Open-loop carrier synchronization and demodulation method for large Doppler PM signal

Abstract

The invention discloses an open-loop carrier synchronization and demodulation method for a large Doppler PM signal, and relates to the field of information transmission and processing. The method comprises the following steps: firstly, carrying out band-pass sampling and digital down-conversion on a received intermediate-frequency large Doppler PM signal, and carrying out integral cleaning and low-pass filtering to obtain I and Q paths of baseband signals; performing Doppler change rate compensation, Doppler frequency acquisition and Doppler frequency compensation to obtain an orthogonal zero intermediate frequency signal; calculating a phase error, performing linear correction, and performing phase compensation to demodulate a subcarrier modulation signal; and finally, down-conversion and low-pass filtering are carried out, and baseband data are demodulated after high frequency doubling is filtered out. Compared with other closed-loop carrier phase extraction methods, the method does not need a feedback loop structure, avoids the hang up problem of the feedback loop structure, has rapid capture performance and high estimation precision, can be suitable for carrier synchronization of burst large Doppler PM signals, and is also suitable for carrier synchronization of a GPU scheme.

Description

Technical Field

[0001] The present invention relates to the field of information transmission and processing, and particularly to an open-loop carrier synchronization and demodulation method for a large Doppler PM signal. Background Art

[0002] Reliable coherent communication depends on good synchronization of carrier frequency and phase. In actual signal demodulation, Doppler frequency shift, the offset of the center frequency of the receiver local oscillator, and phase noise will cause carrier frequency and phase errors. Therefore, actual communication systems need to correct frequency difference and phase difference.

[0003] For large Doppler PM signals, the usual carrier synchronization methods are carrier acquisition based on FFT analysis and residual carrier phase-locked tracking. For a large range of input signal frequency change rates, the frequency change amount within the FFT analysis time is greater than the FFT resolution, and correct analysis and acquisition cannot be performed. For the residual carrier phase-locked loop, carrier locking requires a certain time. If the initial phase difference is near an unstable equilibrium point, the loop will hesitate and take a long time to lock in. This is the hang up problem of the feedback loop structure. Burst signals have relatively high requirements for carrier acquisition time, and the phase-locked loop method is not very suitable for demodulating burst signals. For software basebands implemented by GPUs, different functions can be completed by loading different programs, and software development and debugging are faster and more convenient than hardware implementation. Due to the feedback structure design of the carrier tracking loop, the front and back data have a dependency relationship, which is not suitable for being completed on the GPU and needs to be completed on the CPU. And the data interaction between the CPU and the GPU will cause a long delay. Therefore, the GPU baseband is not suitable for the feedback loop structure. Summary of the Invention

[0004] In view of this, the present invention proposes an open-loop carrier synchronization and demodulation method for a large Doppler PM signal. This method is applicable to carrier synchronization of burst large Doppler PM signals and also applicable to carrier synchronization of GPU solutions.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] An open-loop carrier synchronization and demodulation method for a large Doppler PM signal, comprising the following steps:

[0007] Step 1, perform band-pass sampling on the received intermediate-frequency large Doppler PM signal to obtain an intermediate-frequency PM sampling signal, multiply the local quadrature carrier by the intermediate-frequency PM sampling signal for digital down-conversion, and obtain two baseband signals I(k), Q(k) after integral cleaning and low-pass filtering, including the modulated sub-carrier signal and the phase difference caused by the Doppler frequency and the initial phase;

[0008] Step 2: Respectively perform Doppler rate compensation on the I and Q baseband signals and N groups of orthogonal swept-frequency signals generated locally in a loop at a certain rate of change to obtain N groups of orthogonal Doppler rate compensation signals. Perform FFT analysis on the N groups of Doppler rate compensation signals respectively. By selecting the maximum value from the FFT analysis results of each group and then selecting the maximum value among the N maximum values, the corresponding Doppler frequency and frequency change rate values are obtained. The frequency change amount of the signal after Doppler rate compensation within the single FFT analysis time of the maximum Doppler rate is less than the FFT frequency resolution, so that correct analysis and capture can be performed. Perform Doppler frequency compensation on the Doppler rate compensation signals according to the FFT analysis results to obtain the I and Q static baseband modulation signals;

[0009] Step 3: Perform integration cleaning and narrowband low-pass filtering on the I and Q static baseband modulation signals again to filter out the modulation subcarrier frequencies and obtain the quadrature zero intermediate frequency signals, which include the phase difference caused by the FFT analysis frequency difference and the initial phase. Calculate the phase angle through the I and Q zero intermediate frequency signals, then average the sine and cosine of the phase angle, and calculate the phase error through the arctangent of the average value;

[0010] Step 4: Perform linear correction on the calculated phase error value to make the phase error meet the actual phase value range, and obtain the average carrier phase error of each point. Use the obtained average carrier phase error to perform phase compensation on the two baseband signals including the modulation subcarrier signal and the phase difference caused by the FFT analysis frequency difference and the initial phase to eliminate the phase error and demodulate the subcarrier modulation signal; when performing phase compensation on the two baseband signals including the phase error, the phase error needs to be interpolated to make the sampling rate of the phase error the same as that of the two baseband signals;

[0011] Step 5: Take the 1-fold frequency and 3-fold frequency of the subcarrier of the demodulated subcarrier modulation signal, then multiply by the local orthogonal subcarrier and perform down-conversion and low-pass filtering, filter out the high multiple frequencies, and demodulate the baseband data; complete the open-loop carrier synchronization and demodulation of the large Doppler PM signal.

[0012] Due to the adoption of the above technical solutions, the beneficial effects of the present invention compared with the prior art are as follows:

[0013] 1. Compared with other carrier acquisition methods for large Doppler signals, the present invention adds a Doppler rate compensation method, effectively cancels the Doppler change of the input signal, and reduces the influence of the Doppler rate on acquisition.

[0014] 2. Compared with other closed-loop carrier phase extraction methods, the present invention does not require a feedback loop structure, avoids the hang-up problem of the feedback loop structure, and has the advantages of fast acquisition performance and high estimation accuracy.

[0015] 3. The present invention is applicable to the carrier synchronization of burst Doppler PM signals and also to the carrier synchronization of GPU solutions. Description of the Drawings

[0016] Figure 1 It is a schematic diagram of the working process electrical principle of an open-loop carrier synchronization and demodulation method for a multi-Doppler PM signal in an embodiment of the present invention.

[0017] Figure 1 In the figure: 1 is a band-pass sampling digital down-conversion and low-pass filtering unit, 2 is a Doppler rate-of-change compensation and FFT analysis unit, 3 is a narrow-band filtering phase error extraction unit, 4 is a phase error linear correction and carrier phase compensation unit, and 5 is a sub-carrier demodulation unit. Detailed Embodiment

[0018] The following further describes the content of the present invention in conjunction with the drawings and specific embodiments.

[0019] An open-loop carrier synchronization and demodulation method for a multi-Doppler PM signal, as Figure 1 shown, the embodiment of the present invention is implemented based on a band-pass sampling digital down-conversion and low-pass filtering unit 1, a Doppler rate-of-change compensation and FFT analysis unit 2, a narrow-band filtering phase error extraction unit 3, a phase error linear correction and carrier phase compensation unit 4, and a sub-carrier demodulation unit 5. Figure 1 It is a block diagram of the working process electrical principle of an embodiment of the present invention. The embodiment is connected according to Figure 1 the connection lines.

[0020] As Figure 1 shown, an open-loop carrier synchronization method for a multi-Doppler PM signal includes the steps of:

[0021] 1) The band-pass sampling digital down-conversion and low-pass filtering unit performs band-pass sampling on the received intermediate-frequency PM signal to obtain an intermediate-frequency PM sampling signal. Among them, the received intermediate-frequency PM signal is:

[0022] s(t) = Acos(ω c t + ω d t + ω' d t 2 + θ + m f m(t)sin(Ωt)) + n(t);

[0023] A: Signal amplitude, ω c : Intermediate frequency, ω d : Doppler frequency, ω' d : Doppler frequency rate of change, θ: Initial phase, m f m(t)sin(Ωt): Sub-carrier modulation signal, m(t): Information data, m f: Modulation index, Ω: Sub - carrier modulation frequency, n(t): Noise;

[0024] Multiply the local orthogonal carrier with the intermediate - frequency PM sampled signal for digital down - conversion, and obtain two base - band signals I(k1) and Q(k1) of the I and Q channels through integral cleaning and low - pass filtering.

[0025]

[0026] where k1 is the sampling - point index after the first integral cleaning and low - pass filtering to reduce the sampling rate, and T s1 is the corresponding sampling period;

[0027] The bandwidth of the phase - modulated signal can be approximately calculated by the formula B PM = 2(β PM + 1)f m where f m : Sub - carrier frequency. Here, B PM is taken as 500k, the cut - off frequency of the low - pass filter is selected as 500kHz, and the two base - band signals of the I and Q channels contain the modulated sub - carrier signal and the phase difference caused by the Doppler frequency and the initial phase.

[0028] 2) Doppler rate - of - change compensation and FFT analysis unit for the two base - band signals of the I and Q channels

[0029] and N groups of orthogonal swept - frequency signals generated locally in a loop with a certain rate of change for Doppler rate - of - change compensation:

[0030]

[0031] Obtain N groups of orthogonal Doppler rate - of - change compensation signals. Among them, ω′ dn ={ω′ d1 ,ω′ d2 ,……,ω′ dN}, which is the N groups of Doppler frequency rates of change set locally. Perform FFT analysis on each of the N groups of Doppler rate - of - change compensation signals respectively. By selecting the maximum value from the FFT analysis results of each group, and then selecting the maximum value among the N maximum values, the corresponding Doppler frequency and frequency rate of change values are obtained. The frequency change amount of the maximum Doppler rate of change of the signal after Doppler rate - of - change compensation within the single FFT analysis time is less than the FFT frequency resolution, so that correct analysis and capture can be carried out. Then, perform Doppler frequency compensation on the Doppler rate - of - change compensation signal according to the FFT analysis results:

[0032]

[0033] Obtain two static base - band modulated signals of the I and Q channels. ωdf is the Doppler frequency obtained by FFT analysis, ω Δd is the deviation between the Doppler frequency obtained by FFT analysis and the actual Doppler frequency.

[0034] 3) The narrowband filtering phase error extraction unit further performs integral cleaning on the I and Q static baseband modulation signals to reduce the sampling rate, and then performs narrowband low-pass filtering to make the subcarrier modulation frequency Ω outside the passband of the low-pass filtering characteristic, filtering out the modulation frequency Ω to obtain an orthogonal zero intermediate frequency signal. The cut-off frequency of the low-pass filter is 2k,

[0035]

[0036] k2 is the sampling point index after the second integral cleaning and low-pass filtering to reduce the sampling rate, T s2 is the corresponding sampling period; n I (k2T s2 ) and n Q (k2T s2 ) are the noises of the in-phase and quadrature branches, and they are also narrowband Gaussian processes. Let φ(k2) = ω Δd k2T s2 +θ, which unifies the FFT analysis frequency difference and phase difference into the range of phase error. Carrier recovery needs to obtain the real-time phase error of the carrier.

[0037] The phase difference can be calculated by arctg(Q c (k2) / I c (k2)) = ω Δd k2T s2 +θ. To improve the signal-to-noise ratio, the estimated phase difference is averaged. Since If the phase angle is directly used for moving average, there may be to jumps that cause a large deviation in the average value. The sine and cosine functions are continuous throughout the region, and the situation of mean jump can be completely avoided. Here, the sine and cosine of the phase angle are averaged and then the phase error is calculated through arctangent.

[0038] First, find the phase angle angle(I c (k2)+i*Q c (k2)) = φ(k2), where i represents the imaginary unit. Then, the sine and cosine of the phase angle are averaged and the arctangent is taken to obtain the estimated phase error:

[0039] Taking adjacent 2 M φ(k2) as a group for grouping, calculate the average value of 2 M sin(φ(k2)) within each group and denote it as Calculate 2 within each groupM The average value of cos(φ(k2)) is denoted as Then the phase error is:

[0040] M is an integer greater than 0. The value of M should consider both the accuracy and the influence of the Doppler frequency change rate on the adjustment interval of the phase error estimation.

[0041] 4) The phase error linear correction and carrier phase compensation unit linearly corrects the estimated phase error value. The calculated phase error value is between and While the actual phase value range is from 0 to 2π. Therefore, it is necessary to linearly correct the phase error estimation value:

[0042] If k3 = 1, it is judged whether φ′(k3) < 0 holds. If it holds, then φ′(k3) = φ′(k3) + 2π;

[0043] Otherwise, φ′(k3) = φ′(k3);

[0044] If k3 = 1 does not hold, then:

[0045] When φ′(k3) = φ′(k3) + π;

[0046] When φ′(k3) = φ′(k3) + 2π;

[0047] When neither of the above two formulas is satisfied: it is judged whether φ′(k3) < 0 holds,

[0048] If it holds, then φ′(k3) = φ′(k3) + 2π; otherwise, φ′(k3) = φ′(k3).

[0049] Through the above operations, the average carrier phase error φ′(k3) of each point can be obtained. Using the obtained phase error data to perform phase compensation on the I and Q static baseband modulation signals static_I(k1) and static_Q(k1) can eliminate the phase error and demodulate the subcarrier modulation data. Since the sampling rate of the phase error φ′(k3) is different from that of static_I(k1) and static_Q(k1), when performing phase compensation on static_I(k1) and static_Q(k1), the phase error φ′(k3) needs to be interpolated to obtain the phase error φ′(k1) so that the sampling rate of φ′(k1) is the same as that of static_I(k1) and static_Q(k1).

[0050]

[0051] Among them, J2n+1 is the Bessel function coefficient, and n is the variable; n o (k1T s1 ) is the output noise;

[0052] 5) The subcarrier demodulation unit performs down-conversion low-pass filtering on the demodulated subcarrier signal to demodulate the baseband signal. According to the Bessel function table, when m f = 1, J1 = 0.44, J3 = 0.02. The subcarrier is tripled, and it is attenuated by 26 dB compared to the subcarrier. The higher-order frequency amplitudes above the fifth harmonic of the subcarrier are even lower and can be ignored. The demodulation result is:

[0053] 2A * 0.44m(k1T s1 )sin(Ωk1T s1 ) + 2A * 0.02m(k1T s1 )sin(3Ωk1T s1 ) + n o (k1T s1 ),

[0054] Multiply by the local quadrature subcarrier for down-conversion and then perform low-pass filtering to filter out the high harmonics and demodulate the baseband signal:

[0055] [2A * 0.44m(k1T s1 )sin(Ωk1T s1 ) + 2A

[0056] * 0.02m(k1T s1 )sin(3Ωk1T s1 )

[0057] + n o (k1T s1 )]sin(Ωk1T s1 )

[0058] = 2A * 0.44m(k1T s1 )sin(Ωk1T s1 )sin(Ωk1T s1 ) + 2A

[0059] * 0.02m(k1T s1 )sin(3Ωk1T s1 )sin(Ωk1T s1 )

[0060] + n o (k1T s1 )sin(Ωk1T s1 )

[0061] = 2A * 0.44m(k1T s1) cos(0) - 2A * 0.44m(k1T s1 ) cos(2Ωk1T s1 )

[0062] + 2A * 0.02m(k1T s1 ) cos(2Ωk1T s1 ) - 2A * 0.02m(k1T s1 ) cos(4Ωk1T s1 ) + n o (k1T s1 ) sin(Ωk1T s1 ) ; and:

[0063] [2A * 0.44m(k1T s1 ) sin(Ωk1T s1 ) + 2A

[0064] * 0.02m(k1T s1 ) sin(3Ωk1T s1 )

[0065] + n o (k1T s1 )] cos(Ωk1T s1 )

[0066] = 2A * 0.44m(k1T s1 ) sin(Ωk1T s1 ) cos(Ωk1T s1 ) + 2A

[0067] * 0.02m(k1T s1 ) sin(3Ωk1T s1 ) cos(Ωk1T s1 )

[0068] + n o (k1T s1 ) cos(Ωk1T s1 )

[0069] = 2A * 0.44m(k1T s1 ) sin(0) + 2A * 0.44m(k1T s1 ) sin(2Ωk1T s1 )

[0070] + 2A * 0.02m(k1T s1 ) sin(2Ωk1T s1 ) + 2A * 0.02m(k1T s1 ) sin(4Ωk1T s1 ) + n o(k1T s1 ) cos(Ωk1T s1 ).

[0071] After low-pass filtering, the double subcarrier frequency and quadruple subcarrier frequency are both filtered out, and the baseband data is demodulated.

[0072] Those skilled in the art will realize that the described embodiments are to assist the reader in understanding the principles of the present invention and should be understood that the scope of protection of the present invention is not limited to the described embodiments. For those skilled in the art, various changes and modifications can be made to the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of the claims of the present invention.

Claims

1. An open-loop carrier synchronization and demodulation method for a majority Doppler PM signal, characterized in that, It includes the following steps: Step 1: Perform band-pass sampling on the received intermediate-frequency multi-Doppler PM signal to obtain an intermediate-frequency PM sampled signal. Multiply the local quadrature carrier with the intermediate-frequency PM sampled signal for digital down-conversion, and then obtain two baseband signals I(k) and Q(k) after integration cleaning and low-pass filtering. These signals contain the modulated subcarrier signal and the phase difference caused by the Doppler frequency and the initial phase. Step 2: Perform Doppler rate-of-change compensation on the two baseband signals I and Q respectively with N groups of orthogonal swept-frequency signals generated locally in a cycle at a certain rate of change to obtain N groups of orthogonal Doppler rate-of-change compensation signals. Perform FFT analysis on the N groups of Doppler rate-of-change compensation signals respectively. By selecting the maximum value from the results of each group of FFT analysis, and then selecting the maximum value among the N maximum values, the corresponding Doppler frequency and rate-of-change of frequency values are obtained. The frequency change amount of the maximum Doppler rate-of-change of the signal after Doppler rate-of-change compensation within the single FFT analysis time is less than the FFT frequency resolution, so that correct analysis and capture can be performed. Perform Doppler frequency compensation on the Doppler rate-of-change compensation signals according to the FFT analysis results to obtain two static baseband modulation signals I and Q. Step 3: Perform integration cleaning and narrow-band low-pass filtering on the two static baseband modulation signals I and Q again to filter out the modulated subcarrier frequency and obtain orthogonal zero-intermediate-frequency signals, which contain the phase difference caused by the FFT analysis frequency difference and the initial phase. Calculate the phase angle through the two zero-intermediate-frequency signals I and Q, then average the sine and cosine of the phase angle, and calculate the phase error through the arctangent of the average value. Step 4: Perform linear correction on the calculated phase error value to make the phase error satisfy the actual phase value range, and obtain the average carrier phase error at each point. Use the obtained average carrier phase error to perform phase compensation on the two baseband signals containing the modulated subcarrier signal and the phase difference caused by the FFT analysis frequency difference and the initial phase to eliminate the phase error and demodulate the subcarrier modulation signal. When performing phase compensation on the two baseband signals containing the phase error, the phase error needs to be interpolated so that the sampling rate of the phase error is the same as that of the two baseband signals. Step 5: Take the 1-fold frequency and 3-fold frequency of the subcarrier of the demodulated subcarrier modulation signal, then multiply with the local orthogonal subcarrier and perform down-conversion and low-pass filtering, and filter out the high-fold frequency to demodulate the baseband data; complete the open-loop carrier synchronization and demodulation of the multi-Doppler PM signal.

Images