A nanosecond time difference estimation method combining time difference and phase difference analysis

Abstract

The present invention discloses a nanosecond time difference estimation method integrating time difference and phase difference analysis, belonging to the field of radar target detection signal processing. The present invention respectively performs spectrum analysis and constant false alarm detection on the signals received by two radio detection radar systems; performs cross-correlation operation on the signal matrices of the signals at the same frequency points of the two time-frequency domain signals to obtain a rough time difference estimation value between the two radio detection radar systems; respectively determines the phase differences of the signals at the same frequency points of the two time-frequency domain signals; determines the phase difference values of adjacent frequency points, and combines with the frequency difference values to obtain several accurate time difference estimation values; obtains the final time difference estimation value through filtering processing of multiple accurate time difference estimation values; performs accurate time difference estimation between every two of the signals detected and received by multiple radio detection radar systems to achieve accurate target positioning. The present invention improves the signal-to-noise ratio of the echo signal, improves the detection probability, improves the time delay estimation accuracy, and improves the target positioning accuracy.

Description

Technical Field

[0001] The present invention relates to a nanosecond time difference estimation method that comprehensively analyzes time difference and phase difference, and belongs to the field of radar target detection signal processing. Background Art

[0002] As an important technical means for maintaining national security and achieving victory in the battlefield, radio detection is known as the "fifth-dimensional reconnaissance space" and plays an important role in military intelligence acquisition. Through the organic combination of listening, intercepting, and direction finding, radio detection can effectively and real-time master the intelligence of the enemy army and provide intelligence support for combat operations. Detecting and positioning the frequency-hopping signals emitted by flying targets (such as unmanned aerial vehicles, etc.) or non-cooperative radiation sources in the airspace using radio detection equipment is of great significance in radio reconnaissance.

[0003] In recent years, with the development of information technology, frequency-hopping signals have become a type of communication signal that is key detected. The unique advantages of frequency-hopping communication systems in anti-jamming, anti-interception, anti-attenuation, etc. have been widely applied in civil and military communication systems, such as application scenarios like battlefield reconnaissance, key airspace defense, and air target surveillance.

[0004] Different from active radars that use the method of transmitting electromagnetic wave signals to achieve target detection, radio detection equipment realizes the detection of targets by passively receiving the electromagnetic wave signals emitted by electronic devices on the targets. Since radio detection equipment does not emit signals but only receives signals, it cannot know the time when the electromagnetic wave leaves the target and it is difficult to directly determine the distance between the target and the radio detection radar. Therefore, to achieve positioning and ranging, a radio detection radar system generally requires multiple radio detection observation stations, that is, using multiple radio detection observation stations to simultaneously measure the time difference of arrival of the electromagnetic wave signals emitted by the air target to complete ranging and positioning. The time differences of the electromagnetic wave signals emitted by the target arriving at multiple radio detection observation stations form a hyperboloid (line), and the intersection of multiple hyperboloids (lines) determines the position of the target.

[0005] In the process of detecting airborne electromagnetic wave signals, a radio detection radar system first performs spectral analysis on the received echo signals. If characteristic signals (such as stepped-frequency signals, etc.) are found in the spectrum, it is determined that there are airborne targets. Then, cross-correlation operations are performed on the echo signals received by two observation stations to obtain the time delay difference. Finally, multiple groups of cross-correlation operations are performed on two observation stations to obtain multiple groups of time delay differences, complete the intersection of multiple hyperbolas, and complete the positioning of the target in the air. In the process of the radio detection radar system detecting airborne electromagnetic wave signals, due to the limitations of the bandwidth and signal-to-noise ratio of the received echo signals, there is a problem of low target positioning accuracy. For the frequency-hopping echo signals, only cross-correlation operations are used to estimate the time delay difference to complete the positioning, and its target positioning accuracy is far from the optimal value. Therefore, it is urgent to make the most effective use of the frequency-hopping echo signals received by the radio detection radar system, improve the time delay estimation accuracy, and improve the positioning accuracy of the radio detection radar system for airborne targets. Summary of the Invention

[0006] The purpose of the present invention is to provide a nanosecond-level time delay estimation method that comprehensively analyzes time difference and phase difference, uses cross-correlation operations of frequency-hopping signals and phase differences of adjacent frequency points to improve the signal-to-noise ratio of echo signals, improve the detection probability, combines the phase differences of cross-correlation peaks of adjacent frequency points to obtain the time delay, and realizes nanosecond-level accurate estimation of the time difference between two radio detection radar systems, avoids time delay estimation ambiguity, and further improves the time delay estimation accuracy.

[0007] The purpose of the present invention is achieved through the following technical solutions:

[0008] A nanosecond-level time delay estimation method that comprehensively analyzes time difference and phase difference of the present invention accurately estimates the time difference between two radars at the nanosecond level through cross-correlation operations of frequency-hopping signals and phase differences of adjacent frequency points: when the radio detection radar system detects an effective target, spectral analysis and constant false alarm detection are respectively performed on the signals received by the two radio detection radar systems; cross-correlation operations are performed on the signal matrices of the same frequency points of the two time-frequency domain signals to obtain a rough estimate value of the time difference between the two radio detection radar systems; the phase differences of the signals of the same frequency points of the two time-frequency domain signals are respectively determined, that is, the phases of the peaks corresponding to the frequency points obtained by cross-correlation operations; the phase difference values of adjacent frequency points are determined, and combined with the frequency difference values, several accurate time delay estimation values are obtained; the final time delay estimation value is obtained by filtering multiple accurate time delay estimation values.

[0009] A nanosecond-level time delay estimation method that comprehensively analyzes time difference and phase difference of the present invention includes the following steps:

[0010] Step 1: Perform spectral analysis and constant false alarm detection on the echo signals detected by the two radio detection radar systems:

[0011] Perform spectral analysis on the echo signal based on the Fourier transform to screen out the effective echo signal. If there is a characteristic signal in the echo signal spectrum matrix, it is determined that there is an airborne target. Perform constant false alarm detection on the effective echo signal to reduce the amount of data processing. The data of the echo signal spectrum matrix that exceeds the constant false alarm detection threshold is retained, and the data of the echo signal spectrum matrix that does not exceed the constant false alarm detection threshold is set to zero. Specifically, it includes the following sub-steps:

[0012] Step 1.1: Perform Fourier transform on the echo signal:

[0013] Perform Fourier transform on the echo signals s 01 (m) and s 02 (m) detected by two radio detection radar systems respectively, and obtain the time-frequency domain matrices S 01 (m) and S 02 (m) of the echo signals s 01 (n, f) and S 02 (n, f);

[0014] Step 1.2: Discriminate the characteristic signal of the time-frequency domain matrix:

[0015] If there are characteristic signals in both the time-frequency domain matrices S 01 (n, f) and S 02 (n, f), it is determined that there is an airborne target and the effective echo signal is obtained. If there is no characteristic signal in the time-frequency domain matrix S 01 (n, f) or S 02 (n, f), continue the detection and go to Step 1.1;

[0016] The characteristic signal mentioned above refers to the signal with the property of frequency hopping, including: stepped frequency signal, frequency hopping signal, etc.;

[0017] Step 1.3: Perform constant false alarm detection on the time-frequency domain matrix of the effective echo signal:

[0018] Determine the constant false alarm detection threshold according to the radio detection radar system and the influence of environmental noise. The data of the time-frequency domain matrices S 01 (n, f) and S 02 (n, f) of the effective echo signal that exceed the constant false alarm detection threshold are retained, and the data of the time-frequency domain matrices S 01 (n, f) and S 02 (n, f) of the effective echo signal that do not exceed the constant false alarm detection threshold are set to zero, and the time-frequency domain matrices S 1 (n, f) and S 2 (n, f) of the effective echo signal after constant false alarm detection are obtained;

[0019] Step 2: Perform a cross-correlation operation on the time-frequency domain signal matrix of the effective echo signals obtained after the constant false alarm detection in Step 1:

[0020] Perform a time-frequency domain conjugate multiplication on the signal matrices at the same frequency points in the time-frequency domain signal matrix S 1 (n, f) and S 2 (n, f); perform an inverse Fourier transform on the time-frequency domain signal matrix obtained by the time-frequency domain conjugate multiplication to obtain a rough estimate of the time difference between the two radio detection radar systems; specifically, it includes the following sub-steps:

[0021] Step 2.1: Decompose and correspond the signal matrices at the same frequency points in the time-frequency domain signal matrix of the effective echo signals obtained after the constant false alarm detection in Step 1:

[0022] Decompose the time-frequency domain signal matrix S 1 (n, f) and S 2 (n, f) into multiple single-frequency point signal matrices S 1i (n i , f i ), i = 1, 2, 3,..., k and S 2i (n i , f i ), i = 1, 2, 3,..., k, where n i is the sequence corresponding to the same frequency point f i ;

[0023] Step 2.2: Perform a time-frequency domain conjugate multiplication on the single-frequency point signal matrices at the same frequency points decomposed in Step 2.1:

[0024] Perform a time-frequency domain conjugation on the single-frequency point signal matrix S 1i (n i , f i ) to obtain a conjugate single-frequency point signal matrix Multiply the corresponding position elements of the conjugate single-frequency point signal matrix and the single-frequency point signal matrix S 2i (n i , f i ) to obtain a single-frequency point matched filtering signal matrix S 12i (n i , f i ), as shown in Equation (1):

[0025]

[0026] Step 2.3: Perform an inverse Fourier transform on the single-frequency point matched filtering signal matrix to obtain a rough estimate of the time difference between the two radio detection radar systems:

[0027] Inverse Fourier transform S 12i (n i ,f i ) to obtain sequence r 12i (n i ), and the rough time difference estimation value tc between two radio detection radar systems i is the time corresponding to the peak of sequence r 12i (n i );

[0028] Step 3: Determine the phase corresponding to the peak of the sequence obtained in Step 2 to obtain the phase difference of the effective echo signals at the same frequency points of the two signals; determine the difference between the phase differences of the effective echo signals at adjacent frequency points, and combine the frequency difference to determine several accurate time difference estimation values; obtain the accurate time difference result by performing smoothing filtering on multiple accurate time difference estimation values; specifically, it includes the following sub-steps:

[0029] Step 3.1: Determine the phase of the peak corresponding to the cross-correlation operation at the corresponding frequency point:

[0030] Determine the phase of the peak of sequence r 12i (n i ), i = 1, 2, 3,..., k as shown in Equation (2): As shown in Equation (2):

[0031]

[0032] where n i-max is the abscissa value corresponding to the maximum value of sequence r 12i (n i ), i = 1, 2, 3,..., k, and angle(*) is to determine the phase corresponding to the maximum value;

[0033] Step 3.2: Determine the accurate time difference estimation value:

[0034] Determine the phase difference between adjacent frequency points As shown in Equation (3):

[0035]

[0036] Combine the frequency difference Δf i to determine the accurate time difference estimation value ta i , as shown in Equation (4):

[0037]

[0038] where the frequency difference Δf i is the frequency difference between adjacent frequency points, corresponding to the phase difference between adjacent frequency points, as shown in Equation (5):

[0039] Δf i = f i+1 - f i i = 1, 2, 3, ..., k - 1 (5)

[0040] The time difference estimation accuracy Δt based on the phase difference is shown in Equation (6):

[0041]

[0042] Step 3.3: Determine the final estimated value of the time difference:

[0043] By filtering multiple sets of accurate estimated values of the time difference during the reception and processing of multiple effective echo signals by the radio detection radar system, the final estimated value of the time difference is determined;

[0044] Step Four: Based on Steps One to Three, on the basis of accurately estimating the time difference between the received signals detected by two radio detection radar systems, accurately estimate the time difference between the received signals detected by multiple radio detection radar systems pairwise, determine multiple hyperbolas in space, and accurately locate the target through the intersection positions of the multiple hyperbolas.

[0045] Beneficial effects:

[0046] 1. A nanosecond-level time difference estimation method that comprehensively analyzes the time difference and phase difference of the present invention improves the signal-to-noise ratio of the echo signal, the detection probability, and the time difference estimation accuracy by performing cross-correlation operations on two effective echo signals;

[0047] 2. A nanosecond-level time difference estimation method that comprehensively analyzes the time difference and phase difference of the present invention, on the basis of performing cross-correlation operations on two effective echo signals, accurately estimates the time difference between two radio detection radar systems by determining the phase difference between adjacent frequency points and combining the frequency difference between adjacent frequency points, avoids time difference estimation ambiguity, and improves the time difference estimation accuracy;

[0048] 3. A nanosecond-level time difference estimation method that comprehensively analyzes the time difference and phase difference of the present invention, on the basis of accurately estimating the time difference between the received signals detected by two radio detection radar systems, can achieve accurate target positioning by accurately estimating the time difference between the received signals detected by multiple radio detection radar systems. Description of the Drawings

[0049] Figure 1 is a schematic diagram of a radio detection radar system;

[0050] Figure 2 is a flowchart of a nanosecond-level time difference estimation method that comprehensively analyzes the time difference and phase difference of the present invention;

[0051] Figure 3It is the spectrogram of the received signals of two radio detection radar systems in a nanosecond time difference estimation method that comprehensively analyzes time difference and phase difference according to the present invention;

[0052] Among them, Figure 3 (a) is the time-frequency diagram of the received signal of the No. 1 radio detection radar system, Figure 3 (b) is the time-frequency diagram of the received signal of the No. 2 radio detection radar system;

[0053] Figure 4 It is a schematic diagram of the accurate time difference estimation result;

[0054] Figure 5 It is a schematic diagram of the cross-location of a target by multiple radio detection radar systems. Specific implementation manner

[0055] To better illustrate the purpose and advantages of the present invention, the following further describes the content of the invention in conjunction with the accompanying drawings and examples.

[0056] Example 1:

[0057] In the example, the parameters of the waveform emitted by the airborne target radiation source are as follows: the baseband bandwidth of the signal is 10 MHz, the carrier frequency is changed every 10 ms, and the carrier frequency is uniformly jumped from 30 MHz to 70 MHz, that is, 30 MHz, 40 MHz, 50 MHz, 60 MHz, 70 MHz, the sampling rate is 100 MHz, and the duration of each pulse is 50 ms;

[0058] As Figure 1 shown, the multi-station radio detection radar air detection system of the example includes three radio detection radar receiving systems, namely: receiving station 1, receiving station 2, and receiving station 3, and an airborne target (drone) radiation source. By accurately estimating the time difference between every two of the three radio detection radar receiving systems, multiple hyperbolas are determined in space, and the target is accurately located through the intersection position of the hyperbolas.

[0059] Based on MATLAB simulation analysis, both receiving station 1 and receiving station 2 can completely receive the radio signals emitted by the airborne target (drone) radiation source. The signal of the airborne target (drone) radiation source received by receiving station 2 is delayed by 55.4567 ns relative to the signal of the airborne target (drone) radiation source received by receiving station 1, and 5 pulse signals are collected;

[0060] As Figure 2 shown, the example applies a nanosecond time difference estimation method that comprehensively analyzes time difference and phase difference according to the present invention to accurately estimate the time difference of the echo signal of the airborne target (drone) radiation source, including the following steps:

[0061] Step 1: Conduct spectral analysis and constant false alarm detection on the echo signals detected by two radio detection radar systems:

[0062] Based on Fourier transform, conduct spectral analysis on the echo signals to screen out valid echo signals. If there are characteristic signals in the echo signal spectrum matrix, it is determined that there are aerial targets; conduct constant false alarm detection on the valid echo signals to reduce the data processing volume. The data of the echo signal spectrum matrix that exceeds the constant false alarm detection threshold is retained, and the data of the echo signal spectrum matrix that does not exceed the constant false alarm detection threshold is set to zero. Specifically, it includes the following sub-steps:

[0063] Step 1.1: Conduct Fourier transform on the echo signals:

[0064] Conduct Fourier transform on the echo signals s 01 (m) and s 02 (m) detected by the two radio detection radar systems respectively, and obtain the time-frequency domain matrices S 01 (m) and S 02 (m) of the echo signals s 01 (n,f) and S 02 (n,f);

[0065] Step 1.2: Conduct discrimination of characteristic signals on the time-frequency domain matrix:

[0066] If there are characteristic signals in both the time-frequency domain matrices S 01 (n,f) and S 02 (n,f), it is determined that there are aerial targets and the valid echo signals are obtained; if there are no characteristic signals in the time-frequency domain matrix S 01 (n,f) or S 02 (n,f), continue the detection and go to Step 1.1;

[0067] In the embodiment, the characteristic signal of the time-frequency domain matrix of the echo signal is a stepped frequency signal;

[0068] The time-frequency diagrams of the echo signals detected by receiving station 1 and receiving station 2 are as Figure 3 shown, Figure 3 (a) is the time-frequency diagram of the echo signal detected by receiving station 1, Figure 3 (b) is the time-frequency diagram of the echo signal detected by receiving station 2. Both receiving station 1 and receiving station 2 have completely received the echo signals;

[0069] Step 1.3: Conduct constant false alarm detection on the time-frequency domain matrix of the valid echo signals:

[0070] Determine the constant false alarm detection threshold according to the radio detection radar system and environmental noise influence. The time-frequency domain matrix S 01(n, f) and S 02 The time-frequency domain matrix S of the effective echo signal whose (n, f) data is retained and does not exceed the constant false alarm detection threshold 01 (n, f) and S 02 The (n, f) data is set to zero to obtain the time-frequency domain matrix S of the effective echo signal after constant false alarm detection 1 (n, f) and S 2 (n, f);

[0071] Step 2: Perform a cross-correlation operation on the time-frequency domain signal matrix of the effective echo signal obtained after the constant false alarm detection in Step 1:

[0072] The time-frequency domain signal matrix S of the effective echo signal obtained after the constant false alarm detection in Step 1 1 (n, f) and S 2 The signal matrices at the same frequency points in (n, f) and S are multiplied conjugately in the time-frequency domain; perform an inverse Fourier transform on the time-frequency domain signal matrix obtained by the time-frequency domain conjugate multiplication to obtain a rough estimate of the time difference between two radio detection radar systems; specifically including the following sub-steps:

[0073] Step 2.1: Decompose and correspond the time-frequency domain signal matrix of the effective echo signal obtained after the constant false alarm detection in Step 1 at the same frequency points:

[0074] Decompose the time-frequency domain signal matrix S of the effective echo signal 1 (n, f) and S 2 (n, f) into multiple single-frequency point signal matrices S 1i (n i , f i ), i = 1, 2, 3,..., k and S 2i (n i , f i ), i = 1, 2, 3,..., k, n i is the sequence corresponding to the same frequency point f i ;

[0075] As Figure 3 shown: The carrier frequency of each echo signal in 5 segments of signals jumps uniformly from 30 MHz to 70 MHz. The first pulse signal matrix S 1 (n, f) of receiving station 1 and the first pulse signal matrix S 2 (n, f) of receiving station 2 are respectively decomposed into 5 single-frequency point signal matrices S 1i (n i , f i ), S 2i (n i , f i ), where f i={30, 40, 50, 60, 70} MHz, i = 1, 2, 3, 4, 5, and the single frequency points are 30 MHz, 40 MHz, 50 MHz, 60 MHz, and 70 MHz respectively;

[0076] Step 2.2: Perform time-frequency domain conjugate multiplication on the single frequency point signal matrices of the same frequency points decomposed in Step 2.1:

[0077] For the single frequency point signal matrix S 1i (n i , f i ), perform time-frequency domain conjugation to obtain the conjugate single frequency point signal matrix Multiply the corresponding elements of the conjugate single frequency point signal matrix with the single frequency point signal matrix S 2i (n i , f i ) to obtain the single frequency point matched filtering signal matrix S 12i (n i , f i ), as shown in Equation (7):

[0078]

[0079] Step 2.3: Perform inverse Fourier transform on the single frequency point matched filtering signal matrix to obtain a rough estimate of the time difference between the two radio detection radar systems:

[0080] Perform inverse Fourier transform on S 12i (n i , f i ) to obtain the sequence r 12i (n i ). The rough estimate of the time difference tc i between the two radio detection radar systems is the time corresponding to the peak of the sequence r 12i (n i );

[0081] In the embodiment, the rough estimates of the time differences corresponding to the 5 peaks obtained from the 5 single frequency point signal matrices are all 50 ns;

[0082] Step Three: Determine the phase corresponding to the peak of the sequence obtained in Step Two to obtain the phase difference of the effective echo signals of the same frequency points of the two signals; determine the difference between the phase differences of the effective echo signals of adjacent frequency points, and combine the frequency differences to determine several accurate estimates of the time difference; obtain the accurate result of the time difference through smoothing filtering of multiple accurate estimates of the time difference; specifically, it includes the following sub-steps:

[0083] Step 3.1: Determine the phase of the peak corresponding to the cross-correlation operation for the corresponding frequency point:

[0084] Determine the sequence r12i (n i ), for i = 1, 2, 3, ..., k, the phase of the peak As shown in Equation (8):

[0085]

[0086] For the first pulse, the five phases are -1.0205294 radians, -1.3633871 radians, -1.7062815 radians, -2.0491640 radians, and -2.3920073 radians respectively;

[0087] Step 3.2: Determine the accurate estimated value of the time difference:

[0088] Determine the phase difference between adjacent frequency points As shown in Equation (9):

[0089]

[0090] In the embodiment, for the first pulse, the four phase differences are: -0.3428577 radians, -0.3428944 radians, -0.3428824 radians, and -0.3428433 radians respectively;

[0091] Combined with the frequency difference Δf i , determine the accurate estimated value of the time difference ta i , as shown in Equation (10):

[0092]

[0093] Among them, the frequency difference Δf i is the frequency difference value between adjacent frequency points, corresponding to the phase difference between adjacent frequency points;

[0094] In the embodiment, the frequency difference Δf i is all 10 MHz. Through the phase difference and the frequency difference Δf i , several accurate estimated values of the time difference ta i ={55.4567, 55.4573, 55.4571, 55.4565} ns, i = 1, 2, 3, 4;

[0095] The time difference estimation accuracy Δt based on the phase difference is as shown in Equation (11):

[0096]

[0097] In the embodiment, the estimation accuracy is 1% of the wavelength, Δf is 10 MHz, then the time difference estimation accuracy Δt is 1 ns;

[0098] For comparison, the time difference estimation accuracy of only performing the cross-correlation operation is as shown in Equation (12):

[0099]

[0100] where Δt s is the time difference estimation accuracy, ΔT is the time width of the cross-correlation peak, SNR is the signal-to-noise ratio of the echo signal, and here the time-bandwidth product ΔT·B of the signal is set to 1.

[0101] In the embodiment, the signal bandwidth B of each frequency point is 10 MHz, the corresponding time width of the cross-correlation peak ΔT is 0.1 μs, and the SNR is 50, then the time difference estimation accuracy Δt s is 10 ns;

[0102] The time difference estimation accuracy of the method of the present invention is higher than that of the method of only performing the cross-correlation operation;

[0103] Step 3.3: Determine the final estimated value of the time difference:

[0104] By filtering multiple groups of accurate time difference estimation values during the reception and processing of effective echo signals of the radio detection radar system, the final estimated value of the time difference is determined;

[0105] In the embodiment, through simulation, receiving station 1 and receiving station 2 respectively receive 5 pulse signals, the frequency-hopping signals at 5 frequency points of a single pulse, perform cross-correlation operations and combine the phase differences of adjacent frequency points, and can determine the accurate time difference estimation value between receiving station 1 and receiving station 2 4 times. The accurate time difference estimation value is at the nanosecond level, and a total of 20 accurate time difference estimation values are obtained;

[0106] In the embodiment, the filtering processing method adopts Kalman filtering to correct multiple accurate time difference estimation values at the nanosecond level, as Figure 4 shown. The error of the final estimated value of the time difference after Kalman filtering is smaller than the error of the accurate time difference estimation value at the nanosecond level, and the final estimated value of the time difference is more accurate;

[0107] Step Four: Based on Steps One to Three, further accurately estimate the time difference between receiving station 1 and receiving station 3, and accurately estimate the time difference between receiving station 2 and receiving station 3. By accurately estimating the time difference between two stations pairwise among the three stations, two curves are determined in the two-dimensional space, as Figure 5 shown. The target is accurately located through the intersection position of the two curves.

[0108] The specific description above further elaborates on the purpose, technical solution, and beneficial effects of the invention. It should be understood that the above is only a specific embodiment of the present invention and is not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A nanosecond time difference estimation method based on comprehensive analysis of time difference and phase difference, characterized in that: The time difference between two radars is accurately estimated at the nanosecond level through the cross-correlation operation of the frequency-hopping signals and the phase difference between adjacent frequency points: When the radio detection radar system detects an effective target, the signals received by the two radio detection radar systems are respectively subjected to spectrum analysis and constant false alarm detection; the signal matrices of the same frequency points in the two time-frequency domain signals are subjected to cross-correlation operation to obtain a rough estimate value of the time difference between the two radio detection radar systems; The phase differences of the signals of the same frequency points in the two time-frequency domain signals are respectively determined, that is, the phases of the peaks corresponding to the corresponding frequency points obtained by the cross-correlation operation; the phase difference values between adjacent frequency points are determined, and combined with the frequency difference values, several accurate time difference estimation values are obtained; the final time difference estimation value is obtained by filtering the multiple accurate time difference estimation values, and specifically includes the following steps: Step 1: Perform spectrum analysis and constant false alarm detection on the echo signals detected by the two radio detection radar systems: Based on the Fourier transform, spectrum analysis is performed on the echo signals to screen out effective echo signals. If there are characteristic signals in the echo signal spectrum matrix, it is judged that there are air targets; constant false alarm detection is performed on the effective echo signals to reduce the amount of data processing. The data of the echo signal spectrum matrix that exceeds the constant false alarm detection threshold is retained, and the data of the echo signal spectrum matrix that does not exceed the constant false alarm detection threshold is set to zero; specifically includes the following sub-steps: Step 1.1: Perform Fourier transform on the echo signals: The echo signals s 01 (m) and s 02 (m) detected by two radio detection radar systems are respectively subjected to Fourier transform, and the time-frequency domain matrices S 01 (n,f) and S 02 (m) of the echo signals s 01 (n,f) and S 02 (n,f) are respectively obtained; Step 1.2: Discriminate characteristic signals on the time-frequency domain matrix: If the time-frequency domain matrix S 01 (n,f) and S 02 (n,f) both contain characteristic signals, it is determined that there is an airborne target, and an effective echo signal is obtained; if the time-frequency domain matrix S 01 (n,f) or S 02 (n,f) does not contain characteristic signals, continue detection and go to step 1.1; Step 1.3: Perform constant false alarm detection on the time-frequency domain matrix of the effective echo signals: Determine the constant false alarm detection threshold according to the radio detection radar system and environmental noise impact. Retain the time-frequency domain matrix S 01 (n,f) and S 02 (n,f) data. Zero out the time-frequency domain matrix S 01 (n,f) and S 02 (n,f) data of the effective echo signal that does not exceed the constant false alarm detection threshold, and obtain the time-frequency domain matrix S 1 (n,f) and S 2 (n,f); Step 2: Perform cross-correlation operation on the time-frequency domain signal matrices of the effective echo signals obtained after the constant false alarm detection in Step 1: The time-frequency domain signal matrix S of the effective echo signal obtained after the constant false alarm detection in Step 1 1 (n,f) and the signal matrix at the same frequency point in S 2 (n,f) are subjected to time-frequency domain conjugate multiplication; the time-frequency domain signal matrix obtained by the time-frequency domain conjugate multiplication is subjected to inverse Fourier transform to obtain a rough estimate value of the time difference between two radio detection radar systems; specifically, it includes the following sub-steps: Step 2.1: Decompose and correspond the signal matrices of the same frequency points of the time-frequency domain signal matrices of the effective echo signals obtained after the constant false alarm detection in Step 1: The time-frequency domain signal matrix S of the effective echo signal 1 (n, f) and S 2 (n, f) are respectively decomposed into multiple single-frequency point signal matrices S 1i (n i , f i ), i = 1, 2, 3,..., k and S 2i (n i , f i ), i = 1, 2, 3,..., k, where n i is the sequence corresponding to the same frequency point f i ; Step 2.2: Perform time-frequency domain conjugate multiplication on the single-frequency point signal matrices of the same frequency points decomposed in Step 2.1: For the single-frequency point signal matrix S 1i (n i , f i ), perform time-frequency domain conjugation to obtain the conjugate single-frequency point signal matrix Multiply the conjugate single-frequency point signal matrix element by element at the corresponding positions with the single-frequency point signal matrix S 2i (n i , f i ) to obtain the single-frequency point matched filtering signal matrix S 12i (n i , f i ), as shown in Equation (1): Step 2.3: Perform inverse Fourier transform on the single-frequency point matched filtering signal matrix to obtain a rough estimate value of the time difference between the two radio detection radar systems: Inverse Fourier transform of S 12i (n i ,f i ) gives sequence r 12i (n i ). The rough estimate of the time difference tc between two radio detection radar systems i is the time corresponding to the peak of sequence r 12i (n i ); Step 3: Determine the phase corresponding to the peak of the sequence obtained in Step 2 to obtain the phase difference between the effective echo signals of the same frequency points of the two signals; determine the difference between the phase differences of the effective echo signals of adjacent frequency points, and combine the frequency differences to determine several accurate time difference estimation values; the accurate time difference result is obtained by performing smoothing filtering on the multiple accurate time difference estimation values; specifically includes the following sub-steps: Step 3.1: Determine the phase of the peak corresponding to the corresponding frequency point obtained by the cross-correlation operation: Determine sequence r 12i (n i ), the phases of the peaks of i = 1, 2, 3, ..., k As shown in Equation (2): where n i-max is the abscissa value corresponding to the maximum value of sequence r 12i (n i ), i = 1, 2, 3,..., k, and angle(*) is the phase corresponding to the determined maximum value; Step 3.2: Determine the accurate time difference estimation value: Determine the phase difference between adjacent frequency points As shown in Equation (3): Combined frequency difference Δf i , determine the precise estimated value ta of the time difference i , as shown in Equation (4): Among them, the frequency difference Δf i is the frequency difference value between adjacent frequency points, and corresponds to the phase difference between adjacent frequency points, as shown in Equation (5): Δf i = f i+1 - f i where i = 1, 2, 3, ..., k - 1 (5) The time difference estimation accuracy Δt based on the phase difference is shown in Equation (6): Step 3.3: Determine the final time difference estimation value: The final time difference estimation value is determined by filtering multiple groups of accurate time difference estimation values during multiple effective echo signal reception processes of the radio detection radar system; Step 4: Based on Steps 1 to 3, on the basis of accurately estimating the time difference of the detected received signals of two radio detection radar systems, accurately estimate the time difference between the detected received signals of multiple radio detection radar systems pairwise, determine multiple hyperbolas in space, and accurately locate the target through the intersection positions of the multiple hyperbolas.

2. The nanosecond time difference estimation method for comprehensive time difference and phase difference analysis according to claim 1, characterized in that: the characteristic signal refers to a signal with the property of frequency hopping, including: stepped frequency signal and frequency hopping signal.

Images