A three-channel undersampling-based ultra-wideband digital instantaneous frequency measurement method

Abstract

The application discloses a kind of based on three-channel undersampling technology's ultra-wideband instantaneous frequency measurement method, can realize to the multiple signal in ultra-wideband spectrum range is monitored.Radio frequency signal is directly input to three sampling channels after amplification filtering and is sampled, and each road signal sampling rate is different, uses short time window weighting, and utilizes real signal double FFT point method to calculate the frequency that radio frequency frequency is folded into each sampling channel first Nyquist area in short time window, then using sub-band segmentation demodulation method can be calculated out without ambiguity radio frequency frequency.The application can realize the signal in 0~18GHz range is instantaneous frequency measurement in 200ns time, and frequency measurement accuracy is within 1MHz.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electromagnetic signal reconnaissance. BACKGROUND

[0002] The instantaneous frequency measurement technology has been developed for decades, and the technology is gradually mature. The early instantaneous frequency measurement technology mainly adopts the way of multi-channel analog delay line instantaneous frequency measurement, which is complex and the frequency measurement performance is easily affected by the external environment. With the application of high-performance digital processing chips and the research of corresponding signal processing technology, the instantaneous frequency measurement technology develops towards wideband and digitization.

[0003] Common digital instantaneous frequency measurement technologies include single-bit frequency measurement, optical sampling frequency measurement, and digital channelized frequency measurement. The single-bit instantaneous frequency measurement technology is proposed in ‘Analysis and characterization of a monobit receiver for electronic warfare’ (Grajal J, IEEE Trans. Aerosp. Electron. Syst, 2003, vol. 39, no. 1, pp. 244-258.), which has the advantages of simple structure, low cost, and large instantaneous bandwidth, but the data bit width is small, the instantaneous dynamic is small, and it cannot process multiple simultaneous signals; the optical sampling frequency measurement technology is proposed in ‘Photonics-based broadband microwave instantaneous frequency measurement by frequency-to-phase-slope mapping’ (J.Z. Shi, F.D. Zhang, D. Ben, et al. IEEE Trans. Microw. Theory Tech., 2019, vol. 67, no. 2, pp. 544-552.), and the instantaneous bandwidth can reach dozens of GHz, but the frequency measurement accuracy is poor, which cannot meet the needs of electronic reconnaissance and electronic jamming guidance; a digital channelized frequency measurement technology with superheterodyne structure is proposed in ‘A fast ELINT receiver design. Proceedings of the 13th European Radar Conference’ (A. Alparslan and K. Yegin. London, UK, IEEE, 2016, pp. 217-220), which obtains a larger working bandwidth by using a superheterodyne method. This method has the advantages of high sensitivity of channelization, distinguishing multiple simultaneous signals, and high flexibility of digitization, and is most widely used in electronic reconnaissance, but the maximum instantaneous bandwidth can only reach half of the sampling rate, which cannot realize high probability interception of signals in a wide frequency range.

[0004] With a large number of electronic devices in the battlefield, resulting in signal density is getting higher and higher, battlefield electromagnetic environment is increasingly complex, signal type, power, bandwidth, modulation mode complex and varied, the coverage of the frequency is more and more wide, which requires the frequency receiver to have a large instantaneous bandwidth to obtain a higher probability of interception, while having a high frequency resolution and instantaneous dynamic range, the existing technology is usually difficult to meet these needs. SUMMARY

[0005] To solve the above technical problems, the present application provides a kind of based on three channel undersampling ultra-wideband digital instantaneous frequency measurement method, the present application is realized by the following technical solutions:

[0006] Step S1, the input radio frequency signal is directly undersampled in three ways;

[0007] Step S2, the three-way sampling signal is segmented using a short-time window, and it is assumed that the signal is stationary within the short-time window;

[0008] Step S3, the real signal double-FFT point frequency estimation method is used to estimate the frequency value of the radio frequency frequency folded into each sampling channel first Nyquist region;

[0009] Step S4, the frequency unambiguous radio frequency signal frequency is calculated using the frequency band segmentation frequency unambiguous method.

[0010] Preferably, the three-way sampling signal in step S2 uses a short-time window function of the same length.

[0011] Preferably, the real signal double-FFT point frequency estimation first performs zero-padding FFT on the three-way signal, and the FFT point number is N fft , search peak point, recorded as k p1 , k p2 , k p3 , complete rough frequency measurement;The value of the second largest peak point on both sides of the peak point is used to complete fine frequency measurement, and the frequency measurement formula is as follows:

[0012] m=1,2,3

[0013] m=1,2,3

[0014] In the formula, is the frequency offset value, N m is the number of short-time window sample points of the mth sampling channel, is the N fft point FFT operation result after zero padding, β m is N m , N fftThe relevant constant is obtained in advance through simulation, k pm is the position of the peak point after FFT, f sm is a sampling rate.

[0015] Preferably, the band segmentation deambiguating comprises: first, dividing [0, f s1 , f s2 , f s3 into X sub-bands according to the value of f max , each sub-band corresponding to a unique set of folding parameters, then obtaining the total error value of each sub-band according to the estimated ambiguous frequency f m , and finally searching for the sub-band with the minimum error, assuming that the sub-band number is l, the input signal frequency can be expressed as:

[0016]

[0017] The present application has the advantages of:

[0018] (1) The radio frequency signal is directly sampled, and the sampling is as close to the antenna as possible, avoiding the use of a large number of analog front-end devices, and the structure is simple and flexible.

[0019] (2) A plurality of low-speed ADCs are used to replace high-speed ADCs to realize ultra-wideband instantaneous frequency measurement, which reduces the processing pressure of the digital signal processing unit, and has a high sampling bit number, so that a high instantaneous dynamic range can be obtained.

[0020] (3) Currently, Rife, Jacobsen, Canda and other interpolation DFT algorithms are commonly used to calculate the frequency offset value, and the estimation error of these algorithms fluctuates greatly under different frequency offsets, and the calculation amount is relatively large, the present application adopts a short-time window weighting combined with a real signal double-FFT point frequency estimation method, which is beneficial to improve the frequency measurement stability, and has small calculation amount and high frequency measurement precision.

[0021] (4) Currently, the robust Chinese remainder theorem (RCRT) algorithm is commonly used for deambiguating operation, which has a large calculation amount and a low deambiguating accuracy, the present application proposes a band segmentation deambiguating algorithm, which divides the non-ambiguous frequency range into a plurality of sub-bands, judges the sub-band where the radio frequency is located, and obtains the non-ambiguous radio frequency value, this method can obtain a high deambiguating accuracy and has a small calculation amount.

[0022] The present application can ensure that the frequency measurement time is not more than 200 ns, and the frequency measurement precision is within 1 MHz. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is a structure block diagram of ultra-wideband undersampling instantaneous frequency measurement. Detailed Implementation

[0024] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings.

[0025] This invention employs three sampling channels. For example... Figure 1 As shown, the RF signal, after amplification, power division, and sample-and-hold, is input into three sampling channels for digitization. The digital part mainly consists of frequency measurement and frequency ambiguity resolution. The frequency measurement part first applies a short time window to the input signal, then performs zero-padded FFT on the signal within the window to search for the peak and complete coarse frequency measurement. The frequency offset is estimated using the amplitude values ​​of the two second-largest peak points to the left and right of the peak to complete fine frequency measurement, thus obtaining the frequency of the RF signal folded into the first Nyquist region of each sampling channel. The ambiguity resolution part uses a frequency band segmentation method to divide the frequency band under test into multiple sub-bands. Each sub-band corresponds to a unique set of folding parameters. The error corresponding to each sub-band is calculated based on the sampling rate and ambiguity frequency value of each channel. The sub-band with the smallest error is selected as the sub-band where the RF signal frequency is located. The input signal frequency is calculated using the folding parameters of this sub-band.

[0026] The invention will now be described in detail with reference to specific examples. Assume the input signal frequency range is [0, 18000MHz), the three channel sampling frequencies are 1800MHz, 1900MHz, and 2000MHz respectively, and the short-time window function width is 10ns. Then, the number of sample points contained within the short-time window of the three channels are 18, 19, and 20 respectively, and the FFT point count is set to N. fft =128, and the slope constants corresponding to each channel are β1=0.8741, β2=0.7829, and β3=0.7050, respectively.

[0027] Analysis shows that the input signal frequency range can be divided into 54 sub-bands. Setting the input signal frequency to 13475MHz, after step 3, the interpolation FFT operation yields the folded frequencies of the three sub-channels as 875.5MHz, 174.1MHz, and 525.3MHz. After step 4, the input signal frequency is found to be in the 42nd sub-band, with the folding parameters for the three channels as b1=7, d1=1, b2=7, d2=1, b3=7, d3=-1. According to step 5, the input signal frequency is 13474.7MHz.

Claims

1. A method for ultra-wideband digital instantaneous frequency measurement based on three-channel undersampling, characterized in that: step S1: directly undersampling an input radio frequency signal into three channels; step S2: segmenting the three-channel sampled signals using a short-time window, and assuming that the signals are stationary within the short-time window; step S3: estimating the frequency value of the radio frequency folded into the first Nyquist region of each sampling channel using a real-signal double-FFT point frequency estimation method; wherein the real-signal double-FFT point frequency estimation method in step S3 comprises: step S4: using a frequency band segmentation frequency deambiguating method to solve the unambiguous radio frequency signal frequency; wherein the frequency deambiguating method in step S4 comprises: the relationship between the input signal frequency and the ambiguous frequency folded into the first Nyquist region can be expressed as: step S2: the three-channel sampled signals use the same length of short-time window function. ​ ​ ​ ​ In the formula, is a frequency offset value, N m is the number of short-time window sample points of the mth sampling channel, is the N fft point FFT operation result, β m is a constant related to N m , N fft , which is obtained in advance through simulation, k pm is the position of the peak point after FFT, f sm is a sampling rate; ​ ​ f c = b m f sm = d m f m , 0 ≤ f m ≤ f sm / 2, b m = 0, 1, 2...B m , d m = ±1, m = 1, 2, 3 where b m , d m are folding parameters; where b m is a non-negative integer, and its maximum value f max is the upper limit of the measurable frequency range; according to the values of f s1 , f s2 , f s3 , [0, f max ] is divided into X sub-bands, each of which corresponds to a unique set of folding parameters, and the total error value corresponding to each sub-band is obtained according to the estimated ambiguity frequency f m , and the sub-band with the minimum error is the sub-band where the input signal is located, and assuming that the sub-band number is l, the input signal frequency can be represented as:

2. The method according to claim 1, wherein the method is based on a three-channel under-sampling UWB digital instantaneous frequency measurement. ​

Images