CN102209058A - Low density parity check (LDPC)-coding-assisted mary phase-shift keying (MPSK) system carrier synchronization method - Google Patents

Abstract

The invention discloses a low density parity check (LDPC)-coding-assisted mary phase-shift keying (MPSK) system carrier synchronization method, which belongs to the technical field of modulation and demodulation in digital communication. The method mainly comprises initial carrier synchronization under known frame header condition and carrier tracking synchronization in a modulation process. In an initial carrier synchronization process, initial frequency offset and initial phase offset are estimated, and maximum likelihood estimation values of the frequency offset and the phase offset can be obtained by utilizing a bisection-method-based rapid searching algorithm. In a carrier tracking process, carrier synchronization is determined whether to be accurately tracked or not by utilizing an LDPC coding syndrome. By the method, the initial estimation performance can approach Cramer-Rao lower bound, a receiver can work stably in a low signal to noise ratio environment, LDPCs also exhibit high performance, and carrier synchronization tracking can be ensured to avoid cycle slip and be stably operated if the LDPC decoding can be operated normally and useful information can be provided at a required signal to noise ratio.

Description

A kind of based on the auxiliary MPSK system carrier method for synchronous of LDPC coding

Technical field

The present invention relates to the carrier synchronization method of auxiliary M level phase shift keying (MPSK) system of a kind of employing low density parity check code (LDPC) coding, belong to the modulation-demodulation technique field in the digital communication.

Background technology

Low density parity check code (LDPC) is a kind of coding near shannon limit, and many application are arranged in real life: second generation digital video broadcast-satellite (BVD-S2) system adopts BCH and LDPC concatenated code as its forward error correction; Consultative committee for space data system (CCSDS) has also proposed a series of LDPC codings that near-earth is used and deep space is used that make things convenient for.But the LDPC modulation technique needs operation of receiver under the low signal-to-noise ratio environment, and will bring the threshold effect of carrier frequency recovery this moment.The tracking of estimation that two key issues of carrier frequency recovery are initial frequency deviation, skew and demodulating process medium frequency, phase place.

Initial frequency deviation estimation and skew estimate to be converted into the estimation to a multiple sinusoidal signal frequency and initial phase.Under the environment of low signal-to-noise ratio, for the limited data block of number, this process adopts the MLE method of progressive optimum can access best performance, but because likelihood function is the nonlinear function (this specific character is more outstanding under noisy situation) of frequency, so MLE just need realize effectively search in possible frequency range accurately.

The conventional method that solves demodulating process medium frequency, Phase Tracking is to adopt phase-locked loop (PLL), is modulated to example with BPSK, and the schematic diagram of phase-locked loop as shown in Figure 1.But when signal to noise ratio was lower than 6dB, the cycle-skipping phenomenon of phase-locked loop in tracing process (phase ambiguity appears again in the carrier wave of promptly removing after the phase ambiguity) was just apparent in view, work that can not be stable.Synchronous in order better to realize frequency, Phase Tracking, some systems insert frequency pilot sign in information data between the cycle such as DVB-S2, but frequency pilot sign can bring extra expense, thereby have reduced the availability of frequency spectrum; The other article has proposed to carry out the method that iteration obtains frequency-tracking between the demodulation sign indicating number, but iteration has been brought the increase greatly of computation complexity.The above method all is not suitable for practical engineering application.

At present, more existing method for synchronous based on frequency offset estimating and compensation, it realizes principle as shown in Figure 2 substantially.The intermediate-freuqncy signal that receives at first changes to base band through the local oscillator quadrature frequency conversion, passes through the AD sample varianceization again.Supposing has desirable bit timing, then the baseband signal r after down-conversion _kCan be expressed as

r _k＝exp(jθ _k)exp(j2πf _dkT _s+jφ ₀)+z _k (1)

In the formula (1), θ _kBe the phase place of constellation point symbol of transmission, i.e. modulation intelligence, for MPSK, its value is θ _k∈ 0,2 π/M, and Λ, 2 (M-1) π/M}, M is an order of modulation; f _dIt is residual frequency departure; T _sThe is-symbol cycle; φ ₀It is initial skew; z _kBe noise, be without loss of generality that suppose that it is additivity white complex gaussian noise (AWGN), its average is 0, variance is σ ²J is an imaginary unit; K represents the label of k sampled point.The power normalization of signal is 1.

Remove r _kIn modulation intelligence, obtain a single-carrier signal of only being polluted by additive noise

$h_k=r_k\exp (-j{\mathrm{\&theta;}}_k)=\exp (\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{f}_{d}k{T}_s+j{\mathrm{\&phi;}}_0)+{\hat{z}}_k,$ k＝0，1，ΛL ₀-1(2)

Employing can obtain h based on the searching algorithm of dichotomy _kFrequency f _dWith first phase φ ₀Maximum likelihood estimator

With

And then to baseband signal r _kCompensate, recover original signal and obtain demodulation result d _k:

$d_k=r_k\exp (-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{\hat{f}}_{d}k{T}_s-j{\widehat{\mathrm{\&phi;}}}_0)---\left(3\right)$

Summary of the invention

The objective of the invention is deficiency, proposed a kind of based on the auxiliary carrier synchronization method of LDPC coding at existing carrier synchronization implementation method in the phase modulation system that adopts the LDPC coding.

Communication system comprises two kinds of transmission modes: burst communication and continuous communiction.The data of burst communication are made up of frame head and data, and frame head is used for the beginning of mark burst frame with synchronously, and its content is known for receiving terminal, and the burst communication frame data format as shown in Figure 3.Each frame all is made up of a frame head and n data block, and each data block is the grouping of LDPC sign indicating number.Frame head is expressed as H, comprises L ₀Individual symbol; Data block is expressed as B _i, i=1,2, K, n, each data block comprises L _dIndividual symbol, and

The length of each data block of N, unit is a bit, M is the exponent number of MPSK modulation.

The data of continuous communiction then are to send continuously, middle duties ground is sent out some and is used to remove the unique word of phase ambiguity, the data content of unique word also is known for receiving terminal, but general unique word is very short, can't be used for carrier synchronization under the low signal-to-noise ratio condition, the continuous communiction frame data format as shown in Figure 4.Have only a frame under the continuous communiction pattern, be made up of frame head and numerous data block, the definition of frame head and each data block is with the definition under the logical burst mode.

Two steps that the inventive method comprises:

Step 1, initial frequency deviation and skew are estimated

According to the baseband signal r of following formula removal after down-conversion _kIn modulation intelligence, obtain single-carrier signal h _k:

$h_k=r_k\exp (-j{\mathrm{\&theta;}}_k)=\exp (\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{f}_{d}k{T}_s+j{\mathrm{\&phi;}}_0)+{\hat{z}}_k,$

Wherein, L ₀Number for symbol in the frame head; θ _kBe the phase place of constellation point symbol of transmission, its value is θ _k∈ 0,2 π/M, and 4 π/M, Λ, 2 (M-1) π/M}, M is the MPSK order of modulation; f _dIt is residual frequency departure; T _sThe is-symbol cycle; φ ₀It is initial skew; J is an imaginary unit; K represents the label of k sampled point of this single-carrier signal, and k is a positive integer;

Be the additivity white complex gaussian noise, average is 0, and variance is σ ²z _kIt is the amplitude of additivity white complex gaussian noise; For the burst communication pattern, each frame all is made up of a frame head and n data block, and each data block comprises L _dIndividual coded identification, and

The length of each data block of N.

Obtain single-carrier signal h then _kInitial frequency deviation f _dWith initial skew φ ₀Maximum likelihood estimator

With

And the nothing of the skew of frame head intermediate symbols is estimated partially

As preferably, utilize prior art, can adopt that (please refer to the patent No.: searching algorithm 200910087839.2) obtains single-carrier signal h based on dichotomy _kInitial frequency deviation f _dWith initial skew φ ₀Maximum likelihood estimator

With

Step 2, the auxiliary frequency deviation of decoded result of utilizing LDPC and the tracking of skew, for first data block, adopt directly judgement (Decision-Directed) detector, utilize the Direct Phase backoff algorithm that phase place is followed the tracks of, and after with first data block demodulating and decoding, utilize decoded information to obtain skew estimation more accurately again, and then remaining data is carried out demodulation.

The contrast prior art, beneficial effect of the present invention is: the inventive method is utilized LDPC decoded result subcarrier track algorithm in frequency deviation and skew tracing process, not only do not need extra aiming symbol, and also can steady operation under the low signal-to-noise ratio environment.

Description of drawings

Fig. 1 is the schematic diagram of phase-locked loop;

Fig. 2 is the basic realization schematic diagram based on frequency offset estimating and compensation carrier synchronization of prior art;

Fig. 3 is a burst communication data format schematic diagram;

Fig. 4 is a continuous communiction data format schematic diagram;

Fig. 5 is the discrete probability distribution of skew evaluated error

Fig. 6 is the probability distribution of syndrome and value.

Embodiment

The objective of the invention is deficiency, proposed a kind of based on the auxiliary carrier synchronization method of LDPC coding at existing carrier synchronization implementation method in the phase modulation system that adopts the LDPC coding.Below in conjunction with drawings and Examples the present invention is made an explanation.

Two steps that the inventive method comprises:

Step 1, initial frequency deviation and skew are estimated

For receiving terminal, the frame head symbol is known, utilizes the frame head symbol to carry out initial estimation in conjunction with the MLE algorithm.At first remove the baseband signal r after down-conversion _kIn modulation intelligence, only obtain single-carrier signal h by additive noise polluted _k, can be expressed from the next:

$h_k=r_k\exp (-j{\mathrm{\&theta;}}_k)=\exp (\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{f}_{d}k{T}_s+j{\mathrm{\&phi;}}_0)+{\hat{z}}_k,$

(4)

Wherein, L ₀Number for symbol in the frame head; θ _kBe the phase place of constellation point symbol of transmission, its value is θ _k∈ 0,2 π/M, and 4 π/M, Λ, 2 (M-1) π/M}, M is the MPSK order of modulation; f _dIt is residual frequency departure; T _sThe is-symbol cycle; φ ₀It is initial skew; J is an imaginary unit; K represents the label of k sampled point, and k is a positive integer; Be the additivity white complex gaussian noise, average is 0, and variance is σ ²z _kIt is the amplitude of additivity white complex gaussian noise.For the burst communication pattern, each frame is formed by a frame head and n data block, and each data block comprises L _dIndividual coded identification, and N is the length of each data block.

The likelihood function of formula (4) data model is

$L({\mathrm{\&phi;}}_0,f_d)\&Proportional;p(h;{\mathrm{\&phi;}}_0,f_d)$

$=\frac{1}{{\mathrm{\&pi;}}^{{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0}{\mathrm{\&sigma;}}^{2{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0}}\exp [-\frac{1}{{\mathrm{\&sigma;}}^2}{(h-x)}^H(h-x)]---\left(5\right)$

Wherein,

Be single-carrier signal h _kVector representation,

Be multiple sinusoidal signal x _k=exp (j2 π f _dKT _s+ j φ ₀) vector representation; P (h; φ ₀, f _d) be the probability density function of vectorial h, parameter wherein is φ ₀And f _d, subscript H represents conjugate transpose.φ ₀And f _dMaximum likelihood estimator With Should make likelihood function L (φ ₀, f _d) the value maximum, i.e. (h-x) ^H(h-x) value minimum can get φ in view of the above ₀And f _dThe maximal possibility estimation expression formula be

${\hat{f}}_d=\underset{f}{\arg \max }\left\{\right|\underset{k=0}{\overset{L_0-1}{\mathrm{\&Sigma;}}}{h}_k\exp (-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;fk}{T}_s)\left|\right\}---\left(6\right)$

${\widehat{\mathrm{\&phi;}}}_0=\arg \left\{\underset{k=0}{\overset{L_0-1}{\mathrm{\&Sigma;}}}{h}_k\exp (-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{\hat{f}}_{d}k{T}_s)\right\}---\left(7\right)$

By formula (6) as can be seen, the maximum likelihood estimator of frequency deviation

Analytic solutions can not directly not calculate, and can as preferably, (please refer to the patent No.: 200910087839.2), obtain single-carrier signal h such as dichotomy based on some algorithm search algorithms _kInitial frequency deviation f _dWith initial skew φ ₀Maximum likelihood estimator

With

Because skew estimated value

Be based on the frequency offset estimating value

, so its estimated performance will be subjected to the influence of frequency offset estimating error.When not considering The noise (single carrier in the step 1, promptly the formula in this document (4) is only by the single-carrier signal that additive noise polluted, that says here does not consider The noise, is when handling in the removal formula (4)

Influence), formula (7) can be rewritten as

${\widehat{\mathrm{\&phi;}}}_0=\arg \left\{\underset{k=0}{\overset{L_0-1}{\mathrm{\&Sigma;}}}\exp (\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{f}_{d}k{T}_s+j{\mathrm{\&phi;}}_0)\exp (-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{\hat{f}}_{d}k{T}_s)\right\}$

$=\arg \left\{\underset{k=0}{\overset{L_0-1}{\mathrm{\&Sigma;}}}\exp \left(j{\mathrm{\&phi;}}_0\right)\exp \right[\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}(f_d-{\hat{f}}_d)k{T}_s\left]\right\}$ (8)

$=\arg \left\{\exp \right[j{\mathrm{\&phi;}}_0+\mathrm{j\&pi;}(f_d-{\hat{f}}_d)(L_0-1)T_s]$

$\&times;\underset{k=0}{\overset{L_0-1}{\mathrm{\&Sigma;}}}\exp \left[\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}(f_d-{\hat{f}}_d)(k-(L_0-1)/2)T_s\right]\}$

Obviously, and likes

Be an arithmetic number, so formula (8) can be abbreviated as

${\widehat{\mathrm{\&phi;}}}_0={\mathrm{\&phi;}}_0+\mathrm{\&pi;}(f_d-{\hat{f}}_d)(L_0-1)T_s---\left(9\right)$

When

Depart from f _dWhen too many, the phase estimation error will become greatly, and

Not not have partially to estimate.Convolution (9), the skew of frame head intermediate symbols

Nothing be estimated as partially

${\widehat{\mathrm{\&phi;}}}_{(L_0-1)/2}={\widehat{\mathrm{\&phi;}}}_0+\mathrm{\&pi;}{\hat{f}}_d(L_0-1)T_s={\mathrm{\&phi;}}_0+\mathrm{\&pi;}{f}_d(L_0-1)T_s---\left(10\right)$

The tracking of step 2, frequency deviation and skew utilizes the auxiliary frequency deviation of decoded result of LDPC and the tracking of skew, and detailed process comprises the steps:

The track algorithm of A, first data block

Because first data block is the decoded result that frame head does not have the LDPC sign indicating number, so adopt directly judgement (Decision-Directed) detector and Direct Phase backoff algorithm tracking skew, concrete grammar is as follows: the initial frequency deviation of 1) utilizing step 1 to obtain is estimated

Estimate with skew

And the nothing that convolution (10) obtains the skew of frame head intermediate symbols is estimated partially

Extra accumulation of phase and θ are set _Sum=0; Initialization k=(L ₀-1)/2.For at (L ₀-1)/2≤k≤L ₀+ L _dEach k value in-1 scope, order carry out step 2), L ₀Be the number of symbol in the frame head, L _dNumber for symbol in each remaining data piece.

2) skew of k symbol of compensation obtains its soft demodulation result

$d_k=r_k\exp (-j{\widehat{\mathrm{\&phi;}}}_k)$ (11)

$=r_k\exp \{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{\hat{f}}_d[k-(L_0-1)/2]T_s-j{\widehat{\mathrm{\&phi;}}}_{(L_0-1)/2}\},$ (L ₀-1)/2≤k≤L ₀+L _d-1

R wherein _kBe k the baseband form after the symbol down-conversion, The skew that is k symbol is estimated L ₀Be the number of symbol in the frame head, L _dNumber for symbol in each remaining data.

The phase error that then defines k symbol is

$\left\{\begin{array}{cc}{e}_k=\arg \left(d_k\right)-{\mathrm{\&theta;}}_k,& (L_0-1)/2\&le;k\&le;L_0-1\\ e_k=\arg \left(d_k\right)-\underset{\mathrm{\&theta;}:\mathrm{\&theta;}\&Element;\mathrm{\&Theta;}}{\arg \min }\left[\right|d_k-\exp \left(\mathrm{\&theta;}\right)\left|\right],& {\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0\&le;k\&le;{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+L_d-1\end{array}\right.---\left(12\right)$

In the formula (12), θ is the symbol phase that has the planisphere point of minimum Euclideam distance with soft demodulation result; Θ is the symbol phase set of emission planisphere; For each symbol in the frame head, its phase theta _kFor receiving terminal is known.And, can adopt following direct judgement detection method to obtain θ for non-frame head symbol _k

Make e _kThe integral multiple that adds or deduct 2 π makes it satisfy e _k∈ [π, π); Judge e then _kValue:

Work as e _k〉=0 o'clock, make θ _SumIncrease Δ φ; Otherwise, then make

θ _SumReduce Δ φ; Wherein, Δ φ is default according to actual needs fixed phase value; A kind of preferred version is to be made as π/160.

3) with step 2) soft demodulation result under all k values of obtaining carries out the LDPC decoding.The check matrix of LDPC coding is very sparse, a spot of bit mistake will obtain very little syndrome and value, and vice versa.So determine with thresholding value (relevant) whether frequency deviation and skew are correctly followed the tracks of according to the structure setting of LDPC sign indicating number is a syndrome: if the syndrome of code word is lower than thresholding with value after the LDPC decoding with the factors such as heavy and column weight of going as code length, check matrix, think that then frequency deviation and skew by correct the tracking (signal to noise ratio of suppose this moment greater than the bit error rate that satisfies design or the thresholding signal to noise ratio of frame error ratio demand), change step 5) at this moment over to; Otherwise just think that frequency deviation and skew are not correctly followed the tracks of, change step 4) this moment over to.

4) attempt other frequency offset

Definition

${\hat{f}}_i={\hat{f}}_d+({\mathrm{\&theta;}}_{\mathrm{sum}}+\mathrm{i\&Delta;\&theta;})/\left\{2\mathrm{\&pi;}\right[{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+L_d-(L_0-1)/2\left]T_s\right\},$

i＝±1，±2，Λ，±i _max

Wherein, Δ θ is the stationary phase step-length, 2i _MaxBe the maximum times of attempting, θ _SumGet currency.For each Repeating step 1)～step 3), and

In step 2) in, remember current

Corresponding soft demodulation result is

Then have:

$d_k^i=r_k\exp \{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{\hat{f}}_i[k-(L_0-1)/2]T_s-j{\widehat{\mathrm{\&phi;}}}_{(L_0-1)/2}\},$

(L ₀-1)/2≤k≤L ₀+L _d-1

In step 3), if the code word syndrome be lower than thresholding factors such as (relevant) and column weights heavy with value with the row of the structure of LDPC sign indicating number such as code length, check matrix, then think the carrier synchronization success of first data block to stop circulation, enter step 5) and upgrade

For this moment

And change the tracking that step B carries out other remaining data pieces over to; Otherwise continue to attempt other frequency offset If reached the maximum times 2i that attempts _MaxAnd the syndrome of code word with the value still greater than this threshold value, then change step 5) over to.

5) use

As after the current renewal

Promptly upgrade, change the tracking that step B carries out other remaining data pieces then over to by formula (13).

${\hat{f}}_d={\hat{f}}_d+{\mathrm{\&theta;}}_{\mathrm{sum}}/\left\{2\mathrm{\&pi;}\right[{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+L_d-(L_0-1)/2\left]T_s\right\}---\left(13\right)$

Wherein,

Be initial frequency deviation, extra accumulation of phase and θ _SumBe currency.

The track algorithm of B, remaining data piece

A) utilize the demodulation sign indicating number result of first data block to obtain more accurate phase bias estimation.In the remaining data piece in t data block the maximal possibility estimation of the phase bias of first symbol be:

${\widehat{\mathrm{\&phi;}}}_{{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+(t-1)L_d}=\arg \left\{\underset{k={\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+(t-1){\mathrm{<figure-callout\; id="0"\; label="L\; d\; L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_d}{\overset{L_{}}{\mathrm{\&Sigma;}}}{r}_k\exp \right[-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{\hat{f}}_d(k-L_0-(t-1)L_d)T_s-j{\widehat{\mathrm{\&theta;}}}_k\left]\right\}---\left(14\right)$

Wherein, the t initial value is 1,

It is the phase place of k symbol behind the correct demodulating and decoding of first data block that obtains through process A.Then the skew that can obtain t data block intermediate symbols in the remaining data piece by formula (14) is:

${\widehat{\mathrm{\&phi;}}}_{{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+(t-1)L_d+(L_d-1)/2}={\widehat{\mathrm{\&phi;}}}_{{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+(t-1)L_d}+\mathrm{\&pi;}{\hat{f}}_d(L_d-1)T_s---\left(15\right)$

For at L ₀+ tL _d≤ k≤L ₀+ (t+1) L _dEach k value in-1 scope is carried out step b).

B) the soft demodulation result of k symbol is:

The phase error of definition k symbol this moment is:

$e_k=\arg \left(d_k\right)-\underset{\mathrm{\&theta;}:\mathrm{\&theta;}\&Element;\mathrm{\&Theta;}}{\arg \min }\left[\right|d_k-\exp \left(\mathrm{\&theta;}\right)\left|\right],{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+t{L}_d\&le;k\&le;{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+(t+1)L_d-1$

Adopt directly judgement detector, getting θ is the symbol phase that has the planisphere point of minimum Euclideam distance with soft demodulation result; Θ is the symbol phase set of emission planisphere.

C) the soft demodulation result of all that step b) is obtained is carried out LDPC decoding, if the syndrome of code word be lower than the thresholding (this thresholding is the same with thresholding in the steps A) of setting with value, show that then the carrier wave success of (t+1) individual data block is synchronous, do not upgrade

Change step e) over to; Otherwise change step d) over to.

D) attempt other frequency offset Be defined as:

${\hat{f}}_l={\hat{f}}_d+\mathrm{l\&Delta;f},$ l＝±1，±2，Λ，±l _max(17)

Wherein, Δ f is the frequency step of fixing, 2l _MaxBe the maximum times of attempting, for each

Repeating step a)～step c), and

In step b), remember each

Corresponding soft demodulation result is

Then have:

$d_k^l=r_k\exp \{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="HDA0000065879020000014.png,HDA0000065879020000021.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}{\hat{f}}_l[k-L_0-(t-1)L_d-(L_d-1)/2]T_s-j{\widehat{\mathrm{\&phi;}}}_{{\mathrm{<figure-callout\; id="0"\; label="L"\; filenames="HDA0000065879020000022.png"\; state="\{\{state\}\}">L</figure-callout>}}_0+(j-1)L_d+(L_d-1)/2}\},$ (18)

L ₀+tL _d≤k≤L ₀+(t+1)L _d-1

In step c), if the code word syndrome be lower than thresholding with value, then represent the carrier synchronization success of (t+1) individual data block, stop circulation, upgrade For this moment

And change step e) over to; Otherwise continue to attempt other frequency offset

If reached the maximum times of attempting the syndrome of code word with value still greater than threshold value, then do not upgrade

Change step e) over to.

E) repeat among the B institute in steps, data block demodulation to the last finishes.

Embodiment

The invention will be further described below in conjunction with an example.

Suppose certain QPSK modulating system, the parameter of burst frame is L ₀=128, L _d=512; Only contain a data block.The LDPC code length is 2560, and code rate is 2/5;

Step 1, initial frequency deviation and skew are estimated

For the burst communication pattern, frame head is known for receiving terminal, establishes the local frame head of receiving terminal and is s through QPSK mapping back _k, k=0,1, Λ, L ₀-1, according to (4) formula, this moment θ _k∈ { 0, pi/2, π, 3 pi/2s }, then remove modulation intelligence by following formula and obtain header signal:

$h_k=r_k{s}_k^{*},$ k＝0，1，Λ，L ₀-1

For the continuous communiction pattern, data are unknown for receiving terminal, and then receiving data can remove modulation intelligence by following formula

h _k＝|r _k|exp[j4arg(r _k)]

Employing obtains the single-carrier signal h that obtains through step 1 based on the searching algorithm of dichotomy _kFrequency f _dWith first phase φ ₀Maximum likelihood estimator

With

The tracking of step 2, frequency deviation and skew

In this process, there is Several Parameters to determine: fixed phase value Δ φ, the threshold value of LDPC sign indicating number syndrome and value, frequency step value Δ f, stationary phase step delta θ and maximum attempts 2i in the frequency deviation cut-and-try process _MaxAnd 2l _Max

Fixed phase value Δ φ can estimate mean square error decision by skew, when Δ φ gets π/160, no matter is theory analysis or simulating, verifying, and it is all very good to obtain the skew estimation performance, as shown in Figure 5.

Maximum attempts 2i _MaxThe probability of success of choosing the carrier synchronization that is subjected to design, the influence of the complexity of calculating and the estimated accuracy that will reach, compromise are considered generally to get 3 or 4; 2l _MaxWith 2i _MaxSimilar, but since generally speaking frequency deviation change slow, 2l _MaxCan compare 2i _MaxSmaller, generally get 1.

Factors such as the heavy and column weight of the row of the structure of the thresholding of syndrome and value and LDPC sign indicating number such as code length, check matrix are relevant.Fig. 6 has provided the probability distribution P of syndrome and value s, and (relation of S≤s) and frequency deviation, the initial skew of hypothesis is zero among the figure, Normalized Signal/Noise Ratio E _b/ N ₀Be 2dB, the desirable error rate of whole code modulation system is 10 ^-4

As can be seen from Figure 6, when since frequency deviation and the phase error of in the time of LDPC grouping, accumulating greater than π/4 (2 π f _dTL＞π/4) time, the probability with value s≤400 of syndrome is almost 0, and this moment, the decoding of LDPC sign indicating number fell flat, and the number of bit errors of this section grouping is also a lot; And when the phase error of accumulation during less than π/8, the probability with value s≤400 of syndrome is almost 1, and the number of bit errors of this section grouping seldom at this moment.So the thresholding of the LDPC of this system sign indicating number syndrome and value can be made as 400, Δ θ=π/8, Δ f=1/8L _dT _sWhen the parameter change of LDPC code word, emulation is to determine the new syndrome and the thresholding of value again.

Theory analysis and simulation result show that the performance of initial estimation near a carat Mei-Luo lower bound, receives function steady operation under the low signal-to-noise ratio environment, and the LDPC sign indicating number also shows good performance.This method also can guarantee the energy operate as normal as long as LDPC decodes, and can provide Useful Information under the signal to noise ratio that requires, and carrier synchronization tracking just can be avoided cycle slip, steady operation.

Above-described specific descriptions; purpose, technical scheme and beneficial effect to invention further describe; institute is understood that; the above only is specific embodiments of the invention; and be not intended to limit the scope of the invention; within the spirit and principles in the present invention all, any modification of being made, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (6)

1. one kind based on the auxiliary MPSK system carrier method for synchronous of LDPC coding, it is characterized in that may further comprise the steps:

Step 1, initial frequency deviation and initial skew are estimated:

At first remove the modulation intelligence among the baseband signal rk after down-conversion, obtain single-carrier signal hk, be expressed from the next:

$h_k=r_k\exp (-j{\mathrm{\&theta;}}_k)=\exp (j2\mathrm{\&pi;}{f}_{d}k{T}_s+j{\mathrm{\&phi;}}_0)+{\hat{z}}_k,$

(4)

Wherein, L ₀Number for symbol in the frame head; K represents the label of k sampled point of this single-carrier signal, and k is a positive integer; θ _kBe the phase place of constellation point symbol of transmission, its value is θ _k∈ 0,2 π/M, and 4 π/M, Λ, 2 (M-1) π/M}, M is the MPSK order of modulation; f _dIt is residual frequency departure; T _sThe is-symbol cycle; φ ₀It is initial skew; J is an imaginary unit;

Be the additivity white complex gaussian noise, average is 0, and variance is σ ²z _kIt is the amplitude of additivity white complex gaussian noise; For the burst communication pattern, each frame is formed by a frame head and n data block, and each data block comprises L _dIndividual coded identification, and

N is the length of each data block;

Obtain single-carrier signal h then _kInitial frequency deviation f _dWith initial skew φ ₀Maximum likelihood estimator With And the nothing of the skew of frame head intermediate symbols is estimated partially

Step 2, the auxiliary frequency deviation of decoded result of utilizing LDPC and the tracking of skew comprise the steps:

The track algorithm of A, first data block

1) initial frequency deviation of utilizing step 1 to obtain is estimated

Estimate with skew

And the nothing of the skew of frame head intermediate symbols is estimated partially

Extra accumulation of phase and θ are set _Sum=0; Initialization k=(L ₀-1)/2, at (L ₀-1)/2≤k≤L ₀+ L _dEach k value in-1 scope, order carry out step 2), L ₀Be the number of symbol in the frame head, L _dNumber for symbol in each remaining data piece;

2) skew of k symbol of compensation obtains its soft demodulation result:

$d_k=r_k\exp (-j{\widehat{\mathrm{\&phi;}}}_k)$ (11)

$=r_k\exp \{-j2\mathrm{\&pi;}{\hat{f}}_d[k-(L_0-1)/2]T_s-j{\widehat{\mathrm{\&phi;}}}_{(L_0-1)/2}\},$ (L ₀-1)/2≤k≤L ₀+L _d-1

R wherein _kBe k the baseband form after the symbol down-conversion,

The skew that is k symbol is estimated L ₀Be the number of symbol in the frame head, L _dBe the number of symbol in each remaining data,

Then the phase error of k symbol is

$\left\{\begin{array}{cc}{e}_k=\arg \left(d_k\right)-{\mathrm{\&theta;}}_k,& (L_0-1)/2\&le;k\&le;L_0-1\\ e_k=\arg \left(d_k\right)-\underset{\mathrm{\&theta;}:\mathrm{\&theta;}\&Element;\mathrm{\&Theta;}}{\arg \min }\left[\right|d_k-\exp \left(\mathrm{\&theta;}\right)\left|\right],& L_0\&le;k\&le;L_0+L_d-1\end{array}\right.---\left(12\right)$

In the formula (12), θ is the symbol phase that has the planisphere point of minimum Euclideam distance with soft demodulation result; Θ is the symbol phase set of emission planisphere; For each symbol in the frame head, its phase theta _kFor receiving terminal is known; For non-frame head symbol, adopt following direct judgement detection method to obtain θ _k

Make e _kThe integral multiple that adds or deduct 2 π makes it satisfy e _k∈ [π, π); Judge e then _kValue:

Work as e _k〉=0 o'clock, make

θ _SumIncrease Δ φ; Otherwise, then make θ _SumReduce Δ φ; Wherein, Δ φ is default according to actual needs fixed phase value;

3) with step 2) soft demodulation result under all k values of obtaining carries out the LDPC decoding; Set a syndrome and thresholding value according to the structure of LDPC sign indicating number, and judge in view of the above whether frequency deviation and skew are correctly followed the tracks of: if the syndrome of code word is lower than this thresholding with value after the LDPC decoding, think that then frequency deviation and skew are correctly followed the tracks of, change step 5) this moment over to; Otherwise think that frequency deviation and skew are not correctly followed the tracks of, change step 4) this moment over to;

4) attempt other frequency offset

Definition

${\hat{f}}_i={\hat{f}}_d+({\mathrm{\&theta;}}_{\mathrm{sum}}+\mathrm{i\&Delta;\&theta;})/\left\{2\mathrm{\&pi;}\right[L_0+L_d-(L_0-1)/2\left]T_s\right\},$

i＝±1，±2，Λ，±i _max

Wherein, Δ θ is the stationary phase step-length, 2i _MaxBe the maximum times of attempting, θ _SumGet currency.For each

Repeating step 1)～step 3), and

In step 2) in, remember current

Corresponding soft demodulation result is

Then have:

$d_k^i=r_k\exp \{-j2\mathrm{\&pi;}{\hat{f}}_i[k-(L_0-1)/2]T_s-j{\widehat{\mathrm{\&phi;}}}_{(L_0-1)/2}\},$

(L ₀-1)/2≤k≤L ₀+L _d-1

In step 3), if the code word syndrome be lower than thresholding with value, then think the carrier synchronization success of first data block to stop circulation, enter step 5) and upgrade

For this moment And change the tracking that step B carries out other remaining data pieces over to; Otherwise continue to attempt other frequency offset

If reached the maximum times 2i that attempts _MaxAnd the syndrome of code word with the value still greater than this threshold value, then change step 5) over to;

5) use

As after the current renewal

Promptly upgrade, change the tracking that step B carries out other remaining data pieces then over to by formula (13).

${\hat{f}}_d={\hat{f}}_d+{\mathrm{\&theta;}}_{\mathrm{sum}}/\left\{2\mathrm{\&pi;}\right[L_0+L_d-(L_0-1)/2\left]T_s\right\}---\left(13\right)$

Wherein,

Be initial frequency deviation, extra accumulation of phase and θ _SumBe currency.

The track algorithm of B, remaining data piece

A) in the remaining data piece in t data block the maximal possibility estimation of the phase bias of first symbol be:

${\widehat{\mathrm{\&phi;}}}_{L_0+(t-1)L_d}=\arg \left\{\underset{k=L_0+(t-1)L_d}{\overset{L_0+t{L}_d-1}{\mathrm{\&Sigma;}}}{r}_k\exp \right[-j2\mathrm{\&pi;}{\hat{f}}_d(k-L_0-(t-1)L_d)T_s-j{\widehat{\mathrm{\&theta;}}}_k\left]\right\}---\left(14\right)$

Wherein, the t initial value is 1,

It is the phase place of k symbol behind the correct demodulating and decoding of first data block that obtains through process A; Then the skew that can obtain t data block intermediate symbols in the remaining data piece by formula (14) is:

${\widehat{\mathrm{\&phi;}}}_{L_0+(t-1)L_d+(L_d-1)/2}={\widehat{\mathrm{\&phi;}}}_{L_0+(t-1)L_d}+\mathrm{\&pi;}{\hat{f}}_d(L_d-1)T_s---\left(15\right)$

For at L ₀+ tL _d≤ k≤L ₀+ (t+1) L _dEach k value in-1 scope is carried out step b);

B) the soft demodulation result of k symbol is:

The phase error of definition k symbol this moment is:

$e_k=\arg \left(d_k\right)-\underset{\mathrm{\&theta;}:\mathrm{\&theta;}\&Element;\mathrm{\&Theta;}}{\arg \min }\left[\right|d_k-\exp \left(\mathrm{\&theta;}\right)\left|\right],L_0+t{L}_d\&le;k\&le;L_0+(t+1)L_d-1$

Getting θ is the symbol phase that has the planisphere point of minimum Euclideam distance with soft demodulation result; Θ is the symbol phase set of emission planisphere;

C) the soft demodulation result of all that step b) is obtained is carried out LDPC decoding, if the syndrome of code word be lower than the thresholding of setting with value, the carrier wave success that then shows (t+1) individual data block is not upgraded synchronously

Change step e) over to; Otherwise change step d) over to.

D) attempt other frequency offset

Be defined as:

${\hat{f}}_l={\hat{f}}_d+\mathrm{l\&Delta;f},$ l＝±1，±2，Λ，±lmax(17)

Wherein, Δ f is the frequency step of fixing, 2l _MaxBe the maximum times of attempting, for each

Repeating step a)～step c), and

In step b), remember each

Corresponding soft demodulation result is

Then have:

$d_k^l=r_k\exp \{-j2\mathrm{\&pi;}{\hat{f}}_l[k-L_0-(t-1)L_d-(L_d-1)/2]T_s-j{\widehat{\mathrm{\&phi;}}}_{L_0+(j-1)L_d+(L_d-1)/2}\},$ (18)

L ₀+tL _d≤k≤L ₀+(t+1)L _d-1

In step c), if the code word syndrome be lower than thresholding with value, then think the carrier synchronization success of (t+1) individual data block to stop circulation, upgrade

For this moment

And change step e) over to; Otherwise continue to attempt other frequency offset

If reached the maximum times of attempting the syndrome of code word with value still greater than threshold value, then do not upgrade

Change step e) over to;

E) repeat among the B institute in steps, data block demodulation to the last finishes.

2. according to the described a kind of MPSK system carrier method for synchronous of assisting of claim 1, it is characterized in that based on the LDPC coding, in the step 1, single-carrier signal h _kInitial frequency deviation f _dWith initial skew φ ₀Maximum likelihood estimator

With

Obtain according to following formula:

${\hat{f}}_d=\underset{f}{\arg \max }\left\{\right|\underset{k=0}{\overset{L_0-1}{\mathrm{\&Sigma;}}}{h}_k\exp (-j2\mathrm{\&pi;fk}{T}_s)\left|\right\}---\left(6\right)$

${\widehat{\mathrm{\&phi;}}}_0=\arg \left\{\underset{k=0}{\overset{L_0-1}{\mathrm{\&Sigma;}}}{h}_k\exp (-j2\mathrm{\&pi;}{\hat{f}}_{d}k{T}_s)\right\}---\left(7\right)$

3. described a kind of based on the auxiliary MPSK system carrier method for synchronous of LDPC coding according to claim 1, it is characterized in that employing obtains single-carrier signal h based on the searching algorithm of dichotomy _kInitial frequency deviation f _dWith initial skew φ ₀Maximum likelihood estimator

With

4. according to the described a kind of MPSK system carrier method for synchronous of assisting of claim 1, it is characterized in that based on the LDPC coding, when not considering The noise, single-carrier signal h _kFirst phase φ ₀Maximum likelihood estimator

For:

${\widehat{\mathrm{\&phi;}}}_0={\mathrm{\&phi;}}_0+\mathrm{\&pi;}(f_d-{\hat{f}}_d)(L_0-1)T_s---\left(9\right)$

The skew of frame head intermediate symbols

Nothing be estimated as partially

${\widehat{\mathrm{\&phi;}}}_{(L_0-1)/2}={\widehat{\mathrm{\&phi;}}}_0+\mathrm{\&pi;}{\hat{f}}_d(L_0-1)T_s={\mathrm{\&phi;}}_0+\mathrm{\&pi;}{f}_d(L_0-1)T_s---\left(10\right)$

According to claim 1-4 described any based on the auxiliary MPSK system carrier method for synchronous of LDPC coding, it is characterized in that A step 2 in the step 2) described fixed phase value Δ φ be made as π/160.

6. according to the described MPSK system carrier method for synchronous that any is assisted based on the LDPC coding of claim 1-4, it is characterized in that, the thresholding with value of the syndrome of A step 3) is relevant with the structure of LDPC sign indicating number in the step 2, comprises the heavy and column weight factor of row of code length, check matrix.

Images