CN107579942B

CN107579942B - Demodulation method of APSK (amplitude phase Shift keying) modulation signal

Info

Publication number: CN107579942B
Application number: CN201711066336.8A
Authority: CN
Inventors: 徐晓书; 江涛; 陈绍山
Original assignee: First Research Institute Of Telecommunications Technology Co ltd
Current assignee: First Research Institute Of Telecommunications Technology Co ltd
Priority date: 2017-11-02
Filing date: 2017-11-02
Publication date: 2020-04-10
Anticipated expiration: 2037-11-02
Also published as: CN107579942A

Abstract

The invention provides a demodulation method of an APSK modulation signal. The method comprises the steps of obtaining a signal carrier phase error value and an amplitude value at the time of a header and a tail by calculating a correlation function value between the header and the tail unique code of an APSK modulation signal frame and a local unique code of a receiver; estimating the carrier phase error instantaneous value of each baud of the APSK modulation signal frame by adopting a linear interpolation method, and correcting the phase of each baud of the APSK modulation signal frame; calculating a demodulation amplitude reference value by using a header and a tail correlation function amplitude value; and demodulating data information according to the phase and amplitude of each baud of the corrected APSK modulation signal frame. The method does not need a carrier recovery loop of a traditional coherent demodulator, reduces the carrier recovery time and is suitable for the demodulation of burst signals; the method has good demodulation algorithm performance and can realize data demodulation under the condition of low signal-to-noise ratio; the method can realize the demodulation of the 16APSK and 32APSK modulation signals in a circuit form, and has the advantages of simple algorithm, small calculation amount and easy realization.

Description

Demodulation method of APSK (amplitude phase Shift keying) modulation signal

Technical Field

The invention relates to the field of communication, in particular to a demodulation method of an APSK modulation signal.

Background

APSK modulated signals are a form of signal transmission for satellite communications. Compared with BPSK and QPSK modulation signals, the APSK modulation signals can transmit information with higher rate in the same bandwidth, and the frequency utilization rate can be effectively improved; compared with QAM modulation signals, APSK modulation signals have smaller peak-to-average ratio, and the utilization rate of a satellite transponder power amplifier can be effectively improved. Common APSK modulated signals include 16APSK modulated signals and 32APSK modulated signals, and equation (1) is a complex signal expression thereof:

in equation (1), h (t) is the impulse response of the shaping filter, which is usually a square root raised cosine function; a is_kIs a modulation symbol, which is defined as follows:

in equation (2), K is the length of the data frame. Table (1) is a table of phase transformation rules of the 16APSK signal, where r1 is 0.4182 and r2 is 1.1292 in table (1); table (2) shows the phase transition law of the 32APSK signal, where r1 is 0.2637, r2 is 0.7120, and r3 is 1.2658 in table (2). Fig. 1 is a phase constellation diagram of a 16APSK signal, and fig. 2 is a phase constellation diagram of a 32APSK signal. QPSK modulation can be considered as a special case of APSK modulation; table (3) is a table of the phase transformation rule of the QPSK signal, where r1 is 1 in table (3); fig. 3 is a phase constellation diagram of a QPSK signal. Equation (3) is a real signal expression of the APSK modulated signal.

In formula (3), a (t) is an amplitude function of the complex signal s (t), where a (t) r × h (t), and h (t) is a transfer function of the baseband filter, and a frequency response function thereof is a square root raised cosine function; omega_i＝2πf_iIs the signal carrier angular frequency;

is the instantaneous phase of the signal; r and

from modulation symbols a_kAnd determining to change according to the APSK modulation mode. Table (4) is a frame format table of the APSK modulated signal, and in table 4, HSN is the number of half symbols, and 1 symbol includes 2 HSNs. For convenience of description and without loss of generality, the header unique code and the trailer unique code are assumed to be 14 symbols, and the data transmission portion is assumed to be 335 symbols. The header unique code and the tail unique code adopt QPSK modulation mode, and the data transmission part can select 16APSK modulation mode or 32APSK modulation mode according to the transmission condition. Tables (5) and (6) are definition tables of header unique codes and tail unique codes, and the unique codes can adopt other code lengths and code patterns.

TABLE 1

TABLE 2

TABLE 3

TABLE 4

(HSN25…HSN52)
	00-01-00-10-00-01-00-01-11-10-11-10-11-01

TABLE 5

(HSN723…HSN750)
	01-11-10-11-10-11-01-00-01-00-10-00-01-00

TABLE 6

Disclosure of Invention

In order to solve the demodulation problem of the APSK modulation signal, the invention provides a method for realizing APSK modulation signal demodulation by using a signal frame unique code to carry out parameter estimation, which comprises the following steps:

processing the APSK modulation signal to obtain an orthogonal baseband signal;

the orthogonal baseband signal and the local unique code are subjected to correlation operation to obtain correlation function values corresponding to the header and tail unique codes;

calculating a signal carrier phase error value and a signal amplitude value at the time of the header and the tail according to the correlation function values of the header and tail unique codes;

calculating the carrier instantaneous phase estimation value of each baud of the APSK signal by adopting a linear interpolation method according to the signal carrier phase error values at the header and tail reporting moments;

correcting the phase of each baud of the APSK signal according to the carrier instantaneous phase estimation value of each baud of the APSK signal;

calculating a demodulation amplitude reference value according to the signal amplitude values at the header and tail reporting times; and

and according to the demodulation amplitude reference value and the phase and amplitude value of the corrected APSK signal, demodulating the APSK signal.

Optionally, in the method for demodulating an APSK modulated signal, the step of processing the APSK modulated signal includes: and performing down-conversion, sampling, accumulation and baseband filtering processing on the APSK modulation signal to obtain an orthogonal baseband signal.

Optionally, in the above method for demodulating an APSK modulated signal, the APSK modulated signal is represented by the following formula:

where a (t) is the amplitude function of signal s (t), where a (t) is r × h (t), and h (t) is the baseband filter transfer function whose frequency response function is a square root raised cosine function; omega_i＝2πf_iIs the signal carrier angular frequency;

is the instantaneous phase of the signal; r and

according to the APSK modulation mode.

Optionally, in the above method for demodulating an APSK modulated signal, the down-conversion is represented by the following formula:

in the formula, ω₀For the demodulator to receive the local oscillator angular frequency, its nominal value and omega_iThe same; Δ ω ═ ω_i-ω₀Is the carrier error, Δ ω t is the carrier phase error; s_I(t) and S_QThe former term of the expression (t) is the desired down-conversion component.

Optionally, in the demodulation method of the APSK modulated signal, the sampling frequency of the down-converted signal is 32 times of the baud rate of the signal, and the sampling frequency of 1 baud is 32 sampling points; after 4 samples are accumulated, the number of the samples is reduced to 1 baud and 8 sampling points.

Optionally, in the method for demodulating an APSK modulated signal, in the baseband filtering processing step, the frequency response function of the baseband filter includes a square root raised cosine function, and the quadrature baseband signal generated by the square root raised cosine function is represented by the following formula:

optionally, in the above method for demodulating APSK modulated signals, the orthogonal baseband signal and 2 local unique codes perform correlation operation, and calculate correlation function values of the header and tail unique codes; and then, calculating the signal carrier phase error value and the signal amplitude value of the header and the tail time according to the correlation function value.

Optionally, in the above method for demodulating APSK modulated signal, the first correlator calculates a correlation between the signal header and the first locally unique codeValue of correlation function

The first locally unique code is in a conjugate relationship with the signal header unique code; the second correlator calculates a correlation function value between the signal tail and the second local unique code

The second local unique code is in a conjugate relationship with the signal tail unique code.

Optionally, in the method for demodulating APSK modulated signal, the signal carrier phase error value θ at the time of signal preamble₁Amplitude value ρ₁Are respectively a correlation function value

Phase value and amplitude value of (a), signal carrier phase error value theta at signal tail-in time₂Amplitude value ρ₂Are respectively a correlation function value

Phase values and amplitude values.

Optionally, in the method for demodulating APSK modulated signal, the phase error value θ of the preamble time signal is used₁And tail-reporting time signal phase error value theta₂And carrying out interpolation operation to estimate the carrier instantaneous phase error estimated value delta theta (n) of each baud of the signal.

Optionally, in the method for demodulating the APSK modulated signal, the step of correcting the phase of each baud of the APSK signal is implemented by the following formula:

I″+jQ″＝(I'+jQ')×(A-jB)，

wherein, I 'and Q' are orthogonal baseband vectors after phase correction, I 'and Q' are orthogonal baseband vectors before phase correction, and A, B is a phase correction vector;

the signal phase correction vector is obtained by adopting the following formula:

in the formula, θ (n) is an instantaneous phase estimation value of each baud of the signal.

Optionally, in the above method for demodulating APSK modulated signals, the reference value of demodulation amplitude is calculated by the following formula:

in the formula, ρ₁Is the signal amplitude value at the time of the preamble, p₂Is the signal amplitude value at the time of tail-biting.

The demodulation method of the APSK modulation signal provided by the invention has the following beneficial effects: the method provided by the invention does not need a carrier recovery loop of the traditional coherent demodulator, reduces the carrier recovery time and is suitable for the demodulation of burst signals; the method provided by the invention has good performance and can realize data demodulation under the condition of low signal-to-noise ratio; the invention can realize the demodulation of the 16APSK and 32APSK modulation signals in a circuit form, and has the advantages of simple algorithm, small calculation amount and easy realization.

Drawings

FIG. 1 is a phase constellation diagram for a 16APSK signal;

FIG. 2 is a phase constellation diagram of a 32APSK signal;

fig. 3 is a phase constellation diagram of a QPSK signal;

FIG. 4 is a schematic block diagram of a preferred embodiment of the present invention;

FIG. 5 is a flow chart of a preferred embodiment of the present invention;

FIG. 6 is a functional block diagram of the phase estimation module of FIG. 4;

FIG. 7 is a schematic block diagram of the first correlator of FIG. 6;

FIG. 8 is a functional block diagram of the second correlator of FIG. 6;

FIG. 9 is a functional block diagram of the instantaneous phase estimation module of FIG. 6;

FIG. 10 is a schematic diagram of the instantaneous phase data relationship of FIG. 6;

FIG. 11 is a functional block diagram of the phase error correction module of FIG. 4;

fig. 12 is a schematic block diagram of the data demodulation module of fig. 4.

Detailed Description

The following describes in more detail embodiments of the present invention with reference to the schematic drawings. The advantages and features of the present invention will become more apparent from the following description. It is to be noted that the drawings are in a very simplified form and are not to precise scale, which is merely for the purpose of facilitating and distinctly claiming the embodiments of the present invention.

The invention provides a demodulation method of an APSK modulation signal, a functional block diagram for implementing the invention refers to FIG. 4, an algorithm flow for implementing the invention refers to FIG. 5, and the specific steps are as follows:

the method comprises the following steps: the input signal is down converted.

Specifically, the input signal of the demodulator is s (t), and the mathematical expression thereof is as shown in formula (3). s (t) and quadrature local oscillator signal sin ω_ot and cos omega_ot multiplication generates orthogonal signal s shown in formula (4)_I(t)、s_Q(t) of (d). In the formula (4), the local oscillator signal angular frequency ω_oWith the input carrier signal angular frequency omega_iHas the same nominal value, and has small error delta omega in practical situation, and the delta omega is omega_i-ω_o(ii) a The former term is the baseband component required for demodulation, and the latter high frequency component is filtered out by a 4-sample accumulator and a baseband matched filter.

Step two: and (6) sampling the signal.

In particular, the quadrature mixing signal s_I(t)、s_Q(t) is given a frequency f_sIs sampled to form a digital signal s'_I(m)、s'_Q(m) of the reaction mixture. Preferably, f_sFor the baud rate R of the input signal_iIs 32 times, namely 1 signal wave has 32 sampling points, the invention has the frequency f of the sampling signal_sFor the baud rate R of the input signal_iThe multiple of (c) should be greater than 8.

Step three: producing quadrature baseband signals.

Specifically, the digital quadrature mixing signal s'_I(m)、s'_Q(m) are processed by 4 sampling accumulators and baseband matched filters to form orthogonal baseband signals I (n), Q (n) (see formula 5), and the sampling rate is reduced to 1 baud and 8 sampling points are included.

Preferably, the transmission function of the baseband matched filter is a square root raised cosine function corresponding to the APSK modulated signal; meanwhile, the number of the sampling accumulators is selected to ensure that 1 signal baud contains 8 sampling points.

I (n), q (n) form the real and imaginary parts of the complex baseband signal s (n) of the APSK modulated signal, and equation (6) is a mathematical expression thereof, in which the variation of the signal amplitude is caused by a/D conversion, accumulation, filtering, etc., but the signal properties are not changed.

Due to the frequency error Δ ω, there is a phase error Δ θ (n) ═ Δ ω n Δ t in the complex baseband signal s (n), and in order to realize APSK signal demodulation, it is necessary to accurately estimate and correct Δ θ (n). Fig. 6 is a reference block diagram of the phase extraction circuit of the present invention. The input orthogonal baseband signal is subjected to correlation operation with 2 local unique codes in a first correlator and a second correlator, and correlation function values of signal frame header and tail unique codes are calculated; and then, calculating a signal carrier phase error value and a signal amplitude value at the time of the header and the tail according to the correlation function value.

Step four: header unique code correlation value calculation.

Specifically, a schematic block diagram of the first correlator is shown in fig. 7. The quadrature baseband signals I (n), Q (n) form 14 input signal vectors through a shift register

(equation 7), each vector is separated by 7 samples (one)The baseband symbol takes a vector, and one baseband symbol contains 8 samples).

14 vectors of input signals

Homolocal unique code vector

Performing vector correlation operation to output a modulus rho of the correlation value₁And argument theta₁(equations 8, 9, 10), local unique code vector

Is the conjugate of the unique codevector of the APSK modulated signal frame header (see table 7).

m₁Is a vector of the input signal

Mean value of energy (see formula 11), k₁Is ρ₁And m₁Ratio of (see formula 12), k₁Through 8The bit shift register outputs to the header unique code synchronization decision logic. When the header unique code and the local unique code in the input signal

When matching, the modulus rho of the vector output by the first correlator₁Peak value, i.e. k₁A peak occurs. The decision criteria for header unique code synchronization are: the output value of the 5 th tap in the 8-bit shift register is the largest, and the output value is larger than a specified threshold value. After the first local unique code is synchronized, outputting a synchronization pulse of the first local unique code, and under the action of the synchronization pulse:

a) argument theta of output preamble unique code correlation value₁As the phase error value corresponding to the time at which the preamble unique code is located.

b) Controlling a down-sampling circuit to strip off a header unique code part in an input baseband signal and output a down-sampled baseband signal (one baseband signal code element takes one sample); storing a baseband signal between the header unique code and the tail unique code into a data register, and outputting the tail unique code to a second correlator; and after the down-sampling circuit completely outputs a frame of baseband signal, the output is closed.

c) Outputting a modulo ρ of a header unique code correlation value₁For calculating a demodulation level reference signal.

TABLE 7 local unique code pattern of the first correlator

Step five: and calculating the correlation value of the tail-reported unique code.

Specifically, a functional block diagram of the second correlator refers to fig. 8. The down-sampled quadrature baseband signals I '(n), Q' (n) (1 sampling value of 1 code element) form 14 input signal vectors through a shift register

(formula 13).

14 vectors of input signals

Homolocal unique code vector

Performing vector correlation operation to output a modulus rho of the correlation value₂And argument theta₂(equations 14, 15, 16), local unique code vector

Is the conjugate of the unique code vector of the tail of the transmitted signal (see table 8).

m₂Is a vector of the input signal

Mean value of energy (see formula 17), k₂Is ρ₂And m₂Ratio of (see formula 18), k₂Output to newspaper tailA unique code synchronization decision logic circuit. When the tail part unique code and the local unique code in the input signal

When matching, the modulus rho of the output vector of the second correlator₂Peak value, i.e. k₂A peak occurs and is greater than a specified threshold. After the unique code (2) is synchronized, outputting tail-reporting unique code synchronization pulse, under the action of which:

a) amplitude angle theta of output tail-report unique code correlation value₂And the phase error value is used as the phase error value of the time of the tail-end unique code.

b) Modulo rho of output tail-biting unique code correlation value₂For calculating a demodulation level reference signal.

TABLE 8 second correlator local unique code pattern

Step six: and estimating the instantaneous phase error of the APSK signal.

In particular, a functional block diagram of the instantaneous error phase estimation module is shown in fig. 9. Obtaining the carrier phase error value theta of the corresponding header unique code and the tail unique code time by the vector correlation method₁、θ₂Then, linear interpolation is used to estimate theta₁、θ₂And carrier instantaneous phase estimates of 335 APSK signal headers, wherein fig. 10 is an instantaneous error phase data relationship diagram. At theta₁、θ₂Interval, phase interpolation is estimated by equation (19):

in order to ensure the accuracy of phase estimation, limiting | delta theta | to be less than 2 pi; in phase estimation, the phase continuity at the + -pi boundary should be maintained.

Step seven: and correcting the phase error of the APSK signal.

Specifically, referring to fig. 11, a schematic block diagram of the phase error correction module is shown, in which a data memory is used to temporarily store baseband signals I '(n), Q' (n) (1 sampling value of 1 symbol) that need to be phase corrected, and a delay circuit is used to align data with its corresponding carrier instantaneous phase estimation value in time. In the phase correction circuit 1, the quadrature baseband signals I '(n), Q' (n) constitute a rotation vector I '(n) + jQ' (n), the rotation rate being determined by Δ ω; the instantaneous phase error estimate value Δ θ (n) is transformed by a trigonometric function (with an accuracy of 0.1 degrees), and a ═ cos [ Δ θ (n) ] and B ═ sin [ Δ θ (n) ] are output, and a-jB vector formed by A, B is a rotation vector in the opposite direction to I '(n) + jQ' (n); i "(n), Q" (n) are phase corrected quadrature baseband signals. The phase correction adopts a vector multiplication method, and the calculation formula is shown as the formula (20).

If the estimation of delta theta (n) is accurate, A-jB and I '(n) + jQ' (n) rotate in opposite directions and at the same speed, the phase error can be completely corrected, and the constellation diagram of APSK signals formed by I '(n) and Q' (n) is consistent with the transmitted signals (see fig. 1 and 2).

Step eight: a demodulated amplitude reference value is calculated.

Specifically, the signal amplitude reference value P is an average value of the signal amplitudes of the preamble unique code and the tail unique code, i.e. r in the QPSK signal constellation₁(see table 3, fig. 3), and equation (21) is the calculation equation.

According to the definition of the 16APSK signal (Table 1, FIG. 1), the amplitude reference value P and the signal amplitude r₁、r₂Is expressed by equation (22), and is expressed by equation (23):

according to the definition of 32APSK signal (Table 2, FIG. 2), the amplitude reference value P and the signal amplitude r₁、r₂、r₃Is related to the demodulation amplitude decision value w as shown in equation (24)₁、w₂Is represented by the formula (25):

step nine: and (6) data demodulation.

Specifically, a functional block diagram of the data demodulation module is shown in fig. 12. In the figure, I '(n) and Q' (n) are the data baseband signal (APSK signal) after phase correction, and the amplitude r and the amplitude angle of the I '(n) and the Q' (n) are shown

Calculating according to the formula (26) and the formula (27);

then, according to the amplitude r and the argument

And demodulating amplitude decision values w, w₁、w₂And demodulation of the APSK signal is realized. Table (9) is a data relationship table for 16APSK demodulation, in which d₁、d₂、d₃、d₄Is the demodulated data. Table (10) is a data relationship table for 32APSK demodulation. In table d₁、d₂、d₃、d₄、d₄、d₅Is the demodulated data.

Table 916 APSK modulated signal demodulation data relation table

Table 1032 APSK modulated signal demodulation data relation table

The demodulation method of the APSK modulation signal provided by the invention has the following beneficial effects: the method provided by the invention does not need a carrier recovery loop of the traditional coherent demodulator, reduces the carrier recovery time and is suitable for demodulating burst signals. The method provided by the invention has good performance and can realize data demodulation under the condition of low signal-to-noise ratio. The invention can realize the demodulation of the 16APSK and 32APSK modulation signals in a circuit form, and has the advantages of simple algorithm, small calculation amount and easy realization.

The above description is only a preferred embodiment of the present invention, and does not limit the present invention in any way. It will be understood by those skilled in the art that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A method for demodulating APSK modulated signals, comprising the steps of:

processing the APSK modulation signal to obtain an orthogonal baseband signal;

calculating a demodulation amplitude reference value according to the signal amplitude values at the header and tail reporting times;

and according to the demodulation amplitude reference value, the corrected APSK signal phase and amplitude value, realizing the demodulation of the APSK signal.

2. The method for demodulating an APSK-modulated signal according to claim 1, wherein the step of signal processing the APSK-modulated signal comprises: and performing down-conversion, sampling, accumulation and baseband filtering processing on the APSK modulation signal to obtain an orthogonal baseband signal.

3. The method of demodulating an APSK-modulated signal according to claim 2, wherein the APSK-modulated signal is expressed by the following formula:

is the instantaneous phase of the signal; r and

according to the APSK modulation mode.

4. A method of demodulating an APSK-modulated signal as claimed in claim 3, wherein the down-conversion is represented by the formula:

5. The method for demodulating an APSK-modulated signal according to claim 3, wherein the down-converted signal has a sampling frequency 32 times the baud rate of the signal, and 1 baud comprises 32 sampling points; after 4 samples are accumulated, the number of samples is reduced to 1 baud and 8 sampling points are included.

6. The method for demodulating an APSK-modulated signal according to claim 3, wherein in the baseband filtering processing step, the frequency response function of the baseband filter is a square root raised cosine function, and the quadrature baseband signal generated thereby is expressed by the following formula:

7. the method for demodulating APSK-modulated signals according to claim 1, wherein the orthogonal baseband signals are correlated with 2 local unique codes, and correlation function values of the header and tail unique codes are calculated; and then, calculating the signal carrier phase error value and the signal amplitude value of the header and the tail time according to the correlation function value.

8. The method for demodulating an APSK-modulated signal according to claim 7, wherein the first correlator calculates a correlation function value between the signal header and the first locally unique code

9. The method for demodulating an APSK-modulated signal according to claim 8, wherein the signal carrier phase error value θ at the time of the signal preamble₁Amplitude value ρ₁Are respectively a correlation function value

Phase values and amplitude values.

10. The method for demodulating an APSK-modulated signal according to claim 9, wherein the phase error value θ is determined by the preamble timing signal phase error value θ₁And tail-reporting time signal phase error value theta₂And (4) carrying out interpolation operation to estimate the instantaneous phase error estimated value delta theta (n) of each baud carrier of the signals.

11. The method for demodulating an APSK-modulated signal according to claim 10, wherein the step of correcting the phase of each baud of the APSK signal is implemented by the following equation:

I”+jQ”＝(I'+jQ')×(A-jB)，

wherein, I "and Q" are quadrature baseband vectors after phase correction, I 'and Q' are quadrature baseband vectors before phase correction, and A, B is a phase correction vector; the signal phase correction vector is obtained by adopting the following formula:

in the formula, θ (n) is an instantaneous phase error estimation value of each baud of the signal.

12. The method of demodulating an APSK-modulated signal according to claim 1, wherein the demodulation amplitude reference value is calculated by the following formula: