Method and system for detecting low signal-to-noise ratio (SNR) periodic linear-frequency-modulated continuous-wave (LFMCW) signal

Abstract

Detecting a low signal-to-noise (SNR) periodic linear-frequency-modulated continuous-wave (LFMCW) signal. First the data to be detected is obtained and periodic component detection is performed on the

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of signal detection, radio target monitoring and spectrum management, and in particular, to a method and system for detecting a low signal-to-noise ratio (SNR) periodic linear-frequency-modulated continuous-wave (LFMCW) signal. BACKGROUND

[0002] Efficient detection and parameter estimation of periodic LFMCW signals are necessary means in the field of radio target monitoring and spectrum management and control. In recent years, the rapid development of electronic information technology has brought new problems to the detection of periodic LFMCW signals by an electromagnetic spectrum monitoring system. For example, the increase in frequency-using devices has led to the increasingly congested electromagnetic spectrums, making ambient noise increasingly complex and background noise of a receiver rise, resulting in a decrease in an SNR of a detected signal. In addition, the heavily used high carrier frequency periodic LFMCW signals have larger path losses. However, conventional methods for detecting a low SNR periodic LFMCW signal are complex in calculation and therefore are difficult to be deployed on small-sized devices. These lead to a decrease in a detection distance of the receiver. SUMMARY

[0003] The objective of the present disclosure is to provide a method and system for detecting a low SNR periodic LFMCW signal, which can resolve the problem that an LFMCW signal detection method based on a conventional discrete polynomial transformation is hardly applicable to a low SNR scenario due to energy dispersion, thereby effectively improving a capability of detecting weak periodic LFMCW signals.

[0004] To achieve the above objective, the present disclosure provides the following technical solutions.

[0005] A method for detecting a low SNR periodic LFMCW signal includes:

[0006] obtaining data to be detected, and performing periodic component detection on the data to be detected;

[0007] if there is a periodic component, recording a periodic estimate value according to the periodic component, and setting a delay amount according to the periodic estimate value;

[0008] determining two adjacent observation windows according to the delay amount and a difference obtained by subtracting the delay amount from the periodic estimate value;

[0009] dividing the data to be detected into continuous P groups of data by using the periodic estimate value, where P is a maximum value of the number of groups;

[0010] sliding a start position of an observation window by using a sampling point as a step, and performing segmented discrete polynomial transformation and sum spectrum calculation on the data to be detected corresponding to the sliding observation window at each step;

[0011] detecting a periodic LFMCW signal in the data to be detected according to a maximum value of a sum spectrum at each step and a corresponding frequency; and

[0012] when a detection result is that there is a periodic LFMCW signal, calculating a frequency modulation slope and a frequency modulation inflection point position of the periodic LFMCW signal according to the frequency and the step corresponding to the maximum value of the sum spectrum.

[0013] Optionally, the data to be detected is digitally processed intermediate frequency or baseband data.

[0014] Optionally, the delay amount x<Te / 2, where Te is the periodic estimate value.

[0015] Optionally, after the dividing the data to be detected into continuous P groups of data by using the periodic estimate value, the method further includes:

[0016] performing zero-padding on data obtained by segmenting a shorter observation window in the two adjacent observation windows.

[0017] Optionally, the sliding a start position of an observation window by using a sampling point as a step, and performing segmented discrete polynomial transformation and sum spectrum calculation on the data to be detected corresponding to the sliding observation window at each step specifically includes:

[0018] qth sliding of the observation window: when Tex(p-l)+^<H<Tex(p- / )+g+(Te-T), obtaining a transformation sequence dip by using a formula dlp(Z / )=x(«) where dip =[dlp(l), dlp(2), dlp(3),...], dlp(Zj) is an hA data point of a first segment of the discrete polynomial transformation within a length of a pth periodic estimate value, h=(n mod Te)-g+l, (n mod Te) means dividing n by Te to obtain a remainder, * means performing a conjugation operation, q is a sequence number of sliding, Te is a periodic estimate value, t is a delay amount, and n is an n* sampling point;

[0019] downsampling the transformation sequence dip, and then performing fast Fourier transform and a modulo operation on the downsampled transformation sequence dlp, to obtain a spectrum fl p of the first segment of the discrete polynomial transformation within the length of the pth periodic estimate value;

[0020] when Texp+q-T<n<Texj>, obtaining a transformation sequence d2p by using a formula d2p( / 2)=x(n)-x*(h+t), where d2p=[d2p(l), d2p(2), d2p(3),...], d2p(6) is an / 2th data point of a second segment of the discrete polynomial transformation within the length of the pth periodic estimate value, and h=(n mod Te)+r-^-Te+l);

[0021] padding the transformation sequence d2p with zeros at the end to equalize a data length thereof and a data length of dlp, downsampling the zero-padded transformation sequence d2p, and then performing fast Fourier transform and a modulo operation on the downsampled transformation sequence d2p, to obtain a spectrum f2p of the second segment of the discrete polynomial transformation within the length of the pth periodic estimate value; and

[0022] adding the spectrum of each discrete polynomial transformation within the length of each periodic estimate value to obtain a sum spectrum fq corresponding to the q* sliding of the observation window.

[0023] Optionally, the detecting a periodic LFMCW signal in the data to be detected according to a maximum value of a sum spectrum at each step and a corresponding frequency specifically includes:

[0024] determining a step condition corresponding to the largest one in the maximum values of the sum spectrums;

[0025] determining whether there is a maximum value in sum spectrums of adjacent steps on both sides of the step condition corresponding to the largest one in the maximum values of the sum spectrum; and

[0026] if yes, determining that the data to be detected includes a periodic LFMCW signal; or

[0027] if no, determining that the data to be detected does not include a periodic LFMCW signal.

[0028] A system for detecting a low SNR periodic LFMCW signal includes:

[0029] a periodic component detection module, configured to obtain data to be detected, and perform periodic component detection on the data to be detected;

[0030] a delay amount setting module, configured to: if there is a periodic component, record a periodic estimate value according to the periodic component, and set a delay amount according to the periodic estimate value;

[0031] an observation window determining module, configured to determine two adjacent observation windows according to the delay amount and a difference obtained by subtracting the delay amount from the periodic estimate value;

[0032] a grouping module, configured to divide the data to be detected into continuous P groups of data by using the periodic estimate value, where P is a maximum value of the number of groups;

[0033] a calculation module, configured to slide a start position of an observation window by using a sampling point as a step, and perform segmented discrete polynomial transformation and sum spectrum calculation on the data to be detected corresponding to the sliding observation window at each step;

[0034] a periodic LFMCW signal detection module, configured to detect a periodic LFMCW signal in the data to be detected according to a maximum value of a sum spectrum at each step and a corresponding frequency; and

[0035] a frequency modulation slope and frequency modulation inflection point position determining module, configured to: when a detection result is that there is a periodic LFMCW signal, calculate a frequency modulation slope and a frequency modulation inflection point position of the periodic LFMCW signal according to the frequency and the step corresponding to the maximum value of the sum spectrum.

[0036] According to specific embodiments provided in the present disclosure, the present disclosure has the following technical effects:

[0037] According to the method and system for detecting a low SNR periodic LFMCW signal provided in the present disclosure, periodic component detection and periodic value estimation are first performed on data to be detected. If there is a periodic component, a periodic estimate value is recorded. Second, a delay amount is set according to the periodic estimate value, and two adjacent observation windows whose lengths are respectively the periodic estimate value minus the delay amount and the delay amount are selected. A start position of an observation window is slid by using a sampling point as a step, segmented discrete polynomial transformation is performed on intermediate frequency or baseband data of a required observation length during each sliding of the observation window, and a sum spectrum is calculated. Subsequently, it is determined, according to a maximum value of the sum spectrum under each step condition and a corresponding frequency, whether the data to be detected includes a periodic LFMCW signal. Finally, a frequency modulation slope and a frequency modulation inflection point position of the LFMCW signal are calculated according to the frequency and the step corresponding to the maximum value of the sum spectrum. The present disclosure resolves the problems of hardly applicable to a low-signal-to-noise-ratio scenario due to energy dispersion during LFMCW signal detection based on a conventional discrete polynomial transformation, thereby effectively improving a capability of detecting weak periodic LFMCW signals. Compared with the existing method for detecting a low SNR periodic LFMCW signal, the present disclosure does not involve multi-dimensional domain parallel data transformation and parameter search, and uses downsampling after discrete polynomial transformation to further reduce a data amount, thereby achieving an advantage of small calculation amount. Under the condition of an equivalent amount of calculation, compared with the periodic fractional Fourier transform and the periodic Wigner-Hough transform algorithm, the algorithm of this application has an increase of about 60 dB in a detection SNR gain factor, and an increase of about 400 times in the detection distance of the receiver. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] To describe the technical solutions in embodiments of the present disclosure or in the prior art more clearly, the accompanying drawings required for the embodiments are briefly described below. Apparently, the accompanying drawings in the following description show merely some embodiments of the present disclosure, and those of ordinary skill in the art may still derive other accompanying drawings from these accompanying drawings without creative efforts.

[0039] FIG. 1 is a schematic flowchart of a method for detecting a low SNR periodic LFMCW signal according to the present disclosure.

[0040] FIG. 2 is a schematic principle diagram of a method for detecting a low SNR periodic LFMCW signal according to the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] The technical solutions of the embodiments of the present disclosure are clearly and completely described below with reference to the drawings in the embodiments of the present disclosure. Apparently, the described embodiments are merely a part rather than all of the embodiments of the present disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of the present disclosure without creative efforts shall fall within the protection scope of the present disclosure.

[0042] The objective of the present disclosure is to provide a method and system for detecting a low SNR periodic LFMCW signal, which can resolve the problem that an LFMCW signal detection method based on a conventional discrete polynomial transformation is hardly applicable to a low SNR scenario due to energy dispersion, thereby effectively improving a capability of detecting weak periodic LFMCW signals.

[0043] In order to make the above objective, features and advantages of the present disclosure clearer and more comprehensible, the present disclosure will be further described in detail below in combination with accompanying drawings and particular implementation modes.

[0044] As shown in FIG. 1 and FIG. 2, a method for detecting a low SNR periodic LFMCW signal provided in the present disclosure includes the following steps:

[0045] SI01. Obtain data to be detected, and perform periodic component detection on die data to be detected, where the data to be detected is digitally processed intermediate frequency or baseband data.

[0046] SI 02. If there is a periodic component, record a periodic estimate value according to the periodic component, and set a delay amount according to the periodic estimate value, where the delay amount r<Te / 2, and Te is the periodic estimate value.

[0047] SI03. Determine two adjacent observation windows according to the delay amount (t) and a difference (Te-x) obtained by subtracting the delay amount from the periodic estimate value.

[0048] SI04. Divide the data to be detected into continuous P groups of data by using the periodic estimate value, where P is a maximum value of the number of groups.

[0049] After SI04, the method further includes:

[0050] performing zero-padding on data obtained by segmenting a shorter observation window in the two adjacent observation windows to equalize lengths of the two groups of data, and performing fast Fourier transform and a modulo operation on segmented data to obtain a spectrum of each segment of data.

[0051] SI05. Slide a start position of an observation window by using a sampling point as a step, and perform segmented discrete polynomial transformation and sum spectrum calculation on the data to be detected corresponding to the sliding observation window at each step.

[0052] A start position of an observation window is slid by using a sampling point as a step, and downsampled data is segmented by using the observation window during each sliding of the start position of the observation window. In addition, zero-padding is performed on data obtained by segmenting a shorter window to equalize lengths of the two groups of data length. Fast Fourier transform and a modulo operation are performed on the segmented data to obtain a spectrum of each segment of data. Then, the spectrums of each segment of data are added to obtain a sum spectrum corresponding to each sliding of the observation window, and a maximum spectrum value thereof and its position are recorded. Finally, the foregoing step is repeated. If the maximum values of the spectrums have the same position, it is determined that the data to be detected includes an LFMCW signal, and a frequency modulation slope of the LFMCW signal is calculated according to a discrete polynomial transformation feature.

[0053] SI05 specifically includes:

[0054] q* sliding of the observation window: when TeX(p-l)+^<w<Tex(p-7)+g+(Te-T), obtaining a transformation sequence dlp by using a formula dlp( / 7)=x(n)where dip =[dlp(l), dlp(2), dlp(3),...], dlp(Z / ) is an / 7th data point of a first segment of the discrete polynomial transformation within a length of a p* periodic estimate value, li=(n mod Te)-g+l, (n mod Te) means dividing n by Te to obtain a remainder, * means performing a conjugation operation, q is a sequence number of sliding, Te is a periodic estimate value, t is a delay amount, and n is an nth sampling point;

[0055] downsampling the transformation sequence dlp, and then performing fast Fourier transform and a modulo operation on the downsampled transformation sequence dlp, to obtain a spectrum fl p of the first segment of the discrete polynomial transformation within the length of the p* periodic estimate value;

[0056] when Tex / ?+q-T<«<Texp, obtaining a transformation sequence d2p by using a formula d2p( / 2)=x(«)x*(n+T), where d2p=[d2p(l), d2p(2), d2p(3),...], d2p(h) is an / 2th data point of a second segment of the discrete polynomial transformation within the length of the pth periodic estimate value, and h={n mod Te)+r-^-Te+l);

[0057] padding the transformation sequence d2p with zeros at the end to equalize a data length thereof and a data length of dip, downsampling the zero-padded transformation sequence d2p, and then performing fast Fourier transform and a modulo operation on the downsampled transformation sequence d2p, to obtain a spectrum f2p of the second segment of the discrete polynomial transformation within the length of the pth periodic estimate value; and

[0058] adding the spectrum of each discrete polynomial transformation within the length of each periodic estimate value to obtain a sum spectrum fq corresponding to the q* sliding of the observation window.

[0059] SI06. Detect a periodic LFMCW signal in the data to be detected according to a maximum value of a sum spectrum at each step and a corresponding frequency.

[0060] SI06 specifically includes:

[0061] determining a step condition corresponding to the largest one in the maximum values of the sum spectrums;

[0062] determining whether there is a maximum value in sum spectrums of adjacent steps on both sides of the step condition corresponding to the largest one in the maximum values of the sum spectrum; and

[0063] if yes, determining that the data to be detected includes a periodic LFMCW signal; or

[0064] if no, determining that the data to be detected does not include a periodic LFMCW signal.

[0065] SI07. When a detection result is that there is a periodic LFMCW signal, calculate a frequency modulation slope and a frequency modulation inflection point position of the periodic LFMCW signal according to the frequency and the step corresponding to the maximum value of the sum spectrum.

[0066] Corresponding to the foregoing method, the present disclosure further provides a system for detecting a low SNR periodic LFMCW signal, including:

[0067] a periodic component detection module, configured to obtain data to be detected, and perform periodic component detection on the data to be detected;

[0068] a delay amount setting module, configured to: if there is a periodic component, record a periodic estimate value according to the periodic component, and set a delay amount according to the periodic estimate value;

[0069] an observation window determining module, configured to determine two adjacent observation windows according to the delay amount and a difference obtained by subtracting the delay amount from the periodic estimate value;

[0070] a grouping module, configured to divide the data to be detected into continuous P groups of data by using the periodic estimate value, where P is a maximum value of the number of groups;

[0071] a calculation module, configured to slide a start position of an observation window by using a sampling point as a step, and perform segmented discrete polynomial transformation and sum spectrum calculation on the data to be detected corresponding to the sliding observation window at each step;

[0072] a periodic LFMCW signal detection module, configured to detect a periodic LFMCW signal in the data to be detected according to a maximum value of a sum spectrum at each step and a corresponding frequency; and

[0073] a frequency modulation slope and frequency modulation inflection point position determining module, configured to: when a detection result is that there is a periodic LFMCW signal, calculate a frequency modulation slope and a frequency modulation inflection point position of the periodic LFMCW signal according to the frequency and the step corresponding to the maximum value of the sum spectrum.

[0074] Each embodiment in the description is described in a progressive mode, each embodiment focuses on differences from other embodiments, and references can be made to each other for the same and similar parts between embodiments. Since the system disclosed in an embodiment corresponds to the method disclosed in an embodiment, the description is relatively simple, and for related contents, references can be made to the description of the method.

[0075] Particular examples are used herein for illustration of principles and implementation modes of the present disclosure. The descriptions of the above embodiments are merely used for assisting in understanding the method of the present disclosure and its core ideas. In addition, those of ordinary skill in the art can make various modifications in terms of particular implementation modes and the scope of application in accordance with the ideas of Hie present disclosure. In conclusion, the content of the description shall not be construed as limitations to the present disclosure.

Claims

1. A method for detecting a low signal-to-noise ratio (SNR) periodic linear-frequency-modulated continuous-wave (LFMCW) signal, comprising:obtaining data to be detected, and performing periodic component detection on the data to be detected;if there is a periodic component, recording a periodic estimate value according to the periodic component, and setting a delay amount according to the periodic estimate value;determining two adjacent observation windows according to the delay amount and a difference obtained by subtracting the delay amount from the periodic estimate value;dividing the data to be detected into continuous P groups of data by using the periodic estimate value, wherein P is a maximum value of the number of groups;sliding a start position of an observation window by using a sampling point as a step, and performing segmented discrete polynomial transformation and sum spectrum calculation on the data to be detected corresponding to the sliding observation window at each step;detecting a periodic LFMCW signal in the data to be detected according to a maximum value of a sum spectrum at each step and a corresponding frequency; andwhen a detection result is that there is a periodic LFMCW signal, calculating a frequency modulation slope and a frequency modulation inflection point position of the periodic LFMCW signal according to the frequency and the step corresponding to the maximum value of the sum spectrum.

2. The method for detecting a low SNR periodic LFMCW signal according to claim 1, wherein the data to be detected is digitally processed intermediate frequency or baseband data.

3. The method for detecting a low SNR periodic LFMCW signal according to claim 1, wherein the delay amount r<Te / 2, wherein Te is the periodic estimate value.

4. The method for detecting a low SNR periodic LFMCW signal according to claim 1, wherein after the dividing the data to be detected into continuous P groups of data by using the periodic estimate value, the method further comprises:performing zero-padding on data obtained by segmenting a shorter observation window in the two adjacent observation windows.

5. The method for detecting a low SNR periodic LFMCW signal according to claim 1, wherein the sliding a start position of an observation window by using a sampling point as a step, and performing segmented discrete polynomial transformation and sum spectrum calculation on the data to be detected corresponding to the sliding observation window at each step specifically comprises:q* sliding of the observation window: when Tex(p-l)+^< / 2<Tex(p-7)+^+(Te-T), obtaining a transformation sequence dlp by using a formula dlp(Z / )=x(X)-x*(n-T) •, wherein dlp =[dlp(l), dlP(2), dlp(3),...], dlp(Z / ) is an data point of a first segment of the discrete polynomial transformation within a length of a pth periodic estimate value, h=(n mod Te)-^+l, (n mod Te) means dividing n by Te to obtain a remainder, * means performing a conjugation operation, q is a sequence number of sliding, Te is a periodic estimate value, r is a delay amount, and n is an n* sampling point;downsampling the transformation sequence dlp, and then performing fast Fourier transform and a modulo operation on the downsampled transformation sequence dlp, to obtain a spectrum flp of the first segment of the discrete polynomial transformation within the length of the pth periodic estimate value;when Texp+q-T<w<TeXp, obtaining a transformation sequence d2p by using a formula d2p( / 2)=x(«)-x*(w+t), wherein d2p=[d2p(l), d2p(2), d2p(3),...], d2p(Z2) is an data point of a second segment of the discrete polynomial transformation within the length of the plh periodic estimate value, and h=(n mod Te)+T-g-Te+l);padding the transformation sequence d2p with zeros at the end to equalize a data length thereof and a data length of dlp, downsampling the zero-padded transformation sequence d2p, and then performing fast Fourier transform and a modulo operation on the downsampled transformation sequence d2p, to obtain a spectrum f2p of the second segment of the discrete polynomial transformation within the length of the pth periodic estimate value; andadding the spectrum of each discrete polynomial transformation within the length of each periodic estimate value to obtain a sum spectrum fq corresponding to the qth sliding of the observation window.

6. The method for detecting a low SNR periodic LFMCW signal according to claim 1, wherein the detecting a periodic LFMCW signal in the data to be detected according to a maximum value of a sum spectrum at each step and a corresponding frequency specifically comprises:determining a step condition corresponding to the largest one in the maximum values of thesum spectrums;determining whether there is a maximum value in sum spectrums of adjacent steps on both sides of the step condition corresponding to the largest one in the maximum values of the sum spectrum; andif yes, determining that the data to be detected comprises a periodic LFMCW signal; orif no, determining that the data to be detected does not comprise a periodic LFMCW signal.

7. A system for detecting a low signal-to-noise ratio (SNR) periodic linear-frequency-modulated continuous-wave (LFMCW) signal, comprising:a periodic component detection module, configured to obtain data to be detected, and perform periodic component detection on the data to be detected;a delay amount setting module, configured to: if there is a periodic component, record a periodic estimate value according to the periodic component, and set a delay amount according to the periodic estimate value;an observation window determining module, configured to determine two adjacent observation windows according to the delay amount and a difference obtained by subtracting the delay amount from the periodic estimate value;a grouping module, configured to divide the data to be detected into continuous P groups of data by using the periodic estimate value, wherein P is a maximum value of the number of groups;a calculation module, configured to slide a start position of an observation window by using a sampling point as a step, and perform segmented discrete polynomial transformation and sum spectrum calculation on the data to be detected corresponding to the sliding observation window at each step;a periodic LFMCW signal detection module, configured to detect a periodic LFMCW signal in the data to be detected according to a maximum value of a sum spectrum at each step and a corresponding frequency; anda frequency modulation slope and frequency modulation inflection point position determining module, configured to: when a detection result is that there is a periodic LFMCW signal, calculate a frequency modulation slope and a frequency modulation inflection point position of the periodic LFMCW signal according to the frequency and the step corresponding to the maximum value of the sum spectrum.

Images