CN102711047A - Optimal opportunistic multicast method under mechanism of equal transmission rate in multimedia multicast technology - Google Patents

Abstract

The invention discloses an optimal opportunistic multicast method under a mechanism of an equal transmission rate in multimedia multicast technology. The method comprises the steps of calculating probability Pnk of accurately receiving data for n times by all users under the condition of each fixed transmission rate rk; calculating equivalent transmission rate corresponding to each transmission rate rk after Pnk is calculated; and selecting the transmission rate corresponding to the maximal equivalent transmission rate as a transmission rate of a base station to transmit data. The method gives an opportunistic multicast mechanism at equal transmission rates under the condition of known channel distribution in the case of hardly needing feedback data, and puts forward a scheme acquiring an optimal transmission rate R according to probability theory and related knowledge of stochastic process. The scheme greatly reduces feedback and simultaneously effectively improves system capacitance to be more applicable for practical systems, and the method considers the optimal scheme under various scenes and situations to achieve good generality.

Description

Optimum chance multicasting method under the medium transmission rate mechanism of multi-medium multi-packet broadcasting technology

Technical field

The invention belongs to wireless communication technology field, especially a kind of chance multicast scheme of multi-medium multi-packet broadcasting technology (MBMS, Multimedia Broadcast Multicast Service) medium transmission rate.This scheme can reduce feedback, improves the performance of system.

Background technology

Along with progressively popularizing of large-screen, high-performance, low power consumption portable terminal device; People are also increasing to the professional demand of data; The mobile subscriber has no longer satisfied through these simple data services such as mobile phone reception news, weather, traffic and share prices; They hope to utilize watching television programs by mobile terminal and the streaming media service of enjoying high speed in the hand, like video request program, television broadcasting, video conference, online education, interactive game etc.So realize the multimedia service multimedia broadcast/multicast service of two-forty, be following mobile data Development Trend.

Users different in the wireless communication system are owing to the geographical position, and the influence of factors such as environment makes user's channel status exist than big-difference, and this makes that also system has multi-user diversity gain.In the multi-medium multi-packet broadcasting system; Same data will send to a lot of users, and are so there is the multicast gain again in system, better compromise in order to make that system has between multi-user diversity gain and multicast gain; The notion of chance multicast was proposed in 2004; The good user's receiving multicast traffic data of a part of channel condition are only selected in the base station at every turn, become during owing to wireless channel, so each user has certain chance can correctly receive data.Re-transmission through several times perhaps uses correcting and eleting codes just can make all users can both correctly receive data.Here it is will multicast.

How thin scheme common ground is that all users will feed back to the base station with user's signal to noise ratio at every turn about chance mostly at present; Sort to feeding back the signal to noise ratio of coming in the base station then again, and last base station selects suitable user to dispatch and send data based on correlation theory.The shortcoming of this method is to need a large amount of feedbacks before data are sent in the base station at every turn, and need carry out some to data and handle, and each speed of sending is all different, and this can cause some inconvenience to data processing aspect.Document is also arranged under the condition that known channel distributes; Obtained using the signal-noise ratio threshold under the RS sign indicating number situation through the research of probability theory relevant knowledge; On certain-length, reduced feedback, but document is not considered user's fairness, can cause the certain user to can not get service like this; And document has only been considered all users' channel independent same distribution, lacks the discussion to more general channel model.Based on above consideration, a good chance multicast scheme should be simple and practical, and can reduce the scheme of feeding back and improving systematic function to a certain extent.

Summary of the invention

The objective of the invention is in order to overcome the irrationality of chance multicast scheme in the existing multi-medium multi-packet broadcasting technology; Under the situation that does not almost have feedback; Optimum chance multicast scheme under the medium transmission rate mechanism of a kind of multi-medium multi-packet broadcasting technology is proposed; This scheme can effectively reduce feedback, improves systematic function, is applicable to real system more.

The objective of the invention is to solve through following technical scheme:

Optimum chance multicasting method under the medium transmission rate mechanism of this kind multi-medium multi-packet broadcasting technology specifically may further comprise the steps:

1) at each fixing transmission rate r _kUnder the situation, calculating needs n all user can both correctly receive the probability P of data _n ^k

2) try to achieve P _n ^kAfter, calculate each transmission rate r _kPairing equivalent transmission rate;

3) select the maximum pairing transmission rate of equivalent transmission rate to send data as the transmission rate of base station.

Further, above-mentioned steps 1) concrete grammar following:

Suppose wireless communication system one total M user of single base station, user's channel distribution state is known, and the probability density function of j user's signal to noise ratio is f _i(x), suppose that the transmission rate of single base station disperses, and always total K alternative discrete velocity, constitute set A={ r ₁, r ₂..., r _K, each all with identical speed r _kSend packet, wherein k=l-K is a parameters optimization, all receives up to all users, just carries out the transmission of next packet; The time span of each transmission is T, and it is defined as a time block; Suppose that channel coefficients remains unchanged in a time block, on the different time piece, keep independent, definition snr _kSatisfy following formula:

r _k=log ₂(1+snr _k)(k＝1～K) (1)

Can obtain

${\mathrm{snr}}_k=2^{r_k}-1---\left(2\right)$

P _n ^kCalculating can be divided into following several kinds of condition of different:

(1) homogeneous network

All users' signal to noise ratio independent same distribution under this scene, i user's reception signal indication is:

$y_i=\sqrt{P}{h}_{i}x+n_i---\left(3\right)$

h _iThe expression channel coefficients, corresponding signal to noise ratio does

${\mathrm{SNR}}_i=\frac{P{{|h}_i|}^2}{N_0}---\left(4\right)$

All user's signal to noise ratio probability density functions are identical and be assumed to be f (x), under this network scenarios, below consider P in three kinds of schemes _n ^kFind the solution:

A) general case

P _n ^kObtain by computes:

P _n ^k=[1-(1-p) ⁿ] ^M-[1-(1-p) ^n-1] ^M （5）

Wherein

$p=\underset{{\mathrm{snr}}_k}{\overset{\∞}{\∫}}f\left(x\right)\mathrm{dx}---\left(6\right)$

B) adopt the high specific folding

To carry out high specific from the signal of different time piece and merge, in this case,

${P_n}^k={[1-\underset{i=1}{\overset{n}{\mathrm{\Π}}}(1-{p_i}^{\mathrm{MRC}})]}^M-{[1-\underset{i=1}{\overset{n-1}{\mathrm{\Π}}}(1-{p_i}^{\mathrm{MRC}})]}^M---\left(7\right)$

Wherein

p ₁ ^MRC=p ₁ （8）

${p_i}^{\mathrm{MRC}}=\frac{p_i-p_{i-1}}{1-p_{i-1}}(i\≥2)---\left(9\right)$

p _iCalculating following

p ₁=p

$p_i=\underset{{\mathrm{snr}}_k}{\overset{\∞}{\∫}}{f}_i\left(x\right)\mathrm{dx},(i\≥2)---\left(11\right)$

f _i(y) calculating is following:

f ₁(x)=f(x) （12）

$f_i\left(y\right)=\underset{0}{\overset{\∞}{\∫}}f\left(x\right)f_{i-1}(y-x)\mathrm{dx},(i\≥2)---\left(13\right)$

C) use the fountain sign indicating number

Suppose that it is S that each user needs correct reception size of data at least _mThe information of bit could be recovered initial data, at this moment:

${P_n}^k={(1-\underset{i=0}{\overset{U-1}{\mathrm{\Σ}}}{C_n}^i{p}^i{(1-p)}^{n-i})}^M-{(1-\underset{i=0}{\overset{U-1}{\mathrm{\Σ}}}{C_{n-1}}^i{p}^i{(1-p)}^{n-1-i})}^M---\left(14\right)$

Wherein p calculates gained by (6) formula;

(2) heterogeneous network

Consider user's large scale decline under this scene, all users' signal to noise ratio is independent still obeys different distributions, and i user's received signal is following:

$y_i=\sqrt{{\mathrm{Pd}}_i^{\mathrm{\η}}}{h}_{i}x+n_i---\left(16\right)$

d _iExpression user i is to the distance of base station.η large scale fading coefficients, corresponding signal to noise ratio does

${\mathrm{SNR}}_i=\frac{{\mathrm{Pd}}_i^{-\mathrm{\η}}{\left|h_i\right|}^2}{N_0}---\left(17\right)$

Each user's channel coefficients h _iStill be independent same distribution, but owing to consider the large scale decline, so each user's signal to noise ratio distributes and is inequality.

Under the heterogeneous network scenarios, under the situation of base station distance d, the probability density function of signal to noise ratio is that f (x|d) can know, supposes that the user is evenly distributed in the sub-district in known users, and the radius of sub-district is R _dSo the user is to the probability density function of base station

Also can in the hope of; Therefore the probability density function of all user's signal to noise ratios can be unified to be expressed as:

$f\left(x\right)={\∫}_0^{R_d}f\left(x\right|y)f_{d_i}\left(y\right)\mathrm{dy}---\left(18\right)$

Through after such processing, all users' signal to noise ratio distributes and just can regard the same again as, that is to say that the heterogeneous network just can regard the homogeneous network as, next according to P _n ^kThree kinds of schemes finding the solution.

Under the heterogeneous network scenarios, after the distance of known users and base station, in the heterogeneous network, divide three kinds of situation:

D) general case

At this moment, transmission rate r _kPairing P _n ^k

${P_n}^k=\underset{i=1}{\overset{M}{\mathrm{\Π}}}[1-{\left({1-p}_i\right)}^n]-\underset{i=1}{\overset{M}{\mathrm{\Π}}}[1-{(1-p_i)}^{n-1}]---\left(19\right)$

Wherein

$p_i=\underset{{\mathrm{snr}}_k}{\overset{\∞}{\∫}}{f}_i\left(x\right)\mathrm{dx}---\left(20\right)$

E) adopt the high specific folding

Consider the P after high specific merges _n ^kCan be expressed as

${P_n}^k=\underset{i=1}{\overset{M}{\mathrm{\Π}}}[1-\underset{j=1}{\overset{n}{\mathrm{\Π}}}(1-{p_{\mathrm{ij}}}^{\mathrm{MRC}})]-\underset{i=1}{\overset{M}{\mathrm{\Π}}}[1-\underset{j=1}{\overset{n-1}{\mathrm{\Π}}}(1-{p_{\mathrm{ij}}}^{\mathrm{MRC}})]---\left(21\right)$

Wherein,

p _i1 ^MRC=p _i1 （21）

${p_{\mathrm{ij}}}^{\mathrm{MCR}}=\frac{p_{\mathrm{ij}}-p_{i(j-1)}}{1-p_{i(j-1)}},(j\≥2)---\left(22\right)$

$p_{\mathrm{ij}}={\∫}_{{\mathrm{snr}}_T}^{\∞}{f}_{\mathrm{ij}}\left(x\right)\mathrm{dx}---\left(23\right)$

f _i1(y)=f _i(y) （24）

$f_{\mathrm{ij}}\left(y\right)=\underset{0}{\overset{\∞}{\∫}}{f}_{i1}\left(x\right)f_{i(j-1)}(y-x)\mathrm{dx},(j\≥2)---\left(25\right)$

F) use the fountain sign indicating number

When using the fountain sign indicating number:

${P_n}^k=\underset{i=1}{\overset{M}{\mathrm{\Π}}}(1-\underset{j=0}{\overset{U-1}{\mathrm{\Σ}}}{C_n}^j{p_i}^j{(1-p_i)}^{n-j})-\underset{i=1}{\overset{M}{\mathrm{\Π}}}(1-\underset{j=0}{\overset{U-1}{\mathrm{\Σ}}}{C_{n-1}}^j{p_i}^j{(1-p)}^{n-1-j}).$

Distance when between base station and user is unknown; Suppose that the user can feed back the meeting base station with signal to noise ratio at every turn; At this moment be equivalent to know that signal to noise ratio distributes; Try to achieve approximation parameters according to the parameter that the data prediction that feeds back distributes with the method for data fitting this moment, tries to achieve each user's signal to noise ratio distributed constant.

Further, concrete grammar above step 2) is following:

A, try to achieve P _n ^kAfter, when system adopts general case or uses the high specific folding, transmission rate r _kPairing equivalent transmission rate is:

${R_{\mathrm{eq}}}^k=E\left\{\frac{r_k\×M}{n}\right\}=r_k\×M\×E{\left\{\frac{1}{n}\right\}}_k---\left(26\right)$

Wherein

$E{\left\{\frac{1}{n}\right\}}_k=\underset{n=1}{\overset{\∞}{\mathrm{\Σ}}}\frac{1}{n}{P_n}^k---\left(27\right)$

B, when system uses the fountain sign indicating number, transmission rate r _kPairing equivalent transmission rate is:

${R_{\mathrm{eq}}}^k=E{\left\{\frac{{\mathrm{MS}}_m}{\mathrm{nT}}\right\}}_k=\frac{{\mathrm{MS}}_m}{T}E{\left\{\frac{1}{n}\right\}}_k---\left(28\right)$

$E{\left\{\frac{1}{n}\right\}}_k=\underset{n=1}{\overset{\∞}{\mathrm{\Σ}}}\frac{1}{n}{P_n}^k---\left(29\right)$

Further, the concrete grammar of above step 3) is following:

Optimum transmission rate index can be expressed as:

$k_{\mathrm{opt}}=\arg \underset{k}{\max }\left\{{R_{\mathrm{eq}}}^k\right\},(1\≤k\≤K)---\left(30\right)$

Select

then and carry out the transmission of data as the transmission rate of base station.

The present invention has following beneficial effect:

The present invention needs under the situation of feedback data under the condition that known channel distributes hardly, transmission rate chance multicast mechanism such as has provided, and according to the Probability Theory & Stochastic Process relevant knowledge, has proposed how to obtain the scheme of optimum transmission rate R.This scheme effectively raises the capacity of system when greatly reducing feedback, make it be applicable to real system more, and the present invention considered several scenes, the optimal case under the multiple situation, and the scheme that makes has versatility more.

Description of drawings

Fig. 1 multi-medium multi-packet broadcasting network diagram;

The flow chart that Fig. 2 proposed a plan;

The various situation sketch mapes that Fig. 3 discussed;

Embodiment

Below in conjunction with accompanying drawing and practical implementation instance the present invention is done further detailed description.

As shown in Figure 1, consider the descending wireless communication networks of single sub-district single antenna, radius of society is R _dSame data are sent to M user simultaneously in base station (BS), and channel is the piece wireless fading, and the length of each time block is T, is approximately equal to the correlated time of channel.User's channel distribution is known, h _i(k) channel coefficients of i user of expression on k time block supposed h _i(k) Rayleigh distributed is represented with the multiple Gaussian random variable of zero-mean here, and variance is 1.Suppose that channel coefficients remains unchanged independent variation on the different time piece in a time block.The transmission rate of supposing the base station disperses, and always total K alternative discrete velocity, constitutes set A={ r ₁, r ₂..., r _K.Each all with identical speed r _k(k=1～K is a parameters optimization) sent packet, all receives up to all users, just carries out the transmission of next packet.The time span of each transmission is T, and it is defined as a time block.Definition snr _kSatisfy following formula:

r _k=log ₂(1+snr _k)(k＝1～K) (1)

Can obtain:

${\mathrm{snr}}_k=2^{r_k}-1---\left(2\right)$

The signal that user i receives on j time block can be expressed as:

$y_i\left(j\right)=\sqrt{{\mathrm{Pd}}_i^{\mathrm{\η}}}{h}_i\left(j\right)x+n_i\left(j\right)$

In the formula:

P---transmitted power;

n _i(j)---white Gaussian noise, variance are N ₀

d _i---the distance between user i and the base station;

η---large scale fading coefficients;

Corresponding signal to noise ratio is:

${\mathrm{SNR}}_i=\frac{{\mathrm{Pd}}_i^{-\mathrm{\η}}{{|h}_i|}^2}{N_0}$

Obeying parameter is the exponential distribution of

.If cordless communication network is the homogeneous network, just all users' channel distribution is identical, at this moment, makes d _i=1.The speed that definition user i can reach:

R _i＝log(1+SNR _i)

If r _iGreater than the transmission rate of base station, then user i can correctly receive data, otherwise can not.For traditional multicast scheme, just the base station each all with the poorest user can correctly receive data speed send data.At this moment, transmission rate does

R _t=min{R _i}(i=1,2,...,M)

When the scheduling of the optimum chance multicast of transmission rate scheme such as carrying out, as shown in Figure 2, may further comprise the steps:

1) under each fixing transmission rate rk situation, n all user of needing just who calculates under the various schemes of various scenes can both correctly receive the probability P nk of data; In the computational process of Pnk, as shown in Figure 3, can be divided into following several kinds of situation and discuss respectively:

(1) homogeneous network

All users' signal to noise ratio independent same distribution under this scene, i user's reception signal can be expressed as:

$y_i=\sqrt{P}{h}_{i}x+n_i---\left(3\right)$

h _iThe expression channel coefficients, corresponding signal to noise ratio does

${\mathrm{SNR}}_i=\frac{P{{|h}_i|}^2}{N_0}---\left(4\right)$

Under the channel condition of hypothesis Ruili distribution, the obedience parameter is N ₀The exponential distribution of/P.Its probability density function does

$f\left(x\right)={\mathrm{\λe}}^{-\mathrm{\λx}},(\mathrm{\λ}=\frac{N_0}{P})---\left(5\right)$

Under this network scenarios, below consider P in three kinds of schemes _n ^kFind the solution: a) general case, b) adopt the high specific folding, c) use the fountain sign indicating number.

A) general case

Under the general case, P _n ^kCalculating can obtain by computes:

P _n ^k=[1-(1-p) ⁿ] ^M-[1-(1-p) ^n-1] ^M （6）

Wherein

$p=\underset{{\mathrm{snr}}_k}{\overset{\∞}{\∫}}f\left(x\right)\mathrm{dx}---\left(7\right)$

Wherein f (x) is provided by (5) formula.

B) adopt the high specific folding

The user can carry out the signal from the different time piece the most a large sum of the merging, in this case,

${P_n}^k={[1-\underset{i=1}{\overset{n}{\mathrm{\Π}}}(1-{p_i}^{\mathrm{MRC}})]}^M-{[1-\underset{i=1}{\overset{n-1}{\mathrm{\Π}}}(1-{p_i}^{\mathrm{MRC}})]}^M---\left(8\right)$

Wherein

p ₁ ^MRC=p ₁ （9）

${p_i}^{\mathrm{MRC}}=\frac{p_i-p_{i-1}}{1-p_{i-1}}(i\≥2)---\left(10\right)$

p _iCalculating following

p ₁=p （11）

$p_i=\underset{{\mathrm{snr}}_k}{\overset{\∞}{\∫}}{f}_i\left(x\right)\mathrm{dx},(i\≥2)---\left(12\right)$

f _i(y) calculating is following:

f ₁(x)=f(x) （13）

$f_i\left(y\right)=\underset{0}{\overset{\∞}{\∫}}f\left(x\right)f_{i-1}(y-x)\mathrm{dx},(i\≥2)---\left(14\right)$

C) use the fountain sign indicating number

After data are through fountain sign indicating number coding, will with a) and b) two kinds of situation are different.When using the fountain sign indicating number, suppose that each user needs the correct S of reception at least _mThe information of bit data could be recovered initial data.At this moment:

${P_n}^k={(1-\underset{i=0}{\overset{U-1}{\mathrm{\Σ}}}{C_n}^i{p}^i{(1-p)}^{n-i})}^M-{(1-\underset{i=0}{\overset{U-1}{\mathrm{\Σ}}}{C_{n-1}}^i{p}^i{(1-p)}^{n-1-i})}^M---\left(15\right)$

Wherein p calculates gained by (7) formula.

(2) heterogeneous network

Consider user's large scale decline under this scene, all users' signal to noise ratio is independent still obeys different distributions, and i user's received signal is following:

$y_i=\sqrt{{\mathrm{Pd}}_i^{\mathrm{\η}}}{h}_{i}x+n_i---\left(17\right)$

d _iExpression user i is to the distance of base station.η large scale fading coefficients, corresponding signal to noise ratio does

${\mathrm{SNR}}_i=\frac{{\mathrm{Pd}}_i^{-\mathrm{\η}}{\left|h_i\right|}^2}{N_0}---\left(18\right)$

Each user's channel coefficients h _iStill be independent same distribution, but owing to consider the large scale decline, so each user's signal to noise ratio distributes and is inequality.Provide three kinds of solutions below:

1) similar_homo scheme

Under the situation of base station distance d, the signal to noise ratio probability density function does in known users

$f\left(x\right|d)={\mathrm{\λe}}^{-\mathrm{\λx}},(\mathrm{\λ}=\frac{N_0}{{\mathrm{Pd}}^{-\mathrm{\η}}})---\left(19\right)$

For circular cell, the user to the probability density function of base station distance does

$f_{d_i}=\frac{2d}{{R_d}^2}---\left(20\right)$

Therefore all users newly make than probability density function can unify to be expressed as:

$f\left(x\right)={\∫}_0^{R_d}f\left(x\right|y)f_{d_i}\left(y\right)\mathrm{dy}={\∫}_0^{R_d}{\mathrm{\λe}}^{-\mathrm{\λx}}\frac{2y}{{R_d}^2}\mathrm{dy},(\mathrm{\λ}=\frac{N_0}{{\mathrm{Py}}^{-\mathrm{\η}}})---\left(21\right)$

Through after such processing; All users' signal to noise ratio distributes and just can regard as again equally; That is to say that the heterogeneous network just can regard the homogeneous network as, other step of scheme is also the same with the homogeneous network, can be with reference to processing procedure a) ~ c).

2) optimal algorithm (supposing the distance between known users and base station)

After knowing the distance of user and base station, the probability density function f of each user's signal to noise ratio _i(x) be to be expressed as:

$f_i\left(x\right)={\mathrm{\λ}}_i{e}^{-{\mathrm{\λ}}_{i}x},({\mathrm{\λ}}_i=\frac{N_0}{{\mathrm{Pd}}_i^{-\mathrm{\η}}})---\left(22\right)$

We also can divide three kinds of situation discussion in the heterogeneous network like this: d) general case, e) consider that high specific merges, and f) use the fountain sign indicating number.

D) general case

At this moment, transmission rate r _kPairing P _n ^k

${P_n}^k=\underset{i=1}{\overset{M}{\mathrm{\Π}}}[1-{(1-p_i)}^n]-\underset{i=1}{\overset{M}{\mathrm{\Π}}}[1-{(1-p_i)}^{n-1}]---\left(23\right)$

Wherein

$p_i=\underset{{\mathrm{snr}}_k}{\overset{\∞}{\∫}}{f}_i\left(x\right)\mathrm{dx}---\left(24\right)$

E) adopt the high specific folding

Consider the P after high specific merges _n ^kCan be expressed as

${P_n}^k=\underset{i=1}{\overset{M}{\mathrm{\Π}}}[1-\underset{j=1}{\overset{n}{\mathrm{\Π}}}(1-{p_{\mathrm{ij}}}^{\mathrm{MRC}})]-\underset{i=1}{\overset{M}{\mathrm{\Π}}}[1-\underset{j=1}{\overset{n-1}{\mathrm{\Π}}}(1-{p_{\mathrm{ij}}}^{\mathrm{MRC}})]---\left(25\right)$

Wherein,

p _i1 ^MRC=p _i1 （26）

${p_{\mathrm{ij}}}^{\mathrm{MCR}}=\frac{p_{\mathrm{ij}}-p_{i(j-1)}}{1-p_{i(j-1)}},(j\≥2)---\left(27\right)$

$p_{\mathrm{ij}}={\∫}_{{\mathrm{snr}}_T}^{\∞}{f}_{\mathrm{ij}}\left(x\right)\mathrm{dx}---\left(28\right)$

f _i1(y)=f _i(y) （29）

$f_{\mathrm{ij}}\left(y\right)=\underset{0}{\overset{\∞}{\∫}}{f}_{i1}\left(x\right)f_{i(j-1)}(y-x)\mathrm{dx},(j\≥2)---\left(30\right)$

F) use the fountain sign indicating number

When using the fountain sign indicating number:

${P_n}^k=\underset{i=1}{\overset{M}{\mathrm{\Π}}}(1-\underset{j=0}{\overset{U-1}{\mathrm{\Σ}}}{C_n}^j{p_i}^j{(1-p_i)}^{n-j})-\underset{i=1}{\overset{M}{\mathrm{\Π}}}(1-\underset{j=0}{\overset{U-1}{\mathrm{\Σ}}}{C_{n-1}}^j{p_i}^j{(1-p)}^{n-1-j})---\left(31\right)$

Related definition is seen c)

3) sub-optimal algorithm (distance between base station and user is unknown)

In this case, suppose that the user can feed back the meeting base station with signal to noise ratio at every turn, this moment, we can be according to the parameter of the data prediction distribution that feeds back, and obeying parameter is mathematic expectaion E [x]=1/ λ of the exponential distribution variable x of λ.

Step1: the probability density function of supposing M user's signal to noise ratio is obtained by (21) formula, then

Obtain optimal rate R according to the homogeneous network correlation theories _Opt

Step2: every transmit through K time after, each user's signal to noise ratio probability density function is carried out

Revise, the distributed constant of revising each user of back does

${\mathrm{\λ}}_i\≈\frac{1}{\frac{1}{K}\underset{j=1}{\overset{K}{\mathrm{\Σ}}}{\mathrm{SNR}}_{\mathrm{ij}}},(i=\mathrm{1,2},...,M)---\left(32\right)$

SNR _IjRepresent the signal to noise ratio of i user when the j time transmission.After the correction, calculate the speed R of optimum according to the correlation theory of heterogeneous network _Opt(referring to the step of optimal algorithm).It should be noted that above algorithm also is applicable to the situation that each user slowly moves.

2) try to achieve Pnk after, calculate the pairing equivalent transmission rate of each transmission rate rk, divide two kinds of situation discussion;

A, try to achieve the P under the various schemes of various scenes _n ^kAfter, when system adopts general case or uses the high specific folding, transmission rate r _kPairing equivalent transmission rate is:

${R_{\mathrm{eq}}}^k=E\left\{\frac{r_k\×M}{n}\right\}=r_k\×M\×E{\left\{\frac{1}{n}\right\}}_k---\left(33\right)$

Wherein

$E{\left\{\frac{1}{n}\right\}}_k=\underset{n=1}{\overset{\∞}{\mathrm{\Σ}}}\frac{1}{n}{P_n}^k---\left(34\right)$

B, when system uses the fountain sign indicating number, transmission rate r _kPairing equivalent transmission rate is:

${R_{\mathrm{eq}}}^k=E{\left\{\frac{{\mathrm{MS}}_m}{\mathrm{nT}}\right\}}_k=\frac{{\mathrm{MS}}_m}{T}E{\left\{\frac{1}{n}\right\}}_k---\left(35\right)$

$E{\left\{\frac{1}{n}\right\}}_k=\underset{n=1}{\overset{\∞}{\mathrm{\Σ}}}\frac{1}{n}{P_n}^k---\left(36\right)$

3) under the various schemes of various scenes, select to make the maximum pairing transmission rate of equivalent transmission rate to send data as the transmission rate of base station.Concrete operations are following:

Optimum transmission rate index can be expressed as:

$k_{\mathrm{opt}}=\arg \underset{k}{\max }\left\{{R_{\mathrm{eq}}}^k\right\},(1\≤k\≤K)---\left(37\right)$

Select

then and carry out the transmission of data as the transmission rate of base station.

Claims (7)

1. the optimum chance multicasting method under the medium transmission rate mechanism of multi-medium multi-packet broadcasting technology is characterized in that, may further comprise the steps:

1) at each fixing transmission rate r _kUnder the situation, calculating needs n all user can both correctly receive the probability P of data _n ^k

2) try to achieve P _n ^kAfter, calculate each transmission rate r _kPairing equivalent transmission rate;

3) select the maximum pairing transmission rate of equivalent transmission rate to send data as the transmission rate of base station.

2. the optimum chance multicasting method under the medium transmission rate mechanism of multi-medium multi-packet broadcasting technology according to claim 1 is characterized in that the concrete grammar of step 1) is following:

Suppose wireless communication system one total M user of single base station, user's channel distribution state is known, and the probability density function of i user's signal to noise ratio is f _i(x), suppose that the transmission rate of single base station disperses, and always total K alternative discrete velocity, constitute set A={ r ₁, r ₂..., r _K, each all with identical speed r _kSend packet, wherein k=l～K is a parameters optimization, all receives up to all users, just carries out the transmission of next packet; The time span of each transmission is T, and it is defined as a time block; Suppose that channel coefficients remains unchanged in a time block, on the different time piece, keep independent, definition snr _kSatisfy following formula:

r _k=log ₂(1+snr _k)(k=1～K) (1)

Can obtain

P _n ^kCalculating be divided into following several kinds of condition of different:

(1) homogeneous network

All users' signal to noise ratio independent same distribution under this scene, i user's reception signal indication is:

h _iThe expression channel coefficients, corresponding signal to noise ratio does

All user's signal to noise ratio probability density functions are identical and be assumed to be f (x), under this network scenarios, below consider P in three kinds of schemes _n ^kFind the solution:

A) general case

P _n ^kObtain by computes:

P _n ^k=[1-(1-p) ⁿ] ^M-[1-(1-p) ^n-1] ^M （5）

Wherein

B) adopt the high specific folding

To carry out high specific from the signal of different time piece and merge, in this case,

Wherein

p ₁ ^MRC=p ₁ （8）

p _iCalculating following

p ₁=p （10）

f _i(y) calculating is following:

f ₁(x)=f(x) （12）

C) use the fountain sign indicating number

Suppose that it is S that each user needs correct reception size of data at least _mThe information of bit could be recovered initial data, at this moment:

Wherein p calculates gained by (6) formula;

(2) heterogeneous network

Consider user's large scale decline under this scene, all users' signal to noise ratio is independent still obeys different distributions, and i user's received signal is following:

d _iExpression user i is to the distance of base station.η large scale fading coefficients, corresponding signal to noise ratio does

Each user's channel coefficients h _iStill be independent same distribution, but owing to consider the large scale decline, so each user's signal to noise ratio distributes and is inequality.

3. the optimum chance multicasting method under the medium transmission rate mechanism of multi-medium multi-packet broadcasting technology according to claim 2; It is characterized in that; Under the heterogeneous network scenarios, under the situation of base station distance d, the probability density function of signal to noise ratio is that f (x|d) can know in known users; Suppose that the user is evenly distributed in the sub-district, the radius of sub-district is R _dSo the user is to the probability density function of base station

Also can in the hope of; Therefore the probability density function of all user's signal to noise ratios can be unified to be expressed as:

Through after such processing, all users' signal to noise ratio distributes and just can regard the same again as, that is to say that the heterogeneous network just can regard the homogeneous network as, next according to P _n ^kThree kinds of schemes finding the solution.

4. the optimum chance multicasting method under the medium transmission rate mechanism of multi-medium multi-packet broadcasting technology according to claim 2; It is characterized in that; Under the heterogeneous network scenarios, after the distance of known users and base station, in the heterogeneous network, divide three kinds of situation:

D) general case

At this moment, transmission rate r _kPairing P _n ^k

Wherein

E) adopt the high specific folding

Consider the P after high specific merges _n ^kCan be expressed as

Wherein,

p _i1 ^MRC=p _i1 （21）

f _i1(y)=f _i(y) （24）

F) use the fountain sign indicating number

When using the fountain sign indicating number:

5. the optimum chance multicasting method under the medium transmission rate mechanism of multi-medium multi-packet broadcasting technology according to claim 2; It is characterized in that the distance when between base station and user is unknown, suppose that the user can feed back the meeting base station with signal to noise ratio at every turn; At this moment be equivalent to know that signal to noise ratio distributes; Try to achieve approximation parameters according to the parameter that the data prediction that feeds back distributes with the method for data fitting this moment, tries to achieve each user's signal to noise ratio distributed constant.

6. according to the optimum chance multicasting method under the medium transmission rate mechanism of any described multi-medium multi-packet broadcasting technology of claim 1-5, it is characterized in that step 2) concrete grammar following:

A, try to achieve P _n ^kAfter, when system adopts general case or uses the high specific folding, transmission rate r _kPairing equivalent transmission rate is:

Wherein

B, when system uses the fountain sign indicating number, transmission rate r _kPairing equivalent transmission rate is:

。

7. according to the optimum chance multicasting method under any medium transmission rate mechanism of described multi-medium multi-packet broadcasting technology of claim 1-5, it is characterized in that the concrete grammar of step 3) is following:

Optimum transmission rate index can be expressed as:

Select

then and carry out the transmission of data as the transmission rate of base station.

Images