CN102315876A - Timing synchronization processing method based on code assistance - Google Patents

Abstract

The invention discloses a timing synchronization processing method based on code assistance, belongs to the technical field of wireless communication, and in particular relates to a high-order modulation code-assistance timing synchronization processing method. By adoption of the method, the aims that operation quantity is low and timing synchronization is realized in high-order modulation are fulfilled. The method has the advantages that: the timing synchronization is realized in the high-order modulation; and compared with the operation quantity of the conventional synchronization method, the operation quantity of the timing synchronization in the high-order modulation is greatly reduced.

Description

Based on the auxiliary timing synchronization processing method of sign indicating number

Technical field

The invention belongs to wireless communication technology field, particularly the sign indicating number auxiliary timing synchronization processing method of high order modulation.

Background technology

Under the low signal-to-noise ratio condition, Turbo code and LDPC code system are very strict to regularly requiring synchronously, even exist little timing offset also can make the mis-behave of system.Therefore data must adopt proper technique to make it can be accurately synchronous before getting into decoder.Yet under utmost point low signal-to-noise ratio environment, accurate regularly is a very problem of difficulty synchronously.Traditional Timing Processing method has early door late, Gardner algorithm.The advantage of Gardner algorithm is insensitive to carrier wave frequency deviation and skew, and shortcoming is that this algorithm can not be worked well under the lower situation of signal to noise ratio.What present stage was popular is that the auxiliary timing of sign indicating number is synchronous, but the computational complexity of these relevant algorithms is all than higher.

Summary of the invention

The present invention proposes the auxiliary timing synchronization processing method of sign indicating number, adopt the present invention can make full use of the energy of signal, and can reduce the complexity of computing in a large number.

Innovation part of the present invention is: proposed the auxiliary timing synchronization processing method of sign indicating number, realized that the timing under the low signal-to-noise ratio is synchronous, and reduced regularly synchronous complexity.

The 4 group data of the present invention to receiving; Satisfactory data block merges the entering decoder again after multiply by weights; The decoder dateout is sent into regularly synchronization module again and is carried out iteration after modulation, after last iteration was accomplished, the output of decoder was directly exported after judging module.

1, system configuration

The timing that sign indicating number is assisted is as shown in Figure 1 synchronously, at a symbol period T _SIn, obtain under distinct symbols synchronous error amount 4 groups uniformly-spaced signals through 4 times of oversampler methods, with four groups of metadata caches in buffer; Calculate the weights of every group of data; Weights according to estimating merge, if weights meet some requirements, then satisfactory data block multiply by and get into decoder after remerging into one group of data behind the weights; The value that decoding obtains gets into regularly synchronization module again after modulation; Carry out the iteration second time, after last iteration was accomplished, the output of decoder was directly exported after judging module.

2, the auxiliary timing synchronization processing method of sign indicating number

2.1 the timing synchronization processing method brief introduction that sign indicating number is auxiliary

Existing timing synchronization processing method operand is than higher; We have proposed at first to calculate the merging weights of every group of data based on the auxiliary timing synchronization processing method of sign indicating number, judge then whether weights satisfy the condition of setting; Merge again after afterwards satisfactory data block being multiply by weights; Make full use of the energy of signal like this, and reduce the complexity of system effectively, because have only one group of data to get into decoder.

The signal that receives is designated as r _{K, i}, k refers to the label of every group of data sample point, i refers to the packet of sampling, and i=1, the length of 2,3,4 each data block is N, the weight of each symbol is M _{K, i}, m (i) is the average weight of each group data.The average weight m (i) of every group of data is:

$m\left(i\right)=\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{M}_{k,i}$ (N is the length of every group of data)

Find out maximum average weight and be designated as max{m (i), max{m (i) } corresponding data block is labeled as r ' _Max, make w _l=m _l/ max{m (i) }, so just can obtain normalization weight w _l, l=0,1,2, w _lCorresponding data block is labeled as r ' _lLabel space L={l, l satisfies w _lIf＞constant} is w _l＞constant is then with data r ' _lTake advantage of in weight w _lBack and max{m (i) } corresponding data set r ' _MaxMerge.If w _l＜constant is nonjoinder then; Signal r after then merging _CombFor:

$r_{\mathrm{comb}}=r_{\max }^{\′}+\underset{l\∈L}{\mathrm{\Σ}}{w}_l\·r_l^{\′}$

2.2 the step of timing synchronization processing method

The first step: for the first time the data that receive are carried out hard decision during iteration, obtain and can obtain

from the output of decoder during iteration for the second time

Second step: calculate and be superimposed upon the noise on each symbol, $N_{k,i}={\left|\right|r_{k,i}-{\hat{r}}_{k,i}\left|\right|}^2.$

The 3rd step: calculate the weight of each symbol, $M_{k,i}=\left|\right|\frac{{\left|{\hat{r}}_{k,i}\right|}^2}{N_{k,i}}\left|\right|.$

The 4th step: calculate the average weight of every group of data, $m\left(i\right)\frac{1}{L}\underset{k=1}{\overset{L}{\mathrm{\Σ}}}{M}_{k,i}.$

The 5th step: find out maximum average weight and be designated as max{m (i) }, max{m (i) } corresponding data block is labeled as r ' _Max, make w _l=m _l/ max{m (i) }, l=0,1,2, w _lCorresponding data block is labeled as r ' _l

The 6th step: go on foot the w that draws according to last one _l, merge signal r _Comb:

$r_{\mathrm{comb}}=r_{\max }^{\′}+\underset{l\∈M}{\mathrm{\Σ}}{w}_l\·r_l^{\′}$

Description of drawings

The regularly synchronous sketch map of Fig. 1.

Claims (2)

1. based on the auxiliary timing synchronization processing method of sign indicating number; The thresholding of the data based setting that merges; Satisfactory data block taken advantage of behind weights, merge the entering decoder again; The decoder dateout is sent into regularly synchronization module again and is carried out iteration after modulation, after last iteration was accomplished, the output of decoder was directly exported after judging module.

2. according to the said timing synchronization processing method of claim 1, comprise following step:

Step 1: for the first time the data that receive are carried out hard decision during iteration; During iteration that to obtain

for the second time later, can obtain by the output of decoder

Step 2: calculate the noise of each symbol, $N_{k,i}={\left|\right|r_{k,i}-{\hat{r}}_{k,i}\left|\right|}^2.$

Step 3: calculate the weight of each symbol, $M_{k,i}=\left|\right|\frac{{\left|{\hat{r}}_{k,i}\right|}^2}{N_{k,i}}\left|\right|.$

Step 4: calculate the average weight of every group of data, $m\left(i\right)=\frac{1}{L}\underset{k=1}{\overset{L}{\mathrm{\Σ}}}{M}_{k,i}.$

Step 5: find out maximum average weight and be designated as max{m (i) }, max{m (i) } corresponding data block is labeled as r ' _Max, make m _l=m _l/ max{m (i) }, l=0,1,2, m _lCorresponding data block is labeled as r ' _l

Step 6: go on foot the m that draws according to last one _l, merge signal r _Comb: $r_{\mathrm{Comb}}=r_{\mathrm{Max}}^{\′}+\underset{l\∈M}{\mathrm{\Σ}}{m}_l\·r_l^{\′}.$

Images