CN109379310B - Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method - Google Patents

Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method Download PDF

Info

Publication number
CN109379310B
CN109379310B CN201811066351.7A CN201811066351A CN109379310B CN 109379310 B CN109379310 B CN 109379310B CN 201811066351 A CN201811066351 A CN 201811066351A CN 109379310 B CN109379310 B CN 109379310B
Authority
CN
China
Prior art keywords
signal
frequency
rife
mpsk
interpolation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201811066351.7A
Other languages
Chinese (zh)
Other versions
CN109379310A (en
Inventor
方世良
魏阳杰
刘清宇
王晓燕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southeast University
Original Assignee
Southeast University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southeast University filed Critical Southeast University
Priority to CN201811066351.7A priority Critical patent/CN109379310B/en
Publication of CN109379310A publication Critical patent/CN109379310A/en
Application granted granted Critical
Publication of CN109379310B publication Critical patent/CN109379310B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/0014—Carrier regulation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/18—Phase-modulated carrier systems, i.e. using phase-shift keying
    • H04L27/22—Demodulator circuits; Receiver circuits
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/0014—Carrier regulation
    • H04L2027/0024—Carrier regulation at the receiver end

Abstract

The invention discloses a Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method, which comprises the following steps: s10, acquiring a MPSK signal sampling data sequence to be processed; s20, converting the MPSK signal carrier frequency estimation problem into a CW signal frequency estimation problem by carrying out multiple square transformation on the MPSK signal and removing the direct current component of the MPSK signal; s30, carrying out frequency estimation on the CW signal by using a Rife interpolation algorithm, and correcting the Rife interpolation direction by using a Quinn interpolation algorithm when the frequency spectrum leakage of the estimated signal is large; and S40, calculating the MPSK signal carrier frequency according to the frequency estimation value of the CW signal. Compared with the traditional Rife interpolation and Quinn interpolation, the method can accurately estimate the carrier frequency of the MPSK signal in the environment with low signal data volume and low signal-to-noise ratio, and is suitable for fast and steady estimation of the carrier frequency of the MPSK signal in engineering.

Description

Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method
Technical Field
The invention belongs to the field of signal processing, and particularly relates to a Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method.
Background
Due to good anti-noise performance, MPSK (multiple Phase Shift keying) signals are widely used in the field of aerial and underwater communication, so that the signals have important military significance for detecting and parameter estimation of uncooperative MPSK signals, accurate estimation of MPSK signal carrier frequency is a premise that many other signal parameters can be estimated, and currently, common MPSK signal carrier frequency estimation methods mainly include wavelet transformation methods, spectrum correlation methods, square transformation methods and the like.
The carrier frequency estimation method based on wavelet transformation can simultaneously estimate the carrier frequency and the code rate of signals under the condition of lower signal-to-noise ratio, but the performance difference of the carrier frequency is obviously estimated under different wavelet functions and scale factors by the method based on wavelet transformation, and no solution is found for how to select the optimal wavelet basis function and scale factor. The carrier frequency estimation method based on the spectrum correlation can simultaneously estimate the carrier frequency and the code rate of a signal under the condition of low signal-to-noise ratio and has stronger anti-jamming capability, but the method can only directly estimate the carrier frequency of a BPSK signal, other MPSK signal carrier frequencies can be used for carrier frequency estimation after nonlinear transformation, in addition, the complexity of an algorithm for calculating a signal spectrum correlation function is higher, a spectrum peak search needs to be carried out on a two-dimensional plane of the spectrum correlation function when the signal carrier frequency is estimated, and the operation amount of the algorithm is further increased, so that the algorithm is difficult to be used in engineering practice. The method based on square transformation generally converts an MPSK signal into a CW signal by using square transformation, and then performs carrier frequency estimation through a Rife interpolation or Quinn interpolation algorithm, which has low algorithm complexity, but the interpolation direction of the Rife interpolation or the Quinn interpolation may be wrong under the conditions of low signal-to-noise ratio and low data volume, so that the carrier frequency estimation precision is reduced, and how to reduce the error probability of the interpolation direction to improve the carrier frequency estimation precision is a problem to be solved urgently at present.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the defects of the prior art, the invention provides the MPSK signal carrier frequency estimation method which is small in calculation complexity and high in estimation accuracy under the environment with low signal-to-noise ratio and low signal data volume.
The technical scheme is as follows: in order to achieve the purpose, the invention adopts the following technical scheme:
a carrier frequency estimation method of MPSK signals based on Rife-Quinn synthesis comprises the following steps:
s10, acquiring MPSK signal sampling data sequence to be processed, wherein the signal sampling frequency is fs；
S20, converting the MPSK signal carrier frequency estimation problem into a CW signal frequency estimation problem by carrying out multiple square transformation on the MPSK signal and removing the direct current component of the MPSK signal;
s30, carrying out frequency estimation on the CW signal by using a Rife interpolation algorithm, and correcting the Rife interpolation direction by using a Quinn interpolation algorithm when the frequency spectrum leakage of the estimated signal is large;
and S40, calculating the MPSK signal carrier frequency according to the frequency estimation value of the CW signal.
Preferably, in step S10, the real-time collected data of N sampling points is received from the sensor, or the data of N sampling points starting from the time when the signal is detected is extracted from the memory as the MPSK signal data sequence x (N) to be processed, where N is 0,1, …, N-1.
Preferably, the step S20 includes the steps of:
s21, squaring the signal to obtain a signal x after the signal is squared2(n) is:
x2(n)＝x2(n),n＝0,1,2,…N-1 (1)
s22, squaring signal x2(n) Fourier transform to obtain its frequency spectrum X2(f) Namely:
X2(f)＝FFT{x2(n)} (2)
wherein f is-fs/2,-fs/2+fs/N,…fs/2-fsthe/N, FFT {. is } represents the calculation of Fourier transform; and make X2(0) 0, thereby removing the direct current component in the frequency spectrum; then for X at this time2(f) Performing inverse Fourier transform to obtain x2(n) time domain signal x 'after removal of DC component'2(n) is:
x'2(n)＝IFFT{X2(f)} (3)
wherein IFFT {. denotes inverse Fourier transform;
when the MPSK signal is a BPSK signal, only one square transformation is needed to remove direct current operation, and x 'at the moment'2(n) is a CW signal containing no DC component; for 4PSK signals, two square conversion DC removing operations are needed, namely, time domain signal x'2(n) the operations in steps S21, S22 are performed again, and so on for other MPSK-like signals.
Preferably, the step S30 includes the steps of:
s31 for x 'obtained in step S20'2(n) Fourier transform and taking its frequency spectrum X'2(f) The positive half-frequency part of (a), namely:
X'2(f)＝FFT{x'2(n)},f＞0 (4)
and look for | X'2(f) Maximum frequency at the peak of the spectrumPoint number index k of point correspondences0Where | · | represents a modulo operation, and:
k0＝max(k) (5)
wherein k represents | X'2(f) The discrete point index corresponding to the discrete frequency point corresponding to each spectrum peak, the value range of k is more than or equal to 1 and less than or equal to N/2-1, and max {. cndot } represents the operation of solving the maximum value;
s32, estimating x 'by Rife interpolation'2Frequency of (n)
Figure GDA0002779783040000031
The calculation formula of the Rife interpolation is as follows:
Figure GDA0002779783040000032
when | X'2(k0+1)|＞|X′2(k0-1) | is α ═ 1; when | X'2(k0+1)|≤|X′2(k0-1) |, α ═ 1, α denotes the interpolation direction of the Rife interpolation algorithm; simultaneously determining the ratio r0：
Figure GDA0002779783040000033
r0Represents a frequency spectrum X'2(k) At k0The degree of leakage of (c);
s33, p 'x'2(n) the signal is cut off to obtain the cut-off data x ″2(n) the length of truncation W is:
Figure GDA0002779783040000034
namely:
x″2(n)＝x'2(n),n＝0,1,2,…W-1 (9)
wherein round {. } represents a rounding operation, and for x'2' (n) Fourier transform to obtain its amplitudeThe power spectrum is taken and the positive frequency part is marked as X ″2(f1) Namely:
X″2(f1)＝|FFT{x″2(n)}|,f1＞0 (10)
where f is1＝-fs/2,-fs/2+fs/W,…fs/2-fsW, then look for X ″)2(f1) Discrete point index k 'corresponding to maximum discrete frequency point corresponding to spectral peak'0Namely:
k'0＝max(k') (11)
wherein k' represents X ″)2(f1) The discrete point index corresponding to the corresponding discrete frequency point at each spectral peak, the value range of k 'is more than or equal to 1 and less than or equal to k'. ltoreq.W/2-1, and the ratio r is solved1：
Figure GDA0002779783040000035
r1Denotes the frequency spectrum X ″2(k ') is k'0The degree of leakage of (c);
s34, comparison r0And r1If r is1≥r0X'2(n) is the final frequency estimate
Figure GDA0002779783040000036
If r1＜r0Then, the direction of the Rife interpolation algorithm is judged again by using the Quinn interpolation algorithm, and the specific process is as follows:
according to the Quinn interpolation algorithm, solving a frequency correction term delta in the Quinn interpolation algorithm, and then combining the step S32 to judge that:
if in step S32 | X'2(k0+1)|＞|X′2(k0-1) | and δ > 0, the interpolation direction of the Rife is considered correct, and no correction is needed to the interpolation direction, if | X'2(k0+1)|＞|X′2(k0-1) | and δ < 0, when the direction of the Rife interpolation is considered to be wrong, and let α ═ α, useFormula (6) is newly paired with x'2(n) estimating the frequency;
similarly, if | X 'in step S32'2(k0+1)|＜|X′2(k0-1) | and δ < 0, the direction of Rife interpolation is considered correct, and no correction is needed to the interpolation direction, if | X'2(k0+1)|＜|X′2(k0-1) | and δ > 0, at which point the direction of the Rife interpolation is considered as wrong, let α ═ α, and re-pair x 'using equation (6)'2The frequency of (n) is estimated.
Preferably, the step S40 includes the steps of:
according to the step S30, the frequency estimated after the Rife interpolation algorithm is corrected through the Quinn interpolation algorithm
Figure GDA0002779783040000041
Carrier frequency F of MPSK signalcThe calculation formula is as follows:
Figure GDA0002779783040000042
where M is 2 when x (n) is a BPSK signal, M is 4psk (qpsk), and M is 4 when x (n) is a BPSK signal, and so on.
Has the advantages that: the invention converts the MPSK signal into the CW signal by the square transformation DC-removing method, and estimates the carrier frequency of the MPSK signal by carrying out the Rife-Quinn comprehensive frequency estimation method on the CW signal. The method greatly improves the carrier frequency estimation precision of the CW signal under the environment with low signal-to-noise ratio and low signal data quantity by integrating the Rife interpolation algorithm and the Quinn interpolation algorithm, further improves the carrier frequency estimation precision of the MPSK signal, has small calculated amount, and is suitable for quickly and stably estimating the MPSK signal carrier frequency in engineering.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
fig. 2 is a time domain diagram of a BPSK signal according to an embodiment of the present invention;
fig. 3 is a BPSK signal power spectrum according to an embodiment of the present invention;
fig. 4 is a magnitude spectrogram of a BPSK signal after dc removal by square according to an embodiment of the present invention;
fig. 5 is a graph comparing the performance of the method of the present invention and the prior art method on carrier frequency estimation of BPSK signal.
Detailed Description
The technical scheme of the invention is further explained by combining the attached drawings.
As shown in fig. 1, the method for estimating carrier frequency of MPSK signal based on Rife-Quinn synthesis of the present invention includes the following steps:
and S10, acquiring the MPSK signal sampling data sequence to be processed.
Receiving real-time acquisition data of N sampling points from a sensor as an MPSK signal data sequence x (N) to be processed, wherein N is 0,1, … and N-1, or extracting data of N sampling points starting from the moment of detecting the signal from a memory as the MPSK signal data sequence x (N) to be processed, N is 0,1, … and N-1, and the sampling frequency of the signal is recorded as fs。
S20, converting the MPSK signal carrier frequency estimation problem into the CW signal frequency estimation problem by performing multiple square transform on the MPSK signal and removing the dc component thereof.
The method specifically comprises the following steps:
s21, squaring the signal to obtain a signal x after the signal is squared2(n) that is
x2(n)＝x2(n),n＝0,1,2,…N-1 (1)
S22, then squaring the signal x2(n) Fourier transform to obtain its frequency spectrum X2(f) Namely:
X2(f)＝FFT{x2(n)} (2)
wherein f denotes the signal frequency, f-fs/2,-fs/2+fs/N,…fs/2-fsthe/N, FFT {. cndot } represents the Fourier transform operation. And make X2(0) 0, thereby removing the dc component in the spectrum. Then, for X at this time2(f) Performing inverse Fourier transform to obtain x2(n) removing DC componentPost-quantization time-domain signal x'2(n) is:
x'2(n)＝IFFT{X2(f)} (3)
where IFFT { · } represents an inverse fourier transform.
When the MPSK signal is a BPSK signal, only one square transformation is needed to remove direct current operation, and x 'at the moment'2The CW signal (sinusoidal signal) containing no dc component is (n). For 4PSK signals, two square conversion DC removing operations are needed, namely, time domain signal x'2(n) performing the operations in steps S21, S22 again; three times of square transformation de-direct current operation is needed for 8PSK signals, namely steps S21-S22 need to be performed three times. Other MPSK like signals and so on.
S30, processing the obtained signal x 'in S20 by adopting a Rife-Quinn comprehensive algorithm'2(n) performing frequency estimation.
The Rife-Quinn comprehensive algorithm in the invention means that a Rife interpolation algorithm is used for carrying out frequency estimation on a CW signal, and when the frequency spectrum leakage of the estimated signal becomes large, the Quinn interpolation algorithm is used for correcting the Rife interpolation direction. The method specifically comprises the following steps:
s31, first comparing the x 'obtained in the step S20'2(n) Fourier transform and taking its frequency spectrum X'2(f) The positive half-frequency part of (a), namely:
X'2(f)＝FFT{x'2(n)},f＞0 (4)
and look for | X'2(f) The point index k corresponding to the maximum frequency point corresponding to the peak of the | spectrum (the frequency corresponding to the frequency point is the frequency of the CW signal)0Where | · | represents a modulo operation, and:
k0＝max(k) (5)
wherein k represents | X'2(f) And | discrete point index corresponding to the corresponding discrete frequency point at each spectral peak, k is within the range of 1 to N/2-1, and max {. can } represents the operation of solving the maximum value.
S32, followed by estimating x 'using Rife interpolation'2Frequency of (n)
Figure GDA0002779783040000061
The calculation formula of the Rife interpolation is as follows:
Figure GDA0002779783040000062
when | X'2(k0+1)|＞|X′2(k0-1) | is α ═ 1; when | X'2(k0+1)|≤|X′2(k0-1) |, α ═ 1, α denotes the interpolation direction of the Rife interpolation algorithm. Simultaneously determining the ratio r0：
Figure GDA0002779783040000063
According to the Fourier transform theory, r0Can reflect the frequency spectrum X'2(k) At k0Degree of leakage of (d)0The larger the size, the smaller the degree of spectral leakage, and vice versa0Smaller indicates greater spectral leakage.
S33, then p'2(n) the signal is cut off to obtain the cut-off data x ″2(n) the length of truncation W is:
Figure GDA0002779783040000064
namely:
x″2(n)＝x'2(n),n＝0,1,2,…W-1 (9)
where round {. cndot } represents a rounding operation and is on x ″ "2(n) Fourier transform is carried out to obtain a magnitude spectrum of the frequency spectrum and the positive frequency part of the magnitude spectrum is marked as X ″2(f1) Namely:
X″2(f1)＝|FFT{x″2(n)}|,f1＞0 (10)
where f is1＝-fs/2,-fs/2+fs/W,…fs/2-fs/W，Then look for X ″)2(f1) Discrete point index k 'corresponding to maximum discrete frequency point corresponding to spectral peak'0Namely:
k'0＝max(k') (11)
wherein k' represents X ″)2(f1) The discrete point index corresponding to the corresponding discrete frequency point at each spectral peak, the value range of k 'is more than or equal to 1 and less than or equal to k'. ltoreq.W/2-1, and the ratio r is solved1：
Figure GDA0002779783040000071
For the same reason r1Can reflect the frequency spectrum X ″)2(k ') is k'0The degree of leakage of (c).
S34, comparing the leakage degree at the spectral peak before and after Rife interpolation estimation, if r1≥r0X'2(n) is the final frequency estimate
Figure GDA0002779783040000072
If r1＜r0If so, it is indicated that the direction of the Rife interpolation in the formula (6) may be wrong, and at this time, the Quinn interpolation algorithm is used to re-determine the direction of the Rife interpolation algorithm, and the specific process is as follows:
firstly, according to a Quinn interpolation algorithm, solving a frequency correction term delta in the Quinn interpolation algorithm, wherein a calculation formula of the delta is as follows:
Figure GDA0002779783040000073
wherein:
Figure GDA0002779783040000074
wherein:
Figure GDA0002779783040000075
wherein: re {. is used for representing the real part operation of the data. Then if in step S32, | X'2(k0+1)|＞|X′2(k0-1) | and δ > 0, the interpolation direction of the Rife is considered correct, and no correction is needed to the interpolation direction, if | X'2(k0+1)|＞|X′2(k0-1) | and δ < 0, in which case the direction of the Rife interpolation is considered to be wrong, α ═ α, and x 'is newly corrected using equation (6)'2The frequency of (n) is estimated. Similarly, if in step S32, | X'2(k0+1)|＜|X′2(k0-1) | and δ < 0, the direction of Rife interpolation is considered correct, and no correction is needed to the interpolation direction, if | X'2(k0+1)|＜|X′2(k0-1) | and δ > 0, at which point the direction of the Rife interpolation is considered as wrong, let α ═ α, and re-pair x 'using equation (6)'2The frequency of (n) is estimated.
In the steps S20 and S30, the fourier transform is a fast fourier transform, which improves the computation efficiency.
S40, according to the frequency estimated after the Rife interpolation algorithm is corrected through the Quinn interpolation algorithm in the step S30
Figure GDA0002779783040000076
Calculating MPSK signal carrier frequency FcThe calculation formula is as follows:
Figure GDA0002779783040000081
where M is 2 when x (n) is a BPSK signal, M is 4 when x (n) is 4PSK (qpsk), M is 8 when x (n) is 8PSK, and so on.
According to the above-detailed MPSK signal carrier frequency estimation method based on Rife-Quinn synthesis, the effect of the invention is verified by a simulation experiment. In this embodiment, the mathematical model of the MPSK signal is:
x(t)＝s(t)+v(t)＝Acos(2πfct+θk)+v(t) (17)
where t represents time in seconds, A represents signal amplitude and is constant, v (t) is noise not related to signal s (t), fcRepresenting the carrier frequency of the signal, thetakA set of uniformly spaced modulated phases whose values depend on the values of the baseband symbols, so it can be written as:
Figure GDA0002779783040000082
m is a baseband symbol number, M is usually a power of 2, and x (t) is a BPSK signal when M is 2. In this embodiment, M is 2, that is, the BPSK signal is verified by simulation.
The simulation signal parameters are set as follows: sampling frequency fs10000Hz, carrier frequency f of BPSK signalc1276.85Hz, the code rate is 1000bit/s, and the noise is white Gaussian noise.
Figure 2 is a time domain diagram of a BPSK signal; fig. 3 is a power spectrum of a BPSK signal; fig. 4 shows a fourier-transformed amplitude spectrum of a CW signal, which is a signal obtained by removing dc by BPSK square transformation. Fig. 5 shows the comparison between the performance of the carrier frequency of BPSK signal estimated by the method of the present invention and the performance of the Rife interpolation and the Quinn interpolation estimation and the CRLB (cramer-mello boundary, a lower limit of the parameter estimation performance), where the data length is 0.423s and the number of monte carlo trials is 1000, and it can be seen from the figure that the performance of the method of the present invention is better than that of the conventional Rife interpolation and the Quinn interpolation in the low signal-to-noise ratio and low signal data volume environments, and the frequency performance estimated by the method is closer to the CRLB in the low signal-to-noise ratio environments.
The result of the embodiment shows that the error between the signal carrier frequency estimated by the MPSK signal carrier frequency estimation method and the real carrier frequency is very small in the low signal-to-noise ratio environment, and the method is suitable for the occasion of quickly and accurately estimating the MPSK signal carrier frequency.

Claims (4)

1. A carrier frequency estimation method of MPSK signals based on Rife-Quinn synthesis is characterized by comprising the following steps:
s10, acquiring MPSK signal sample data sequence to be processedNumber sampling frequency of fs；
S20, converting the MPSK signal carrier frequency estimation problem into the CW signal frequency estimation problem by performing multiple square transform on the MPSK signal and removing the dc component thereof, including:
s21, squaring the signal to obtain a signal x after the signal is squared2(n) is:
x2(n)＝x2(n),n＝0,1,2,…N-1 (1)
s22, squaring signal x2(n) Fourier transform to obtain its frequency spectrum X2(f) Namely:
X2(f)＝FFT{x2(n)} (2)
wherein f is-fs/2,-fs/2+fs/N,…fs/2-fsthe/N, FFT {. is } represents the calculation of Fourier transform; and make X2(0) 0, thereby removing the direct current component in the frequency spectrum; then for X at this time2(f) Performing inverse Fourier transform to obtain x2(n) time domain signal x 'after removal of DC component'2(n) is:
x'2(n)＝IFFT{X2(f)} (3)
wherein IFFT {. denotes inverse Fourier transform;
when the MPSK signal is a BPSK signal, only one square transformation is needed to remove direct current operation, and x 'at the moment'2(n) is a CW signal containing no DC component; for 4PSK signals, two square conversion DC removing operations are needed, namely, time domain signal x'2(n) performing the operations in steps S21, S22 again, and so on for other MPSK-like signals;
s30, performing frequency estimation on the CW signal by using a Rife interpolation algorithm, and correcting the Rife interpolation direction by using a Quinn interpolation algorithm when the estimated signal spectrum leakage becomes large, the method includes:
s31 for x 'obtained in step S20'2(n) Fourier transform and taking its frequency spectrum X'2(f) The positive half-frequency part of (a), namely:
X'2(f)＝FFT{x'2(n)},f＞0 (4)
and searching a point index k corresponding to the maximum frequency point corresponding to the spectrum peak of the | X' 2(f) |0Where | · | represents a modulo operation, and:
k0＝max(k) (5)
wherein k represents | X'2(f) The discrete point index corresponding to the discrete frequency point corresponding to each spectrum peak, the value range of k is more than or equal to 1 and less than or equal to N/2-1, and max {. cndot } represents the operation of solving the maximum value;
s32, estimating x 'by Rife interpolation'2Frequency of (n)
Figure FDA0002779783030000021
The calculation formula of the Rife interpolation is as follows:
Figure FDA0002779783030000022
when | X'2(k0+1)|＞|X′2(k0-1) | is α ═ 1; when | X'2(k0+1)|≤|X′2(k0-1) |, α ═ 1, α denotes the interpolation direction of the Rife interpolation algorithm; simultaneously determining the ratio r0：
Figure FDA0002779783030000023
r0Represents a frequency spectrum X'2(k) At k0The degree of leakage of (c);
s33, p 'x'2(n) the signal is cut off to obtain the cut-off data x ″2(n) the length of truncation W is:
Figure FDA0002779783030000024
namely:
x″2(n)＝x'2(n),n＝0,1,2,…W-1 (9)
where round {. cndot } represents a rounding operation and is on x ″ "2(n) Fourier transform is carried out to obtain a magnitude spectrum of the frequency spectrum and the positive frequency part of the magnitude spectrum is marked as X ″2(f1) Namely:
X″2(f1)＝|FFT{x″2(n)}|,f1＞0 (10)
where f is1＝-fs/2,-fs/2+fs/W,…fs/2-fsW, then look for X ″)2(f1) Discrete point index k 'corresponding to maximum discrete frequency point corresponding to spectral peak'0Namely:
k'0＝max(k') (11)
wherein k' represents X ″)2(f1) The discrete point index corresponding to the corresponding discrete frequency point at each spectral peak, the value range of k 'is more than or equal to 1 and less than or equal to k'. ltoreq.W/2-1, and the ratio r is solved1：
Figure FDA0002779783030000025
r1Denotes the frequency spectrum X ″2(k ') is k'0The degree of leakage of (c);
s34, comparison r0And r1If r is1≥r0X'2(n) is the final frequency estimate
Figure FDA0002779783030000026
If r1＜r0Then, the direction of the Rife interpolation algorithm is judged again by using the Quinn interpolation algorithm, and the specific process is as follows:
according to the Quinn interpolation algorithm, solving a frequency correction term delta in the Quinn interpolation algorithm, and then combining the step S32 to judge that:
if in step S32 | X'2(k0+1)|＞|X′2(k0-1) | and δ > 0, when the direction of the Rife interpolation is considered correct, without having to do soCorrecting the interpolation direction to obtain a value of | X'2(k0+1)|＞|X′2(k0-1) | and δ < 0, in which case the direction of the Rife interpolation is considered to be wrong, α ═ α, and x 'is newly corrected using equation (6)'2(n) estimating the frequency;
similarly, if | X 'in step S32'2(k0+1)|＜|X′2(k0-1) | and δ < 0, the direction of Rife interpolation is considered correct, and no correction is needed to the interpolation direction, if | X'2(k0+1)|＜|X′2(k0-1) | and δ > 0, at which point the direction of the Rife interpolation is considered as wrong, let α ═ α, and re-pair x 'using equation (6)'2(n) estimating the frequency;
and S40, calculating the MPSK signal carrier frequency according to the frequency estimation value of the CW signal.
2. The method for estimating carrier frequency of MPSK signal based on Rife-Quinn synthesis as claimed in claim 1, wherein in step S10, the real-time collected data of N sampling points is received from the sensor, or the data of N sampling points starting from the moment of detecting the signal is extracted from the memory as the MPSK signal data sequence x (N) to be processed, where N is 0,1, …, N-1.
3. The method for estimating carrier frequency of MPSK signal based on Rife-Quinn synthesis as claimed in claim 1, wherein the step S40 comprises the following steps:
according to the step S30, the frequency estimated after the Rife interpolation algorithm is corrected through the Quinn interpolation algorithm
Figure FDA0002779783030000031
Carrier frequency F of MPSK signalcThe calculation formula is as follows:
Figure FDA0002779783030000032
where M is 2 when x (n) is a BPSK signal, M is 4psk (qpsk), and M is 4 when x (n) is a BPSK signal, and so on.
4. The method for carrier frequency estimation of MPSK signals based on Rife-Quinn synthesis as claimed in claim 1, wherein the Fourier transform is fast Fourier transform.
CN201811066351.7A 2018-09-13 2018-09-13 Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method Active CN109379310B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811066351.7A CN109379310B (en) 2018-09-13 2018-09-13 Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811066351.7A CN109379310B (en) 2018-09-13 2018-09-13 Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method

Publications (2)

Publication Number Publication Date
CN109379310A CN109379310A (en) 2019-02-22
CN109379310B true CN109379310B (en) 2021-01-26

Family

ID=65404541

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811066351.7A Active CN109379310B (en) 2018-09-13 2018-09-13 Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method

Country Status (1)

Country Link
CN (1) CN109379310B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110007148B (en) * 2019-03-28 2021-03-16 东南大学 Single-frequency signal frequency estimation method based on comprehensive interpolation of discrete spectrum phase and amplitude
CN110851781B (en) * 2019-11-01 2020-08-11 东南大学 Channel impulse response window estimation method based on frequency domain energy and group delay screening
CN112333123B (en) * 2020-10-10 2022-06-03 南京世海声学科技有限公司 Non-cooperative PSK underwater acoustic communication signal multi-wavelet-based automatic preferred baseband demodulation method
CN114280366B (en) * 2021-12-21 2023-10-31 中国航天科工集团八五一一研究所 Sinusoidal signal frequency estimation method based on improved frequency interpolation algorithm
CN114268341B (en) * 2021-12-21 2024-02-13 中国航天科工集团八五一一研究所 Direct sequence spread spectrum signal pseudo code rate estimation method based on cyclostationary characteristic

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101179549A (en) * 2007-12-14 2008-05-14 清华大学 Communication signal carrier frequency estimation method employing three-point weight interpolation algorithm
CN106443178A (en) * 2016-09-08 2017-02-22 东南大学 IQuinn-Rife integration based sinusoidal signal frequency estimation method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8675744B1 (en) * 2007-09-24 2014-03-18 Atmel Corporation Channel tracking in an orthogonal frequency-division multiplexing system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101179549A (en) * 2007-12-14 2008-05-14 清华大学 Communication signal carrier frequency estimation method employing three-point weight interpolation algorithm
CN106443178A (en) * 2016-09-08 2017-02-22 东南大学 IQuinn-Rife integration based sinusoidal signal frequency estimation method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
非合作通信信号侦察算法研究与实现;靳鑫;《中国优秀硕士论文全文数据库 信息科技辑》;20180315;18-19页 *

Also Published As

Publication number Publication date
CN109379310A (en) 2019-02-22

Similar Documents

Publication Publication Date Title
CN109379310B (en) Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method
JP4388118B2 (en) Modulation method estimation apparatus and method
CN110007148B (en) Single-frequency signal frequency estimation method based on comprehensive interpolation of discrete spectrum phase and amplitude
CN103941089B (en) Sinusoidal signal frequency method of estimation based on DFT
CN109285561B (en) Ship propeller cavitation noise modulation spectrum feature fidelity enhancement method based on self-adaptive window length
CN108845973B (en) Doppler frequency estimation method based on improved Quinn algorithm
CN110068727B (en) Single-frequency signal frequency estimation method based on Candan-Rife comprehensive interpolation
CN104901909B (en) The method for parameter estimation of chirp signals under a kind of α non-Gaussian noises
CN110691051A (en) GMSK signal frequency offset estimation algorithm based on FFT
CN110690931A (en) Digital signal adaptive code rate estimation method and device based on multi-wavelet-base combination
CN114280366B (en) Sinusoidal signal frequency estimation method based on improved frequency interpolation algorithm
CN110196407B (en) Single-vector hydrophone signal incoming wave direction estimation method based on frequency estimation
CN109359633B (en) Signal joint classification method based on Hilbert-Huang transform and wavelet ridge line
CN103152294B (en) The method and system carrying out noise estimation are eliminated based on signal
CN108900445B (en) Method and device for estimating signal symbol rate
CN109660475B (en) A kind of non-cooperation phase code water sound communication signal autonomous identifying method
CN102087313B (en) Frequency estimation method for satellite search and rescue signal
CN104270328B (en) A kind of signal to noise ratio real-time estimation method
CN113468474B (en) Power grid frequency estimation method based on root Mini-Norm
CN110198196B (en) Amplitude estimation method based on signal intensity in communication system
CN114268341B (en) Direct sequence spread spectrum signal pseudo code rate estimation method based on cyclostationary characteristic
CN111611686A (en) Detection method for communication signal time-frequency domain
Tang et al. A modified blind OFDM systems parameters estimation method
CN110224769B (en) Estimation method for joint amplitude and noise variance in communication system
Wang et al. Non-data-aided frequency offset estimation for binary CPM signals

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant