CN104065455A - Method for MPPM constellation selection based on average symbol error rates in X-ray communication - Google Patents

Abstract

The invention discloses a method for MPPM constellation selection based on average symbol error rates in X-ray communication. The method includes the following steps of establishing an average symbol error rate model based on constellation selection, calculating the average symbol error rates under different MPPM constellation combination conditions for different MPPM constellations according to the obtained average symbol error rate model based on constellation selection, converting the constellation selection problem into the conditional constraint optimization problem, and obtaining the constellation selection scheme based on the minimum average symbol error rate condition by selecting the search strategy according to the average symbol error rate model based on constellation selection and according to the conditional constraint optimization problem. By means of the method, on one hand, the constellation selection problem is converted into the optimization problem and simplified, calculation amount is small and speed is high when search is conducted through the search strategy, and the method is suitable for engineering realization; on the other hand, constellation selection is conducted with the average symbol error rates as the standard, channel transmission accuracy is easily improved, and channel transmission errors are reduced.

Description

A kind of MPPM constellation system of selection based on average error sign ratio in X ray communication

Technical field

The present invention relates to communication technical field, be specifically related to a kind of MPPM constellation system of selection based on average error sign ratio in X ray communication.

Background technology

X ray pulsar navigation (X-ray Pulsar-based Navigation, XPNAV) is to utilize the X ray signal of pulsar periodic transmission so that the new airmanship of navigation Service to be provided.It can provide navigation Service for near-earth orbit task and deep space aerial mission, and this will be the important air navigation aid that following deep space is explored.Along with the requirement of increasing exploration activity and transfer of data, the progress of the communication technology is further urgent.In recent years, free space optical communication has demonstrated it and has been better than traditional communication mode.X ray communication (X-raycommunications, XCOM), as the new mode of optical communication, demonstrates very large potentiality in deep space communication field.The most frequently used free-space communication method is radio frequency (RadioFrequency, RF) communication and microwave communication now.The shortcoming of these methods is that they all need large-scale antenna and have compared with low transmission rate.Compared with traditional communication mode, X ray communication has higher transmission rate, and quality is lighter, and lower power consumption, can meet the communication requirement that deep space aerial mission increases day by day.

Because XPNAV and XCOM utilize X ray to carry out their function as medium, there is a kind of possibility, XPNAV and XCOM are integrated into an independent system, and this makes system less, lighter, and more energy-conservation.In the application of XCOM and XPNAV, still have a lot of key technical problems that need solution.For example, need to design XCOM signal, avoid and the mutual interference of navigation signal phase.The research signal processing algorithm to navigation and signal of communication processing simultaneously.For XCOM, the performance how research appropriately selects modulator approach to improve system is extremely necessary.

As the conventional modulator approach of XCOM, multiple-pulse pulse position modulation (Multi-pulsePulse Position Modulation, MPPM) conveys a message by the position of MPPM pulse.MPPM can be regarded as the mapping between binary sequence and the MPPM code word of input.In practice, the size of mapping can not be mated with MPPM constellation size, and so, the applicable constellation of How to choose shines upon and will be a problem to be solved.For the selection of MPPM constellation, also there is different standards.Criterion is during from two code words of MPPM constellation, to minimize continuous zero maximum possible number, the cascade of this applicable iteration Soft decision decoding and outer error correcting code at series winding.Along with the exploitation of the approaching code of modern capacity, the channel capacity of selecting based on constellation also can be used for selecting the subset of MPPM constellation.But above-mentioned two kinds of algorithm complexity, amount of calculation is large, inapplicable engineering application.

Summary of the invention

For the deficiencies in the prior art, the present invention aims to provide a kind of MPPM constellation system of selection based on average error sign ratio in X ray communication, the present invention derives the expression of the average error sign ratio of selecting based on constellation by constraint, problem is converted into the constrained optimization problem in mathematics, has also proposed discrete search strategy simultaneously and obtained the MPPM constellation selection scheme under minimum average B configuration error sign ratio condition.

To achieve these goals, the present invention adopts following technical scheme:

A kind of MPPM constellation system of selection based on average error sign ratio in X ray communication, is characterized in that, described method comprises the steps:

Step 1, the average error sign ratio model that structure is selected based on constellation;

Step 2, the average error sign ratio model of selecting based on constellation drawing according to described step 1, for different MPPM constellations, calculates the average error sign ratio under different MPPM constellation combination conditions, and constellation selection problem is converted into constraint optimization problem;

Step 3, the described average error sign ratio model of selecting based on constellation drawing according to step 1, and the described constraint optimization problem that draws of step 2, by selecting search strategy, obtain the MPPM constellation selection scheme based under minimum average B configuration error sign ratio condition.

It should be noted that, described step 1 specifically comprises the steps:

Step 1.1, is n for number of time slots, and the MPPM signal frame that pulse number is w, can be designated as (n, w) MPPM, wherein in w described time slot, comprises inclusion information, and this w described time slot is called signal slot; By the various combination of signal slot position, one is total plant different MPPM symbols, and the set of these different described MPPM symbols forms MPPM constellation, can be expressed as X (x ₁, x ₂..., x _m), wherein x _ibe a MPPM symbol of described MPPM constellation, can be expressed as

x _i＝(x _i1,x _i2,…,x _in)；

Wherein x _inthe value that represents a signal slot, value is 0 or 1; x _iduty ratio be w/n;

Step 1.2, uses P _irepresent x _iprobability tables be shown, i.e. P _i=P (x _i), vector P=(P ₁, P ₂..., P _m) represent X (x ₁, x ₂..., x _m) in the selecteed probability of each element; For completing the MPPM modulation that length is the binary sequence of L, need K=2 ^lindividual MPPM symbol, therefore need to be from X (x ₁, x ₂..., x _m) select K MPPM symbol, i.e. P=(P with equiprobability ₁, P ₂..., P _m) in to have K element value be 1/K, all the other element values are 0; Therefore have plant MPPM symbol selection mode, the MPPM constellation combination that Z kind is different;

Step 1.3, makes λ _sin the representation unit time, arrive the par of the signal photon of each time slot, λ _bin the representation unit time, arrive the par of the noise photon of each time slot, (ξ _i1, ξ _i2..., ξ _iw) for working as x _ix during by transmission _ithe position of a middle w signal slot, (c _i1, c _i2..., c _in) be x _ithe signal photon that during by channel, maximum likelihood receiver device detects in each time slot and the total quantity of noise photon; Definition N _ifor x _iphoton number summation in the w signal slot that during by channel, maximum likelihood detector detects, N _ican be expressed as

$N_i=\underset{k=1}{\overset{w}{\mathrm{\Σ}}}{c}_{{i,\mathrm{\ξ}}_{\mathrm{ik}}};$

Step 1.4, under noisy condition, channel is input as x _i, the upper limit of error sign ratio is defined as at transmission x _iunder condition, obtain x at channel output _j(j ∈ N ⁺, j≤M, j ≠ i), and x _jsignal slot in the number of photons that observes be more than or equal to x _ithe conditional probability density of number of photons of signal slot

$\begin{array}{c}{P}_{\mathrm{SER}}\left(x_i\right)\≤P\left\{\underset{j\≠i}{\overset{M}{\∪}}(N_i\≤N_j,x_j|x_i)\right\}\\ \≤\underset{j\≠i}{\overset{M}{\mathrm{\Σ}}}P\left(x_j\right)P(N_i\≤N_j|x_i)\end{array};$

Wherein P (N _i≤ N _j| x _i) be transmission x _itime symbol error be received as x _jprobability, be to define by the Hamming distance of two vectors, d _h(x _i, x _j) expression x _iand x _jbetween Hamming distance, λ is to be signal strength signal intensity λ by the par of the signal photon that arrives each time slot in the unit interval _swith the par of the noise photon that arrives each time slot in the unit interval be noise intensity λ _bderivation draws, can be expressed as ${\mathrm{\λ}}^2={(\sqrt{{\mathrm{\λ}}_s+{\mathrm{\λ}}_b}-\sqrt{{\mathrm{\λ}}_b})}^2;$

Step 1.5, the upper limit of the error sign ratio defining according to step 1.4, while passing through transmission after symbol in MPPM constellation is selected with different probability, each symbol correspondence has error sign ratio, and the probability that this error sign ratio occurs is based on the selecteed probability of this symbol, the average error sign ratio model of selecting based on constellation described in can defining is thus:

$\begin{array}{c}{P}_{\mathrm{SER}\_\mathrm{avr}}=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}P\left(x_i\right)\underset{j\≠i}{\overset{M}{\mathrm{\Σ}}}P\left(x_j\right)P(N_i\≤N_j|x_i)\\ =\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{P}_i\underset{j\≠i}{\overset{M}{\mathrm{\Σ}}}{P}_{j}P(N_i\≤N_j|x_i)\end{array}.$

It should be noted that, the concrete steps of described step 2 are:

Step 2.1, the average error sign ratio model that utilizes the system of selection of the MPPM constellation combination drawing in step 1 and select based on constellation, for different MPPM constellation combination, calculates average error sign ratio;

Step 2.2, utilizes the average error sign ratio of the different MPPM constellation combination that step 2.1 draws, constellation selection problem is converted into constraint optimal problem,

$P_{\mathrm{SER}\_\min }\left(P\right)=\underset{P}{\arg \max }{P}_{\mathrm{SER}\_\mathrm{avr}}$

$s.t.\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{P}_i=1;$

P _i∈{0,1/K}

Γ(P)＝K

Wherein, P=(P (x ₁), P (x ₂) ..., P (x _m)) represent that the selecteed probability of symbol of MPPM constellation, Γ () are illustrated in the number that certain vector median is 0 element, K=2 ^l, L is the length that needs the binary sequence of transmission.

It should be noted that, in described step 3, both can adopt direct stochastic search methods, also adopt traversal search method.

In the time adopting direct stochastic search methods, described step 3 specifically comprises the steps:

Step 3.1, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ ₀(ξ _0,1, ξ _0,2..., ξ _{0, K}); If the starting point ξ of search procedure ₀probability tables be shown P ₀(ξ _{0, i})=1/K, i=1,2 ..., K, calculates the average error sign ratio P of initial constraint _{sER_min}(P ₀), making total searching times is N _total, iterations k=1;

Step 3.2, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ (ξ ₁, ξ ₂..., ξ _k), and make P (ξ _i)=1/K, i=1,2 ..., K, calculates the average error sign ratio P of constraint _{sER_min}(P);

Step 3.3, judges P _{sER_min}(P) <P _{sER_min}(P ₀) whether set up, if set up P ₀=P, P _{sER_min}(P ₀)=P _{sER_min}(P); Otherwise P ₀remain unchanged;

Step 3.4, makes k=k+1, if k>N _total, stop search procedure and obtain the MPPM constellation selection scheme P based on minimum average B configuration error sign ratio ^*=P ₀, and the corresponding average error sign ratio of least commitment is P _{sER_opt}=P _{sER_min}(P ₀), otherwise, return to step 3.2.

In the time adopting traversal search method, the concrete implementation step of described step 3 is as follows:

Step 3.1, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ ₀(ξ _0,1, ξ _0,2..., ξ _{0, K}); If the starting point ξ of search procedure ₀probability tables be shown P ₀(ξ _{0, i})=1/K, i=1,2 ..., K, utilizes the described constraint optimal problem that step 2 draws to calculate the average error sign ratio P of initial constraint _{sER_min}(P ₀); Make iterations k=1;

Step 3.2, according to the described average error sign ratio model of selecting based on constellation, P _ivalue 0 or 1/K, for convenience of description, be referred to as two values the other side's " opposition value "; For the P in step 3.1 ₀a certain element P _k, its current value is changed into its opposition value, obtain P ' _k, and P ₀in other values remain unchanged, thereby make P ₀change into P ' ₀;

Step 3.3, for P ' ₀in except P ' _kelement in addition, finds its value to equal P _kall elements, make at P ' ₀one group of sequence of middle composition

Step 3.4, for P ' ₀, change and belong to P _cin the value of one of them element be its opposition value P ' ₀in other elements remain unchanged, obtain one group of new sequence, wherein two elements and P ₀in difference, calculate the average error sign ratio of this constraint of organizing new sequence;

Step 3.5, for P ' ₀, change and belong to P _cin one of them element, and this element be different from before any element once, and P ' ₀in other elements remain unchanged, again obtain one group of new sequence and calculate the average error sign ratio of constraint of the sequence that this group is new; Repeating said steps 3.5t-1 time;

Step 3.6, according to step 3.4 and step 3.5, obtains t the average error sign ratio of constraint, by relatively drawing the average error sign ratio of minimum constraint, is designated as P _{sER_min}(P' _min), corresponding sequence is designated as P' _min;

Step 3.7, judges P _{sER_min}(P' _min) <P _{sER_min}(P ₀) whether set up: if be false, make k=k+1, jump to step 3.2; If set up, make P _{sER_min}(P ₀)=P _{sER_min}(P' _min), k=k+1, and jump to step 3.8;

Step 3.8, judges whether k>M sets up, if set up, stops search procedure, obtains the MPPM constellation selection scheme P based under minimum average B configuration error sign ratio ^*=P ₀with the corresponding average error sign ratio P of least commitment _{sER_min}(P ^*); If be false, jump to step 3.2.

Beneficial effect of the present invention is: one aspect of the present invention selects to be converted to optimization problem by constellation, simplify problem, utilize search strategy search, amount of calculation is little, speed is fast, is applicable to Project Realization, carries out on the other hand constellation selection taking average error sign ratio as standard, be conducive to improve transmission accuracy, reduce transmission mistake.

Brief description of the drawings

Fig. 1 is implementing procedure schematic diagram of the present invention.

Embodiment

Below with reference to accompanying drawing, the invention will be further described, it should be noted that, the present embodiment, taking the technical program as prerequisite, has provided detailed implementation step, but and be limited to the present embodiment.

Shown in Fig. 1, a kind of MPPM constellation system of selection based on average error sign ratio in X ray communication, is characterized in that, described method comprises the steps:

Step 1, the average error sign ratio model that structure is selected based on constellation, concrete steps are as follows;

Step 1.1, for (n, w) MPPM, has n time slot, wherein has inclusion information in w described time slot, and this w described time slot is called signal slot; , by the various combination of signal slot position, one is total plant different MPPM symbols; And the set of these different described MPPM symbols forms MPPM constellation, can be expressed as X (x ₁, x ₂..., x _m), wherein x _ibe a MPPM symbol from MPPM constellation, it can be expressed as

x _i＝(x _i1,x _i2,…,x _in)；

Wherein x _inthe value that represents a signal slot, value is 0 or 1; x _iduty ratio be w/n;

Step 1.2, uses P _irepresent to select x _iprobability, i.e. P _i=P (x _i), vector P=(P ₁, P ₂..., P _m) represent X (x ₁, x ₂..., x _m) in the selecteed probability of each element; For completing the MPPM modulation that length is the binary sequence of L, need K=2 ^lindividual MPPM symbol, therefore need to be from X (x ₁, x ₂..., x _m) select K MPPM symbol, i.e. P=(P with equiprobability ₁, P ₂..., P _m) in to have K element value be 1/K, all the other element values are 0; Therefore have plant MPPM symbol selection mode, the MPPM constellation combination that Z kind is different.

Step 1.3, makes λ _sin the representation unit time, arrive the par of the signal photon of each time slot, λ _bin the representation unit time, arrive the par of the noise photon of each time slot, (ξ _i1, ξ _i2..., ξ _iw) for working as x _ix during by transmission _ithe position of a middle w signal slot, (c _i1, c _i2..., c _in) be x after detecting by a maximum likelihood receiver device _ithe signal photon that time slot detects in each time slot and the total quantity of noise, definition N _ifor x _iphoton number summation in the w signal slot that during by channel, maximum likelihood detector detects, N so _ican be expressed as

$N_i=\underset{k=1}{\overset{w}{\mathrm{\&Sigma;}}}{c}_{{i,\mathrm{\&xi;}}_{\mathrm{ik}}};$

Step 1.4, channel is input as x _i, the upper limit of error sign ratio is defined as at transmission x _iunder condition, obtain x at channel output _j(j ∈ N ⁺, j≤M, j ≠ i), and x _jsignal slot in the number of photons that observes be more than or equal to x _ithe conditional probability density of number of photons of signal slot

$\begin{array}{c}{P}_{\mathrm{SER}}\left(x_i\right)\&le;P\left\{\underset{j\&NotEqual;i}{\overset{M}{\&cup;}}(N_i\&le;N_j,x_j|x_i)\right\}\\ \&le;\underset{j\&NotEqual;i}{\overset{M}{\mathrm{\&Sigma;}}}P\left(x_j\right)P(N_i\&le;N_j|x_i)\end{array};$

Wherein P (N _i≤ N _j| x _i) be transmission x _itime symbol error be received as x _jprobability, be by the Hamming distance of two vectors $P(N_i\&le;N_i|x_i)\&le;\frac{1}{2}{e}^{[-\frac{1}{2}{\mathrm{\&lambda;}}^2{d}_H\left(x_i,x_j\right)]}$ Define d _h(x _i, x _j) expression x _iand x _jbetween Hamming distance, λ is by signal strength signal intensity λ _swith noise intensity λ _bderivation draws, can be expressed as ${\mathrm{\&lambda;}}^2={(\sqrt{{\mathrm{\&lambda;}}_s+{\mathrm{\&lambda;}}_b}-\sqrt{{\mathrm{\&lambda;}}_b})}^2;$

Step 1.5, the upper limit of the error sign ratio defining according to step 1.4, while passing through transmission after symbol in MPPM constellation is selected with different probability, each symbol correspondence has error sign ratio, and the probability that this error sign ratio occurs is based on the selecteed probability of this symbol, the average error sign ratio model of selecting based on constellation described in can defining is thus:

$\begin{array}{c}{P}_{\mathrm{SER}\_\mathrm{avr}}=\underset{i=1}{\overset{M}{\mathrm{\&Sigma;}}}P\left(x_i\right)\underset{j\&NotEqual;i}{\overset{M}{\mathrm{\&Sigma;}}}P\left(x_j\right)P(N_i\&le;N_j|x_i)\\ =\underset{i=1}{\overset{M}{\mathrm{\&Sigma;}}}{P}_i\underset{j\&NotEqual;i}{\overset{M}{\mathrm{\&Sigma;}}}{P}_{j}P(N_i\&le;N_j|x_i)\end{array};$

Step 2, the average error sign ratio model that the constellation drawing according to described step 1 is selected, for different MPPM constellations, calculates the average error sign ratio under different MPPM constellation combination conditions, and constellation selection problem is converted into constraint optimization problem, concrete steps are:

Step 2.1, the average error sign ratio model that utilizes the system of selection of the MPPM constellation combination drawing in step 1 and select based on constellation, for different MPPM constellation combination, calculates average error sign ratio;

Step 2.2, utilizes the average error sign ratio of the different MPPM constellation combination that step 2.1 draws, constellation selection problem is converted into constraint optimal problem,

$P_{\mathrm{SER}\_\min }\left(P\right)=\underset{P}{\arg \max }{P}_{\mathrm{SER}\_\mathrm{avr}}$

$s.t.\underset{i=1}{\overset{M}{\mathrm{\&Sigma;}}}{P}_i=1;$

P _i∈{0,1/K}

Γ(P)＝K

Wherein, P=(P (x ₁), P (x ₂) ..., P (x _m)) represent to select the probability of symbol in MPPM constellation, Γ () to be illustrated in the number that certain vector median is 0 element, K=2 ^l, the length that L is list entries.

Step 3, the described average error sign ratio model of selecting based on constellation drawing according to step 1, and the described constraint optimization problem that draws of step 2, by selecting search strategy, obtain the MPPM constellation selection scheme based under minimum average B configuration error sign ratio condition.

Can be direct stochastic search methods for one of the search strategy adopting, the feasible set that this method is randomly drawed be as selecting candidate value, and whether check result is favourable to search procedure, if so, and using this value as new starting point and repeat search procedure.Otherwise, abandon this value, still adopt present worth as starting point.Because direct search method is selected candidate value at random, therefore need to arrange under larger iterations and can ensure to contain whole Candidate Set with higher probability.

While adopting direct stochastic search methods, described step 3 may be summarized to be following step:

Step 3.1, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ ₀(ξ _0,1, ξ _0,2..., ξ _{0, K}); If the starting point ξ of search procedure ₀probability tables be shown P ₀(ξ _{0, i})=1/K, i=1,2 ..., K, calculates the average error sign ratio P of initial constraint _{sER_min}(P ₀), making total searching times is N _total, iterations k=1;

Step 3.2, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ (ξ ₁, ξ ₂..., ξ _k), and make P (ξ _i)=1/K, i=1,2 ..., K, calculates the average error sign ratio P selecting based on constellation _{sER_min}(P);

Step 3.3, judges P _{sER_min}(P) <P _{sER_min}(P ₀) whether set up, if set up P ₀=P, P _{sER_min}(P ₀)=P _{sER_min}(P); Otherwise P ₀remain unchanged;

Step 3.4, makes k=k+1, if k>N _total, stop search procedure and obtain the MPPM constellation selection scheme P based on minimum average B configuration error sign ratio ^*=P ₀, and the corresponding average error sign ratio of least commitment is P _{sER_opt}=P _{sER_min}(P ₀), otherwise, return to step 3.2.

Another can be traversal search method for the search strategy adopting, and in the time adopting described traversal search method, described step 3 specifically comprises the steps:

Step 3.1, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ ₀(ξ _0,1, ξ _0,2..., ξ _{0, K}); If the starting point ξ of search procedure ₀probability tables be shown P ₀(ξ _{0, i})=1/K, i=1,2 ..., K, utilizes the described constraint optimal problem that step 2 draws to calculate the average error sign ratio P of initial constraint _{sER_min}(P ₀); Make iterations k=1;

Step 3.2, according to the described average error sign ratio model of selecting based on constellation, P _ivalue 0 or 1/K, for convenience of description, be referred to as two values the other side's " opposition value "; For the P in step 3.1 ₀a certain element P _k, its current value is changed into its opposition value, obtain P ' _k, and P ₀in other values remain unchanged, thereby make P ₀change into P ' ₀;

Step 3.3, for P ' ₀in except P ' _kelement in addition, finds its value to equal P _kall elements, make at P ' ₀one group of sequence of middle composition

Step 3.4, for P ' ₀, change and belong to P _cin the value of one of them element be its opposition value P ' ₀in other elements remain unchanged, obtain one group of new sequence, wherein two elements and P ₀in difference, calculate the average error sign ratio of this constraint of organizing new sequence;

Step 3.5, for P ' ₀, change and belong to P _cin one of them element, and this element be different from before any element once, and P ' ₀in other elements remain unchanged, again obtain one group of new sequence and calculate the average error sign ratio of constraint of the sequence that this group is new; Repeating said steps 3.5t-1 time;

Step 3.6, according to step 3.4 and step 3.5, obtains t the average error sign ratio of constraint, by relatively drawing the average error sign ratio of minimum constraint, is designated as P _{sER_min}(P' _min), corresponding sequence is designated as P' _min;

Step 3.7, judges P _{sER_min}(P' _min) <P _{sER_min}(P ₀) whether set up: if be false, make k=k+1, jump to step 3.2; If set up, make P _{sER_min}(P ₀)=P _{sER_min}(P' _min), k=k+1, and jump to step 3.8;

Step 3.8, judges whether k>M sets up, if set up, stops search procedure, obtains the MPPM constellation selection scheme P based under minimum average B configuration error sign ratio ^*=P ₀with the corresponding average error sign ratio P of least commitment _{sER_min}(P ^*); If be false, jump to step 3.2.

For a person skilled in the art, can be according to above technical scheme and design, make various corresponding changes and distortion, and all these changes and distortion, within all should being included in the protection range of the claims in the present invention.

Claims (5)

1. A kind of MPPM constellation system of selection based on average error sign ratio in the communication of 1.X ray, is characterized in that, described method comprises the steps:

   Step 1, the average error sign ratio model that structure is selected based on constellation;

   Step 2, the average error sign ratio model of selecting based on constellation drawing according to described step 1, for different MPPM constellation combination, calculate the average error sign ratio under different MPPM constellation combination conditions, and constellation selection problem is converted into constraint optimization problem;

   Step 3, the described average error sign ratio model of selecting based on constellation drawing according to described step 1, and the described constraint optimization problem that draws of described step 2, by selecting search strategy, obtain the MPPM constellation selection scheme based under minimum average B configuration error sign ratio condition.
2. 2. a kind of MPPM constellation system of selection based on average error sign ratio in X ray communication according to claim 1, is characterized in that, described step 1 specifically comprises the steps:

   Step 1.1, is n for number of time slots, and the MPPM signal frame that pulse number is w, can be designated as (n, w) MPPM, the wherein position inclusion information of w described time slot, and this w described time slot is called signal slot; By the various combination of signal slot position, one is total plant different MPPM symbols, and the set of these different described MPPM symbols forms MPPM constellation, can be expressed as X (x ₁, x ₂..., x _m), wherein x _ia MPPM symbol that is described MPPM constellation can be expressed as

   x _i＝(x _i1,x _i2,…,x _in)；

   Wherein x _inthe value that represents a time slot, value is 0 or 1, if value is 1, is signal slot; x _iduty ratio be w/n;

   Step 1.2, uses P _irepresent to select x _iprobability, i.e. P _i=P (x _i), vector P=(P ₁, P ₂..., P _m) represent X (x ₁, x ₂..., x _m) in the selecteed probability of each element; For completing the MPPM modulation that length is the binary sequence of L, need K=2 ^lindividual MPPM symbol, therefore need to be from X (x ₁, x ₂..., x _m) select K MPPM symbol, i.e. P=(P with equiprobability ₁, P ₂..., P _m) in to have K element value be 1/K, all the other element values are 0; Therefore have plant MPPM symbol selection mode, the MPPM constellation combination that Z kind is different;

   Step 1.3, makes λ _sin the representation unit time, arrive the par of the signal photon of each time slot, λ _bin the representation unit time, arrive the par of the noise photon of each time slot, (ξ _i1, ξ _i2..., ξ _iw) for working as x _ix during by transmission _ithe position of a middle w signal slot, (c _i1, c _i2..., c _in) be x _ithe noise photon that during by channel, maximum likelihood receiver device detects in each time slot and the total quantity of signal photon; Definition N _ifor x _iphoton number summation in the w signal slot that during by channel, maximum likelihood detector detects, N _ican be expressed as

   $N_i=\underset{k=1}{\overset{w}{\mathrm{\&Sigma;}}}{c}_{{i,\mathrm{\&xi;}}_{\mathrm{ik}}};$

   Step 1.4, under noisy condition, channel is input as x _i, the upper limit of error sign ratio is defined as at transmission x _iunder condition, obtain x at channel output _j(j ∈ N ⁺, j≤M, j ≠ i), and x _jsignal slot in the number of photons that observes be more than or equal to x _ithe conditional probability density of number of photons of signal slot

   $\begin{array}{c}{P}_{\mathrm{SER}}\left(x_i\right)\&le;P\left\{\underset{j\&NotEqual;i}{\overset{M}{\&cup;}}(N_i\&le;N_j,x_j|x_i)\right\}\\ \&le;\underset{j\&NotEqual;i}{\overset{M}{\mathrm{\&Sigma;}}}P\left(x_j\right)P(N_i\&le;N_j|x_i)\end{array};$

   Wherein P (N _i≤ N _j| x _i) be transmission x _itime symbol error be received as x _jprobability, be to define by the Hamming distance of two vectors, d _h(x _i, x _j) expression x _iand x _jbetween Hamming distance, λ is to be signal strength signal intensity λ by the par of the signal photon that arrives each time slot in the unit interval _swith the par of the noise photon that arrives each time slot in the unit interval be noise intensity λ _bderivation draws, can be expressed as ${\mathrm{\&lambda;}}^2={(\sqrt{{\mathrm{\&lambda;}}_s+{\mathrm{\&lambda;}}_b}-\sqrt{{\mathrm{\&lambda;}}_b})}^2;$

   Step 1.5, the upper limit of the error sign ratio defining according to step 1.4, while passing through transmission after symbol in MPPM constellation is selected with different probability, each symbol correspondence has error sign ratio, and the probability that this error sign ratio occurs is based on the selecteed probability of this symbol, the average error sign ratio model of selecting based on constellation described in can defining is thus:

   $\begin{array}{c}{P}_{\mathrm{SER}\_\mathrm{avr}}=\underset{i=1}{\overset{M}{\mathrm{\&Sigma;}}}P\left(x_i\right)\underset{j\&NotEqual;i}{\overset{M}{\mathrm{\&Sigma;}}}P\left(x_j\right)P(N_i\&le;N_j|x_i)\\ =\underset{i=1}{\overset{M}{\mathrm{\&Sigma;}}}{P}_i\underset{j\&NotEqual;i}{\overset{M}{\mathrm{\&Sigma;}}}{P}_{j}P(N_i\&le;N_j|x_i)\end{array}.$
3. 3. a kind of MPPM constellation system of selection based on average error sign ratio in X ray communication according to claim 1, is characterized in that, the concrete steps of described step 2 are:

   Step 2.1, the average error sign ratio model that utilizes the system of selection of the MPPM constellation combination drawing in step 1 and select based on constellation, for different MPPM constellation combination, calculates average error sign ratio;

   Step 2.2, utilizes the average error sign ratio of the different MPPM constellation combination that step 2.1 draws, constellation selection problem is converted into constraint optimal problem,

   $P_{\mathrm{SER}\_\min }\left(P\right)=\underset{P}{\arg \max }{P}_{\mathrm{SER}\_\mathrm{avr}}$

   $s.t.\underset{i=1}{\overset{M}{\mathrm{\&Sigma;}}}{P}_i=1;$

   P _i∈{0,1/K}

   Γ(P)＝K

   Wherein, P=(P (x ₁), P (x ₂) ..., P (x _m)) represent that the selecteed probability of symbol of MPPM constellation, Γ () are illustrated in the number that certain vector median is 0 element, K=2 ^l, L is the length that needs the binary sequence of transmission.
4. 4. a kind of MPPM constellation system of selection based on average error sign ratio in X ray communication according to claim 1, is characterized in that, in described step 3, adopting direct stochastic search methods is search strategy, specifically comprises the steps:

   Step 3.1, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ ₀(ξ _0,1, ξ _0,2..., ξ _{0, K}); If the starting point ξ of search procedure ₀probability tables be shown P ₀(ξ _{0, i})=1/K, i=1,2 ..., K, utilizes the described constraint optimal problem that step 2 draws to calculate the average error sign ratio P of initial constraint _{sER_min}(P ₀), making total searching times is N _total, iterations k=1;

   Step 3.2, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ (ξ ₁, ξ ₂..., ξ _k), and make P (ξ _i)=1/K, i=1,2 ..., K, utilizes the described constraint optimal problem that step 2 draws to calculate the average error sign ratio P of constraint _{sER_min}(P);

   Step 3.3, judges P _{sER_min}(P) <P _{sER_min}(P ₀) whether set up, if set up P ₀=P, P _{sER_min}(P ₀)=P _{sER_min}(P); Otherwise P ₀remain unchanged;

   Step 3.4, makes k=k+1, if k>N _total, stop search procedure and draw the MPPM constellation selection scheme P based on minimum average B configuration error sign ratio ^*=P ₀, and the corresponding average error sign ratio P of least commitment _{sER_opt}=P _{sER_min}(P ₀), otherwise return to step 3.2.
5. 5. a kind of MPPM constellation system of selection based on average error sign ratio in X ray communication according to claim 1, is characterized in that, in described step 3, can also adopt traversal search method is search strategy, and concrete implementation step is as follows:

   Step 3.1, utilizes being uniformly distributed on [1M] to produce one group of integer sequence ξ ₀(ξ _0,1, ξ _0,2..., ξ _{0, K}); If the starting point ξ of search procedure ₀probability tables be shown P ₀(ξ _{0, i})=1/K, i=1,2 ..., K, utilizes the described constraint optimal problem that step 2 draws to calculate the average error sign ratio P of initial constraint _{sER_min}(P ₀); Make iterations k=1;

   Step 3.2, according to the described average error sign ratio model of selecting based on constellation, P _ivalue 0 or 1/K, for convenience of description, be referred to as two values the other side's " opposition value "; For the P in step 3.1 ₀a certain element P _k, its current value is changed into its opposition value, obtain P ' _k, and P ₀in other values remain unchanged, thereby make P ₀change into P ' ₀;

   Step 3.3, for P ' ₀in except P ' _kelement in addition, finds its value to equal P _kall elements, make at P ' ₀one group of sequence of middle composition

   Step 3.4, for P ' ₀, change and belong to P _cin the value of one of them element be its opposition value P ' ₀in other elements remain unchanged, obtain one group of new sequence, wherein two elements and P ₀in difference, calculate the average error sign ratio of this constraint of organizing new sequence;

   Step 3.5, for P ' ₀, change and belong to P _cin one of them element, and this element be different from before any element once, and P ' ₀in other elements remain unchanged, again obtain one group of new sequence and calculate the average error sign ratio of constraint of the sequence that this group is new; Repeating said steps 3.5t-1 time;

   Step 3.6, according to step 3.4 and step 3.5, obtains t the average error sign ratio of constraint, by relatively drawing the average error sign ratio of minimum constraint, is designated as P _{sER_min}(P ' _min), corresponding sequence is designated as P' _min;

   Step 3.7, judges P _{sER_min}(P' _min) <P _{sER_min}(P ₀) whether set up: if be false, make k=k+1, jump to step 3.2; If set up, make P _{sER_min}(P ₀)=P _{sER_min}(P' _min), k=k+1, and jump to step 3.8;

   Step 3.8, judges whether k>M sets up, if set up, stops search procedure, obtains the MPPM constellation selection scheme P based under minimum average B configuration error sign ratio ^*=P ₀with the corresponding average error sign ratio P of least commitment _{sER_min}(P ^*); If be false, jump to step 3.2.