Improved nyfr structure and signal estimation method

Abstract

This invention discloses an improved NYFR structure and signal estimation method. The improved NYFR structure uses a bandwidth RF filter as input to intercept the non-cooperative radar signal x(t); the LO generates a non-uniform RF sampling pulse train p at the zero-crossing rising edge (ZCR) of a periodic chirp modulated signal. LO (t); the signal x(t) after passing through the ultra-wideband RF filter and the non-uniform RF sampling pulse train p with periodic chirp modulation. LO (t) After mixing, the signal y(t) is obtained through a low-pass filter; the signal y(t) is then discretized / digitized by an ADC to obtain the output discrete signal y(n); the RF sampling clock sends synchronization timing information T0 to the output signal processing module at the start of LO operation; the output signal processing module combines the discrete signal y(n) and the synchronization timing information T0 to achieve ultra-wideband reception and processing of non-cooperative radar signals that are insensitive to output signal truncation. This invention improves the NZ label estimation of the NYFR structure, making it insensitive to output signal truncation, resulting in better broadband reception robustness and enhancing the broadband reconnaissance effectiveness of NYFR.

Description

Technical Field

[0001] This invention belongs to the field of electronic reconnaissance technology, specifically relating to a novel Nyquist folding receiver (NYFR) structure. Compared to the prototype NYFR structure, this structure is insensitive to NYFR output signal truncation in non-cooperative reception, eliminating the impact of NYFR output signal truncation during signal detection on Nyquist zone (NZ) label estimation, thus achieving robust interception and efficient processing of unknown radar signals over an ultra-wideband frequency range. This invention also relates to a signal estimation method based on the aforementioned improved Nyquist folding receiver structure. Background Technology

[0002] Currently, the operating frequency bands used by radar systems have become increasingly wider. In radar reconnaissance, to ensure a high probability of intercepting non-cooperative radar signals in ultra-wideband frequencies, radar reconnaissance receivers need to have ultra-wide instantaneous monitoring bandwidth. Constrained by the Nyquist sampling theorem, existing constant-scale-to-digital converters (ADCs) cannot meet the ultra-wideband reception requirements of radar reconnaissance. Therefore, new technologies are needed to overcome the limitations of the Nyquist sampling theorem and achieve instantaneous ultra-wideband reception even with low-speed ADCs.

[0003] NYFR is a novel ultra-wideband receiver architecture that utilizes a monolithic ADC to achieve high-probability interception of ultra-wideband signals. NYFR employs RF analog non-uniform sampling and low-speed ADC sampling to achieve wideband signal interception. In the prototype NYFR architecture, the non-uniform sampling of the RF local oscillator (LO) uses sinusoidal frequency modulation (SFM). Therefore, the modulation component of the NYFR output signal becomes a composite of the intercepted signal modulation and SFM. In the added SFM component of the output composite signal, the NZ label is used to identify the original carrier frequency of the intercepted signal. Since the carrier frequency of the intercepted signal is unknown, the NZ label of the output signal is also unknown. By estimating the NZ label, the intercepted non-cooperative radar signal can be reconstructed.

[0004] In practical electronic reconnaissance, the signals intercepted by radar reconnaissance receivers in the radio frequency space are real signals. In this case, the NZ label in the NYFR output signal will contain both positive and negative values. In the prototype NYFR structure, the estimated value of the NZ label is affected by the positive or negative coefficient of the SFM component added to the output signal. However, in non-cooperative reception, due to limitations in the detection algorithm performance and unknown time of arrival (TOA) and pulse width (PW), the NYFR output data may experience signal truncation due to TOA estimation errors. In this case, the coefficients of the SFM component in the actual NYFR output signal will differ from the theoretical values. This means that the NZ label estimation is incorrect, leading to de-LO modulation failure of the output signal, making it impossible to recover the unknown radar signal intercepted by the NYFR, and causing radar reconnaissance to fail.

[0005] Therefore, in response to the problem that the NZ label estimation of the prototype NYFR structure is sensitive to the truncation of the output signal in radar reconnaissance, this invention proposes an improved receiving structure to overcome the above problem, realize robust non-cooperative reception that is not sensitive to the truncation of the output signal, and improve the reconnaissance receiver's reconnaissance processing capability. Summary of the Invention

[0006] This invention proposes an improved technique to overcome the sensitivity of NZ label estimation to output signal truncation in the prototype NYFR structure. This improves the robustness of NYFR broadband reception by eliminating the impact of NYFR output signal truncation on NZ label estimation. Taking a pulsed single-carrier radar signal intercepted by NYFR as an example, the invention combines the proposed periodic chirp modulation (LO) and synchronization timing structure to transform the positive / negative estimation of the output signal's NZ label into an estimation of the frequency modulation slope of the output periodic chirp modulation. This achieves robust NYFR output signal NZ label estimation that is insensitive to truncation.

[0007] The present invention adopts the following technical solution:

[0008] An improved NYFR structure includes a bandwidth RF filter, an RF sampling clock, a low-pass filter, an ADC, and an output signal processing module; the bandwidth RF filter receives the intercepted non-cooperative radar signal x(t); the LO (local oscillator) generates a non-uniform RF sampling pulse train p at the zero-crossing rising edge (ZCR) of a periodic chirp (linear frequency modulation) modulated signal. LO (t); the signal x(t) after passing through the ultra-wideband RF filter and the non-uniform RF sampling pulse train p with periodic chirp modulation. LO (t) After mixing, the signal y(t) is obtained by passing through a low-pass filter; the signal y(t) is then processed by the ADC for discrete / digital processing (here, / means or), resulting in the output discrete signal y(n); the RF sampling clock sends the synchronization timing information T0 to the output signal processing module at the start of LO operation; the output signal processing module combines the discrete signal y(n) and the synchronization timing information T0 to achieve ultra-wideband reception and processing of non-cooperative radar signals that are insensitive to output signal truncation.

[0009] This invention also discloses a signal estimation method, which is performed according to the following steps:

[0010] S1. Construct an improved NYFR structure with periodic chirp modulation LO and synchronous timing as described above;

[0011] S2. Based on the improved NYFR structure in step S1, intercept the non-cooperative single-frequency pulse radar signal and obtain the composite signal output by the NYFR.

[0012] S3. Based on periodic chirp modulation LO and synchronization timing information, construct an alignment search signal containing NZ search index modulation, wherein the NZ search index range includes all NZ of the instantaneous monitoring video segment and has positive and negative values;

[0013] S4. Using the search signal containing all NZ labels constructed in step S3, perform deLO modulation and fast Fourier transform, and use the maximum value of the peak value of the spectrum of the deLO-modulated signal corresponding to each NZ search label to obtain the NZ label estimation result of the NYFR output signal.

[0014] S5. Using the NZ label estimated in step S4, perform deLO modulation and FFT on the complex signal output by the NYFR structure. Estimate the carrier frequency of the folded signal by the frequency corresponding to the maximum value of the spectrum of the deLO-modulated signal, and further calculate the carrier frequency of the intercepted unknown radar signal to complete the signal estimation.

[0015] Preferably, as an improved NYFR structure, step S1 specifically includes the following: constructing the improved NYFR, such as... Figure 1 As shown, the LO modulation θ(t) is a periodic chirp, and the modulation model is as follows:

[0016] θ(t)=πμ0mod(t,T LO ) 2 (1)

[0017] Where μ0 is the modulation frequency of the periodic chirp modulation, and T LO is the modulation period of the periodic chirp, and mod(·) is the modulo operation. Figure 1 In the improved NYFR structure shown, x(t) represents the intercepted non-cooperative radar signal, ZCR represents the time of the zero-crossing rising edge, and f LO For the carrier frequency of LO, t k At the zero-crossing rising edge, p(t) is the sampling pulse train template. For simplified analysis, the sampling pulse train template is assumed to be a unit impulse function, i.e., p(t) = δ(t). y(t) is the intercepted signal x(t) after passing through the ultra-wideband RF filter and the non-uniform RF sampling pulse train p with periodic chirp modulation. LO (t) The result after mixing and passing through a low-pass filter, where the cutoff frequency of the low-pass filter is f. LO / 2. Subsequently, the signal y(t) is discretized / digitized by a low-speed ADC to obtain the improved NYFR output discrete signal y(n). Simultaneously, at the start of LO operation, synchronization timing information T0 is sent to the output signal processing module. Combining the discrete signal y(n) and the synchronization timing information T0, this invention enables ultra-wideband reception and processing of non-cooperative radar signals that are insensitive to output signal truncation.

[0018] Preferably, step S2 is as follows: the intercepted non-cooperative single-frequency pulse radar signal is

[0019] x(t)=Acos(2πf c t)[ε(t)-ε(tT)] (2)

[0020] Where A is the amplitude, f c The signal carrier frequency is denoted by ε, which is the parameter to be estimated in the non-cooperative signal. ε(·) is the step function, and T is the pulse width. After interception, the NYFR outputs a composite signal.

[0021] y(t)=Acos[2π|f c -k NZ f LO |t-sgn(f c -k NZ f LO )k NZ πμ0mod(t,T LO ) 2 ][ε(t)-ε(tT)](3)

[0022] Where, k NZ =round(f c / f LO ), sgn(·) is the sign function, and the NZ label is defined as sgn(f c -k NZ f LO )k NZ For non-cooperative reception, the TOA estimation may be inaccurate, resulting in truncation of the NYFR output signal. In this case, equation (3) will become...

[0023]

[0024] Among them, T TOA The known TOA time is output by NYFR after detection.

[0025] Preferably, step S3 is as follows: the constructed search signal is

[0026]

[0027] Where n is an integer, T ADC The ADC sampling interval is k, and the NZ search index is k. s ∈[-L,L],k s ∈Z, where L is the total number of NZs monitored by NYFR. Since the synchronization timing information T0 is known, a time-aligned search signal can be directly constructed at the detected TOA time, i.e., T0 = T TOA Then equation (5) can be rewritten as

[0028]

[0029] Preferably, step S4 is as follows: NZ label estimation result for

[0030]

[0031] Among them, F f For the fast Fourier transform operator, y c (n) is the y in equation (4) c Discrete complex signal after ADC (t)

[0032] Preferably, step S5 is as follows: based on the estimated NZ label Demodulating the composite signal output by NYFR, the carrier frequency estimation result for the non-cooperative monopulse radar signal is as follows:

[0033]

[0034] Finally, the ultra-wideband interception and parameter estimation of NYFR's non-cooperative single-carrier pulse radar signal was completed.

[0035] This invention proposes an improved NYFR structure and corresponding signal estimation method that are insensitive to output signal truncation, thereby enhancing the robustness of NYFR-based broadband radar reconnaissance. Attached Figure Description

[0036] Figure 1 This is a structural diagram of the improved NYFR proposed in this invention.

[0037] Figure 2 This is the time-frequency diagram of the output signal of the prototype NYFR using SFM local oscillator.

[0038] Figure 3 The result is the estimated NZ label of the prototype NYFR output signal using the SFM local oscillator.

[0039] Figure 4 This is the time-frequency diagram of the prototype NYFR with SFM local oscillator after the output signal is truncated.

[0040] Figure 5 The result is the estimated NZ label by truncating the output signal of the prototype NYFR using the SFM local oscillator.

[0041] Figure 6 This is the time-frequency diagram of the output signal of the improved NYFR using a periodic chirp local oscillator.

[0042] Figure 7 The result is the NZ label estimation of the improved NYFR output signal using a periodic chirp local oscillator.

[0043] Figure 8 It is a time-frequency diagram of the improved NYFR output signal after truncation using a periodic chirp local oscillator.

[0044] Figure 9 The result is the estimated NZ label of the improved NYFR output signal with periodic chirp local oscillator.

[0045] Figure 10 This is the carrier frequency estimation result of the NYFR output signal after demodulation. Detailed Implementation

[0046] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0047] The NYFR uses a modulated RF LO to generate an analog non-uniform sampling pulse train. By combining analog non-uniform sampling with ADC uniform sampling, it achieves ultra-wideband reception of a single low-speed ADC, breaking through the limitations of the Nyquist sampling theorem. This embodiment improves upon the prototype NYFR structure, making the NZ label estimation and reconstruction of the output signal no longer affected by output signal truncation.

[0048] like Figure 1 As shown, this embodiment presents an improved NYFR structure, characterized by including a bandwidth RF filter, an RF sampling clock, a low-pass filter, an ADC, and an output signal processing module; the bandwidth RF filter receives the intercepted non-cooperative radar signal x(t); the improved NYFR's LO uses the zero-crossing rising edge (ZCR) of a periodic chirp modulated signal to generate a non-uniform RF sampling pulse train p. LO (t); the signal x(t) after passing through the ultra-wideband RF filter and the non-uniform RF sampling pulse train p with periodic chirp modulation. LO (t) After mixing, the signal y(t) is obtained by passing through a low-pass filter; the signal y(t) is then processed by the ADC to obtain the output discrete signal y(n); the RF sampling clock sends the synchronization timing information T0 to the output signal processing module at the start of LO operation; the output signal processing module combines the discrete signal y(n) and the synchronization timing information T0 to achieve ultra-wideband reception and processing of non-cooperative radar signals that are insensitive to output signal truncation.

[0049] This embodiment presents a signal estimation method, the steps of which are as follows:

[0050] S1. Construct the improved NYFR structure described above, as follows: Figure 1 As shown, the improved NYFR structure has periodic chirp modulation LO and synchronous timing functions;

[0051] S2. Based on the improved NYFR structure in step S1, intercept the non-cooperative single-frequency pulse radar signal and obtain the composite signal output by the NYFR.

[0052] S3. Based on periodic chirp modulation LO and synchronization timing information, construct an alignment search signal containing NZ search index modulation, wherein the range of NZ search index includes all NZ of the instantaneous monitoring video segment and has positive and negative values.

[0053] S4. Using the search signal containing all NZ labels constructed in step S3, perform deLO modulation and fast Fourier transform, and use the maximum value of the peak value of the spectrum of the deLO-modulated signal corresponding to each NZ search label to obtain the NZ label estimation result of the NYFR output signal.

[0054] S5. Using the NZ label estimated in step S4, perform deLO modulation and FFT on the complex signal output by the NYFR structure. Estimate the carrier frequency of the folded signal by the frequency corresponding to the maximum value of the spectrum of the deLO-modulated signal, calculate the intercepted unknown radar signal carrier frequency, and complete the signal estimation.

[0055] In step S1 of this embodiment, the construction of the improved NYFR structure is as follows: the LO modulation θ(t) is a periodic chirp, and the modulation model is...

[0056] θ(t)=πμ0mod(t,T LO ) 2 (1)

[0057] Where μ0 is the modulation frequency of the periodic chirp modulation, and T LO The modulation period is the periodic chirp, and mod(·) is the modulo operation; in the NYFR structure, x(t) is the intercepted non-cooperative radar signal, ZCR represents the time of the zero-crossing rising edge, and f LO For the carrier frequency of LO, t k Let p(t) be the sampling pulse train template at the zero-crossing rising edge time; let the sampling pulse train template be a unit impulse function, i.e., p(t) = δ(t), and y(t) be the intercepted signal x(t) after passing through the ultra-wideband RF filter and the non-uniform RF sampling pulse train p with periodic chirp modulation. LO (t) The result after mixing and passing through a low-pass filter, where the cutoff frequency of the low-pass filter is f. LO / 2; then, the signal y(t) is discretized / digitized by the low-speed ADC to obtain the output discrete signal y(n) of the improved NYFR; at the same time, the synchronization timing information T0 is sent to the output signal processing module at the start of LO operation; combining the discrete signal y(n) and the synchronization timing information T0, the ultra-wideband reception and processing of non-cooperative radar signals that are insensitive to the truncation of the output signal is realized.

[0058] In this embodiment, step S2 is as follows:

[0059] The intercepted non-cooperative single-frequency pulse radar signal is

[0060] x(t)=Acos(2πf c t)[ε(t)-ε(tT)] (2)

[0061] Where A is the amplitude, f c The signal carrier frequency is the parameter to be estimated in the non-cooperative signal, ε(·) is the step function, and T is the pulse width; the NYFR output composite signal after interception is

[0062] y(t)=Acos[2π|f c -k NZ f LO |t-sgn(f c -k NZ f LO )k NZ πμ0mod(t,T LO ) 2 ][ε(t)-ε(tT)](3)

[0063] Where, k NZ =round(f c / f LO ), sgn(·) is the sign function, and the NZ label is defined as sgn(f c -k NZ f LO )k NZ For non-cooperative reception, if the TOA estimation is inaccurate during signal detection, the NYFR output signal will be truncated, and equation (3) will become...

[0064]

[0065] Among them, T TOA The known TOA time is output by NYFR after detection.

[0066] In this embodiment, step S3 is as follows:

[0067] The constructed search signal is

[0068]

[0069] Where n is an integer, T ADC The ADC sampling interval is k, and the NZ search index is k. s ∈[-L,L],k s∈Z, where L is the total number of NZs monitored by NYFR; since the synchronization timing information T0 is known, a time-aligned search signal is directly constructed at the detected TOA time, i.e., T0 = T TOA Then equation (5) is rewritten as

[0070]

[0071] In this embodiment, step S4 is as follows:

[0072] NZ labeling estimation results for

[0073]

[0074] Among them, F f For the fast Fourier transform operator, y c (n) is the y in equation (4) c Discrete complex signal after ADC (t)

[0075] In this embodiment, step S5 is as follows:

[0076] Based on the estimated NZ designation Demodulating the composite signal output by NYFR, the carrier frequency estimation result for the non-cooperative monopulse radar signal is as follows:

[0077]

[0078] Finally, the ultra-wideband interception and parameter estimation of NYFR's non-cooperative single-carrier pulse radar signal was completed.

[0079] In this embodiment:

[0080] The non-uniform sampling pulse train generated by the improved modulation LO is

[0081]

[0082] Furthermore, equation (9) can be calculated as follows:

[0083]

[0084] The Fourier transform of equation (10) is:

[0085]

[0086] Where F is the Fourier transform operator, and * is the convolution operator.

[0087] according to Figure 1 The frequency domain of the NYFR output analog signal y(t) can be calculated using the improved NYFR structure and the intercepted signal in equation (2).

[0088]

[0089] Where X(f)=F[x(t)], H W (f) is a broadband radio frequency filter, H LPF (f) is a low-pass filter. Equation (12) can be further calculated as follows:

[0090]

[0091] Among them, X L (f) represents the left band of the intercepted signal's spectrum, X R (f) represents the right-hand band of the intercepted signal's spectrum. Therefore, the output signal of the improved NYFR is...

[0092]

[0093] In the formula, F -1 This is the Fourier inverse transform operator.

[0094] The prototype NYFR uses sinusoidal frequency modulation (SFM) as the LO modulation method. The prototype NYFR output corresponding to the intercepted radar signal in equation (2) is:

[0095] y o (t)=Acos[2π|f c -k NZ f LO |t-sgn(f c -k NZ f LO )k NZ πsin(2πf s t)][ε(t)-ε(tT)] (15)

[0096] Among them, f s This is the modulation frequency of SFM.

[0097] Considering the signal truncation caused by TOA estimation in equation (15), the NZ label estimation will become sensitive to the truncation of the output signal. Specifically, with f c -k NZ f LO Taking >0 as an example, let the detected cutoff TOA time be T. TOA Equation (15) can be written as

[0098]

[0099] At this point, the NZ label is simplified to k. NZ In equation (16), if T TOA It exactly satisfies half a cycle of SFM, then the NZ label k in equation (16) NZThis will result in a negative value, further causing errors in the NZ label estimation. Therefore, the sign of the estimated NZ label value of the prototype NYFR output signal is sensitive to output signal truncation; that is, signal truncation may cause errors in the NZ label estimation.

[0100] For the improved NYFR output, still in f c -k NZ f LO For example, if the signal in equation (14) is truncated due to TOA estimation, it can be written as:

[0101]

[0102] From equation (17), it can be seen that in this case, the NZ label k NZ The truncation of TOA estimation does not affect the sign of the FM slope due to the influence of the LO modulation section of the output signal. Therefore, it does not affect the estimated value of NZ label, realizing the advantage of NZ label estimation being insensitive to NYFR output signal truncation.

[0103] Based on equation (17), the improved NYFR output, after passing through a low-speed ADC, allows f to... ic =f c -k NZ f LO The complex form of the output discrete signal can be written as:

[0104]

[0105] Among them, T ADC Let k be the ADC sampling interval. In equation (18), the simplified NZ label k NZ and the intermediate frequency f of the folded signal ic These are the parameters to be estimated.

[0106] Next, the NZ label is estimated. Since the improved NYFR structure introduces synchronization timing information T0, the TOA estimate T can be obtained after signal detection is complete. TOA At this point, the synchronization time becomes T0 = T TOA Based on LO parameter information and synchronization information, a search signal can be constructed.

[0107]

[0108] The NZ search label is k. s ∈[-L,L] and k s ∈Z, where L is the total number of NZs monitored by NYFR. Combining equations (18) and (19), we have

[0109]

[0110] When NZ search label k s =k NZ Then, equation (20) can be further calculated as

[0111]

[0112] at this time, The maximum value is obtained. Therefore, combining the FFT, the NZ label estimation result is:

[0113]

[0114] After completing the NZ label estimation, the search signal in equation (19) can be rewritten as follows:

[0115]

[0116] Equation (23) can be used to demodulate the LO modulation component contained in the NYFR output signal, and further estimate the intermediate frequency carrier frequency f. ic ,have

[0117]

[0118] Ultimately, the carrier frequency of the intercepted signal was estimated to be...

[0119]

[0120] Because this invention employs periodic chirp modulation and introduces synchronization information, the NZ label estimation of the output signal is no longer affected by truncation conditions such as TOA estimation. Simultaneously, the synchronization information ensures that the search signal is aligned with the NYFR output signal, guaranteeing the correctness of NZ label estimation and LO demodulation. Furthermore, the estimation method of this invention has a computational complexity on the order of FFT, making it suitable for engineering implementation.

[0121] The effectiveness of this invention was verified through simulation experiments. The prototype NYFR uses SFM for its LO, where the LO carrier frequency f... LO The SFM modulation frequency is 2GHz. s The frequency is 10MHz, the low-pass filter cutoff frequency is 1GHz, and the ADC sampling rate is 2GHz; the intercepted signal is a single-carrier pulse radar signal with a carrier frequency f. c The signal frequency is 4.3 GHz, the signal duration is 0.5 μs, and when this signal is intercepted by NYFR, its NZ designation is sgn(f). c -k NZ f LO )round(f c / f LO =2.

[0122] Figure 2The time-frequency diagram (TFR) of the output signal after the signal is intercepted by the prototype NYFR is given. Figure 3 This is the result of estimating the NZ label of the output signal using the search signal. Figure 2 It can be seen that the TFR of the prototype NYFR output signal contains SFM modulation. From Figure 3 It can be seen that when the NZ search label is 2, equation (22) reaches its maximum value. Therefore, the NZ label estimation result is 2, and the data segment output by the prototype NYFR can obtain the correct NZ label estimation.

[0123] In non-cooperative reception, the TOA estimation for signal detection will have errors. In this case, the output data may be truncated. Based on the simulation conditions above, considering a TOA estimation error of 0.05 μs, the NYFR output truncated data duration is 0.45 μs. Figure 4 The TFR for the NYFR output truncation signal is given. Figure 5 The NZ labeling results for the NYFR output truncation signal are displayed. Figure 4 It can be seen that the output signal is truncated. Figure 2 From the TFR, we can know Figure 4 The initial phase of the signal changes by half a cycle due to truncation; from Figure 5 It can be seen that when the NZ search index is -2, equation (22) reaches its maximum value. At this time, the estimated NZ index of the output signal is -2, which is incorrect. This proves that the NZ index estimation of the prototype NYFR output signal is sensitive to the truncation of the output signal, and the corresponding NZ index estimation will be incorrect.

[0124] Next, an improved NYFR structure is adopted, in which the LO uses periodic chirp modulation with a modulation period of 0.1 μs and a chirp modulation frequency of 400 MHz / μs. 2 The remaining parameters are the same as Figure 2 The prototype used in the simulation is the same as the NYFR, and the intercepted signal parameters are also the same. Figure 2 The simulation uses the same methods. Figure 6 The TFR of the output signal after the signal is intercepted by the improved NYFR is given. Figure 7 This is the result of labeling the output signal NZ. (From...) Figure 6 It can be seen that the output signal contains a periodic chirp modulation component; from Figure 7 It can be seen that when the NZ search label is 2, equation (22) reaches its maximum value. Therefore, the estimated value of the NZ label is 2, and the estimation result is correct.

[0125] Considering that the NYFR output data duration is truncated to 0.45μs, Figure 8 The TFR for the NYFR output truncation signal is given. Figure 9 The NZ label result for the NYFR output truncation signal. (From...) Figure 8 It can be seen that the output signal is truncated, and its starting frequency is... Figure 6 Different; from Figure 9 As can be seen, the NZ estimate label of the output signal is still 2, and the result is correct.

[0126] After obtaining the correct NZ designation, the LO modulation of the output signal can be demodulated, and the intermediate frequency of the folded signal can be estimated using FFT. The carrier frequency of the intercepted signal can then be calculated by combining the estimated NZ designation, thus completing the ultra-wideband interception and estimation of non-cooperative radar signals. Figure 10 Given Figure 9 After estimating the NZ designation, the signal spectrum after de-LO modulation and FFT of the NYFR output signal is obtained using the estimated NZ designation. It can be seen that the estimated carrier frequency is approximately 0.3 GHz, which is close to the theoretical value of f. c -k NZ f LO =0.3GHz consistent. Using equation (25), the carrier frequency of the intercepted original radar signal can be estimated. The above simulation proves that the improved NYFR proposed in this invention can effectively solve the problem of signal truncation sensitivity when the prototype NYFR estimates the NZ label of the output signal. At the same time, the proposed fast algorithm can effectively estimate the carrier frequency of the intercepted non-cooperative radar signal.

[0127] This invention addresses the issue that the NZ label estimation of the NYFR output signal is sensitive to output signal truncation in the prototype NYFR structure. It proposes an improved NYFR structure that is insensitive to output signal truncation, has better broadband interception robustness, and improves the broadband reconnaissance performance of NYFR.

[0128] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. An improved NYFR structure, characterized in that... Includes a bandwidth RF filter, RF sampling clock, low-pass filter, ADC, and output signal processing module; the bandwidth RF filter input intercepts non-cooperative radar signals. The LO uses the zero-crossing rising edge of the periodic chirp modulated signal to generate a non-uniform radio frequency sampling pulse train. ;Signal With non-uniform radio frequency sampling pulse train The signal is obtained after mixing and passing through a low-pass filter. ;Signal After discrete / digital processing by an ADC, the output discrete signal is obtained. ; The RF sampling clock sends synchronization timing information to the output signal processing module at the moment the LO starts operating. ; The output signal processing module combines discrete signals and synchronization timing information To achieve ultra-wideband reception and processing of non-cooperative radar signals that are insensitive to output signal truncation.

2. A signal estimation method, characterized in that: Follow these steps: S1. Construct the improved NYFR structure as described in claim 1, wherein the improved NYFR structure has periodic chirp modulation (LO) and synchronization timing functions; S2. Based on the improved NYFR structure in step S1, intercept the non-cooperative single-frequency pulse radar signal and obtain the composite signal output by the NYFR structure. S3. Based on periodic chirp modulation LO and synchronization timing information, construct an alignment search signal containing NZ search index modulation, wherein the range of NZ search index includes all NZ of the instantaneous monitoring video segment and has positive and negative values. S4. Using the search signal containing all NZ labels constructed in step S3, perform deLO modulation and fast Fourier transform, and use the maximum value of the peak value of the spectrum of the deLO-modulated signal corresponding to each NZ search label to obtain the NZ label estimation result of the NYFR structure output signal. S5. Using the NZ label estimated in step S4, perform deLO modulation and FFT on the complex signal output by the NYFR structure. Estimate the carrier frequency of the folded signal by the frequency corresponding to the maximum value of the spectrum of the deLO-modulated signal, calculate the intercepted unknown radar signal carrier frequency, and complete the signal estimation.

3. The signal estimation method according to claim 2, characterized in that, The construction of the improved NYFR structure in step S1 is as follows: LO modulation For periodic chirps, the modulation model is: (1) in, The frequency modulation of periodic chirp modulation, The modulation period of the periodic chirp. For modulo operation; in the NYFR structure, For intercepted non-cooperative radar signals, ZCR represents the time of the rising edge of the zero point. For the carrier frequency of LO, For the time when the rising edge of zero is crossed, Let the sampling pulse train template be a unit impulse function, i.e. , The intercepted signal after passing through the ultra-wideband radio frequency filter With non-uniform radio frequency sampling pulse trains with periodic chirp modulation The result after mixing and passing through a low-pass filter has a cutoff frequency of 1000 Hz. After that, the signal The signal is discretized / digitized by a low-speed ADC to obtain the output discrete signal of the improved NYFR structure. Simultaneously, at the moment the LO starts working, a synchronization timing information is sent to the output signal processing module. Combined with discrete signals and synchronization timing information This enables ultra-wideband reception and processing of non-cooperative radar signals that are insensitive to output signal truncation.

4. The signal estimation method according to claim 3, characterized in that, Step S2 is as follows: The intercepted non-cooperative single-frequency pulse radar signal is as follows: (2) in, For amplitude, The carrier frequency is the parameter to be estimated in the non-cooperative signal. It is a step function. The pulse width is [value]; the composite signal output by the intercepted NYFR structure is: (3) in, , For symbolic functions, the NZ label is defined as follows: For non-cooperative reception, if the TOA estimation is inaccurate during signal detection, the output signal of the NYFR structure will be truncated, and equation (3) will become: (4) in, The known TOA time is the output of the NYFR structure after detection.

5. The signal estimation method according to claim 4, characterized in that, Step S3 is as follows: The constructed search signal is: (5) in, It is an integer. The ADC sampling interval is [value], and the NZ search index is [value]. , , The total number of NZs monitored for the NYFR structure; due to synchronization timing information. It is known that a time-aligned search signal can be directly constructed at the detected TOA time, i.e. Then equation (5) can be rewritten as: (6)。 6. The signal estimation method according to claim 5, characterized in that, Step S4 is as follows: NZ labeling estimation results for: (7) in, For the fast Fourier transform operator, In equation (4) The discrete complex signal after ADC.

7. The signal estimation method according to claim 6, characterized in that, Step S5 is as follows: Based on the estimated NZ designation Demodulating the composite signal output by the NYFR structure, the carrier frequency estimation result for the non-cooperative monopulse radar signal is as follows: (8) Finally, the ultra-wideband interception and parameter estimation of non-cooperative single-carrier pulse radar signals of the NYFR structure were completed.

Images