CN110311759B - A design method and communication system of magnetic induction communication system based on quasi-cyclic LDPC code - Google Patents

Abstract

Aiming at the technical defects of poor anti-interference ability of signal transmission in existing magnetic induction communication and serious error propagation in wireless communication, the present invention proposes a magnetic induction communication system design method and communication system based on quasi-cyclic LDPC codes. The present invention adopts discrete particles The group optimization algorithm and the PEXIT algorithm generate the basis matrix of the quasi-cyclic LDPC code and calculate the setting distance of the transmitting coil and the receiving coil in the magnetic induction communication system. The magnetic induction communication system designed based on this design method can effectively improve the magnetic induction communication system in the underground complex channel environment. performance.

Description

A design method and communication system of magnetic induction communication system based on quasi-cyclic LDPC code

technical field

The invention relates to the technical field of magnetic induction communication, in particular to a design method and communication system of a magnetic induction communication system based on a quasi-cyclic LDPC code.

Background technique

In the complex underground environment, magnetic induction communication technology has significant advantages compared with traditional electromagnetic wave communication. The reason is that electromagnetic wave communication is greatly attenuated under the influence of various underground media such as different types of rocks, soils with different water contents, etc., which seriously affects the quality of signal propagation, and is not suitable for signal transmission in complex underground environments; while magnetic induction Communication technology, whose signal transmission is little affected by the medium in the underground environment, has obvious advantages.

Magnetic induction communication uses the coupling between coils to transmit signals wirelessly. According to the deployment method of the coils, it can be divided into direct magnetic induction communication and magnetic induction waveguide communication. Channel coding plays an important role in wireless communication. By channel coding the transmission signal, the reliability of the wireless transmission of the signal can be effectively improved, and the bit error rate (BER: Bit error rate) of the wireless communication system can be reduced.

As an important member of channel coding, LDPC (Low Density Parity Check) codes have received more and more attention. Recently, LDPC codes have been selected as one of the channel coding schemes in the 5G mobile communication standard. Quasi-cyclic LDPC codes ( QC-LDPC code) uses cyclic shift matrix and zero matrix as sub-matrix, which can be easily realized by shift register. It is the most important LDPC code and has been used in IEEE 802.16e, IEEE 802.11n and DVB-S2. and other international standards. There are few studies involving channel coding in the existing magnetic induction communication research, and the application research of LDPC codes in magnetic induction communication has not been found yet. However, in the existing magnetic induction communication research, no channel coding optimization design research dedicated to magnetic induction communication has been found. This is because the traditional LDPC code encoding needs to be converted from the H matrix to the generator matrix G, and then encoded, and the coding complexity is relatively high. Large is a bottleneck for the application of LDPC codes.

SUMMARY OF THE INVENTION

Purpose of the invention: In order to make up for the blank of the prior art, and for the magnetic induction communication requirements in the complex underground environment, the present invention utilizes a quasi-cyclic LDPC code with a relatively simple code structure and a relatively small matrix storage space, which is very suitable for underground wireless sensor network applications. For the characteristics of low energy consumption requirements of sensing nodes, a magnetic induction communication system based on quasi-cyclic LDPC codes and its design method are proposed for the complex underground environment, which can effectively improve the performance of the magnetic induction communication system in the complex underground channel environment.

Technical scheme: In order to realize the above-mentioned technical effect, the present invention proposes the following technical scheme:

A method for designing a magnetic induction communication system based on a quasi-cyclic LDPC code, the method uses a discrete particle swarm optimization algorithm and a PEXIT algorithm to generate a basis matrix of the quasi-cyclic LDPC code and calculates the setting distance of the transmitting coil and the receiving coil in the magnetic induction communication system, including: step:

(1) Set the number of iterations to r, initialize r=1, initialize the local optimal value pbest ^r to a small value at this time; set the maximum number of iterations I _max and the number of particles P _num ;

每个二元矢量

对应基矩阵B的信息部分，即B(H_I)，B(H_I)大小为M×K，B(H_I)与大小为K×K的基矩阵的校验部分B(H_p)组合生成大小为M×N的二元基矩阵：B(H)＝[B(H_I)|B(H_P)]，其中，N＝M+K；B(H_p)如下式所示：(2) Randomly generate P _num binary vectors with dimensions M×K

every binary vector

Corresponding to the information part of the base matrix B, that is, B(H _I ), the size of B(H _I ) is M×K, and B(H _I ) is combined with the check part B(H _p ) of the base matrix of size K×K A binary basis matrix of size M×N is generated: B(H)=[B(H _I )|B(H _P )], where N=M+K; B(H _p ) is as follows:

Among them, the position of "1" in the first column and the last two columns is fixed, the weight of the first column is 3, the weight of the other columns is 2, and the upper right diagonal is "1", except for the first column and the last two columns. , the position of another "1" in each of the remaining columns is random, but the column weight must be 2;

共构建P_num个基矩阵，得到每个基矩阵后，再构造其对应的H矩阵，即有P_num个H矩阵；P _num binary vectors

A total of P _num base matrices are constructed, and after each base matrix is obtained, its corresponding H matrix is constructed, that is, there are P _num H matrices;

为第p个粒子对应的二元矢量

的第t个比特；Taking P _num binary vectors as P _num particles, initialize the particle swarm;

is the binary vector corresponding to the p-th particle

The t-th bit of ;

(3) For the H matrix of each quasi-cyclic LDPC code, calculate the distance between the transmitting and receiving coils in the magnetic induction communication system by the PEXIT algorithm, and the specific steps include:

(31) Initialize the distance d=d ₀ between the transmitting and receiving coils; initialize the likelihood information transmitted to v _j by each edge associated with the j-th variable point v _j and the i-th check point c _i and v _j The previous prior mutual information I _Av (i, j) is 0; the initial likelihood information for calculating v _j is:

in,

σ ^* = 1.6363

a _{J, 1} = -0.0421061, b _{J, 1} = 0.209252

c _{J, 1} = -0.00640081,

a _{J, 2} = 0.00181491, b _{J, 2} = -0.142675

c _{J, 2} = -0.0822054, d _{J, 2} = 0.0549608

R表示准循环LDPC码的码率，f(d)＝P_t-L_MI-P_n，P_t为发射功率，P_n为噪声功率，L_MI为磁感应通信的路径损耗，L_MI的表达式为：in,

R represents the code rate of the quasi-cyclic LDPC code, f(d)=P _t -L _MI -P _n , P _t is the transmit power, P _n is the noise power, L _MI is the path loss of the magnetic induction communication, the expression of L _MI for:

N _t and N _r are the number of turns of the transmitting coil and the receiving coil, respectively, and at and a _r are the _radii of the transmitting coil and the receiving coil, respectively;

(32) Update:

σ_Ev(i，j)的表达式为：Update the likelihood information passed by the variable point v _j to the check point c _i and the relevant mutual information I _Ev (i, j) between v _j :

The expression of σ _Ev (i, j) is:

I ^* =0.3646

a _{σ, 1} = 1.09542, b _{σ, 1} = 0.214217, c _{σ, 1} = 2.33727

a _{σ, 2} = 0.706692, b _{σ, 2} = 0.386013, c _{σ, 2} = -1.75017

Among them, b _{i, j} is the bipartite graph corresponding to the H matrix of the quasi-cyclic LDPC code, connecting the edge between the variable node v _j and the check node c _i , when the element at the position (i, j) in the H matrix is 1 , indicating that there is an edge connected between v _j and c _i , then its corresponding b _i,j =1, otherwise b _i,j =0,

Update the prior mutual information between the likelihood information of _{ci and v j that} each edge associated with the checkpoint ci and the variable point v _j transmits: I _Ac ( _i , j)=I _Ev (i, j) ;

σ_Ec(i，j)的表达式为：Update the likelihood information passed by the checkpoint _{ci to the variable point v j and} the external mutual information between v _j :

The expression of σ _Ec (i, j) is:

The likelihood information passed to v _j and the prior mutual information I _Av ( _i , _j ) before v _j : I _Av (i, j)=I _Ec (i,j)

Update the posterior likelihood information of variable point v _j and the posterior mutual information between variable point v _j :

in,

(33) Judge whether the iteration stopping criterion is satisfied:

若不满足，则更新d＝d+Δd，Δd为预设的增量步长，然后返回步骤(32)；If it is satisfied, end step (3), output d, and then use d as the distance threshold of the current round of the corresponding protograph

If not satisfied, update d=d+Δd, where Δd is the preset incremental step size, and then return to step (32);

从这P_num个距离阈值

中找出本轮迭代的全局最优值

(4) After step (3) is performed on the P _num H matrices obtained in step (2), P _num distance thresholds are obtained in total

from this P _num distance thresholds

Find the global optimal value of this iteration in

(5) Binary vector update:

Update particle velocity:

in,

为(0，1)之间均匀分布的随机数；Among them, λ, c ₁ , c ₂ , η ₁ , η ₂ are coefficients, where λ = 1, c ₁ =c ₂ =2, η ₁ , η ₂ are random uniformly distributed in the interval (0, 1). number;

is a random number uniformly distributed between (0, 1);

赋值给d；否则，根据步骤(5)更新后的粒子重新生成P_num个H矩阵，返回第(3)步。(6) Update the number of iterations r=r+1, and judge whether the number of iterations satisfies r>I _max ; if so, stop the iteration, construct the optimal QC-LDPC code matrix according to the binary vector corresponding to the current particle, and set the The obtained global optimum

Assign the value to d; otherwise, regenerate P _num H matrices according to the updated particles in step (5), and return to step (3).

The present invention also provides a magnetic induction communication system designed by using the design method, including:

The transmitting end and the receiving end; the transmitting end includes an encoder, a modulator and a transmitting coil; the receiving end includes a receiving coil, a demodulator and a decoder; the distance d between the transmitting coil and the receiving coil adopts the design method described in claim 1 get;

The transmitting end performs the following steps: the encoder utilizes the H matrix of the quasi-cyclic LDPC code constructed in claim 1 to encode the source data, and the encoded data is sent to the modulator for modulation, and then to the transmitting coil, and the transmitting coil generates an induced current;

The receiving end performs the following steps: exciting the receiving coil to generate an induced current, then sending the relevant data to the demodulator for demodulation, and then sending the data string to the decoder after demodulation, and the decoder obtains the reconstructed data after decoding.

Beneficial effect: Compared with the prior art, the present invention has the following advantages:

The magnetic induction communication system constructed by the invention effectively improves the communication performance of the system by adopting the quasi-cyclic LDPC code, and the quasi-cyclic LDPC code obtained by the optimized design can be directly encoded by the H matrix, with low coding complexity and low energy consumption, and is very suitable for Low energy consumption requirements for underground wireless sensor network nodes.

In addition, the proposed algorithm can also predict the maximum communication distance of the coil in direct magnetic induction communication, which provides a good guide for coil deployment in the design of magnetic induction communication system.

Description of drawings

Fig. 1 shows the flow chart of the design method of the present invention;

Fig. 2 is the framework diagram of the magnetic induction communication system designed by the design method of the present invention;

Fig. 3 is the base matrix schematic diagram of three kinds of A codes described in the embodiment;

Figure 4 is a schematic diagram of the BER performance comparison between A2 code and 802.16e code;

Fig. 5 is the H matrix schematic diagram of B1 code and B2 code;

Fig. 6 is the base matrix and H matrix schematic diagram of B3 code;

FIG. 7 is a schematic diagram showing the BER performance comparison of three B codes.

Detailed ways

The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

The present invention proposes a method for designing a magnetic induction communication system based on a quasi-cyclic LDPC code. The method adopts the discrete particle swarm optimization algorithm and the PEXIT algorithm to generate the basis matrix of the quasi-cyclic LDPC code and calculates the settings of the transmitting coil and the receiving coil in the magnetic induction communication system. Distance, including steps:

(1) Set the number of iterations to r, initialize r=1, initialize the local optimal value pbest ^r to a small value at this time; set the maximum number of iterations I _max and the number of particles P _num ;

每个二元矢量

对应基矩阵B的信息部分，即B(H_I)，B(H_I)大小为M×K，B(H_I)与大小为K×K的基矩阵的校验部分B(H_p)组合生成大小为M×N的二元基矩阵：B(H)＝[B(H_I)|B(H_p)]，其中，N＝M+K；B(H_p)如下式所示：(2) Randomly generate P _num binary vectors with dimensions M×K

every binary vector

Corresponding to the information part of the base matrix B, that is, B(H _I ), the size of B(H _I ) is M×K, and B(H _I ) is combined with the check part B(H _p ) of the base matrix of size K×K A binary basis matrix of size M×N is generated: B(H)=[B(H _I )|B(H _p )], where N=M+K; B(H _p ) is as follows:

Among them, the position of "1" in the first column and the last two columns is fixed, the weight of the first column is 3, the weight of the other columns is 2, and the upper right diagonal is "1", except for the first column and the last two columns. , the position of another "1" in each of the remaining columns is random, but the column weight must be 2;

共构建P_num个基矩阵，得到每个基矩阵后，再构造其对应的H矩阵，即有P_num个H矩阵；P _num binary vectors

A total of P _num base matrices are constructed, and after each base matrix is obtained, its corresponding H matrix is constructed, that is, there are P _num H matrices;

为第p个粒子对应的二元矢量

的第t个比特；Taking P _num binary vectors as P _num particles, initialize the particle swarm;

is the binary vector corresponding to the p-th particle

The t-th bit of ;

(3) For the H matrix of each quasi-cyclic LDPC code, calculate the distance between the transmitting and receiving coils in the magnetic induction communication system by the PEXIT algorithm, and the specific steps include:

(31) Initialize the distance d=d ₀ between the transmitting and receiving coils; initialize the likelihood information transmitted to v _j by each edge associated with the j-th variable point v _j and the i-th check point c _i and v _j The previous prior mutual information I _Av (i, j) is 0; the initial likelihood information for calculating v _j is:

in,

σ ^* = 1.6363

a _{J, 1} = -0.0421061, b _{J, 1} = 0.209252

c _{J, 1} = -0.00640081,

a _{J, 2} = 0.00181491, b _{J, 2} = -0.142675

c _{J, 2} = -0.0822054, d _{J, 2} = 0.0549608

R表示准循环LDPC码的码率，f(d)＝P_t-L_MI-P_n，P_t为发射功率，P_n为噪声功率，L_MI为磁感应通信的路径损耗，L_MI的表达式为：in,

R represents the code rate of the quasi-cyclic LDPC code, f(d)=P _t -L _MI -P _n , P _t is the transmit power, P _n is the noise power, L _MI is the path loss of the magnetic induction communication, the expression of L _MI for:

N _t and N _r are the number of turns of the transmitting coil and the receiving coil, respectively, and at and a _r are the _radii of the transmitting coil and the receiving coil, respectively;

(32) Update:

σ_Ev(i，j)的表达式为：Update the likelihood information passed by the variable point v _j to the check point c _i and the relevant mutual information I _Ev (i, j) between v _j :

The expression of σ _Ev (i, j) is:

I ^* =0.3646

a _{σ, 1} = 1.09542, b _{σ, 1} = 0.214217, c _{σ, 1} = 2.33727

a _{σ, 2} = 0.706692, b _{σ, 2} = 0.386013, c _{σ, 2} = -1.75017

Among them, b _{i, j} is the bipartite graph corresponding to the H matrix of the quasi-cyclic LDPC code, connecting the edge between the variable node v _j and the check node c _i , when the element at the position (i, j) in the H matrix is 1 , indicating that there is an edge connected between v _j and c _i , then its corresponding b _i,j =1, otherwise b _i,j =0,

Update the prior mutual information between the likelihood information of _{ci and v j that} each edge associated with the checkpoint ci and the variable point v _j transmits: I _Ac ( _i , j)=I _Ev (i, j) ;

σ_Ec(i，j)的表达式为：Update the likelihood information passed by the checkpoint _{ci to the variable point v j and} the external mutual information between v _j :

The expression of σ _Ec (i, j) is:

The likelihood information passed to v _j and the prior mutual information I _Av ( _i , _j ) before v _j : I _Av (i, j)=I _Ec (i,j)

Update the posterior likelihood information of variable point v _j and the posterior mutual information between variable point v _j :

in,

(33) Judge whether the iteration stopping criterion is satisfied:

若不满足，则更新d＝d+Δd，Δd为预设的增量步长，然后返回步骤(32)；If it is satisfied, end step (3), output d, and then use d as the distance threshold of the current round of the corresponding protograph

If not satisfied, update d=d+Δd, where Δd is the preset incremental step size, and then return to step (32);

从这P_num个距离阈值

中找出本轮迭代的全局最优值

(4) After step (3) is performed on the P _num H matrices obtained in step (2), P _num distance thresholds are obtained in total

from this P _num distance thresholds

Find the global optimal value of this iteration in

(5) Binary vector update:

Update particle velocity:

in,

为(0，1)之间均匀分布的随机数；Among them, λ, c ₁ , c ₂ , η ₁ , η ₂ are coefficients, where λ = 1, c ₁ =c ₂ =2, η ₁ , η ₂ are random uniformly distributed in the interval (0, 1). number;

is a random number uniformly distributed between (0, 1);

赋值给d；否则，根据步骤(5)更新后的粒子重新生成P_num个H矩阵，返回第(3)步。(6) Update the number of iterations r=r+1, and judge whether the number of iterations satisfies r>I _max ; if so, stop the iteration, construct the optimal QC-LDPC code matrix according to the binary vector corresponding to the current particle, and set the The obtained global optimum

Assign the value to d; otherwise, regenerate P _num H matrices according to the updated particles in step (5), and return to step (3).

The present invention also provides a magnetic induction communication system designed by using the design method, including:

a transmitting end and a receiving end; the transmitting end includes an encoder, a modulator and a transmitting coil; the receiving end includes a receiving coil, a demodulator and a decoder; the distance d between the transmitting coil and the receiving coil is obtained by the design method;

The transmitting end performs the following steps: the encoder uses the H matrix of the quasi-cyclic LDPC code constructed by the design method to encode the source data, and the encoded data is sent to the modulator for modulation, and then to the transmitting coil, and the transmitting coil generates an induced current;

The receiving end performs the following steps: exciting the receiving coil to generate an induced current, then sending the relevant data to the demodulator for demodulation, and then sending the data string to the decoder after demodulation, and the decoder obtains the reconstructed data after decoding.

Different from the traditional electromagnetic wave communication, the coded and modulated data of the system is sent to the receiving coil by means of coil coupling, and then demodulated and decoded. The distance between the coils, the related parameter settings of the coils, and the transmit power all have an important impact on the signal transmission.

Different from the existing magnetic induction communication system, the present invention considers the quasi-cyclic LDPC code as a channel coding scheme, and encodes it before signal transmission. In other studies, there is no magnetic induction communication system that adopts the quasi-cyclic LDPC code. plan.

The technical effects of the present invention are further described below through specific embodiments.

Embodiment: This embodiment adopts the described design method to construct 2 groups of quasi-cyclic LDPC codes (A code and B code), a total of 6 codes, Table 1 and Table 3 respectively provide the relevant specific parameters of the A code and the B code. , Table 2 and Table 4 show the maximum transmission distance threshold between the system coils corresponding to the quasi-cyclic LDPC code constructed by the method proposed by the present invention, that is, the maximum distance between the transmitting coil and the receiving coil, the base matrix of the code and the diagram of the H matrix shown in Figure 3, Figure 5 and Figure 6.

Table 1 Related parameters of A code

Table 2 The maximum transmission distance threshold of A code

name A1 size A2 size A3 size Maximum transmission distance (meters) 8.4 8.42 8.37

Table 3 Related parameters of B code

Table 4 Maximum transmission distance threshold of B code

name Code B1 B2 code B3 code Maximum transmission distance (meters) 17.16 18.48 36.91

Figure 4 shows the BER performance obtained by the simulation of the A2 code and the quasi-cyclic LDPC code in the 802.16 standard (parameters such as code rate and code length are the same as the A2 code) in the magnetic induction communication system. As can be seen from the figure, the quasi-cyclic LDPC code constructed by the method proposed in the present invention has a transmission distance of about 4 meters higher than that of the 802.16e code when the BER is 10^(-4).

Figure 7 shows the B1, B2 and B3 codes. It can be seen from the figure that the transmission distance of the three quasi-cyclic LDPC codes in the magnetic induction communication system has been greatly improved compared with that of the A code. greater impact.

It can also be seen from Fig. 4 and Fig. 7 that the maximum transmission distance threshold predicted by the method of the present invention in Table 2 and Table 4 is in good agreement with the maximum transmission distance obtained by BER simulation.

The above drawings and data illustrate that the present invention can predict the maximum transmission distance based on quasi-cyclic LDPC codes on the one hand, and can optimize the construction of quasi-cyclic LDPC codes suitable for magnetic induction communication systems on the other hand, which is very suitable for the design of underground wireless sensor networks. Options.

The above is only the preferred embodiment of the present invention, it should be pointed out that: for those skilled in the art, without departing from the principle of the present invention, several improvements and modifications can also be made, and these improvements and modifications are also It should be regarded as the protection scope of the present invention.

Claims (2)

1. A magnetic induction communication system design method based on quasi-cyclic LDPC codes is characterized in that the method adopts a discrete particle swarm optimization algorithm and a PEXIT algorithm to generate a base matrix of the quasi-cyclic LDPC codes and calculate the setting distance of a transmitting coil and a receiving coil in the magnetic induction communication system, and comprises the following steps:

(1) setting the iteration number as r, initializing r to 1, and initializing the local optimal value pbest at the moment^rIs a constant value; setting the maximum number of iterations I_maxNumber of sum particles P_num；

(2) Randomly generating P_numBinary vector with dimension of M multiplied by K

Each binary vector

Corresponding to the information part of the base matrix B, i.e. B (H)_I)，B(H_I) Size of M × K, B (H)_I) Check portion B (H) with a base matrix of size K_p) Combining to generate a binary basis matrix of size mxn: b (H) ═ B (H)_I)|B(H_P)]Wherein N ═ M + K; b (H)_p) As shown in the following formula:

the positions of '1' of the first column and the last column are fixed, the column weight of the first column is 3, the column weights of the other columns are 2, the diagonal line on the upper right is '1', the position of another '1' of each column except the first column and the last column is random, but the column weight is ensured to be 2;

P_numa binary vector

Co-construction of P_numObtaining each base matrix, and then constructing the corresponding H matrix, namely P_numAn H matrix;

with P_numA binary vector of P_numInitializing a particle swarm; note the book

Binary vector corresponding to p-th particle

The t-th bit of (1);

(3) for each H matrix of the quasi-cyclic LDPC codes, calculating the distance between transmitting and receiving coils in the magnetic induction communication system by a PEXIT algorithm, and the specific steps comprise:

(31) initializing the distance d between the transmitting and receiving coils₀(ii) a Initializing the jth variable point v_jAnd the ith check point c_iEach edge of the association is passed to v_jLikelihood information of v_jA priori mutual information I_Av(i, j) is 0; calculating v_jThe initial likelihood information of (a) is:

wherein,

σ^*＝1.6363

a_J，1＝-0.0421061，b_J，1＝0.209252

c_J，1＝-0.00640081，

a_J，2＝0.00181491，b_J，2＝-0.142675

c_J，2＝-0.0822054，d_J，2＝0.0549608

wherein,

r represents the code rate of the quasi-cyclic LDPC code, and f (d) is P_t-L_MI-P_n，P_tTo transmit power, P_nIs the noise power, L_MIPath loss for magnetic induction communication, L_MIThe expression of (a) is:

N_t、N_rnumber of turns of the transmitting coil and receiving coil, respectively, a_t、a_rThe radii of the transmitting coil and the receiving coil respectively;

(32) updating:

updating variable point v_jIs transmitted to a check point c_iLikelihood information of v_jRelated mutual information I between_Ev(i，j)：

σ_EvThe expression of (i, j) is:

I^*＝0.3646

a_σ，1＝1.09542，b_σ，1＝0.214217，c_σ，1＝2.33727

a_σ，2＝0.706692，b_σ，2＝0.386013，c_σ，2＝-1.75017

wherein b is_i，jIn bipartite graph corresponding to H matrix of quasi-cyclic LDPC code, variable nodes v are connected_jAnd check node c_iThe edge in between, when the element with (i, j) in the H matrix is 1, indicates v_jAnd c_iAre connected with each other by edges, then the corresponding b_i，j1, otherwise b_i，j＝0，

Updating check point c_iAnd variable point v_jEach associated edge is passed to c_iLikelihood information of v_jPrior mutual information between: i is_Ac(i，j)＝I_Ev(i，j)；

Updating check point c_iTo a variable point v_jLikelihood information of v_jExternal mutual information between:

σ_Ecthe expression of (i, j) is:

updating variable point v_jAnd check point c_iEach edge of the association is passed to v_jLikelihood information of v_jA priori mutual information I_Av(i，j)：I_Av(i，j)＝I_Ec(i，j)

Updating variable point v_jA posteriori likelihood information and variable points v_jPosterior mutual information between:

wherein,

(33) judging whether an iteration stop criterion is met:

if yes, ending the step (3), outputting d, and then taking d as the distance threshold value p of the current turn of the corresponding original pattern diagram^r _best(ii) a If not, updating d to d + Δ d, and Δ d to be a preset increment step length, and then returning to the step (32);

(4) for P obtained in step (2)_numAfter the H matrixes respectively execute the step (3), P is obtained in total_numA distance threshold value

From this P_numA distance threshold value

To find out the global optimum value of the iteration of the current round

(5) And (3) updating a binary vector:

updating the particle speed:

wherein,

wherein, λ, c₁、c₂、η₁、η₂Is a coefficient, where λ ═ 1, c₁＝c₂＝2，η₁、η₂Are random numbers uniformly distributed in the interval (0, 1);

random numbers uniformly distributed among (0, 1);

(6) updating the iteration number r to r +1, and judging whether the iteration number satisfies r > I_max(ii) a If so, stopping iteration, constructing an optimal QC-LDPC code matrix according to the binary vector corresponding to the current particle, and obtaining an overall optimal value

Assigning a value to d; otherwise, regenerating P according to the updated particles in the step (5)_numAnd (4) returning to the step (3) by the H matrix.

2. A magnetic induction communication system designed using the design method of claim 1, comprising:

a sending end and a receiving end; the transmitting end comprises an encoder, a modulator and a transmitting coil; the receiving end comprises a receiving coil, a demodulator and a decoder; the distance d between the transmitting coil and the receiving coil is obtained by the design method of claim 1;

the sending end executes the following steps: an encoder encodes source data by using an H matrix of a quasi-cyclic LDPC code constructed in claim 1, the encoded data is sent to a modulator for modulation and then sent to a transmitting coil, and the transmitting coil generates induced current;

the receiving end executes the following steps: exciting the receiving coil to generate induced current, sending the related data to a demodulator for demodulation, sending the data string to a decoder after demodulation, and obtaining the reconstructed data after the decoder decodes the data.

Images