invention is applied to semi-active pulse radar. Its basic principle is consistent with the common NUDFT (so it can achieve a coherent accumulation gain consistent with the traditional NUDFT), but due to the random agility of the carrier frequency, the Doppler frequencies carried by different echo pulse signals are difficult to align, resulting in difficulty in signal accumulation. Therefore, in terms of the representation form of nonlinear phase, the present invention considers the situation where the carrier frequency and repetition frequency change randomly between pulses at the same time. In this case, how to effectively accumulate signals and accurately identify the speed and acceleration information of the target is the important value of the present invention.

[0043] Fourth, the waveform transmission scheme provided by the invention with random and agile carrier frequency and repetition frequency between pulses can solve the problem of speed ambiguity under single carrier frequency and single repetition frequency of traditional pulse radar. Compared with the traditional design idea of ​​using staggered repetition frequency strategy to resolve speed ambiguity, which is prone to speed ambiguity resolution errors at high speed ambiguity, the present invention does not use the traditional technical solution for speed ambiguity resolution, but uses the random characteristics of carrier frequency and repetition frequency to match and search for speed, avoiding the problem of easy error in resolution caused by traditional speed resolution schemes. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 This is a schematic diagram of the working scenario of the semi-active pulse radar;

[0046] Figure 2 Schematic diagram of parameter search sequence for coherent accumulation process;

[0047] Figure 3 This is a signal processing flow chart of the present invention;

[0048] Figure 4 The distance, speed and acceleration measurement accuracy of the present invention varies with the accumulated pulse number;

[0049] Figure 5 This is the variation of the signal-to-noise ratio with the number of accumulated pulses after NUDFT coherent integration. DETAILED DESCRIPTION

[0050] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0051] It should be noted that the following embodiments and features in the embodiments may be combined with each other in the absence of conflict; and, based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making any creative work are within the scope of protection of the present disclosure.

[0052] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0053] First, the parameter model involved in the present invention is introduced.

[0054] Consider a practical scenario where the radiation source and receiver of the semi-active pulse radar are located on different platforms and are relatively stationary, and their directions relative to the target are approximately the same, such as Figure 1 Assume that the number of accumulated pulses is M, then the carrier frequency of the mth (m=0,1,2,…,M-1) pulse is recorded as f m ,and The pulse signal radiated by the radiation source and the reference pulse signal received by the reference receiving channel of the receiver have similar signal parameters. For the convenience of explanation, it is assumed here that a simple pulse signal without modulation is considered, and the pulse signal width is T p . When the target is irradiated by the first radiation source pulse signal, the initial distance of the target relative to the radar receiver is R, and the speed is v (defined as positive relative to the radar), and the target dynamic model of the highest second order in the radar radial direction is considered, and its acceleration is a. When the radiation source echo signal characteristics within the pulse width are insensitive to changes in velocity and acceleration, the speed and distance of the target relative to the radar receiver during the period of being irradiated by the radiation source pulse signal are considered to be independent of time. Consider that the minimum time interval between two consecutive radiation source emission pulses is T rmin , the electromagnetic wave propagation speed is c. In any pulse repetition interval, when the target moves, the delay change is much smaller than T rmin When the target speed and Doppler information are not affected by the movement of the target within the pulse repetition interval, but only by the pulse repetition interval T r (m). That is, for any m∈0,1,2,…,M-1, we have

[0055]

[0056] When the target is irradiated by the pulse signal of the mth radiation source, the approximate speed is

[0057]

[0058] Among them, T r(i) represents the time interval from the emission of the i-1th radiation source pulse to the emission of the i-th radiation source pulse, and defines T r (0) = 0, v(-1) = 0. Then the distance of the target relative to the radar receiver when the target is irradiated by the pulse signal of the mth radiation source is

[0059]

[0060] Consider that the pulse signal emitted by the radiation source has the following form

[0061]

[0062] Among them, rect(·) is the rectangular window function, is the random phase carried by the pulse signal emitted by the mth radiation source. The reference pulse signal received by the radar receiver has the following form:

[0063]

[0064] Among them, τ TR is the link delay of the reference signal from the radiation source transmitter to the radar receiver, represents the thermal noise of the receiver or reference pulse signal coupling link. Then the echo pulse signal reflected by the target can be expressed as

[0065]

[0066] Where L represents the path and scattering attenuation caused by signal transmission and reflection, represents the thermal noise received by the receiver for the echo pulse signal, and the time delay τ of the mth echo pulse signal m (t) can be expressed as

[0067]

[0068] It is worth noting that when the radiation source and receiver of the semi-active pulse radar are located on different platforms and are relatively stationary, and their directions relative to the target are approximately the same, formula (7) holds true.

[0069] The following describes a coherent accumulation method of a semi-active pulse radar with random carrier frequency and repetition frequency agility of the present invention. Figure 3 As shown, the signal processing flow is:

[0070] The first step is to perform carrier frequency identification on the reference pulse signal received by the semi-active pulse radar receiver, perform digital orthogonal down-conversion processing on the residual intermediate frequency component identified in the carrier frequency identification process to obtain a zero intermediate frequency reference pulse signal, and perform repetition frequency identification on the zero intermediate frequency reference pulse signal;

[0071] Carrier frequency identification: Carrier frequency identification of the reference pulse signal received by the reference receiving channel of the semi-active pulse radar receiver. The specific process is: the RF echo pulse signal received by the radar receiver is identified at the RF frequency. Perform down-conversion processing to obtain the residual intermediate frequency According to the residual intermediate frequency distribution interval The length of the carrier frequency identification method is determined. When the residual intermediate frequency frequency distribution interval is small, that is, less than the set threshold, DFT can be used directly to accurately calculate the residual intermediate frequency frequency. For the case where the carrier frequency is stable and the Doppler is zero, the phase difference can also be used to calculate the residual intermediate frequency frequency. For the residual intermediate frequency frequency distribution interval is large, that is, greater than the set threshold, in order to obtain an accurate residual intermediate frequency frequency estimate, the direct DFT calculation complexity is large. DFT coarse step search can be used, combined with the idea of ​​accurate locking of the phase-locked loop (or frequency-locked loop) to obtain the residual intermediate frequency f I The accurate estimation of the residual IF frequency also means the accurate estimation of the carrier frequency.

[0072] The residual intermediate frequency component in the intermediate frequency reference pulse signal identified during the carrier frequency re-frequency identification process is digitally orthogonally down-converted to obtain the zero intermediate frequency form of the reference pulse signal.

[0073]

[0074] in, represents the zero intermediate frequency reference pulse signal, A represents the pulse signal amplitude, τ TR represents the link delay of the reference signal from the radiation source transmitter to the radar receiver, T p Indicates the pulse signal width, f m represents the carrier frequency of the pulse, Indicates the random phase carried by the pulse signal emitted by the mth radiation source, Represents the noise after the corresponding digital orthogonal down-conversion.

[0075] Repetition frequency identification: detect the reference pulse signal with high signal-to-noise ratio, and time the time interval between two adjacent reference pulse signals to obtain T r (i) count value, that is, the repetition frequency estimation is realized (the pulse repetition frequency is T r (i) is the inverse of r The schematic diagram of (i) is shown in Figure 1 .

[0076] In the second step, the RF receiving component within the echo pulse signal receiving time is calculated according to the RF frequency f used in the first step. R Perform down-conversion processing, and then use the residual intermediate frequency f determined in the first step to IPerform digital orthogonal down-conversion processing, which converts the part of the received component containing the echo pulse signal into a zero intermediate frequency form:

[0077]

[0078] in, Indicates zero intermediate frequency echo pulse signal, represents the noise after the corresponding digital orthogonal down-conversion, L represents the path and scattering attenuation caused by signal transmission and reflection, τ(m) represents the time delay of the mth echo pulse signal, and R(m) represents the distance of the target relative to the radar receiver when the target is irradiated by the mth radiation source pulse signal.

[0079] The third step is to use the reference pulse signal in the form of zero intermediate frequency (i.e., as shown in equation (8)) to perform correlation matching processing on the received component containing the echo pulse signal in the form of zero intermediate frequency (i.e., as shown in equation (9)).

[0080] To specifically illustrate the correlation matching processing method, the zero intermediate frequency reference pulse signal and zero intermediate frequency echo pulse signal Expressed in discrete sampling form, they are:

[0081]

[0082] In the specific implementation, by detecting the reference pulse signal with high signal-to-noise ratio Extracting a zero-IF reference signal

[0083]

[0084] For zero-IF reference signals The finite impulse response (FIR) filter is constructed using its sampling value to transform the zero intermediate frequency echo pulse signal Sampling sequence of receiving interval Input FIR filter to complete the correlation matching process.

[0085] The calculation result of the correlation matching processing of the echo pulse signal part is:

[0086]

[0087] Where N is the number of sampling points within the pulse width, and * indicates the calculation of conjugate.

[0088] Obviously, when kT s =τ(m), the correlation matching result has the maximum envelope peak, and the corresponding signal power is A 2 L r L sN, reasonably ignoring the influence of the noise in the reference pulse signal on the noise power of the correlation matching result, the noise power is (This assumes without loss of generality that and Stably obeys zero mean and variance and Gaussian distribution), ignoring the minimum value term The signal-to-noise ratio of the peak point of the signal after correlation matching is approximately

[0089] The fourth step is signal peak position detection and gate data alignment cache.

[0090] Considering that the signal-to-noise ratio of the echo pulse signal after accumulation is large enough to reach the detection threshold, when the signal c after correlation matching m (k) Reach the detection threshold T h When c m (k) Perform peak selection and determine the peak selection position as the position of the echo pulse signal.

[0091] Therefore, the signal peak position detection is equivalent to solving the following problem

[0092]

[0093] With a sampling rate of f s , the pulse width is T p For example, the number of effective sampling points within the echo pulse width is ( Indicates rounding up).

[0094] When the distance gate number (i.e., sampling point index value) corresponding to the peak selection position is k=k max When cache[k max -n,k max +n] range to complete the extraction of the target signal and peak alignment cache.

[0095] It is worth mentioning that although accurate signal peak position detection is difficult to achieve in the case of a small signal-to-noise ratio, it is still possible to perform the relevant matching peak point movement alignment in advance in a certain wave gate dimension area according to the time delay shown in formula (7) when the speed and acceleration are matched in the subsequent fifth step, and the subsequent steps can also be used to achieve signal accumulation. This situation will not be introduced in detail here.

[0096] In the fifth step, the non-uniform Fourier transform target signal is used for signal accumulation to improve the signal-to-noise ratio and estimate the Doppler frequency.

[0097] From the analysis in the fourth step, it can be seen that the attenuation caused by the echo pulse signal transmission link and target scattering is large (corresponding to Ls When the receiver noise is high, the signal after correlation matching is greatly affected by the noise, making it difficult to accurately extract and estimate the target distance and speed information with high precision.

[0098] Therefore, the present invention further considers using non-uniform discrete Fourier transform NUDFT to achieve signal accumulation, so as to enhance the waveform characteristics of the signal and facilitate accurate extraction of the distance, speed and acceleration characteristics of the target.

[0099] Note that at the theoretical peak point of the correlation match (i.e., kT s =τ(m), we have

[0100]

[0101] Among them, w m The approximate mean is 0 and the variance is Observe formula (15), the phase of the pure signal part of the signal after correlation matching is rotated, aligned and accumulated, which can be similar to the traditional coherent accumulation method to achieve signal accumulation and improve the signal-to-noise ratio of the accumulated signal.

[0102] The signal accumulation of multiple related matched signals using non-uniform Fourier transform is expressed as:

[0103]

[0104] To achieve phase-aligned accumulation, the phase rotation factor is:

[0105]

[0106] According to formula (3), but

[0107]

[0108] Therefore, the phase rotation factor calculated by NUDFT is given in the recursive form of formula (18). Obviously, formula (18) also gives the phase rotation factor from f m to f m-1 In order to reduce the influence of error propagation caused by multiple recursions due to inaccurate parameter estimation when accumulating multiple pulse signals on the phase rotation alignment, it is possible to choose m=M / 2 as the base point and recurse to both sides respectively.

[0109] Observe formula (18), in order to effectively perform phase rotation, the speed and acceleration need to be searched in two dimensions to determine the speed and acceleration information of the completed signal accumulation and the extracted echo signal. When the speed change Δv=v(M-1)-v(0) during the accumulation time is less than the speed search interval, it can be considered that the target speed is approximately unchanged during the accumulation time, otherwise, the target speed v(m) is given by formula (2).

[0110] Generally speaking, according to the paper “Range-Doppler reconstruction for frequency agile and PRF-jittering radar” published by Jia Xu et al., IEEE Transactions on Aerospace and Electronic Systems, Vol. 47, No. 2, pp. 1186-1202, and the doctoral dissertation “Study on long-term accumulation signal processing algorithm for radar maneuvering targets” by Tian Jing of Beijing Institute of Technology, the lower bound of the target velocity resolution can be given by formula (19).

[0111]

[0112] In order to ensure that the velocity effect caused by acceleration does not exceed the velocity resolution, when the residual acceleration Δa satisfies formula (20), the effect of the residual acceleration on the coherent accumulation can be ignored.

[0113]

[0114] The maximum value of the target acceleration search interval is

[0115]

[0116] Note that in practice, the speed search interval and acceleration search interval can be given according to formula (19) and formula (21), but the given search intervals are all relaxed search intervals. Through simulation, a tighter search interval (satisfying the effect that the speed search and residual acceleration have almost no effect on signal accumulation, and having a wider search interval) can be given to reduce the number of speed and acceleration searches.

[0117] On the other hand, the distance difference method is used to pre-estimate the velocity value to reduce the number of velocity searches. Assume that the target has an acceleration range a∈[a min ,a max ], then the target speed estimation interval is [v min ,v max ],in,

[0118]

[0119] in, and They represent the range gate numbers corresponding to the Mth and 1st echo pulse signals selected by correlation matching, k p is the number of distance protection units, and k p ≥1. When the accumulated pulse number M increases, the speed search interval decreases according to formula (19), and the speed search times increase when the speed search interval is fixed. However, through speed interval pre-estimation, when there is a specific relationship between related parameters such as speed and acceleration (for example, the speed search range is large and the acceleration range is small), the speed search range will also be contracted by formulas (22) and (23) to reduce the speed search times.

[0120] Step 6: Complete the distance gate, speed and acceleration search.

[0121] according to Figure 2 The search order shown is to search all the cached gate dimension data in turn. When the cached data of the distance gate where the echo pulse signal peak is located is matched by NUDFT with the correct speed and acceleration search values, the signal accumulation result will have a globally unique maximum power value, completing the signal accumulation process and sending it to the subsequent signal detection and judgment program.

[0122] It is worth noting that the NUDFT used in the present invention performs accumulation processing on the signal with non-uniform phase change after correlation matching, which is essentially the coherent accumulation of the signal, so the signal-to-noise ratio gain of the NUDFT accumulation processing is consistent with the coherent accumulation gain.

[0123] The embodiment of the present application provides a signal coherent accumulation method for a semi-active pulse radar in which the carrier frequency repetition rate randomly changes between pulses, which solves the problem that the signal-to-noise ratio of a single pulse is low and it is difficult to achieve target detection.

[0124] The embodiment of the present application is aimed at a waveform transmission scheme in which the carrier frequency repetition rate randomly changes between pulses, which can solve the problem of speed ambiguity under a single carrier frequency and a single repetition rate of a traditional pulse radar. Compared with the traditional design idea of ​​using a staggered repetition rate strategy to resolve speed ambiguity, which is prone to errors in resolving speed ambiguity at high speed ambiguity, the present invention does not have such problems.

[0125] The embodiment of the present application only gives a waveform representation of a simple pulse form, but in fact, the NUDFT coherent accumulation method of the present invention is also applicable to other waveforms such as frequency modulation signals. Therefore, in order to improve the radar detection capability or anti-interference, the method is also applicable to the radiation source not listed in the above embodiment.

[0126] The signal coherent accumulation scheme proposed in the embodiment of the present application in which the carrier frequency repetition rate randomly changes between pulses is also applicable to situations in which the carrier frequency repetition rate does not change randomly (regularly), the carrier frequency repetition rate changes stably between accumulation segments within the accumulation time, or only a single (carrier frequency or repetition frequency) changes.

[0127] The embodiment of the present application considers the highest second-order dynamics of the target, that is, excluding the influence of acceleration and higher-order dynamics. However, in fact, regardless of whether acceleration and higher-order dynamic characteristics are considered, the processing idea of ​​NUDFT is consistent, and the rotation phase can be given in the form of formula (18). Therefore, the technical solution of the present invention can also process the target echo pulse signal containing high-order dynamic characteristics.

[0128] The random agility of carrier frequency repetition rate proposed in the present invention has the following characteristics: first, the radar receiver and processor do not need accurate a priori information of the carrier frequency repetition rate of the signal transmitted by the radiation source; second, the carrier frequency repetition rate agility does not need to change according to a specific rule; third, the carrier frequency repetition rate agility does not need to remain consistent within the accumulation time of multiple pulse signals, but can be randomly agile between pulses.

[0129] The technical scheme of the present invention is described in detail below in conjunction with the embodiments and drawings.

[0130] Table 1 Example parameters

[0131]

[0132] Based on the specific parameters given in Table 1 above, the steps of the specific implementation are as follows:

[0133] Step 1: Carrier re-frequency identification.

[0134] Step 1.1, perform RF down-conversion processing on the received reference pulse signal. The reference pulse signal is mixed with the received RF and then low-pass filtered to obtain an intermediate frequency reference pulse signal with a residual intermediate frequency range of [-100,100] MHz. Residual intermediate frequency frequency identification can be achieved through fast Fourier transform (FFT). When the reference signal-to-noise ratio is relatively large, the phase difference can also be used to identify the carrier frequency. Due to the complexity of FFT calculation, in order to obtain a high-precision carrier frequency estimate, the idea of ​​combining FFT coarse identification with phase-locked loop (or frequency-locked loop) fine identification can also be used to identify the carrier frequency. In a certain embodiment, the carrier frequency of the first pulse (m=0) is 14.05 GHz. After carrier frequency identification, the residual intermediate frequency of the received reference pulse signal is obtained to be f I =50MHz, then according to f I =50MHz corresponding intermediate frequency word is used to perform digital orthogonal down-conversion processing on the reference pulse signal to obtain a zero intermediate frequency receiving reference pulse signal with an intermediate frequency of 0.

[0135] Step 1.2: Detect the zero intermediate frequency reference pulse signal and calculate the pulse repetition interval T r (i) Perform timing. Perform signal-to-noise ratio detection on the reference pulse signal and calculate the signal-to-noise ratio of the reference signal. When the reference signal-to-noise ratio is greater than the reference pulse signal detection threshold, it is determined to be a reference signal in the form of (12). Record the time interval between adjacent reference signals, i.e., the pulse repetition interval T r (m). In a certain embodiment, T r (0) = 0.09s.

[0136] Step 2, down-conversion processing of echo pulse signal: The radar receives the echo pulse signal, performs down-conversion processing according to the carrier frequency (the carrier frequency is the synthesis of the RF frequency and the residual intermediate frequency) output by the reference pulse signal carrier frequency identification, and obtains the echo pulse signal of the relevant matching input. Since the carrier frequency is known and the RF frequency is fixed, the residual intermediate frequency frequency is equivalently known, so the two-step down-conversion can be completed. So here it is directly expressed as down-conversion according to the carrier frequency.

[0137] Step 2.1, the RF echo received by the radar is down-mixed and filtered according to the receiver carrier frequency 14 GHz to obtain an intermediate frequency received echo pulse signal;

[0138] Step 2.2: Perform digital orthogonal down-conversion on the echo pulse. Taking m = 0 as an example, using f I = The intermediate frequency word corresponding to 50MHz performs digital orthogonal down-conversion on the receiving components in the echo pulse signal receiving area to obtain an echo pulse signal with zero intermediate frequency;

[0139] Step 3: Correlation matching processing.

[0140] Step 3.1: Construct an FIR correlation matched filter with the reference pulse signal sampling value. Extract the sampling value of the reference pulse signal detected in step 1.2. In the embodiment, the reference pulse width T p =25ns, sampling rate f s =300MHz, then the number of reference signal sampling points is 7 or 8 sampling points (depending on the sampling position), and these sampling values ​​are conjugated and sequentially flipped to construct the FIR correlation matching filter coefficients. Taking the correlation matching process of the first pulse signal in the embodiment as an example, the reference pulse signal length is L ref =8, construct a 7th-order FIR correlation matched filter.

[0141] Step 3.2, perform correlation matching processing on the echo pulse signal. Considering that the echo pulse signal is weak and it is difficult to directly detect the echo delay, the echo pulse signal sampling values ​​within the echo pulse signal receiving time or the receiving delay time corresponding to the characteristic area are selected and input into the FIR correlation matching filter constructed in step 3.1 in the sampling order.

[0142] Step 4: Signal peak position detection and gate data buffering. In the embodiment, after the first (m=0) echo pulse signal is correlated and matched, the peak value is searched in the time domain of the received signal and the corresponding distance gate number (i.e., sampling point index number) is selected as k. max = 10018. According to the wave gate data selection method described in step 4 above, Then the gate data within the range of [10010,10026] is cached in DDR (or other cache). At the same time, in order to accumulate the signal later, the carrier frequency f corresponding to the pulse corresponding to the current pulse is m and the pulse repetition interval T r (m) is also cached for standby. The gate data corresponding to the M pulses are cached in sequence.

[0143] Step 5: NUDFT coherent accumulation.

[0144] Step 5.1: Determine the acceleration search value and the speed search value according to the requirements. In the embodiment, the speed search range is [-15, 15] m / s and the acceleration search range is [-0.1, 0.1] m / s. 2 , under the premise of meeting the requirements of formulas (19) and (21), the speed search interval of a certain test is 0.0126m / s and the acceleration search interval is 0.0149m / s 2 .

[0145] Step 5.2: Calculate the rotation phase of the currently searched range gate, velocity and acceleration search values ​​according to formula (18).

[0146] Step 5.3: Calculate the accumulated value of NUDFT according to formula (16) and cache it.

[0147] Step 6: Complete the search of all distance gates, speeds, and acceleration values ​​and select the maximum value. After completing the search of all cached distance gates, speeds, and acceleration values, perform power selection (or signal-to-noise ratio selection) on the signal of the NUDFT coherent accumulation output result to obtain the distance gate, speed, and acceleration corresponding to the maximum accumulated power value. In a certain test of the embodiment, when M=8, the gate selection result is k max =10018, then the distance of the target relative to the radar receiver can be expressed as R = [(k max -L ref )Ts -τ TR ]c / 2=5000m, the velocity search result is v(0)=-9.9978m / s, the acceleration search result is a=0.004m / s 2 As M changes, the root mean square error of velocity and acceleration measurements is as follows: Figure 4 As shown in the figure, when the distance measurement is located at the sampling grid point, the RMSE is 0, and the maximum RMSE is 0.25m. The RMSE at different distances obeys the theoretical results under uniform distribution.

[0148] To compare the NUDFT coherent accumulation signal-to-noise ratio gain of the present invention with the theoretical gain 10log 10 The relationship between M, Figure 5 The change of NUDFT coherent integration signal-to-noise ratio is given when log2M=1, 2, …, 7. It can be seen that the signal-to-noise ratio gain after NUDFT accumulation is basically consistent with the theoretical gain of coherent integration signal-to-noise ratio.

[0149] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A coherent integration method for semi-active pulse radar with random carrier frequency and repetition frequency agility, characterized in that: The specific process is: The first step is to perform carrier frequency identification on the reference pulse signal received by the semi-active pulse radar receiver, perform digital orthogonal down-conversion on the residual intermediate frequency component identified in the carrier frequency identification process to obtain a zero intermediate frequency reference pulse signal, and perform repetition frequency identification on the zero intermediate frequency reference pulse signal; The second step is to down-convert the RF signal within the echo pulse signal reception time using the RF frequency, and then perform digital orthogonal down-conversion processing using the residual intermediate frequency to obtain a zero intermediate frequency echo pulse signal; The third step is to use the zero intermediate frequency reference pulse signal to perform correlation matching processing on the received component of the zero intermediate frequency echo pulse signal; The fourth step is to detect the matched signal according to the set detection threshold and select the target signal according to the peak position; The fifth step is to perform signal coherent accumulation of the target signal using non-uniform Fourier transform based on the repetition frequency identification result; Step 6: Complete the search for distance gates, speed, and acceleration.

2. The coherent integration method of semi-active pulse radar with random agility of carrier frequency and repetition frequency according to claim 1 is characterized in that: The carrier frequency identification in the first step is: the radio frequency echo pulse signal received by the radar receiver is detected at the radio frequency Perform down-conversion processing to obtain the residual intermediate frequency The intermediate frequency echo signal; according to the residual intermediate frequency frequency distribution interval The length of the carrier frequency is determined by the identification method, and the residual intermediate frequency f is obtained. I , Indicates the carrier frequency range; according to the intermediate frequency f I Realize carrier frequency f m Estimation; the zero intermediate frequency reference pulse signal is subjected to re-frequency identification as follows: timing the time interval between two adjacent reference pulse signals to obtain T r The count value of (i) realizes the repetition frequency estimation.

3. The coherent integration method of semi-active pulse radar with random agility of carrier frequency and repetition frequency according to claim 1 is characterized in that: The specific process of the third step is: Zero intermediate frequency reference pulse signal Discrete sampling is performed to obtain from Extract the zero-IF reference signal from Construct a finite impulse response (FIR) filter; Zero intermediate frequency echo pulse signal Discrete sampling is performed to obtain Input it into the FIR filter to complete the relevant matching process and obtain the matching result c m (k).

4. The coherent integration method of semi-active pulse radar with random carrier frequency and repetition frequency agility according to claim 3 is characterized in that: The matching results are:

5. The coherent integration method of semi-active pulse radar with random agility of carrier frequency and repetition frequency according to claim 4 is characterized in that: The specific process of the fourth step is: When the signal c after correlation matching m (k) Reach the detection threshold T h When c m (k) performing peak selection, and determining the peak selection position as the position of the echo pulse signal; Let the distance gate number corresponding to the peak selection position be k=k max When cache [k max -n,k max +n] range, complete the extraction of the target signal and peak alignment cache, n is the number of effective sampling points within the echo pulse width, f s is the sampling rate, T p is the pulse width, Indicates rounding up.

6. The coherent integration method of semi-active pulse radar with random agility of carrier frequency and repetition frequency according to claim 1 is characterized in that: In the fifth step, the non-uniform Fourier transform is used to transform the multiple correlated matched signals c m (k) Perform signal coherent accumulation, and the accumulated result is expressed as: Among them, T s represents the sampling interval, τ(m) represents the time delay of the echo pulse signal, and M is the number of accumulated pulses; φ m is the phase rotation factor.

7. The coherent integration method of semi-active pulse radar with random agility of carrier frequency and repetition frequency according to claim 6, characterized in that: Phase rotation factor φ calculated in non-uniform Fourier transform m Based on the following recursive formula: Among them, f m represents the carrier frequency of the mth pulse, v(m) represents the speed of the target when it is irradiated by the pulse signal of the radiation source, T r (m+1) represents the time interval between the emission of the radiation source pulse, a represents the target's velocity when it is irradiated by the radiation source pulse signal, and c represents the speed of light.

8. The coherent integration method of semi-active pulse radar with random carrier frequency and repetition frequency agility according to claim 1 is characterized in that: The speed search interval is Δv, 9. The coherent integration method of semi-active pulse radar with random agility of carrier frequency and repetition frequency according to claim 8, characterized in that: The acceleration search interval Δa is:

10. The coherent integration method of semi-active pulse radar with random agility of carrier frequency and repetition frequency according to claim 9, characterized in that: The target's speed estimation interval [v min ,v max ]for: in, and They represent the range gate number corresponding to the Mth and 1st echo pulse signals after correlation matching, k p is the number of distance protection units, [a min ,a max ] indicates the range of acceleration.

Images