CN101753264B - Acquisition method and device for evaluating threshold value of perforation pattern performance - Google Patents

Abstract

The invention discloses an acquisition method and a device for evaluating a threshold value of perforation pattern performance. The method comprises the following steps of acquiring a perforation pattern used when encoding information bits according to an expansion matrix of a basic matrix, wherein the expansion matrix is obtained according to the quasi-cyclic basic matrix with low density and an odd-even check code; obtaining a perforation factor and a non perforation factor according to the perforation pattern; obtaining a first mutual information value according to a set signal noise ratio; obtaining a fourth mutual information value according to a conversion matrix, the perforation factor, the non perforation factor and the first mutual information value of the basic matrix; and obtaining the threshold value used for evaluating the perforation pattern performance according to the fourth mutual information value. The invention can enhance the working efficiency and is suitable for perforation pattern performance evaluation of perforations with arbitrary amount and perforation positions.

Description

Method and device for acquiring threshold value for evaluating punching pattern performance

Technical Field

The present invention relates to mobile communication technologies, and in particular, to a method and an apparatus for obtaining a threshold value for evaluating a puncturing pattern performance.

Background

Low Density Parity Check (LDPC) codes have unique advantages in the field of channel coding and decoding, and Hybrid Automatic Repeat Request (HARQ) techniques can achieve higher gain spectral efficiency. LDPC codes typically employ a belief propagation algorithm, which is an iterative decoding algorithm. The LDPC original in the decoding process (for the quasi-cyclic LDPC code, the LDPC original may be obtained by base matrix transformation thereof) is punctured (processed), so that the LDPC technique and the HARQ technique can be effectively combined. In order to improve the system performance, it is necessary to obtain the perforation patterns with optimal performance, and currently, a simulation method is usually adopted to evaluate the performance of each perforation pattern.

However, the simulation method consumes a lot of time and resources, and for this reason, there is a method in the prior art that an algorithm is used to obtain a threshold value, and the threshold value is used to evaluate the performance of the puncturing pattern used when encoding the information bits. The algorithm obtains a critical signal-to-noise ratio according to the mutual information quantity of the quasi-cyclic LDPC code in the iterative decoding process, and the punching performance is evaluated by using the critical signal-to-noise ratio (namely a threshold value). The mutual information quantity includes a channel output probability log-likelihood ratio and a mutual information value of a sending sequence (hereinafter referred to as a first mutual information value), a variable node decoder output probability log-likelihood ratio and a mutual information value of a sending sequence (hereinafter referred to as a second mutual information value), a check node decoder output probability log-likelihood ratio and a mutual information value of a sending sequence (hereinafter referred to as a third mutual information value), and a decoder output posterior probability log-likelihood ratio and a mutual information value of a sending sequence (hereinafter referred to as a fourth mutual information value). The calculation formula of the first mutual information value is as follows:

transformation matrix H of quasi-cyclic LDPC code_i，jWhen the signal is not equal to 0, the signal is transmitted,

the calculation formula of the second mutual information value is as follows: $I_{\mathrm{Ev}}(i,j)=J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

the calculation formula of the third mutual information value is as follows: $I_{\mathrm{Ec}}(i,j)=1-J\left(\sqrt{\underset{s\≠j}{\mathrm{\Σ}}{H}_{i,s}{\left[J^{-1}(1-I_{\mathrm{Ac}}(i,s))\right]}^2}\right)$

the calculation formula of the fourth mutual information value is as follows: $I_{\mathrm{APP}}\left(j\right)=J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right).$

finally, obtaining a threshold value according to the fourth mutual information value, namely if the threshold value is met, I_APP(j)＝1， $\∀j=0...N-1$ Then the explanation is at the present E_b/N₀Can correctly decode, so will E_b/N₀Decreasing, starting with calculating the first mutual information value again; if not, indicating that the current preset E is_b/N₀Cannot decode correctly, so will E_b/N₀And increasing, and starting from calculating the first mutual information value again. A critical E is finally obtained_b/N₀Fail to decode correctly when the SNR is greater than the threshold, E_b/N₀The critical value is the puncturing performance threshold of the quasi-cyclic LDPC code, and the smaller the threshold is, the better the performance is.

The inventor finds that the prior art has at least the following problems in the process of implementing the invention: since the parameters (first mutual information values) related to the puncturing pattern are obtained before the spreading, the prior art can only estimate the situation that the number of bits to be punctured is exactly an integral multiple of the spreading factor, and can only estimate the situation that the puncturing is performed by taking the block matrix as a unit, that is, the applicable scenarios are limited.

Disclosure of Invention

The invention provides a method and a device for acquiring a threshold value for evaluating the performance of a punching pattern, which solve the problem of limited applicable scenes.

The embodiment of the invention provides a method for acquiring a threshold value for evaluating the performance of a punching pattern, which comprises the following steps:

acquiring a punching pattern used when information bits are coded according to an expansion matrix of a base matrix, wherein the expansion matrix is obtained according to the base matrix of a quasi-cyclic low-density parity check code;

obtaining a punching factor and a non-punching factor according to the punching pattern;

obtaining a first mutual information value according to the set signal-to-noise ratio;

obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix;

obtaining a threshold value for evaluating the performance of the punched pattern according to the fourth mutual information value;

the first mutual information value is a channel output probability log-likelihood ratio and a sending sequence, the second mutual information value is a variable node decoder output probability log-likelihood ratio and a sending sequence, the third mutual information value is a check node decoder output probability log-likelihood ratio and a sending sequence, and the fourth mutual information value is a decoder output posterior probability log-likelihood ratio and a sending sequence;

the calculation formula for obtaining the punching factor and the non-punching factor is as follows:

λ_j＝z_j/Z

wherein λ is_jIs a puncturing factor, z_jThe number of the expanded punched holes corresponding to each column of the base matrix, and Z is an expansion factor; the non-puncturing factor is 1-lambda_j；

The calculation formula for obtaining the first mutual information value according to the set signal-to-noise ratio is as follows:

$I_{\mathrm{ch}}^{\left(j\right)}=J\left(\sqrt{8R\frac{E_b}{N_0}}\right)$

wherein,

is the value of the j-th column of the first mutual information value, j =0 … N-1, N is the number of columns of the transformation matrix, R is the code rate of the LDPC code, E is the code rate of the LDPC code_b/N₀For a set signal-to-noise ratio, the expression of the J function is:

$J\left(\mathrm{\σ}\right)=1-{\∫}_{-\∞}^{+\∞}\frac{1}{\sqrt{2{\mathrm{\π\σ}}^2}}{e}^{-\frac{{(y-\frac{{\mathrm{\σ}}^2}{2})}^2}{2{\mathrm{\σ}}^2}}\×{\log }_2(1+e^{-y})\mathrm{dy};$

the obtaining of the fourth mutual information value includes:

obtaining a second mutual information value and a third mutual information value through iterative operation according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix;

obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor, the first mutual information value and the third mutual information value of the base matrix;

the obtaining of the second mutual information value and the third mutual information value includes:

initializing a third mutual information value, obtaining a non-punching mutual information value according to the initialized third mutual information value, the first mutual information value and the transformation matrix, and obtaining a punching mutual information value according to the initialized third mutual information value and the transformation matrix;

weighting the non-perforation mutual information value by using a non-perforation factor, and weighting the perforation mutual information value by using a perforation factor to obtain a second mutual information value;

obtaining a third mutual information value according to the second mutual information value;

repeating the steps of initializing a third mutual information value according to a preset iteration number, obtaining a non-punching mutual information value according to the initialized third mutual information value, the first mutual information value and the transformation matrix, obtaining a punching mutual information value according to the initialized third mutual information value and the transformation matrix, weighting the non-punching mutual information value by using a non-punching factor, weighting the punching mutual information value by using a punching factor, obtaining a second mutual information value, and obtaining the third mutual information value according to the second mutual information value, so as to obtain the second mutual information value and the third mutual information value after iteration;

the non-punctured mutual information value is $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right),$ The value of the perforation mutual information is $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right);$

The calculation formula for obtaining the fourth mutual information value is as follows:

for j =0 … N-1:

$I_{\mathrm{APP}}\left(j\right)=(1-{\mathrm{\λ}}_j)\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+{\mathrm{\λ}}_j\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

or,

$I_{\mathrm{APP}}\left(j\right)=a_3\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+a_4\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

or,

$I_{\mathrm{APP}}\left(j\right)=J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{[a\×J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)]}^2}\right);$

wherein, I_APP(j) Is the fourth mutual information value of the j column, H_s,jFor transforming the values of the elements of the matrix at the s-th row and j-th column, I_Av(s, j) is the third mutual information value, | a, of the s-th row and j-th column₃-(1-λ_j)|≤0.5,|a₄-λ_j|≤0.5，|a-(1-λ_j)|≤0.5；

The calculation formula for obtaining the third mutual information value is as follows:

for j =0 … N-1 and i =0 … M-1:

if H is present_i,j≠0，

$I_{\mathrm{Ec}}(i,j)=1-J\left(\sqrt{\underset{s\≠j}{\mathrm{\Σ}}{H}_{i,s}{\left[J^{-1}(1-I_{\mathrm{Ac}}(i,s))\right]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ec(i,j)=0；

Wherein, I_Ec(i, j) is the third mutual information value of ith row and jth column, H_i,sFor the values of the elements of the ith row and the s column of the transformation matrix, M is the number of rows of the transformation matrix, I_Ac(i, s) is the second mutual information value of the ith row and the ith column.

The embodiment of the invention provides an acquisition device for evaluating a threshold value of punching pattern performance, which comprises the following steps:

the pattern acquisition module is used for acquiring a punching pattern used when information bits are coded according to an extended matrix of a base matrix, wherein the extended matrix of the base matrix is obtained according to the base matrix of the quasi-cyclic low-density parity check code;

the factor acquisition module is used for acquiring a punching factor and a non-punching factor according to the punching pattern;

the first module is used for obtaining a first mutual information value according to the set signal-to-noise ratio;

the fourth module is used for obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix;

the threshold module is used for obtaining a threshold value for evaluating the performance of the punching pattern according to the fourth mutual information value;

the fourth module includes:

the node module is used for obtaining a second mutual information value and a third mutual information value through iterative operation according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix;

the probability module is used for obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor, the first mutual information value and the third mutual information value of the base matrix;

the node module includes:

the second module is used for initializing a third mutual information value, obtaining a non-punching mutual information value according to the initialized third mutual information value, the first mutual information value and the transformation matrix, and obtaining a punching mutual information value according to the initialized third mutual information value and the transformation matrix; weighting the non-punching mutual information value by using a non-punching factor, weighting the punching mutual information value by using a punching factor, and obtaining a second mutual information value according to the preset iteration number;

the third module is used for obtaining a third mutual information value according to the second mutual information value and the preset iteration times;

the first mutual information value is a channel output probability log-likelihood ratio and a sending sequence, the second mutual information value is a variable node decoder output probability log-likelihood ratio and a sending sequence, the third mutual information value is a check node decoder output probability log-likelihood ratio and a sending sequence, and the fourth mutual information value is a decoder output posterior probability log-likelihood ratio and a sending sequence;

the calculation formula for obtaining the punching factor and the non-punching factor is as follows:

λ_j＝z_j/Z

wherein λ is_jIs a puncturing factor, z_jThe number of the expanded punched holes corresponding to each column of the base matrix, and Z is an expansion factor; the non-puncturing factor is 1-lambda_j；

The calculation formula for obtaining the first mutual information value according to the set signal-to-noise ratio is as follows:

$I_{\mathrm{ch}}^{\left(j\right)}=J\left(\sqrt{8R\frac{E_b}{N_0}}\right)$

wherein,is the value of the j-th column of the first mutual information value, j =0 … N-1, N is the number of columns of the transformation matrix, R is the code rate of the LDPC code, E is the code rate of the LDPC code_b/N₀For a set signal-to-noise ratio, the expression of the J function is:

$J\left(\mathrm{\σ}\right)=1-{\∫}_{-\∞}^{+\∞}\frac{1}{\sqrt{2{\mathrm{\π\σ}}^2}}{e}^{-\frac{{(y-\frac{{\mathrm{\σ}}^2}{2})}^2}{2{\mathrm{\σ}}^2}}\×{\log }_2(1+e^{-y})\mathrm{dy};$

the non-punctured mutual information value is $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right),$ The value of the perforation mutual information is $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right);$

The calculation formula for obtaining the fourth mutual information value is as follows:

for j =0 … N-1:

$I_{\mathrm{APP}}\left(j\right)=(1-{\mathrm{\λ}}_j)\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+{\mathrm{\λ}}_j\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

or,

$I_{\mathrm{APP}}\left(j\right)=a_3\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+a_4\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

or,

$I_{\mathrm{APP}}\left(j\right)=J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{[a\×J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)]}^2}\right);$

wherein, I_APP(j) Is the fourth mutual information value of the j column, H_s,jFor transforming the values of the elements of the matrix at the s-th row and j-th column, I_Av(s, j) is the third mutual information value, | a, of the s-th row and j-th column₃-(1-λ_j)|≤0.5,|a₄-λ_j|≤0.5，|a-(1-λ_j)|≤0.5；

The calculation formula for obtaining the third mutual information value is as follows:

for j =0 … N-1 and i =0 … M-1:

if H is present_i,j≠0，

$I_{\mathrm{Ec}}(i,j)=1-J\left(\sqrt{\underset{s\≠j}{\mathrm{\Σ}}{H}_{i,s}{\left[J^{-1}(1-I_{\mathrm{Ac}}(i,s))\right]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ec(i,j)=0；

Wherein, I_Ec(i, j) is the third mutual information value of ith row and jth column, H_i,sFor the values of the elements of the ith row and the s column of the transformation matrix, M is the number of rows of the transformation matrix, I_Ac(i, s) is the second mutual information value of the ith row and the ith column.

According to the technical scheme, the embodiment of the invention adopts the expanded punching pattern to obtain the punching factor, so that the performance evaluation of the punching pattern in any shape is realized.

Drawings

FIG. 1 is a schematic flow chart of an embodiment of the method of the present invention;

fig. 2 is a schematic diagram of a detailed process of obtaining a fourth mutual information value in the embodiment of the method of the present invention;

fig. 3 is a schematic diagram of a specific process of obtaining a second mutual information value and a third mutual information value in the embodiment of the method of the present invention;

fig. 4 is a schematic diagram illustrating a specific process of obtaining a threshold value according to a fourth mutual information value in the embodiment of the method of the present invention;

FIG. 5 is a diagram of a first embodiment of a verification process in an embodiment of a method of the present invention;

FIG. 6 is a diagram of a second embodiment of a verification process in an embodiment of a method of the invention;

FIG. 7 is a diagram of a third embodiment of a verification process in an embodiment of a method of the invention;

FIG. 8 is a schematic structural diagram of a first embodiment of the apparatus of the present invention;

fig. 9 is a schematic structural diagram of a second embodiment of the apparatus of the present invention.

Detailed Description

The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.

In the embodiment of the invention, a punching pattern used when information bits are coded is obtained according to an expansion matrix of a base matrix, wherein the expansion matrix is obtained according to the base matrix of a quasi-cyclic low-density parity check code; obtaining a punching factor and a non-punching factor according to the punching pattern; obtaining a first mutual information value according to the set signal-to-noise ratio; obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix; and obtaining a threshold value for evaluating the performance of the punching pattern according to the fourth mutual information value. By the method, threshold values for evaluating the performance of various punching patterns can be obtained, and the application range is expanded.

Fig. 1 is a schematic flow chart of an embodiment of the method of the present invention, which includes:

step 11: and according to the basic matrix of the quasi-cyclic LDPC code, obtaining a transformation matrix of the basic matrix and an extension matrix of the basic matrix, and punching the extension matrix of the basic matrix to obtain a punching pattern.

There is one base matrix H for each quasi-cyclic LDPC code_b(M N size), the base matrix H_bThe non-zero element in (1) is replaced by "1", the zero element is replaced by "0", and the transformation matrix of the base matrix, namely a binary matrix H (M multiplied by N size), is obtained. For the base matrix H_bSpreading with a spreading factor Z results in a spreading matrix (MZ × NZ size). Different puncturing patterns can be obtained according to different puncturing schemes for the spreading matrix.

Step 12: and obtaining a punching factor and a non-punching factor according to the punching pattern.

The extended number of punctures zj corresponding to each column in the base matrix is different for different puncture patterns.

Setting the spreading factor of the matrix as Z, the puncturing factor lambda_jThe calculation formula of (2) is as follows:

for j =0 … N-1, λ_j=z_j/Z。

Accordingly, the non-puncturing factor is: 1-lambda_j。

Step 13: obtaining a first mutual information value I according to the set signal-to-noise ratio_ch. In particular, the amount of the solvent to be used,

setting a signal-to-noise ratio E_b/N₀For j =0 … N-1, the first mutual information value is calculated as:

$I_{\mathrm{ch}}^{\left(j\right)}=J\left(\sqrt{8R\frac{E_b}{N_0}}\right)$

wherein, R is the code rate of the LDPC code, and the expression of the J function is as follows:

$J\left(\mathrm{\σ}\right)=1-{\∫}_{-\∞}^{+\∞}\frac{1}{\sqrt{2{\mathrm{\π\σ}}^2}}{e}^{-\frac{{(y-\frac{{\mathrm{\σ}}^2}{2})}^2}{2{\mathrm{\σ}}^2}}\×{\log }_2(1+e^{-y})\mathrm{dy}$

this step is different from the conventional algorithm step, i.e. the conventional first mutual information value is related to the puncturing pattern, and when J columns are punctured, the first mutual information value is the value represented by the above-mentioned J function, and when J columns are not punctured, the first mutual information value is 0.

There is no timing constraint relationship between the above steps 12, 13.

Step 14: and obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix.

Step 15: and obtaining a threshold value according to the fourth mutual information value, wherein the threshold value is used for evaluating the performance of the punching pattern.

Fig. 2 is a schematic diagram of a specific process for obtaining a fourth mutual information value in the embodiment of the method of the present invention. Referring to fig. 2, the above step 14 includes:

step 21: and obtaining a second mutual information value and a third mutual information value through iterative operation according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix.

Step 22: and obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor, the first mutual information value and the third mutual information value after iteration of the base matrix.

Fig. 3 is a schematic diagram of a specific process for obtaining the second mutual information value and the third mutual information value in the embodiment of the method of the present invention. Referring to fig. 3, the above step 21 includes:

step 31: obtaining a non-punching mutual information value according to the third mutual information value, the first mutual information value and the transformation matrix, and obtaining a punching mutual information value according to the third mutual information value and the transformation matrix; and weighting the non-perforation mutual information value by using the non-perforation factor, and weighting the perforation mutual information value by using the perforation factor to obtain a second mutual information value. In the first iteration, the third mutual information value needs to be initialized, for example, the initial value of the third mutual information value is set to 0.

For j =0 … N-1 and i =0 … M-1, if H is present_i,jNot equal to 0, second mutual information value I_EvThe calculation formula of (i, j) is:

$I_{\mathrm{Ev}}(i,j)=(1-{\mathrm{\λ}}_j)\×J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+{\mathrm{\λ}}_j\×J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

if H is present_i,jIf not than 0, then I_Ev(i,j)=0。

Then for j =0 … N-1 and i =0 … M-1, the following are calculated:

[0063] I_Ac(i,j)=I_Ev(i,j)。

[0064] wherein, I_Ac(i,j)=I_Ev(I, j) is a second mutual information value, I_Av(i,j)=I_Ec(i, j) is a third mutual information value, $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$ is a non-punctured mutual information value, $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$ is a punctured mutual information value.

This step is different from the existing algorithm step, i.e. the existing second mutual information value and λ_jIrrespective of the prior art $I_{\mathrm{Ev}}(i,j)=\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)(H_{i,j}\≠0).$

Step 32: and obtaining a third mutual information value according to the second mutual information value.

For j =0 … N-1 and i =0 … M-1, if H is present_i,jNot equal to 0, third mutual information value I_Ec(i, j) calculation formula:

$I_{\mathrm{Ec}}(i,j)=1-J\left(\sqrt{\underset{s\≠j}{\mathrm{\Σ}}{H}_{i,s}{\left[J^{-1}(1-I_{\mathrm{Ac}}(i,s))\right]}^2}\right)$

if H is present_i,jIf not than 0, then I_Ec(i,j)=0。

Then for j-0 … N-1 and i-0 … M-1, calculate:

[0071] I_Av(i,j)=I_Ec(i,j)。

[0072] step 33: and judging whether the iteration times reach the preset iteration times, if not, repeatedly executing the steps 31-33, otherwise, executing the step 34.

Step 34: and outputting the iterated third mutual information value.

Then, the specific calculation formula of the fourth mutual information value in step 22 may be:

fourth mutual information value I for j =0 … N-1_APP(j) The calculation formula of (2) is as follows:

$I_{\mathrm{APP}}\left(j\right)=(1-{\mathrm{\λ}}_j)\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+{\mathrm{\λ}}_j\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

this step is different from the existing algorithm step, i.e. the existing fourth mutual information value and λ_jIrrespective of the prior art $I_{\mathrm{APP}}\left(j\right)=J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right).$

Fig. 4 is a schematic diagram of a specific process of obtaining a threshold value according to a fourth mutual information value in the embodiment of the method of the present invention, where the specific process includes:

step 41: judging whether the fourth mutual information values are all 1 or not

(i.e., whether:

) If so, go to step 42, otherwise, go to step 43.

Step 42: the set snr is decreased and then steps 13-15 are repeated until a threshold snr is obtained, which is a threshold value used to evaluate the performance of the perforation pattern.

Step 43: the set snr is increased and then steps 13-15 are repeated until a threshold snr is obtained, which is a threshold value used to evaluate the performance of the perforation pattern.

In fig. 4, steps 11-15 of fig. 1 are shown to illustrate the flow of the loop.

That is, the above steps 41-43 indicate that: if it is satisfied with

Indicate that E is currently preset_b/N₀Can correctly decode, so will E_b/N₀Decreasing, starting again with step 13; if not, indicating that the current preset E is_b/N₀Cannot decode correctly, so will E_b/N₀And increased, starting again with step 13. A critical E is finally obtained_b/N₀When the signal-to-noise ratio is larger than the critical value, the decoding cannot be correctly performed, that is, the puncturing performance threshold of the quasi-cyclic LDPC code is estimated.

Fig. 5 is a schematic diagram of a first embodiment of a verification process in an embodiment of the method of the present invention, fig. 6 is a schematic diagram of a second embodiment of the verification process in the embodiment of the method of the present invention, and fig. 7 is a schematic diagram of a third embodiment of the verification process in the embodiment of the method of the present invention. Fig. 5 is a diagram comparing punching patterns with different numbers of bits punched in the same column, fig. 6 is a diagram comparing punching patterns with the same number of bits punched in different columns, and fig. 7 is a diagram comparing punching patterns with the same total number of bits punched in different columns.

Fig. 5-7 are method verifications using a quasi-cyclic LDPC code with N24, M12, and Z96.

The number X (Y) in fig. 5-7 means that the X-th column in the base matrix is truncated by Y bits. For example, 3(60) indicates that 60 bits are dropped in the third column of the base matrix; 3, (48)24 and (48) indicate that 48 bits are dropped in the third column of the base matrix, and 48 bits are dropped in the 24 th column. In the figure, BER represents bit error rate, BLER represents frame error rate, and threshold represents a performance threshold estimated by an embodiment of the method of the present invention.

The abscissas of FIGS. 5-7 illustrate the signal-to-noise ratio E_b/N₀And the ordinate indicates the bit error rate. The solid line in fig. 5-7 indicates BER and the dotted line indicates BLER obtained by simulation.

As can be seen from fig. 5, the estimation result of the embodiment of the method of the present invention is very accurate. That is, by means of simulation, it can be seen from the simulation curve that 3(30) the puncturing pattern has the best performance (BER, BLER is the best); the threshold value of the puncturing pattern in the embodiment 3(30) of the present invention is the minimum (at the leftmost side), and the smaller the threshold value, the better the performance, therefore, the best performance of the puncturing pattern in the embodiment 3(30) can be obtained by the present invention, which is consistent with the simulation result.

As can be seen from fig. 6, the estimation result of the embodiment of the method of the present invention is very accurate. Namely, by means of simulation, it can be seen from the simulation curve that 10(48) the puncturing pattern has the best performance (BER, BLER is best); the threshold value of the puncturing pattern is minimum (at the leftmost side) in the embodiment 10(48) of the present invention, and the smaller the threshold value is, the better the performance is, therefore, the best performance of the puncturing pattern in the embodiment 10(48) can be obtained by the present invention, which is consistent with the simulation result.

As can be seen from fig. 7, the estimation result of the embodiment of the method of the present invention is very accurate. That is, by means of simulation, it can be seen from the simulation curve that 3(96) the puncturing pattern has the best performance (BER, BLER is best); the threshold value of the puncturing pattern is minimum (on the leftmost side) in the embodiment 3(96) of the present invention, and the smaller the threshold value is, the better the performance is, therefore, the best performance of the puncturing pattern in the embodiment 3(96) can be obtained by the present invention, and the result is consistent with the simulation result.

Therefore, in various cases of fig. 5-7, the results obtained by the embodiment of the present invention are consistent with the actual simulation results, and the performance difference between different punching patterns can be estimated very accurately.

The difference from the existing algorithm is that: the existing algorithm is to determine a first mutual information value according to a puncturing pattern, fix a certain value for a second mutual information value and a fourth mutual information value, and match the extended puncturing factor lambda_jRegardless, the embodiment of the present invention is that the first mutual information value is fixed to a certain value, and then the second mutual information value and the fourth mutual information value are related to the puncturing pattern, i.e. related to the puncturing factor. That is, the existing algorithm performs puncturing and then expansion, and the embodiment of the present invention performs puncturing after expansion. Therefore, the existing algorithm can only evaluate the situation where the number of punctured bits is an integer multiple of the spreading factor due to post-spreading, and puncturing is performed according to a block (if the spreading factor is 5, the block size is 5 × 5) matrix. For example, in fig. 5, the two puncturing patterns 3(30), 3(60) are not integer multiples of the spreading factor 96 due to the number of puncturing bits 30, 60, so the threshold values corresponding to the two puncturing patterns 3(30), 3(60) cannot be calculated by the conventional algorithm, but can be calculated by the embodiment of the present invention; similarly, in fig. 6, the two puncturing patterns 13(48), 10(48) are not integer multiples of the spreading factor 96, so the threshold values corresponding to the two puncturing patterns 13(48), 10(48) cannot be calculated by the conventional algorithm, but can be calculated by the embodiment of the present invention; in fig. 7, the puncturing patterns 3, 24, and 48 are not 96 bits in the entire block matrix because the number of bits to be punctured is located in the 3 rd and 24 th columns, respectively, and the conventional algorithm can only process the block matrix unit, and therefore, the conventional algorithm cannot calculate the puncturing pattern3, (48)24, and (48) the threshold corresponding to the puncturing pattern can be calculated according to the embodiment of the present invention.

In this embodiment, the second mutual information value and the fourth mutual information value are weighted by calculating the puncturing factor and the non-puncturing factor, so that a performance threshold of any puncturing pattern can be obtained, and the performance of each puncturing pattern can be evaluated according to the threshold. Compared with a simulation method, the method can save time and resources, and compared with the existing algorithm, the method can be applied to various punching patterns.

In the above embodiment, the weighting factor of the second mutual information value and the fourth mutual information value is the puncturing factor λ_jAlternatively, the weighting factor may also be the remaining value, i.e.,

the calculation formula of the second mutual information value in step 31 may also be:

for j =0 … N-1 and i =0 … M-1, if H is present_i,j≠0：

$I_{\mathrm{Ev}}(i,j)=a_1\×J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+a_2\×J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ev(i,j)=0。

The formula for calculating the fourth mutual information value in step 22 may also be:

for j =0 … N-1:

$I_{\mathrm{APP}}\left(j\right)=a_3\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+a_4\×\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

wherein，a₁And a₃Is and (1-lambda)_j) Number of like a₂And a₄Is and λ_jA similar number. By close is meant that a threshold is set according to the actual need, e.g. | a₁-(1-λ_j)|≤0.5,|a₂-λ_j|≤0.5。

The remaining parameter calculations are the same as in the above embodiment.

Alternatively, the calculation formula of the second mutual information value in step 31 may also be:

for j =0 … N-1 and i =0 … M-1, if H is present_i,j≠0：

$I_{\mathrm{Ev}}(i,j)=J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{[a\×J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ev(i,j)=0。

The formula for calculating the fourth mutual information value in step 22 may also be:

for j =0 … N-1:

$I_{\mathrm{APP}}\left(j\right)=J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{[a\×J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)]}^2}\right)$

wherein a is (1-lambda)_j) Similar numbers, e.g. | a- (1- λ)_j)|≤0.5。

The remaining parameter calculations are the same as in the above embodiment.

Similarly, other algorithms for modifying the second mutual information value and the fourth mutual information value using the relationship between the number of punctured bits and the number of unpunctured bits are also within the scope of the present invention.

The threshold value is obtained by adopting the algorithm, so that the waste of time and resources by adopting a simulation mode can be avoided, and the working efficiency is improved; in this embodiment, a puncturing factor and a non-puncturing factor are obtained according to the extended puncturing pattern, and the puncturing factor and the non-puncturing factor are used to weight the mutual information value. The accurate estimation evaluation of the puncturing performance under the conditions of any puncturing bit number, any puncturing position, any combination of the number and the position and the like is realized.

Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.

Fig. 8 is a schematic structural diagram of a first embodiment of the apparatus of the present invention, which includes a pattern obtaining module 81, a factor obtaining module 82, a first module 83, a fourth module 84, and a threshold module 85. The pattern obtaining module 81 is configured to obtain a puncturing pattern used when encoding information bits according to an extended matrix of a base matrix, where the extended matrix of the base matrix is obtained according to a base matrix of a quasi-cyclic LDPC code; the factor obtaining module 82 is configured to obtain a puncturing factor and a non-puncturing factor according to the puncturing pattern obtained by the pattern obtaining module 81; the first module 83 is configured to obtain a first mutual information value according to the set signal-to-noise ratio; the fourth module 84 is configured to obtain a transform matrix of the basis matrix according to the basis matrix obtained by the pattern obtaining module 81, the puncturing factor and the non-puncturing factor obtained by the factor obtaining module 82, and the first mutual information value obtained by the first module 83 to obtain a fourth mutual information value; the threshold module 85 is configured to obtain a threshold value according to the fourth mutual information value obtained by the fourth module 84, where the threshold value is used to evaluate the performance of the puncturing pattern.

The threshold value is obtained by adopting the algorithm, so that the waste of time and resources by adopting a simulation mode can be avoided, and the working efficiency is improved; in this embodiment, a puncturing factor and a non-puncturing factor are obtained according to the extended puncturing pattern, and the puncturing factor and the non-puncturing factor are used to weight the mutual information value. The accurate estimation evaluation of the puncturing performance under the conditions of any puncturing bit number, any puncturing position, any combination of the number and the position and the like is realized.

Fig. 9 is a schematic structural diagram of a second apparatus embodiment of the present invention, which includes a pattern obtaining module 91, a factor obtaining module 92, a first module 93, a fourth module 94, and a threshold module 95, where the modules may implement functions of corresponding modules in fig. 8.

The fourth module 94 specifically includes a node module 941 and a probability module 942. The node module 941 obtains a second mutual information value and a third mutual information value through iterative operation according to the transformation matrix of the basis matrix obtained by the pattern obtaining module 91, the punching factor and the non-punching factor obtained by the factor obtaining module 92, and the first mutual information value obtained by the first module 93; the probability module 942 obtains a fourth mutual information value according to the transformation matrix of the base matrix obtained by the pattern obtaining module 91, the puncturing factor and the non-puncturing factor obtained by the factor obtaining module 92, the first mutual information value obtained by the first module 93, and the third mutual information value obtained by the node module 941.

Further, the node module 941 specifically includes a second module 9411 and a third module 9412, where the second module 9411 initializes a third mutual information value, obtains a non-punching mutual information value according to the initialized third mutual information value, the first mutual information value obtained by the first module 93, and the transformation matrix obtained by the pattern obtaining module 91, and obtains a punching mutual information value according to the initialized third mutual information value and the transformation matrix; weighting the non-perforation mutual information value by using the non-perforation factor obtained by the factor obtaining module 92, weighting the perforation mutual information value by using the perforation factor obtained by the factor obtaining module 92, and obtaining a second mutual information value according to the preset iteration number; the third module 9412 obtains a third mutual information value according to a preset iteration number according to the second mutual information value obtained by the second module 9411.

The threshold module 95 includes a determination module 951, an increase module 952, and a decrease module 953. The determining module 951 determines whether the fourth mutual information value is 1, if so, the decreasing module 953 is triggered, otherwise, the increasing module 952 is triggered; after the reducing module 953 is triggered, the set signal-to-noise ratio is reduced, the first module 93 is triggered according to the reduced signal-to-noise ratio to obtain a new first mutual information value, then the subsequent second module 9411 is triggered to obtain a new second mutual information value, the third module 9412 obtains a new third mutual information value, and the fourth module 94 obtains a new fourth mutual information value, the above process is repeated until the obtained fourth mutual information value is not 1, and the signal-to-noise ratio is the threshold value at this time; after the increasing module 952 is triggered, the set signal-to-noise ratio is increased, the first module 93 is triggered to obtain a new first mutual information value according to the increased signal-to-noise ratio, then the subsequent second module 9411 is triggered to obtain a new second mutual information value, the third module 9412 obtains a new third mutual information value, and the fourth module 94 obtains a new fourth mutual information value, the above processes are repeated until the obtained fourth mutual information value is 1, and the signal-to-noise ratio at this time is the threshold value.

The calculation of the parameters such as the puncturing factor, the first mutual information value, the second mutual information value, the third mutual information value, the fourth mutual information value, etc. is the same as the calculation formula in the method embodiment, and is not repeated.

The threshold value is obtained by adopting the algorithm, so that the waste of time and resources by adopting a simulation mode can be avoided, and the working efficiency is improved; in this embodiment, a puncturing factor and a non-puncturing factor are obtained according to the extended puncturing pattern, and the puncturing factor and the non-puncturing factor are used to weight the mutual information value. The accurate estimation evaluation of the puncturing performance under the conditions of any puncturing bit number, any puncturing position, any combination of the number and the position and the like is realized.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the invention without departing from the spirit and scope of the invention.

Claims (5)

1. An obtaining method of a threshold value for evaluating the performance of a punching pattern is characterized by comprising the following steps:

acquiring a punching pattern used when information bits are coded according to an expansion matrix of a base matrix, wherein the expansion matrix is obtained according to the base matrix of a quasi-cyclic low-density parity check code;

obtaining a punching factor and a non-punching factor according to the punching pattern;

obtaining a first mutual information value according to the set signal-to-noise ratio;

obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix;

obtaining a threshold value for evaluating the performance of the punched pattern according to the fourth mutual information value;

the first mutual information value is a channel output probability log-likelihood ratio and a mutual information value of a sending sequence, and the fourth mutual information value is a posterior probability log-likelihood ratio and a mutual information value of the sending sequence output by a decoder;

the calculation formula for obtaining the punching factor and the non-punching factor is as follows:

λ_j=z_j/Z

wherein λ is_jIs a puncturing factor, z_jThe number of the expanded punched holes corresponding to each column of the base matrix, and Z is an expansion factor; the non-puncturing factor is 1-lambda_j；

The calculation formula for obtaining the first mutual information value according to the set signal-to-noise ratio is as follows:

$I_{\mathrm{ch}}^{\left(j\right)}=J\left(\sqrt{8R\frac{E_b}{N_0}}\right)$

wherein,

is the value of the j-th column of the first mutual information value, j =0 … N-1, N is the number of columns of the transformation matrix, R is the code rate of the LDPC code, E is the code rate of the LDPC code_b/N₀For a set signal-to-noise ratio, the expression of the J function is:

$J\left(\mathrm{\σ}\right)=1-{\∫}_{-\∞}^{+\∞}\frac{1}{\sqrt{2\mathrm{\π}{\mathrm{\σ}}^2}}{e}^{-\frac{{(y-\frac{{\mathrm{\σ}}^2}{2})}^2}{2{\mathrm{\σ}}^2}}\×{\log }_2(1+e^{-y})\mathrm{dy};$

the obtaining of the fourth mutual information value includes:

obtaining a second mutual information value and a third mutual information value through iterative operation according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix;

the second mutual information value is a mutual information value of a variable node decoder output probability log-likelihood ratio and a sending sequence, and the third mutual information value is a mutual information value of a check node decoder output probability log-likelihood ratio and a sending sequence;

obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor, the first mutual information value and the third mutual information value of the base matrix;

the obtaining of the second mutual information value and the third mutual information value includes:

initializing a third mutual information value, obtaining a non-punching mutual information value according to the initialized third mutual information value, the first mutual information value and the transformation matrix, and obtaining a punching mutual information value according to the initialized third mutual information value and the transformation matrix;

weighting the non-perforation mutual information value by using a non-perforation factor, and weighting the perforation mutual information value by using a perforation factor to obtain a second mutual information value;

obtaining a third mutual information value according to the second mutual information value;

repeating the steps of initializing a third mutual information value according to a preset iteration number, obtaining a non-punching mutual information value according to the initialized third mutual information value, the first mutual information value and the transformation matrix, obtaining a punching mutual information value according to the initialized third mutual information value and the transformation matrix, weighting the non-punching mutual information value by using a non-punching factor, weighting the punching mutual information value by using a punching factor, obtaining a second mutual information value, and obtaining the third mutual information value according to the second mutual information value, so as to obtain the second mutual information value and the third mutual information value after iteration;

the non-punctured mutual information value is $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right),$ The value of the perforation mutual information is $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right);$

The calculation formula for obtaining the fourth mutual information value is as follows:

for j =0 … N-1:

$I_{\mathrm{APP}}\left(j\right)=(1-{\mathrm{\λ}}_j)\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+{\mathrm{\λ}}_j\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

or,

$I_{\mathrm{APP}}\left(j\right)=a_3\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+a_4\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

or,

$I_{\mathrm{APP}}\left(j\right)=J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{[a\×J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)]}^2}\right);$

wherein, I_APP(j) Is the fourth mutual information value of the j column, H_s,jFor transforming the values of the elements of the matrix at the s-th row and j-th column, I_Av(s, j) is the third mutual information value, | a, of the s-th row and j-th column₃-(1-λ_j)|≤0.5,|a₄-λ_j|≤0.5，|a-(1-λ_j)|≤0.5；

The calculation formula for obtaining the third mutual information value is as follows:

for j =0 … N-1 and i =0 … M-1:

if H is present_i,j≠0，

$I_{\mathrm{Ec}}(i,j)=1-J\left(\sqrt{\underset{s\≠j}{\mathrm{\Σ}}{H}_{i,s}{\left[J^{-1}(1-I_{\mathrm{Ac}}(i,s))\right]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ec(i,j)=0；

Wherein, I_Ec(i, j) is the third mutual information value of ith row and jth column, H_i,sFor the values of the elements of the ith row and the s column of the transformation matrix, M is the number of rows of the transformation matrix, I_Ac(i, s) is the second mutual information value of the ith row and the ith column.

2. The method of claim 1, wherein obtaining the threshold value for evaluating the puncturing pattern performance according to the fourth mutual information value comprises:

when the fourth mutual information value is 1, reducing the set signal-to-noise ratio, and obtaining a new first mutual information value, a new second mutual information value, a new third mutual information value and a new fourth mutual information value according to the reduced signal-to-noise ratio until the obtained fourth mutual information value is not 1, wherein the signal-to-noise ratio is a threshold value;

and when the fourth mutual information value is not 1, increasing the set signal-to-noise ratio, and obtaining a new first mutual information value, a new second mutual information value, a new third mutual information value and a new fourth mutual information value according to the increased signal-to-noise ratio until the obtained fourth mutual information value is 1, wherein the signal-to-noise ratio is the threshold value.

3. The method according to claim 1, wherein the calculation formula for obtaining the second mutual information value is:

for j =0 … N-1 and i =0 … M-1:

if H is present_i,j≠0，

$I_{\mathrm{Ev}}(i,j)=(1-{\mathrm{\λ}}_j)\×J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}(I_{\mathrm{Av}}(s,j))\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+{\mathrm{\λ}}_j\×J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[I_{\mathrm{Av}}(s,j)\right]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ev(i, j) = 0; or,

if H is present_i,j≠0，

$I_{\mathrm{Ev}}(i,j)=a_1\×J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+a_2\×J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ev(i, j) = 0; or,

if H is present_i,j≠0，

$I_{\mathrm{Ev}}(i,j)=J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}(I_{\mathrm{Av}}(s,j))\right]}^2+{[a\×J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ev(i,j)=0；

Wherein, I_Ev(i, j) is the second mutual information value of ith row and jth column, H_i,jFor transforming the values of the elements of the matrix at row i and column j, H_s,jFor transforming the values of the elements of the matrix at the s-th row and j-th column, I_Av(s, j) is the third mutual information value, | a, of the s-th row and j-th column₁-(1-λ_j)≤0.5,|a₂-λ_j|≤0.5，|a-(1-λ_j)|≤0.5。

4. An apparatus for obtaining a threshold value for evaluating a performance of a perforation pattern, comprising:

the pattern acquisition module is used for acquiring a punching pattern used when information bits are coded according to an extended matrix of a base matrix, wherein the extended matrix of the base matrix is obtained according to the base matrix of the quasi-cyclic low-density parity check code;

the factor acquisition module is used for acquiring a punching factor and a non-punching factor according to the punching pattern;

the first module is used for obtaining a first mutual information value according to the set signal-to-noise ratio;

the fourth module is used for obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix;

the threshold module is used for obtaining a threshold value for evaluating the performance of the punching pattern according to the fourth mutual information value;

the fourth module includes:

the node module is used for obtaining a second mutual information value and a third mutual information value through iterative operation according to the transformation matrix, the punching factor, the non-punching factor and the first mutual information value of the base matrix;

the probability module is used for obtaining a fourth mutual information value according to the transformation matrix, the punching factor, the non-punching factor, the first mutual information value and the third mutual information value of the base matrix;

the node module includes:

the second module is used for initializing a third mutual information value, obtaining a non-punching mutual information value according to the initialized third mutual information value, the first mutual information value and the transformation matrix, and obtaining a punching mutual information value according to the initialized third mutual information value and the transformation matrix; weighting the non-punching mutual information value by using a non-punching factor, weighting the punching mutual information value by using a punching factor, and obtaining a second mutual information value according to the preset iteration number;

the third module is used for obtaining a third mutual information value according to the second mutual information value and the preset iteration times;

the first mutual information value is a channel output probability log-likelihood ratio and a sending sequence, the second mutual information value is a variable node decoder output probability log-likelihood ratio and a sending sequence, the third mutual information value is a check node decoder output probability log-likelihood ratio and a sending sequence, and the fourth mutual information value is a decoder output posterior probability log-likelihood ratio and a sending sequence;

the calculation formula for obtaining the punching factor and the non-punching factor is as follows:

λ_j＝z_j/Z

wherein λ is_jIs a puncturing factor, z_jThe number of the expanded punched holes corresponding to each column of the base matrix, and Z is an expansion factor; the non-puncturing factor is 1-lambda_j；

The calculation formula for obtaining the first mutual information value according to the set signal-to-noise ratio is as follows:

$I_{\mathrm{ch}}^{\left(j\right)}=J\left(\sqrt{8R\frac{E_b}{N_0}}\right)$

wherein,

is the value of the j-th column of the first mutual information value, j =0 … N-1, N is the number of columns of the transformation matrix, R is the code rate of the LDPC code, E is the code rate of the LDPC code_b/N₀For a set signal-to-noise ratio, the expression of the J function is:

$J\left(\mathrm{\σ}\right)=1-{\∫}_{-\∞}^{+\∞}\frac{1}{\sqrt{2\mathrm{\π}{\mathrm{\σ}}^2}}{e}^{-\frac{{(y-\frac{{\mathrm{\σ}}^2}{2})}^2}{2{\mathrm{\σ}}^2}}\×{\log }_2(1+e^{-y})\mathrm{dy};$

the non-punctured mutual information value is $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right),$ The value of the perforation mutual information is $J\left(\sqrt{\underset{s\≠i}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right);$

The calculation formula for obtaining the fourth mutual information value is as follows:

for j =0 … N-1:

$I_{\mathrm{APP}}\left(j\right)=(1-{\mathrm{\λ}}_j)\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+{\mathrm{\λ}}_j\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

or,

$I_{\mathrm{APP}}\left(j\right)=a_3\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{\left[J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)\right]}^2}\right)$

$+a_4\×J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2}\right)$

or,

$I_{\mathrm{APP}}\left(j\right)=J\left(\sqrt{\underset{s}{\mathrm{\Σ}}{H}_{s,j}{\left[J^{-1}\left(I_{\mathrm{Av}}(s,j)\right)\right]}^2+{[a\×J^{-1}\left(I_{\mathrm{ch}}^{\left(j\right)}\right)]}^2}\right);$

wherein, I_APP(j) Is the fourth mutual information value of the j column, H_s,jFor transforming the values of the elements of the matrix at the s-th row and j-th column, I_Av(s, j) is the third mutual information value, | a, of the s-th row and j-th column₃-(1-λ_j)|≤0.5,|a₄-λ_j|≤0.5，|a-(1-λ_j)|≤0.5；

The calculation formula for obtaining the third mutual information value is as follows:

for j =0 … N-1 and i =0 … M-1:

if H is present_i,j≠0，

$I_{\mathrm{Ec}}(i,j)=1-J\left(\sqrt{\underset{s\≠j}{\mathrm{\Σ}}{H}_{i,s}{\left[J^{-1}(1-I_{\mathrm{Ac}}(i,s))\right]}^2}\right)$

If H is present_i,jIf not than 0, then I_Ec(i,j)=0；

Wherein, I_Ec(i, j) is the third mutual information value of ith row and jth column, H_i,sFor the values of the elements of the ith row and the s column of the transformation matrix, M is the number of rows of the transformation matrix, I_Ac(i, s) is the second mutual information value of the ith row and the ith column.

5. The apparatus of claim 4, wherein the threshold module comprises:

the judging module is used for judging whether the fourth mutual information value is 1, if so, the reducing module is triggered, and if not, the increasing module is triggered;

the reduction module is used for reducing the set signal-to-noise ratio when the fourth mutual information value is 1, triggering the first module to obtain a new first mutual information value and the fourth module to obtain a new fourth mutual information value according to the reduced signal-to-noise ratio until the obtained fourth mutual information value is not 1, and the signal-to-noise ratio is a threshold value at the moment;

and the increasing module is used for increasing the set signal-to-noise ratio when the fourth mutual information value is not 1, triggering the first module to obtain a new first mutual information value and the fourth module to obtain a new fourth mutual information value according to the increased signal-to-noise ratio until the obtained fourth mutual information value is 1, and the signal-to-noise ratio is the threshold value at the moment.

Images