CN106230760A - A kind of Soft Inform ation acquisition methods being applicable to pi/4_DQPSK and device - Google Patents

Abstract

The invention provides a kind of Soft Inform ation acquisition methods being applicable to pi/4_DQPSK, described method includes: before demodulated signal, the correction process being normalized the amplitude of two adjacent signals received；According to two adjacent signals after normalized correction process, calculate corresponding Soft Inform ation, decode according to described Soft Inform ation.Soft Inform ation acquisition methods of the present invention is independent of channel quality, has the feature of low complex degree and low latency, beneficially hardware and realizes.

Description

A kind of Soft Inform ation acquisition methods being applicable to pi/4_DQPSK and device

Technical field

The invention belongs to field of information transmission, particularly relate to a kind of Soft Inform ation acquisition methods being applicable to pi/4_DQPSK and Device.

Background technology

When the signal received is demodulated, including hard decision and two kinds of methods of soft-decision.Soft-decision need not sentence immediately The bit that disconnecting is received is 0 or 1.To based on log-likelihood ratio, (English full name is Log likelihood Ratio, English abbreviation For LLR) decoder, receiver assign one the to represent multilevel scale value of probability that this bit is 1 or 0 to each bit, i.e. Log-likelihood ratio (LLR) probability.The multilevel scale value of each bit is represented with certain scope integer, i.e. Soft Inform ation. The numerical range of the Soft Inform ation of input decoder is the biggest, and the performance of decoder is the best, and the bit error rate is the lowest.

But, for fixed-point implementation, if it is desired to it is enough that Soft Inform ation retains all of data message, i.e. numerical range Greatly, then need the number of bits that comparison is many, but so decoding time will be long, thus affect decoding efficiency.So be Guarantee decoding efficiency, typically can limit the input range of decoder, to improve the decoding efficiency of decoder.Actual feelings In condition, the numerical range of the Soft Inform ation that demodulation obtains is typically greater than the input range of decoder, i.e. demodulates the Soft Inform ation obtained Number of bits allow the number of bits of input more than decoder, cut so needing that the Soft Inform ation that obtains of demodulation is carried out bit Take, to meet the requirement of decoding efficiency.

Pi/4 phase shift-difference quadrature phase shift keying (pi/4-DQPSK) is a kind of common modulation system, owing to it is to frequency The most insensitive, non-coherent demodulation (differential ference spiral) can be used to reduce the complexity of receiving terminal.In order to improve demodulation performance, After pi/4-DQPSK differential ference spiral, need to extract Soft Inform ation and deliver to subsequent module (such as low-density checksum (English abbreviation For LDPC, English full name is Low Density Parity Check Code) decoding module) process.Subsequent module is to this The bit wide of Soft Inform ation may require that in limiting certain scope, common such as 6～10 bit bit wides, it is therefore necessary to pass through LLR Carry out cut position to adapt to the demand of subsequent module.

Due to influence of noise and the imperfection of automatic growth control (AGC), the signal amplitude received can be caused to become Change, thus cause LLR meansigma methods and variance also to fluctuate in certain limit.Owing to follow-up LDPC decoding module needs the most defeated Entering amplitude, this fluctuation causes the difficulty of LLR cut position.Conventional treatment mode is the LLR average amplitude of statistics a period of time, then It is normalized to be suitable for the meansigma methods of LDPC decoding module.So operation can bring certain time delay, and statistics is the most abundant, needs The data wanted are the most, and time delay is the biggest.

Summary of the invention

It is an object of the invention to provide a kind of Soft Inform ation acquisition methods being applicable to pi/4_DQPSK, to solve existing skill Art is by adding up the LLR average amplitude of a period of time, then when being normalized to the meansigma methods being suitable for LDPC decoding module, can carry Carrying out certain time delay, and statistics is the most abundant, the data of needs are the most, the problem that time delay is the biggest.

First aspect, embodiments provides a kind of Soft Inform ation acquisition methods being applicable to pi/4_DQPSK, described Method includes:

Before demodulated signal, the correction process that the amplitude of two adjacent signals received is normalized；

According to two adjacent signals after normalized correction process, calculate corresponding Soft Inform ation, according to described soft letter Breath decodes.

In conjunction with first aspect, may be in implementation in the first of first aspect, described to receiving two are adjacent The correction process step that the amplitude of signal is normalized includes:

According to formulaAt the correction that the amplitude of two adjacent signals of calculating is normalized Reason, wherein,For the signal in the n-th moment after normalized correction process, the signal of the n-th reception is:And| | it is delivery Operator.

In conjunction with first aspect, first aspect the second may in implementation, described according to normalized correction at Two adjacent signals after reason, calculate corresponding Soft Inform ation step and include:

According to formula: Calculate corresponding Soft Inform ation, wherein: LLR_2n-1、LLR_2nIt is two adjacent signals after correction processWithCorrelation, Re{} is treating excess syndrome portion operator, and Im{} is for taking imaginary-part operation symbol, and * is conjugate operation symbol, and the signal of the n-th reception is:And| | it is modulo operator.

In conjunction with first aspect, may implementation described be carried out according to described Soft Inform ation at the third of first aspect Decoding procedure particularly as follows:

According to the described Soft Inform ation of the information retrieval after the correction process that adjacent signals is normalized, reach LDPC and translate Code module carries out decoding process.

Second aspect, embodiments provides a kind of Soft Inform ation acquisition device being applicable to pi/4_DQPSK, described Device includes:

Normalization unit, for before demodulated signal, is normalized the amplitude of two the adjacent signals received Correction process；

Soft Inform ation computing unit, for according to two adjacent signals after normalized correction process, calculating correspondence Soft Inform ation, decodes according to described Soft Inform ation.

In conjunction with second aspect, second aspect the first may in implementation, described normalization unit specifically for:

According to formulaAt the correction that the amplitude of two adjacent signals of calculating is normalized Reason, wherein,For the signal in the n-th moment after normalized correction process, the signal of the n-th reception is:And| | it is delivery Operator.

In conjunction with second aspect, in the possible implementation of the second of second aspect, described Soft Inform ation computing unit is concrete For:

According to formula:Meter Calculate corresponding Soft Inform ation, wherein: LLR_2n-1、LLR_2nIt is two adjacent signals after correction processWithCorrelation, Re { } is treating excess syndrome portion operator, and Im{} is for taking imaginary-part operation symbol, and * is conjugate operation symbol, and the signal of the n-th reception is:And| | it is modulo operator.

In conjunction with second aspect, in the third possible implementation of second aspect, described Soft Inform ation computing unit is also used In:

According to the described Soft Inform ation of the information retrieval after the correction process that adjacent signals is normalized, reach LDPC and translate Code module carries out decoding process.

In the present invention, before demodulation module, the signal received is carried out dynamic normalized correction process, then The signal after correction process is utilized to be calculated Soft Inform ation.The amplitude of Soft Inform ation now is less than or equal to 1, before and after only remaining The difference of two symbols, and pi/4-DQPSK demodulation itself is also to seek former and later two intersymbol differences, therefore, both principles Identical, and owing to Soft Inform ation acquisition methods of the present invention is independent of channel quality, there is the spy of low complex degree and low latency Point, beneficially hardware realize.

Accompanying drawing explanation

Fig. 1 is the flowchart of the Soft Inform ation acquisition methods being applicable to pi/4_DQPSK that the embodiment of the present invention provides；

Fig. 2 is the structural representation of the Soft Inform ation acquisition device being applicable to pi/4_DQPSK that the embodiment of the present invention provides.

Detailed description of the invention

In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, right The present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, and It is not used in the restriction present invention.

The purpose of the embodiment of the present invention is to provide a kind of Soft Inform ation acquisition methods being applicable to pi/4_DQPSK, to solve In prior art, due to influence of noise and the imperfection of AGC automatic gain, the amplitude of variation of the signal of reception can be caused, Thus cause the meansigma methods of log-likelihood ratio LLR and variance to fluctuate the most within the specific limits.Owing to decoding module is suitable to needs Amplitude, this fluctuation causes the difficulty of the cut position of log-likelihood ratio LLR.In order to LLR carries out cut position, usual processing mode is The average amplitude of the LLR in statistics a period of time, then it is normalized to be suitable for the meansigma methods of decoding module.This mode of operation Can bring certain time delay, and data are the most, time delay is the biggest.Below in conjunction with the accompanying drawings, the present invention is further illustrated.

Fig. 1 shows the realization stream of the Soft Inform ation acquisition methods being applicable to pi/4_DQPSK that the embodiment of the present invention provides Journey, details are as follows:

In step S101, before demodulated signal, the amplitude of two adjacent signals received is normalized Correction process.

Specifically, the embodiment of the present invention introduces the correction coefficient docking collection of letters number dynamically changed and carries out at amplitude normalization Reason.Wherein, normalization formula can be:Wherein s_nIt is the reception signal in the n-th moment, and| | it is modulo operator,It is through returning Signal after one change, a_n=Re{s_n, b_n=Im{s_n, Re{ } it is treating excess syndrome portion operator, Im{ } for taking imaginary-part operation symbol.

In step s 102, according to two adjacent signals after normalized correction process, calculate corresponding Soft Inform ation, Decode according to described Soft Inform ation.

Concrete, the Soft Inform ation of decoder output, for representing the logarithm ratio of probability that each bit is zero or 1, it may be assumed that

$soft\_bit=log\left(\frac{p}{1-p}\right)---\left(1\right)$

Wherein, be p bit be the probability of zero, soft_bit be decoder output Soft Inform ation.

When noiseless and channel change (channel does not cause intersymbol interference) slowly, define d_n-1And d_nDuring difference (n-1)th Signal after quarter and the demodulation of the n-th reception machine, n=1,2 .., then

$\begin{array}{cc}{d}_n=w^{k_n}{d}_{n-1}& k_n\∈\{1,3,5,7\}\end{array}---\left(2\right)$

Wherein,

Assume that noise is to d_n-1And d_nInterference obey average be zero, variance beGauss distribution, the most now, the n-th moment The signal s being an actually-received_nConditional probability function be:

$f\left(s_n\right|d_n)=\frac{1}{2{\mathrm{\π\σ}}_N^2}{e}^{-\left(\frac{1}{2{\mathrm{\σ}}_N^2}{\left|s_n-d_n\right|}^2\right)}---\left(3\right)$

OrderWithIt is respectively d_n-1Real part and imaginary part, as two the symbol s receiving adjacent moment_n-1And s_n Time, it is known that k_nConditional probability function be:

$\begin{array}{c}p\left(k_n\right|s_n,s_{n-1})=\underset{-\mathrm{\∞}}{\overset{\mathrm{\∞}}{\∫}}\underset{-\mathrm{\∞}}{\overset{\mathrm{\∞}}{\∫}}f(k_n,d_{n-1}|s_n,s_{n-1}){\mathrm{dx}}_{d_{n-1}}{\mathrm{dy}}_{d_{n-1}}\\ =\frac{p\left(k_n\right)}{f(s_n,s_{n-1})}\underset{-\mathrm{\∞}}{\overset{\mathrm{\∞}}{\∫}}\underset{-\mathrm{\∞}}{\overset{\mathrm{\∞}}{\∫}}f(s_n,s_{n-1}|k_n,d_{n-1})f\left(d_{n-1}\right){\mathrm{dx}}_{d_{n-1}}{\mathrm{dy}}_{d_{n-1}}\end{array}---\left(4\right)$

Wherein, f (s_n,s_n-1) it is s_n-1And s_nJoint probability distribution function, p (k_n) it is k_nProbability density function, f (d_n-1) it is during noiseless, to receive signal d_n-1Probability density function, f (s_n,s_n-1|k_n,d_n-1) it is s_n-1And s_nCombination condition Probability function.

Assuming that noise jamming independence and k_nEach value occur probability equal, i.e. p (k_n)=1/4, by formula (3) and (4) Can obtain

$\begin{array}{c}p\left(k_n\right|s_n,s_{n-1})=\frac{1/4}{{\left(2\mathrm{\π}\right)}^2{\mathrm{\σ}}_N^{2}f(s_n,s_{n-1})}\×\\ =\underset{-\mathrm{\∞}}{\overset{\mathrm{\∞}}{\∫}}\underset{-\mathrm{\∞}}{\overset{\mathrm{\∞}}{\∫}}{e}^{-\left(\frac{1}{2{\mathrm{\σ}}_N^2}\right({\left|s_{n-1}-d_{n-1}\right|}^2+{\left|w^{-k_n}{s}_n-d_{n-1}\right|}^2\left)\right)}f\left(d_{n-1}\right){\mathrm{dx}}_{d_{n-1}}{\mathrm{dy}}_{d_{n-1}}\end{array}---\left(5\right)$

Under rayleigh fading channel,WithObedience average is zero, variance isGauss distribution, now, nothing Reception signal probability density function f (d during noise_n-1) it is:

$f\left(d_{n-1}\right)=\frac{1}{2{\mathrm{\π\σ}}_s^2}{e}^{-\left(\frac{{\left|d_{n-1}\right|}^2}{2{\mathrm{\σ}}_s^2}\right)}---\left(6\right)$

Formula (6) is substituted in (5), can obtain

$p\left(k_n\right|s_n,s_{n-1})=-{\mathrm{Ke}}^{(w^{-k_n}{s}_n{s}_{n-1}^{*}+w^{k_n}{s}_n^{*}{s}_{n-1})/\left(2{\mathrm{\σ}}_N^2\right(2+{\mathrm{SNR}}^{-1}\left)\right)}---\left(7\right)$

Wherein,K is constant.

From the definition of Soft Inform ation, the probability of first bit is

$b_1=ln\left(\frac{p\left(1\right|s_n,s_{n-1})+p\left(7\right|s_n,s_{n-1})}{p\left(3\right|s_n,s_{n-1})+p\left(5\right|s_n,s_{n-1})}\right)---\left(8\right)$

Similar conclusion is also applied for second bit b₂。

Formula (7) is substituted in (8) and can obtain:

$\begin{array}{cc}{b}_1=\frac{\sqrt{2}\mathrm{Re}\left\{s_n{s}_{n-1}^{*}\right\}}{{\mathrm{\σ}}_N^2({\mathrm{SNR}}^{-1}+2)}& b_2=\frac{\sqrt{2}\mathrm{Im}\left\{s_n{s}_{n-1}^{*}\right\}}{{\mathrm{\σ}}_N^2({\mathrm{SNR}}^{-1}+2)}\end{array}---\left(9\right)$

Wherein, Re{} is treating excess syndrome portion operator, and Im{} is for taking imaginary-part operation symbol, and * is for taking conjugate operation symbol.

If for all of DQPSK symbolBe all constant with SNR, then formula (9) can be reduced to

$\begin{array}{cc}{b}_1=\mathrm{Re}\left\{s_n{s}_{n-1}^{*}\right\}& b_2=\mathrm{Im}\left\{s_n{s}_{n-1}^{*}\right\}\end{array}---\left(10\right)$

It follows that the Soft Inform ation of pi/4-DQPSK is based on adjacent two symbols s_n-1And s_nCorrelation, computing formula As follows:

$\begin{array}{cc}{\mathrm{LLR}}_{2n-1}=\mathrm{Re}\left\{s_{n-1}{s}_n^{*}\right\}& {\mathrm{LLR}}_{2n}=\mathrm{Im}\left\{s_{n-1}{s}_n^{*}\right\}\end{array},n=1,2,\mathrm{..}---\left(11\right)$

Due to influence of noise and the imperfection of AGC (automatic growth control), the signal s received can be caused_nAmplitude Change, thus cause LLR meansigma methods and variance also to fluctuate in certain limit.Due to follow-up decoding module, such as LDPC decodes Module needs suitable input range, and this fluctuation causes the difficulty of LLR cut position.

Conventional treatment mode is the LLR average amplitude of statistics a period of time, then is normalized to be suitable for LDPC decoding mould The meansigma methods of block.So operation can bring certain time delay, and statistics is the most abundant, and the data of needs are the most, and time delay is the biggest.

In order to reduce time delay, simplifying design, the present invention proposes a kind of new Soft Inform ation normalization being applicable to pi/4DQSPK Mode, at s_nOn process, directly to s_nAmplitude normalization.IfWherein, a_n=Re{s_n, b_n =Im{s_n,ThenAfter substituting into formula (11), Obtain:

$\begin{array}{c}{\mathrm{LLR}}_{2n-1}=\mathrm{Re}\left\{{\tilde{s}}_{n-1}{\tilde{s}}_n^{*}\right\}=\mathrm{Re}\left\{e^{j({\mathrm{\θ}}_{n-1}-{\mathrm{\θ}}_n)}\right\}=\cos ({\mathrm{\θ}}_{n-1}-{\mathrm{\θ}}_n)\\ {\mathrm{LLR}}_{2n}=\mathrm{Im}\left\{{\tilde{s}}_{n-1}{\tilde{s}}_n^{*}\right\}=\mathrm{Im}\left\{e^{j({\mathrm{\θ}}_{n-1}-{\mathrm{\θ}}_n)}\right\}=\sin ({\mathrm{\θ}}_{n-1}-{\mathrm{\θ}}_n)\end{array},n=1,2,\mathrm{..}---\left(12\right)$

LLR value amplitude is less than or equal to 1, and simple process just can be sent directly into LDPC decoding module.Visible LLR value now Only remain the difference of former and later two symbols, and pi/4-DQPSK demodulation itself is also to seek former and later two intersymbol differences, two Person's ultimate principle is identical.

Further, all shown by emulation and realization, Soft Inform ation acquisition methods of the present invention, and the mistake of conventional treatment method Rate curve essentially coincides, and performance will not deteriorate.And when realizing, the time delay that the method for the invention introduces is by removing The time delay of several clocks that method introduces, can ignore.

The structural representation of the Soft Inform ation acquisition device being applicable to pi/4_DQPSK that Fig. 2 provides for the embodiment of the present invention, Details are as follows:

The Soft Inform ation acquisition device of pi/4_DQPSK it is applicable to described in the embodiment of the present invention, including:

Normalization unit 201, for before demodulated signal, carries out normalizing to the amplitude of two the adjacent signals received The correction process changed；

Soft Inform ation computing unit 202, for according to two adjacent signals after normalized correction process, calculates correspondence Soft Inform ation, decode according to described Soft Inform ation.

Preferably, described normalization unit specifically for:

According to formulaAt the correction that the amplitude of two adjacent signals of calculating is normalized Reason, wherein,For the signal in the n-th moment after normalized correction process, the signal of the n-th reception is:And| | it is delivery Operator.

Preferably, described Soft Inform ation computing unit specifically for:

According to formula:Meter Calculate corresponding Soft Inform ation, wherein: LLR_2n-1、LLR_2nIt is two adjacent signals after correction processWithCorrelation, Re { } is treating excess syndrome portion operator, and Im{} is for taking imaginary-part operation symbol, and * is conjugate operation symbol, and the signal of the n-th reception is:And| | it is modulo operator.

Preferably, described Soft Inform ation computing unit is additionally operable to:

According to the described Soft Inform ation of the information retrieval after the correction process that adjacent signals is normalized, reach LDPC and translate Code module carries out decoding process.

It is applicable to the Soft Inform ation acquisition device of pi/4_DQPSK described in the embodiment of the present invention, is applicable to pi/4_ with above-mentioned The Soft Inform ation acquisition methods of DQPSK is corresponding, is not repeated at this and repeats.

In several embodiments provided by the present invention, it should be understood that disclosed apparatus and method, can be passed through it Its mode realizes.Such as, device embodiment described above is only schematically, such as, and the division of described unit, only Being only a kind of logic function to divide, actual can have other dividing mode, the most multiple unit or assembly to tie when realizing Close or be desirably integrated into another system, or some features can be ignored, or not performing.Another point, shown or discussed Coupling each other or direct-coupling or communication connection can be the INDIRECT COUPLING by some interfaces, device or unit or logical Letter connects, and can be electrical, machinery or other form.

The described unit illustrated as separating component can be or may not be physically separate, shows as unit The parts shown can be or may not be physical location, i.e. may be located at a place, or can also be distributed to multiple On NE.Some or all of unit therein can be selected according to the actual needs to realize the mesh of the present embodiment scheme 's.

It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to It is that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.Above-mentioned integrated list Unit both can realize to use the form of hardware, it would however also be possible to employ the form of SFU software functional unit realizes.

If described integrated unit realizes and as independent production marketing or use using the form of SFU software functional unit Time, can be stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially The part that in other words prior art contributed or this technical scheme completely or partially can be with the form of software product Embodying, this computer software product is stored in a storage medium, including some instructions with so that a computer Equipment (can be personal computer, server, or the network equipment etc.) performs the complete of method described in each embodiment of the present invention Portion or part.And aforesaid storage medium includes: USB flash disk, portable hard drive, read only memory (ROM, Read-Only Memory), Random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can store program code Medium.

The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all essences in the present invention Any amendment, equivalent and the improvement etc. made within god and principle, should be included within the scope of the present invention.

Claims (8)

1. the Soft Inform ation acquisition methods being applicable to pi/4_DQPSK, it is characterised in that described method includes:

Before demodulated signal, the correction process that the amplitude of two adjacent signals received is normalized；

According to two adjacent signals after normalized correction process, calculate corresponding Soft Inform ation, enter according to described Soft Inform ation Row decoding.

Method the most according to claim 1, it is characterised in that the amplitude of described two adjacent signals to receiving is returned The one correction process step changed includes:

According to formulaCalculate the correction process that the amplitude of two adjacent signals is normalized, its In,For the signal in the n-th moment after normalized correction process, the signal of the n-th reception is:And| | it is delivery Operator.

Method the most according to claim 1, it is characterised in that described adjacent according to two after normalized correction process Signal, calculates corresponding Soft Inform ation step and includes:

According to formula:It is right to calculate The Soft Inform ation answered, wherein: LLR_2n-1、LLR_2nIt is two adjacent signals after correction processWithCorrelation, Re{} is Treating excess syndrome portion operator, Im{} is for taking imaginary-part operation symbol, and * is conjugate operation symbol, and the signal of the n-th reception is:And| | it is modulo operator.

Method the most according to claim 1, it is characterised in that described carry out decoding procedure according to described Soft Inform ation particularly as follows:

According to the described Soft Inform ation of the information retrieval after the correction process that adjacent signals is normalized, reach LDPC and decode mould Block carries out decoding process.

5. the Soft Inform ation acquisition device being applicable to pi/4_DQPSK, it is characterised in that described device includes:

Normalization unit, for before demodulated signal, the school that the amplitude of two the adjacent signals received is normalized Just process；

Soft Inform ation computing unit, for according to two adjacent signals after normalized correction process, calculating corresponding soft letter Breath, decodes according to described Soft Inform ation.

Device the most according to claim 5, it is characterised in that described normalization unit specifically for:

According to formulaCalculate the correction process that the amplitude of two adjacent signals is normalized, Wherein,For the signal in the n-th moment after normalized correction process, the signal of the n-th reception is:And| | it is delivery Operator.

Device the most according to claim 5, it is characterised in that described Soft Inform ation computing unit specifically for:

According to formula:It is right to calculate The Soft Inform ation answered, wherein: LLR_2n-1、LLR_2nIt is two adjacent signals after correction processWithCorrelation, Re{} is Treating excess syndrome portion operator, Im{} is for taking imaginary-part operation symbol, and * is conjugate operation symbol, and the signal of the n-th reception is:And| | it is modulo operator.

Device the most according to claim 6, it is characterised in that described Soft Inform ation computing unit is additionally operable to:

According to the described Soft Inform ation of the information retrieval after the correction process that adjacent signals is normalized, reach LDPC and decode mould Block carries out decoding process.