CN101854229B - Iteration demodulation decoding method of encoded modulation signals based on climax frequency deviation compensation - Google Patents

Abstract

The invention relates to an iteration demodulation decoding method of encoded modulation signals based on climax frequency deviation compensation, effectively solving the problem of iteration demodulation decoding of an LDPC (Low Density Parity Check Codes) encoded BPSK (Binary Phase Shift Keying) modulation system under frequency deviation and an unknown-phase channel. The method comprises the following steps of: receiving one frame of a full frame of encoded modulation signals, and dividing the full frame into a plurality of signal subblocks; initializing the external information of encoded bits; carrying out frequency deviation estimation by adopting a revised M&M arithmetic according to the external information of the encoded bits and the sample value of the received one-frame signals; carrying out a Tikhonov demodulation arithmetic to complete signal demodulation according to the structure of the signal subblocks and the frequency deviation estimation, and outputting a demodulation soft value of the encoded bits; receiving the demodulation soft value by a decoder, completing decoding once according to a confidence coefficient transmission decoding arithmetic, and outputting the external information of the encoded bits; and outputting a decoded value when the iteration demodulation times reach the preset times, and otherwise, increasing the length of the signal subblocks and repeating the operation until decoding is successful or iteration times achieve the regulated value.

Description

Modulation signals advances the iterative demodulation coding/decoding method of compensate of frequency deviation based on layer

Technical field

The present invention relates to wireless and technical field of satellite communication, relate in particular to iterative demodulation coding/decoding method and the device of modulation signals under a kind of utmost point low signal-to-noise ratio.

Background technology

In the wireless and satellite communication system, transmitted bit is subjected to the impact of channel random noise and produces random error.Theory and practice proves that providing the error correction/encoding method of transmission reliability by the introducing redundancy is the effective means of a class.And one of the Turbo code of introducing in recent years and LDPC code encoding scheme that to be the error correcting capability found so far the strongest.

Turbo code equals proposition [C.Berrou in 1993 by C.Berrou, A.Glavieux, andP.Thitimajshima, " Near Shannon limit error-correcting coding and decoding:Turbo-codes, " in ICC ' 93, Geneva, Swithzerland, May, 1993, pp.1064-1070], it is considered to one of maximum progress that coding theory in recent years obtains.Turbo code is 10 in the error rate under white Gaussian noise (Additive White GaussianNoise-AWGN) channel ^-5The time approach the channel capacity limit with the signal to noise ratio of 0.7dB.LDPC (Low-Density Parity Check) code was at first proposed in 1961 by Gallager, but was forgotten by people for a long time until rediscovered by Mackay etc. in 1996.Well-designed LDPC long code performance even be better than Turbo code.For satellite channel, because power limited, the work signal to noise ratio that reduces satellite communication system is vital beyond doubt, introduces the shortcoming that Turbo (or LDPC) encoding and decoding technique can solve the satellite channel power limited well.

Relevant knowledge about the LDPC encoding and decoding sees [T.J.Richardson and R ü diger L.Urbanke for details, " TheCapacity of Low-Density Parity-Check Codes Under Message-Passing Decoding; " IEEE Trans.Inform.Theory, vol.47, Feb.2001, pp.599-618], the LDPC decoding adopts confidence level to relay decoding algorithm, carry out iterative decoding based on the bipartite graph structure, and output class is similar to the external information of Turbo decoding.

Than Turbo code, the design of LDPC code more flexible (different code check/code length), the full parallel organization of LDPC code decoding algorithm is so that the LDPC decoder of design high-throughput is easier.Therefore, the LDPC code is mostly selected in the standardization of relevant chnnel coding in the future communications system.

According to 1 distribution character in the check matrix, the LDPC code can be divided into regular LDPC code and irregular LDPC codes.(the d of a rule _v, d _c)-LDPC code means that check matrix has following characteristic: it is constant that every delegation contains 1 number, and its number is d _cIt is also constant that each row contains 1 number, and its number is d _vOtherwise, then be irregular LDPC codes.

Present most of Turbo/LDPC coding/decoding method has only been considered white Gaussian noise (AWGN) channel, and this is inadequate in actual wireless and satellite communication.Under actual channel, after generally need modulating, coded digital signal sends again, and this just requires transmitting-receiving two-end that identical frequency generator is arranged.But because the non-ideal characteristic of practical devices, particularly transmission medium the time become the characteristic such as decline, so that frequency deviation and the phase place of the modulation signal of transmitting-receiving are unknown in fact to a great extent, this just requires receiver to carry out channel estimation and tracking to eliminate unknown frequency deviation and phase place to the impact of transmitted signal.Like this, thus the Turbo/LDPC code is decoded is the study hotspot that paid close attention to by people to various parameters how to estimate Unknown Channel.The difficulty of this problem is: the signal to noise ratio of Turbo/LDPC code work will be well below the signal to noise ratio under the normal condition, and in fact how accurately to estimate and to follow the tracks of channel parameter itself to be exactly a great problem under low signal-to-noise ratio.

In satellite and radio communication, typical channel unknown parameter is frequency shift (FS) (frequency deviation) and carrier phase.Traditional channel estimation technique is to adopt PHASE-LOCKED LOOP PLL TECHNIQUE.Because the particularity of communicating by letter under the utmost point low signal-to-noise ratio, traditional algorithm does not re-use.Possible solution is channel estimating and channel decoding characteristic and replace the Joint iteration demodulating and decoding independently separately before abandoning, so that channel estimating can take full advantage of the redundancy that coding provides.

Summary of the invention

Technical problem: the purpose of this invention is to provide a kind of for the simple and practical iterative demodulation coding/decoding method of LDPC coding BPSK communication system under the utmost point low signal-to-noise ratio.

Technical scheme: modulation signals of the present invention advances the iterative demodulation coding/decoding method of compensate of frequency deviation based on layer, and its step comprises:

1) receives a frame modulation signals whole frame, and be cut to a plurality of signal sub-blocks;

2) external information of initialization codes bit;

3) according to the external information of coded-bit and a frame signal sample value that receives, adopt a kind of correction M﹠amp; The M algorithm carries out frequency deviation and estimates;

4) estimate to carry out the demodulation of a kind of correction Tikhonov demodulating algorithm settling signal, the soft value of the demodulation of output encoder bit according to current demand signal sub-block structure and above-mentioned frequency deviation;

Decoder receives the above-mentioned soft value of demodulation, finishes once decoding according to the degree of confidence propagation decoding algorithm, and the external information of output encoder bit;

5) when the iterative demodulation number of times reaches predetermined number of times, the output decode value, otherwise, increase the length of signal sub-block, change the next iteration demodulation over to, also be repeating step 3)-5).

Described 3) in based on a kind of correction M﹠amp; The M algorithm calculates frequency deviation, can pass through M﹠amp; M algorithm [M.Morelliand U.Mengali, " Feedforward frequency estimation for psk:A tutorial review; " Eur.Trans.Commun.Related Technol., vol.9, Mar./Apr.2004, pp.103-116] carry out correction measure of the present invention and carry out.

Described 4) signal of receive encoded and modulation is carried out the soft value of demodulation of a kind of Tikhonov of correction demodulating algorithm settling signal demodulation and output encoder bit, can pass through Tikhonov algorithm [G.Colavolpe, A.Barbieri, and G.Caire, " Algorithms for iterative decoding in the presence of strongphase noise; " IEEE J.Select.Areas Commun., vol.23, Sept.2005, pp.1748-1757] carry out correction measure of the present invention and carry out.。

The described LDPC code that is encoded to also extends to the TURBO code; The described BPSK of being modulated to also extends to the MPSK modulation.

The invention provides the iterative demodulation decoding algorithm of the unknown lower LDPC coding BPSK of channel parameter (channel phase, inherent spurious frequency deviation) modulation.Based on the Tikhonov algorithm in conjunction with the distinctive correction of a kind of this patent M﹠amp; M frequency deviation algorithm for estimating has proposed the piecemeal startup, block size recovers and demodulating algorithm with the carrier wave that iteration constantly increases, and this algorithm can take full advantage of the iteration external information and finish the recovery of carrier wave frequency deviation and phase place and obtaining of demodulating data.If the carrier wave that utilizes this patent to propose recovers to carry out to received signal demodulation (" demodulation " word in fact is the meaning of parameter correction) here, then the signal after the demodulation can be regarded desirable awgn channel as, thereby can adopt the standard degree of confidence propagation iterative decoding algorithm under the awgn channel to carry out efficient decoding.

The model of communication system that the present invention relates to is as follows: message bit stream passes through first the LDPC encoder, and modulation sends on the channel coded-bit through BPSK again.Suppose channel phase, channel magnitude, noise variance, inherent spurious frequency deviation the unknown, under the bit synchronous hypothesis of ideal, the equivalent low pass complex signal that receives can be expressed as

$r_k=c_k{e}^{j{\mathrm{\θ}}_k}+w_k,k=0,...,N_s-1,$

Wherein, c _kThe bpsk signal that expression sends, its value in set of signals+1 ,-1}, θ _kThe channel unknown phase,

w _kThe expression average is that 0 variance is 2 σ ²White complex gaussian noise, N _sIt is the total length of a frame.

Symbol definition:

Iterations: l;

Coding code length in one frame: L;

The block count that coding codeword begins in iteration, size (below be called initial sub-block): B, D and L=BD;

Leading code is distributed in each initial sub-block, the leading code number in each sub-block: P;

Block count, the size of the coding codeword data of the l time iteration: B _l, D _l

The piecemeal symbol size of the l time iteration (comprising coded identification and leading code): N _l

Frame signal total sample number a: N who receives _s=(D+P) B;

Frame signal sample subscript a: k who receives;

The subscript position at leading code place set: I in the frame signal that receives _p

The subscript position at coded data place set: I in the frame signal that receives _d

A frame signal sample that receives:

Revise M﹠amp; Segmentation accumulation length in the M frequency deviation algorithm for estimating: F;

One frame data are carried out the sample length of segmentation after cumulative: N _f=N _s/ F.

The present invention relates to the summary of the invention of following four aspects:

One, layer advances the formula partition strategy

The layer of coded modulation small-signal of the present invention advances compensate of frequency deviation and demodulation method, its step 2) and 7) need to advance the formula partition strategy according to iteration enforcement layer, the core that layer advances the formula partition strategy is a minute block size N _lConstantly enlarge (correspondingly, piecemeal number B with iteration _lThen constantly reduce), initially piecemeal is less can guarantee that then the cumulative effect of frequency deviation in this little piecemeal is not obvious, thereby can start normally carrying out of demodulation.Thereby layer advances the formula partition strategy and will guarantee: N _L+1〉=N _lPerhaps B _L+1≤ B _lUntil the piecemeal number is 1.This layer advances the formula partition strategy can be referring to Fig. 1.

Two, revise M﹠amp; The M algorithm

Revise M﹠amp; The M algorithm flow (is designated as )

Step 1) to a frame signal

Utilize leading code information (k ∈ I _p) and external information go modulation operations:

$x_k=\left\{\begin{array}{cc}{r}_k\·\tanh \left(L_d\left(c_k\right)\right),& k\∈I_d;\\ r_k\·c_k,& k\∈I_p.\end{array}\right.$

Step 2) based on the signal that goes to modulate

Carry out the cumulative N that obtains of segmentation _f=N _s/ F sample z _k, k=0 ..., N _f-1:

$z_k=\underset{j=0}{\overset{F-1}{\mathrm{\Σ}}}{x}_{\mathrm{kF}+j},k=0,...,N_f-1.$

Step 3) based on the cumulative sample z of segmentation _k, k=0 ..., N _f-1 calculates its autocorrelation value:

$R_s\left(m\right)=\underset{k=0}{\overset{N_f-1-m}{\mathrm{\Σ}}}{z}_{k+m}{z}_k^{*},m=0,...,E.$

Step 4) calculates final frequency deviation estimated value according to following weighting scheme

Wherein: $w_m=\frac{3({(2E+1)}^2-{(2m+1)}^2)}{({(2E+1)}^2-1)(2E+1)},$

Δ(m)＝mod[arg(R(m+1))-arg(R(m))，2π].

Here mod[x, 2 π] expression is to x delivery 2 π.

Three, revise the Tikhonov algorithm

If the signal sub-block is whole signal frame (r _k, k=0 ..., N _s-1) and known frequency offsets be estimated as

In, then revise the Tikhonov algorithm has mainly been considered frequency deviation on the basis of Tikhnov algorithm phase rotating effect, its concrete calculation process is as follows:

Step 1) front and back are to measuring iteration initialization:

$a_{f,0}=0,a_{b,N_s-1}=0.$

Step 2) forward metrics iterative computation:

K-1 constantly declares α based on the symbol of external information is soft _K-1=tanh (L (c _K-1)) and variance β _K-1=1,

K constantly forward metrics calculates

Step 3) backward tolerance iterative computation:

Step 4) the demodulation bit soft value calculates:

Perhaps adopt approximate calculation

$y_k=|a_{f,k}+a_{b,k}+\frac{r_k\·(+1)}{{\mathrm{\σ}}^2}|-|a_{f,k}+a_{b,k}+\frac{r_k\·(-1)}{{\mathrm{\σ}}^2}|,\mathrm{if},k\∈I_d$

I wherein ₀(x) expression zeroth order Bei Sheer function, for the purpose of simplifying calculating, we adopt approximate representation logI ₀(x)=and x, can get approximate calculation, greatly simplified calculating.

Four, modulation signals advances the iterative demodulation coding/decoding method of compensate of frequency deviation based on layer

Step 1) parameter initialization: L (c _kThe ∞ of)=+, k ∈ I _pL (c _k)=0, k ∈ I _d, l=0.

Step 2) degree of confidence propagation decoding algorithm initialization;

Step 3) carries out correction M﹠amp; The M algorithm calculates frequency deviation:

Step 4) execution is finished carrier wave recovery and demodulation based on the Tikhonov algorithm of signal sub-block:

1) to g (g=0 ..., B _l-1) before and after the individual sub-block carrier wave to iteration initialization:

$a_{f,0}^g=0,a_{b,N_l-1}^g=0.$

2) in g sub-block, carry out carrier wave recover before and after to iterative computation:

The forward direction iterative computation:

k＝g(N _l-P)+n，

α _k-1＝tanh(L(c _k-1))，β _k-1＝1，

Backward iterative computation:

k＝g(N _l-P)+n，

The demodulation bit soft value calculates:

$y_k=|a_{f,n}^g+a_{b,n}^g+\frac{r_k\·(+1)}{{\mathrm{\σ}}^2}|-|a_{f,n}^g+a_{b,n}^g+\frac{r_k\·(-1)}{{\mathrm{\σ}}^2}|,\mathrm{if},k\∈I_d.$

Step 5) based on LDPC coding bound confidence propagation decoding algorithm:

Input: $y^{\left(l\right)}={\left\{y_k^{\left(l\right)}\right\}}_{k\∈I_d},$

Output:

And decoding output codons c.

Step 6) if the decoding code word can satisfy the verification of LDPC check matrix or arrive maximum decoding number of times, then decoding stops, otherwise forwards step 3 to) continue to carry out.

Wherein, be that this coding bound can utilize bipartite graph effectively to represent, such as Fig. 2 according to LDPC coding bound confidence propagation decoding algorithm step 5).

The present invention proposes a kind of correction M﹠amp; M frequency deviation algorithm for estimating, its characteristic feature are can take full advantage of the external information that iterative decoding provides and segmentation is cumulative after going to modulate sample improves the work thresholding.The present invention is based on the Tikhonov algorithm in conjunction with revising M﹠amp; M frequency deviation algorithm for estimating has proposed the piecemeal startup, block size recovers and demodulating algorithm with the carrier wave that iteration constantly increases, this algorithm can take full advantage of the iteration external information and finish the recovery of carrier wave frequency deviation and phase place and obtaining of demodulating data, can effectively solve frequency deviation estimation and the mutual difficult problem that restricts of decoding startup under the utmost point low signal-to-noise ratio.

Beneficial effect:

1. the characteristics of iterative demodulation decoding scheme provided by the invention are that the demodulation code is substantially independent, and both carry out information interaction by external information and improve constantly performance;

2. scheme provided by the invention is simple and practical, is easy to Digital Implementation;

3. correction M﹠amp provided by the invention; The M frequency deviation estimating modules can provide in decoding and can take full advantage of this external information under the condition of external information and improve performance;

4. correction M﹠amp provided by the invention; The M frequency deviation estimating modules is carried out the cumulative operation of sample based on segmentation, has greatly improved work thresholding disease and has effectively reduced implementation complexity;

5. the carrier wave that piecemeal provided by the invention starts, block size constantly increases with iteration recovers and demodulating algorithm can take full advantage of the iteration external information and finishes the recovery of carrier wave frequency deviation and phase place and obtaining of demodulating data, can effectively solve under the utmost point low signal-to-noise ratio frequency deviation and estimate to start the mutually difficult problem of restriction with decoding.

6. irrelevant with concrete channel coding method in carrier wave recovery provided by the invention and demodulation module and the system nature, applicable to the good code of the typical cases such as LDPC code, Turbo code, convolution code;

7. the present invention is applied to typical permanent ginseng satellite channel, can obtain to approach the performance of decoding under the desirable awgn channel.

Description of drawings

Fig. 1 is the demodulating and decoding factor graph that minute block size constantly increases with iteration;

Fig. 2 is the structure chart of a LDPC code bipartite graph, i.e. the connection diagram of check-node and variable node, and variable node and check-node are designated as respectively v and c;

The BPSK receiver system performance of Fig. 3 example;

Abscissa is bit signal to noise ratio Eb/N0 among the figure, ordinate is bit error rate (BER), 3 from top to bottom respectively expressions of curve among the figure: the performance that the Tikhonov algorithm iteration is 60 times, the performance that algorithm iteration of the present invention is 60 times, 60 decoded performances of desirable coherent reception iteration.

All explanation of symbols:

v _n: n variable node in the LDPC code bipartite graph;

c _m: m check-node in the LDPC code bipartite graph;

L: iterations;

L: the coding code length in the frame;

B: the block count (initial piecemeal number) of coding codeword when iteration begins;

D: initial sub-block size;

P: the leading code number in each initial sub-block;

B _l: the block count of the coding codeword data of the l time iteration;

D _l: minute block size of the coding codeword data of the l time iteration;

N _l: the piecemeal symbol size of the l time iteration (comprising coded identification and leading code);

N _s: a frame signal total sample number that receives;

F: revise M﹠amp; Segmentation accumulation length in the M frequency deviation algorithm for estimating;

N _f: frame data are carried out the sample length of segmentation after cumulative;

M﹠amp; M algorithm: document [M.Morelli and U.Mengali, " Feedforward frequency estimationfor psk:A tutorial review; " Eur.Trans.Commun.Related Technol., vol.9, Mar./Apr.2004, pp.103-116] in a kind of frequency deviation algorithm for estimating of providing;

Tikhonov algorithm: document [G.Colavolpe, A.Barbieri, and G.Caire, " Algorithms for iterativedecoding in the presence of strong phase noise; " IEEE J.Select.Areas Commun., vol.23, Sept.2005, pp.1748-1757] in a kind of carrier wave of providing recover and demodulating algorithm;

AWGN: additivity white Gaussian Profile noise.

Embodiment

It is 1/2 rule (3,6)-LDPC code that this example is selected code check, and code length is L=4096.The primary data sub-block number of coding codeword is B=128, and primary data sub-block size D=32 and leading code are distributed in each initial sub-block, and the leading code in each sub-block is counted P=1.The piecemeal number of the coded data of the l time iteration, a minute block size are chosen as respectively B _l=min (2 ^7-l, 128), D _l=min (2 ^lD, 4096), wherein min (x, y) represents the minimum number among peek x and the y.The piecemeal symbol size of the l time iteration (comprising coded identification and leading code) is N _l=D _l+ min (2 ^l+ 1,128+1).Revise M﹠amp; Segmentation accumulation length F=32 in the M frequency deviation algorithm for estimating, parameter E=32.

The iterative demodulation coding/decoding method that the present invention proposes is as follows: receive a frame modulation signals whole frame, and be cut to a plurality of signal sub-blocks; The external information of initialization codes bit; According to the external information of coded-bit and a frame signal sample value that receives, adopt a kind of correction M﹠amp; The M algorithm carries out frequency deviation and estimates; Estimate to carry out the carrier wave recovery according to signal subspace block structure and above-mentioned frequency deviation, and calculate accordingly the soft value of demodulation of coded-bit; Decoder receives the above-mentioned soft value of demodulation, finishes once decoding according to the degree of confidence propagation decoding algorithm, and the external information of output encoder bit; When the iterative decoding number of times reaches predetermined number of times, the output decode value, otherwise, increase the length of signal sub-block, repeat above operation until successfully decoded or iterations reach stipulated number.

Its concrete steps are as follows:

Step 1) parameter initialization: when sample position k is leading code position k ∈ I _p, put L (c _kThe ∞ of)=+ is when sample position k is Data Position k ∈ I _p, L (c _k)=0; Initialization iterations l=0.

Step 2) degree of confidence propagation decoding algorithm initialization: check-node is made as 0 to the transmission of information of variable node.

Step 3) based on receiving sample r and decoding external information L _dCarry out and revise M﹠amp; The M algoritic module calculates frequency deviation:

Step 4) execution is finished carrier wave recovery and demodulation based on the Tikhonov algorithm of signal sub-block:

1) to g (g=0 ..., B _l-1) before and after the individual sub-block carrier wave to iteration initialization:

$a_{f,0}^g=0,a_{b,N_l-1}^g=0.$

2) in g sub-block, carry out carrier wave recover before and after to iterative computation:

The forward direction iteration is calculated successively:

k＝g(N _l-P)+n，

α _k-1＝tanh(L(c _k-1))，β _k-1＝1，

Backward iteration is calculated successively:

k＝g(N _l-P)+n，

The demodulation bit soft value calculates:

$y_k=|a_{f,n}^g+a_{b,n}^g+\frac{r_k\·(+1)}{{\mathrm{\σ}}^2}|-|a_{f,n}^g+a_{b,n}^g+\frac{r_k\·(-1)}{{\mathrm{\σ}}^2}|,\mathrm{if},k\∈I_d.$

Step 5) based on LDPC coding bound confidence propagation decoding algorithm:

Input: $y^{\left(l\right)}={\left\{y_k^{\left(l\right)}\right\}}_{k\∈I_d},$

Output:

And decoding output codons c.

Step 6) if the decoding code word can satisfy the verification of LDPC check matrix or arrive maximum decoding number of times, then decoding stops, otherwise forwards step 3 to) continue to carry out.

The hypothesis channel is typical satellite constant-parameter channel in this example, and the initial normalization frequency deviation (with respect to symbol transmission speed) of channel is positioned at Δ fT=[-0.01,0.01 at random], and the channel unknown phase is the phase place between (0,2 π) that choose at random.This example is seen Fig. 3 in the performance under 60 iteration under the typical satellite constant-parameter channel.

Although reference wherein particular specific embodiment illustrates and illustrates the present invention, but, those of ordinary skill in the art will be understood that, in the situation that does not break away from the defined spirit and scope of claim that the present invention adds, can carry out various modifications on form and the details to the present invention.

Claims (5)

1. a modulation signals advances the iterative demodulation coding/decoding method of compensate of frequency deviation based on layer, it is characterized in that the method comprises the steps:

A. receive a frame modulation signals whole frame, and be cut to a plurality of signal sub-blocks;

B. the external information of initialization codes bit;

C. according to the external information of coded-bit and a frame signal sample value that receives, adopt a kind of correction M﹠amp; The M algorithm carries out frequency deviation and estimates; This revises M﹠amp; The M algorithm at first utilizes the external information of coded-bit to go modulation to received signal, and is then cumulative to going modulated pattern to carry out segmentation, and the sample value after segmentation is cumulative is again according to M﹠amp; M frequency deviation algorithm for estimating calculates frequency deviation value;

D. estimate to carry out the demodulation of a kind of correction Tikhonov demodulating algorithm settling signal according to current demand signal sub-block structure and above-mentioned frequency deviation, the demodulation quantized value of output encoder bit is the soft value of demodulation; This is revised the Tikhonov demodulating algorithm and operates based on the independent signal sub-block; In each independent sub-block, at first calculate forward metrics, and sample-by-sample is rotated correction according to the frequency deviation estimated value to the forward direction sample, then calculate backward tolerance, and sample-by-sample is rotated correction according to the frequency deviation estimated value to backward sample, last according to front and back to the soft value of the demodulation of metric calculation coded-bit;

E. decoder receives the above-mentioned soft value of demodulation, finishes once decoding according to the degree of confidence propagation decoding algorithm, and the external information of output encoder bit;

F. when the iterative demodulation number of times reaches in advance predetermined number of times, the output decode value, otherwise, increase the length of signal sub-block, change the next iteration demodulation over to, also i.e. execution in step c again)～step e).

2. modulation signals as claimed in claim 1 advances the iterative demodulation coding/decoding method of compensate of frequency deviation based on layer, it is characterized in that described step a) is encoded to the LDPC code.

3. modulation signals as claimed in claim 1 advances the iterative demodulation coding/decoding method of compensate of frequency deviation based on layer, it is characterized in that described step a) is modulated to the B-PSK modulation.

4. modulation signals as claimed in claim 1 advances the iterative demodulation coding/decoding method of compensate of frequency deviation based on layer, it is characterized in that described steps d) the signal demodulation is based on the signal subspace block structure, calculate to the soft value of metric calculation and coded-bit the front and back of different sub-block Tikhonov algorithms is independently, the accumulation of phase effect that this has effectively been avoided the frequency deviation evaluated error is conducive to the decoder startup work after the demodulation.

5. modulation signals as claimed in claim 1 advances the iterative demodulation coding/decoding method of compensate of frequency deviation based on layer, it is characterized in that described step f) next iteration solution timing, the signal subspace agllutination is really modulated, the length of the signal sub-block after the modulation will guarantee greater than a front iteration, this will effectively strengthen the precision that carrier wave recovers, and along with iteration improves constantly the Demodulation Systems performance.

Images