CN104320174A - Satellite multi-beam collaborative transmission method based on partial channel information - Google Patents

Abstract

The invention relates to a satellite multi-beam collaborative transmission method. The effective utilization problem of inter-beam interference rejection and spectrum resource in a multi-beam satellite mobile communication system and the limitation that the multi-beam satellite mobile communication system can only be used for fixing the satellite system are solved. The method comprises the following steps: a, randomly grouping users in a collaborative beam coverage area, wherein the users in each collaborative user group are respectively from different beam coverage areas, and a frequency band is distributed to users of each group; b, computing a link effective transmission gain from the satellite to the user according to partial channel information of the known satellite user; c, computing an optimal grouping pre-coding matrix; d, pre-coding an emitting signal of the collaborative beam according to the computed optimal pre-coding matrix. The method disclosed by the invention has the advantages that the limitation of the prior art is solved, and the method can be applied to the satellite mobile communication system. The power gain is acquired, and the system performance is greatly promoted in comparison with the traditional frequency multiplexing method.

Description

A kind of satellite multi-beam cooperation transmission method based on partial channel knowledge

Technical field

The present invention relates to a kind of technical field of satellite communication, especially relate to a kind of satellite multi-beam cooperation transmission method based on partial channel knowledge relating to multibeam satellite system.

Background technology

Multi-beam antenna technology provide not only ground communication flexibly and covers, and can realize higher data transfer rate, obtains a wide range of applications in modern satellite communications system.But, in multi-beam satellite system, due to the cochannel interference between wave beam and the restriction of rare frequency spectrum resource, hinder the further lifting of multibeam satellite system performance.

Suppress and resource allocation problem for inter-beam interference, the multibeam satellite system solution of at present application is, utilizes traditional frequency multiplexing method; But, low based on the system spectral efficiency of this method.In order to address this problem, up-to-date investigative technique mainly have employed the method suppressing presence of intercell interference in land honeycomb mobile communication, comprises cooperation transmission technology and interference coordination technique.But a basic assumption condition of these methods is, satellite transmitter or control juncture station need accurate terrestrial user channel information; For mobile satellite communication system, particularly based on the mobile communication system of synchronous satellite, due to propagation delay, this condition is very inappeasable.Therefore, these techniques and methods proposed at present can only be used for fixed satellite system.

Summary of the invention

The present invention mainly solves effective Utilizing question of multi-beam satellite mobile communication system medium wave interfascicular AF panel and frequency spectrum resource, and the confinement problems of fixed satellite system can only be used for, provide a kind of suppression interference, raising transmission rate, be not limited only to the satellite multi-beam cooperation transmission method based on partial channel knowledge of fixed satellite system.

Above-mentioned technical problem of the present invention is mainly solved by following technical proposals: a kind of satellite multi-beam cooperation transmission method based on partial channel knowledge, comprises the following steps:

A. carry out random packet to user in the cooperative beam area of coverage, the user in each collaboration user group, respectively from the different beams area of coverage, often organizes user and distributes a frequency range;

B. the link effective transmission gain of satellite to user is calculated according to known satellite subscriber channel partial information: adopt this shade channel model of Lay to derive the outage probability expression formula of satellite user link, according to User Part channel information and the requirement of user link outage probability, try to achieve the effective transmission gain of subscriber channel;

C. calculate optimal group pre-coding matrix: the optimization problem structure obtaining optimum precoding and power division target function according to partial channel knowledge, system power constraints and user rate request, Solve problems structure obtains optimum pre-coding matrix;

D. according to the optimum pre-coding matrix calculated, precoding is carried out to transmitting of cooperative beam.The present invention is based on the multi-beam cooperation transmission method of partial channel knowledge, according to impaction of partial channel state information and the quality of service requirement of user, the precoding vectors of optimal control signal and the distribution of resource, overcome the impact of interference between wave beam and channel fading effectively.Due to the slow time-varying characteristic of partial channel knowledge, put forward the methods efficiently solves the limitation of prior art, and the improvement in performance for multi-beam satellite mobile communication system provides a kind of solution.

As a kind of preferred version, step a detailed process is: multi-beam cooperation transmission method is realized by juncture station, juncture station manages K wave beam, N number of user is uniform-distribution with in K footprint of a beam, adopt the method for Stochastic choice by user grouping, often organize K user respectively from K the different beams area of coverage.

As a kind of preferred version, step b detailed process is:

The signal often organizing user adopts multi-beam cooperation to send, then the Received signal strength of terrestrial user k can be expressed as:

$y_k=a_k{b}_{k,\max }{g}_k^T\underset{j=1}{\overset{K}{\mathrm{\Σ}}}{w}_j{s}_j+n_k---\left(1\right)$

Wherein k is a kth user, a _kit is the fading channel coefficients of user k; w _jfor the transmitting precoding weight vector of user k; S _jit is the predetermined data-signal sending to user; be the direction vector of user, K represents K the wave beam that juncture station manages, or is G=[g by radiation gain matrix notation ₁g ₂g _k], it reflects the radiation gain characteristic of satellite antenna to terrestrial user; n _kit is the Gaussian noise in Received signal strength; for the beam gain of a kth beam center, it contains the impact of transmitter antenna gain (dBi), path loss, receiving antenna gain and noise power, is defined as

$b_{k,\max }^2=\frac{G_R^2{G}_T^2\left(\mathrm{\α}\right)}{{\left(4\mathrm{\π}\frac{L_k}{\mathrm{\λ}}\right)}^2{k}_B{T}_R{B}_W}---\left(2\right)$

L in formula (2) _kfor the distance between user k and satellite, λ is carrier frequency, k _bfor Boltzmann constant, user's receiving antenna gain, be maximum satellite transmitting antenna gain, depend on the mode parameter α of transmitting antenna, T _rand B _wbe respectively the bandwidth of receiver noise temperature and transmission link;

Then the Received signal strength Signal to Interference plus Noise Ratio of cooperative transmission system downlink user k can be expressed as

${\mathrm{\Γ}}_k=\frac{{\left|a_k\right|}^2{b}_{k,\max }^2{\left|g_k^T{w}_k\right|}^2}{{\left|a_k\right|}^2{b}_{k,\max }^2\underset{j=1,j\≠k}{\overset{K}{\mathrm{\Σ}}}{\left|g_k^T{w}_j\right|}^2+1},---\left(3\right)$

Therefore, the channel capacity of user k is C _k=log ₂(1+ Γ _k);

Calculate satellite to user link outage probability:

$P_{\mathrm{out}}=\mathrm{Pr}(C_k<C_{\mathrm{out},k})=\mathrm{Pr}({\mathrm{\&gamma;}}_k<{\mathrm{\&gamma;}}_k^{\mathrm{thr}})---\left(4\right)$

Wherein C _{out, k}be the target capacity meeting outage probability, be outage capacity;

Adopt this shadow fading channel model of Lay, the probability density function of fading channel is:

$f_{{\mathrm{\&gamma;}}_k}\left({\mathrm{\&gamma;}}_k\right)=K_0\exp (-\frac{{\mathrm{\&gamma;}}_k}{2b_0})F_1^1(m,1;c_0{\mathrm{\&gamma;}}_k)---\left(5\right)$

Wherein channel parameter $K_0=\frac{1}{2b_0}{\left(\frac{2b_{0}m}{2b_{0}m+\mathrm{\&Omega;}}\right)}^m,c_0=\frac{1}{2b_0}\frac{\mathrm{\&Omega;}}{2b_{0}m+\mathrm{\&Omega;}},$ Ω is the average power of sight line component, and m is Nakagami distributed constant, 2b ₀the average power of scattering component, ₁f ₁(m, 1; c ₀γ _k) be confluent hypergeometric function, thresholding for:

${\mathrm{\&gamma;}}_k^{\mathrm{thr}}=\frac{(2^{C_{\mathrm{out},k}}-1)/b_{k,\max }^2}{({\left|g_k^T{w}_k\right|}^2-(2^{C_{t\arg \mathrm{et},k}}-1)\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2)}---\left(6\right)$

Solve formula (4) to obtain

P _out＝Φ(u _k)＝1-exp(-u _k)Q(u _k) (7)

In formula (7) q (u _k) be a unlimited item value of series of restraining,

$Q\left(u_k\right)=\frac{K_0}{(1/{2b}_0-c_0)}\underset{n=0}{\overset{\&infin;}{\mathrm{\&Sigma;}}}(\underset{j=0}{\overset{n-1}{\mathrm{\&Pi;}}}(m-1-j)/n!){\left(\frac{c_0}{1/{2b}_0-c_0}\right)}^n\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{u}_k^i/k!---\left(8\right)$

Apply limited item number with approximant (8) in value of series, obtain

$P_{\mathrm{out}}\&ap;1-(1+u(1-\frac{K_0}{(1/{2b}_0-c_0)})+\frac{u^2}{2}(1-\frac{K_0}{(1/{2b}_0-c_0)}-\frac{(m-1)K_0{c}_0}{{(1/{2b}_0-c_0)}^2}))\exp (-u)---\left(9\right)$

Satellite user partial channel knowledge comprises the directional information { g of subscriber channel _j, j=1,2 ..., K} sum test statistics information (wherein directional information { the g of subscriber channel under User Part channel information known case _j, j=1,2 ..., K} can utilize the channel estimating of up link to try to achieve, statistic information Ω ', b ' ₀, m} can estimate at user side, and feeds back to satellite launch pole or juncture station), based on the user terminal probability P determined _outrequirement, the dichotomy calculating formula (9) of using iterative, obtains the effective transmission gain of subscriber channel

According to u _kdefinition, user's outage capacity can be expressed as

$C_{\mathrm{out},k}\&ap;{\log }_2(1+\frac{u_k^{*}\left(P_{\mathrm{out}}\right){\left|g_k^T{w}_k\right|}^2}{(1/{2b}_0-c_0)/b_{k,\max }^2+u_k^{*}\left(P_{\mathrm{out}}\right)\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2})---\left(10\right)$

Suppose the band bandwidth B often organizing user's distribution _w, then the signal transmission rate of user k can be expressed as R _k=B _wc _{out, k}.

As a kind of preferred version, step c detailed process is:

Based on User Part channel information, system power constraint and the rate request of user, the optimization problem of structure precoding and power division, the target of this optimization problem is: between guarantee user under fairness condition, the speed of user is maximized, and problem is:

$\begin{array}{c}\underset{\{w_k,p_k\}}{\min }\max (1-\frac{p_k}{F_k})\\ s.t.C_1:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left|w_k\right|}^2{p}_k\&le;P_{\max },C_2:R_k\&le;F_k\end{array}---\left(11\right)$

P in formula (11) _maxfor the constraint of system maximum power, F _kit is the rate request of user k;

Adopt ZF pre-coding power optimized algorithm, order and by ZF pre-coding matrix be applied in (11), formula (11) is rewritten as

$\begin{array}{c}\underset{\{t,p_k\}}{\min }\mathrm{t\; s}.t.C_1:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left[{\left({\mathrm{GG}}^T\right)}^{-1}\right]}_{k,k}{p}_k\&le;P_{\max },\\ C_2:\frac{b_{k,\max }^2{u}_k^{*}\left(P_{\mathrm{out}}\right)p_k}{(1/{2b}_0-c_0)}\&GreaterEqual;\left(2^{\frac{(1-t)F_{}k}{B_w}}\right),C_3:t\&le;1\end{array}---\left(12\right)$

Solve formula (12), obtain optimal power allocation { p _k, k=1,2 ..., K}, wherein the distribution power of user k is [(GG ^t) ^-1] _{k, k}p _k;

Formula (11) is the feasibility Solve problems of a convex optimization problem, has the fast algorithm of many maturations to adopt.

Adopt precoding optimized algorithm, formula (11) is rewritten as

$\begin{array}{c}\underset{\{w_k,t\}}{\min }\mathrm{t\; s}.t.C_1:t\&le;1,C_2:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left|w_k\right|}^2\&le;P_{\max }\\ C_3:\frac{{\left|g_K^T{w}_k\right|}^2}{\frac{(1/{2b}_0-c_0)}{b_{k,\max }^2{u}_k^{*}\left(P_{\mathrm{out}}\right)}+\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2}\&GreaterEqual;(2^{\frac{(1-t)F_k}{B_W}}-1)\end{array}---\left(\text{13}\right)$

C in formula (13) ₂a convex function, C ₃be a convex Second-order cone programming constraint, adopt the dichotomy of iteration to solve and obtain optimum precoding vectors { w _k, k=1,2 ..., K}.

As a kind of preferred version, the detailed process of steps d is:

According to the precoding vectors of the optimum obtained cooperative system transmit for:

$x=\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{w}_k^{*}{s}_k,k=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,K---\left(14\right).$

Therefore, advantage of the present invention is:

1., due to the slowly varying behavior of partial channel knowledge relative user position change, the method for proposition efficiently solves the limitation of prior art, can be applied in satellite mobile communication system.

2. although, the complete correlation of satellite multiple antennas, system cannot obtain fading diversity gain; But owing to adopting cooperation precoding transmissions, the signal of each user is sent by multiple wave beam, system can obtain a kind of ' array gain ' of equivalence, obtain a kind of power gain in other words, compare traditional frequency multiplexing method, systematic function is greatly improved.

Accompanying drawing explanation

Accompanying drawing 1 is a kind of structural representation of multibeam satellite system of the present invention;

Accompanying drawing 2 is a kind of method flow schematic diagrames of the present invention;

Accompanying drawing 3 is the first position distribution schematic diagram of collaboration user in the present invention;

Accompanying drawing 4 is collaboration user the second position distribution schematic diagrames in the present invention;

Accompanying drawing 5 is the third position distribution schematic diagrames of collaboration user in the present invention;

Accompanying drawing 6 is cooperation transmission method and legacy frequencies multiplexing method Performance comparision schematic diagrames in the first position distribution of collaboration user;

Accompanying drawing 7 is cooperation transmission method and legacy frequencies multiplexing method Performance comparision schematic diagrames in collaboration user the second position distribution;

Accompanying drawing 8 is cooperation transmission method and legacy frequencies multiplexing method Performance comparision schematic diagrames in the third position distribution of collaboration user.

Embodiment

Below by embodiment, and by reference to the accompanying drawings, technical scheme of the present invention is described in further detail.

Embodiment:

A kind of satellite multi-beam cooperation transmission method based on partial channel knowledge of the present embodiment, run in system, multibeam satellite system as shown in Figure 1, includes juncture station, multi-beam satellite and ground based mobile subscriber, and wherein black round dot represents mobile subscriber.Setting juncture station has obtained the partial channel knowledge of user, and these channel informations are slow time-varying, and this partial channel knowledge comprises the directional information { g of user _j, j=1,2 ..., K} sum test statistics information wherein direction main information relies on the position of user and the radiation mode of antenna, and the impact of frequency is very little, and the channel estimating of up link therefore can be utilized to obtain; The average power Ω ' of sight line component, the average power b ' of scattering component ₀can estimate at user side with Nakagami distributed constant m, and feed back to juncture station; The flow process of the method as shown in Figure 2.

The present embodiment method includes following steps:

A. carry out random packet to user in the cooperative beam area of coverage, the user in each collaboration user group, respectively from the different beams area of coverage, often organizes user and distributes a frequency range; Suppose there is N number of user in all footprint of a beams that juncture station manages, adopt random packet that user is divided into M group, often group includes K user, and each user is from different footprint of a beams, and then, often organize user and distribute a frequency range, the bandwidth of each frequency range is B _w.Suppose each satellite juncture station management K=7 wave beam in the present embodiment, in each footprint of a beam, have user, therefore a N=7.

B. the effective transmission gain of link is calculated according to known satellite subscriber channel partial information.Effective transmission gain coefficient u ^*what describe is a kind of Statistical Parameters that satellite arrives user link, and it relies on association's fading characteristic of channel and the outage probability level of link, and in the present embodiment, we consider a kind of simple scene: the user of cooperation is in identical shade environment.According to this shade channel model of Lay, derive the relation of effective transmission gain and system parameters and channel parameter, process is:

The signal often organizing user adopts multi-beam cooperation to send, then the Received signal strength of terrestrial user k can be expressed as:

$y_k=a_k{b}_{k,\max }{g}_k^T\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{w}_j{s}_j+n_k---\left(1\right)$

Wherein k is a kth user, a _kit is the fading channel coefficients of user k; w _jfor the transmitting precoding weight vector of user k; S _jit is the predetermined data-signal sending to user; be the direction vector of user, K represents K the wave beam that juncture station manages, or is G=[g by radiation gain matrix notation ₁g ₂g _k], it reflects the radiation gain characteristic of satellite antenna to terrestrial user; n _kit is the Gaussian noise in Received signal strength; for the beam gain of a kth beam center, it contains the impact of transmitter antenna gain (dBi), path loss, receiving antenna gain and noise power, is defined as

$b_{k,\max }^2=\frac{G_R^2{G}_T^2\left(\mathrm{\&alpha;}\right)}{{\left(4\mathrm{\&pi;}\frac{L_k}{\mathrm{\&lambda;}}\right)}^2{k}_B{T}_R{B}_W}---\left(2\right)$

L in formula (2) _kfor the distance between user k and satellite, λ is carrier frequency, k _bfor Boltzmann constant, user's receiving antenna gain, be maximum satellite transmitting antenna gain, depend on the mode parameter α of transmitting antenna, T _rand B _wbe respectively the bandwidth of receiver noise temperature and transmission link;

Then the Received signal strength Signal to Interference plus Noise Ratio of cooperative transmission system downlink user k can be expressed as

${\mathrm{\&Gamma;}}_k=\frac{{\left|a_k\right|}^2{b}_{k,\max }^2{\left|g_k^T{w}_k\right|}^2}{{\left|a_k\right|}^2{b}_{k,\max }^2\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2+1},---\left(3\right)$

Therefore, the channel capacity of user k is C _k=log ₂(1+ Γ _k);

Calculate satellite to user link outage probability:

$P_{\mathrm{out}}=\mathrm{Pr}(C_k<C_{\mathrm{out},k})=\mathrm{Pr}({\mathrm{\&gamma;}}_k<{\mathrm{\&gamma;}}_k^{\mathrm{thr}})---\left(4\right)$

Wherein C _{out, k}be the target capacity meeting outage probability, be outage capacity;

Adopt this shadow fading channel model of Lay, the probability density function of fading channel is:

$f_{{\mathrm{\&gamma;}}_k}\left({\mathrm{\&gamma;}}_k\right)=K_0\exp (-\frac{{\mathrm{\&gamma;}}_k}{2b_0})F_1^1(m,1;c_0{\mathrm{\&gamma;}}_k)---\left(5\right)$

Wherein channel parameter $K_0=\frac{1}{2b_0}{\left(\frac{2b_{0}m}{2b_{0}m+\mathrm{\&Omega;}}\right)}^m,c_0=\frac{1}{2b_0}\frac{\mathrm{\&Omega;}}{2b_{0}m+\mathrm{\&Omega;}},$ Ω is the average power of sight line component, and m is Nakagami distributed constant, 2b ₀the average power of scattering component, ₁f ₁(m, 1; c ₀γ _k) be confluent hypergeometric function, thresholding for:

${\mathrm{\&gamma;}}_k^{\mathrm{thr}}=\frac{(2^{C_{\mathrm{out},k}}-1)/b_{k,\max }^2}{({\left|g_k^T{w}_k\right|}^2-(2^{C_{t\arg \mathrm{et},k}}-1)\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2)}---\left(6\right)$

Solve formula (4) to obtain

P _out＝Φ(u _k)＝1-exp(-u _k)Q(u _k) (7)

In formula (7) q (u _k) be a unlimited item value of series of restraining,

$Q\left(u_k\right)=\frac{K_0}{(1/{2b}_0-c_0)}\underset{n=0}{\overset{\&infin;}{\mathrm{\&Sigma;}}}(\underset{j=0}{\overset{n-1}{\mathrm{\&Pi;}}}(m-1-j)/n!){\left(\frac{c_0}{1/{2b}_0-c_0}\right)}^n\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{u}_k^i/k!---\left(8\right)$

Apply limited item number with approximant (8) in value of series, obtain the relation of effective transmission gain and system parameters and channel parameter

$P_{\mathrm{out}}\&ap;1-(1+u(1-\frac{K_0}{(1/{2b}_0-c_0)})+\frac{u^2}{2}(1-\frac{K_0}{(1/{2b}_0-c_0)}-\frac{(m-1)K_0{c}_0}{{(1/{2b}_0-c_0)}^2}))\exp (-u)---\left(9\right)$

In the present embodiment, suppose that outage probability requires as P _out=0.01, shade channel circumstance have employed Loo ' s average shadow ambient measurements: average power Ω=0.835 of sight line component, m=10.1, the average power 2b of scattering component ₀=2.52.According to (9) formula, and adopt dichotomy, the equivalent gain u of link can be tried to achieve ^*=0.01511.

For multi-beam cooperation transmission, the outage capacity of each user is

$C_{\mathrm{out},k}\&ap;{\log }_2(1+\frac{u_k^{*}\left(P_{\mathrm{out}}\right){\left|g_k^T{w}_k\right|}^2}{(1/{2b}_0-c_0)/b_{k,\max }^2+u_k^{*}\left(P_{\mathrm{out}}\right)\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2})---\left(10\right)$

Its transmission rate can be expressed as R _k=B _wc _{out, k}; Obviously, the rate dependent of user is in the selection of the radiation mode of the position distribution of user, satellite antenna, channel statistics and precoder.We suppose α=0.5 in this example, by the matrix element g of radiation gain matrix G _{k, n}be modeled as wherein R _beamterrestrial beam covering radius (-3dB profile), d _{m, n}be on ground m wave beam user from the distance of the n-th beam center.Further, we consider three kinds of different user location distribution, analyze the performance characteristics of put forward the methods;

As shown in Figure 3, the radiation gain matrix of collaboration user can be modeled as the first user location distribution:

$G=\left[\begin{array}{ccccccc}0.5470& 0.0096& 0.0001& 0.0001& 0.0066& 0.4541& 0.1277\\ 0.0897& 0.9094& 0.0204& 0.0000& 0.0000& 0.0002& 0.0322\\ 0.0024& 0.0993& 0.7422& 0.1325& 0.00032& 0.0004& 0.2837\\ 0.0004& 0.0006& 0.0254& 0.6009& 0.3479& 0.0085& 0.2351\\ 0.0000& 0.0000& 0.0000& 0.0182& 0.7867& 0.0263& 0.0098\\ 0.01675& 0.0011& 0.0000& 0.0002& 0.0324& 0.8978& 0.0966\\ 0.0895& 0.2768& 0.1443& 0.0243& 0.0079& 0.0151& 0.7462\end{array}\right]$

The second user location distribution is as shown in Figure 4: the radiation gain matrix of collaboration user can be modeled as:

$G=\left[\begin{array}{ccccccc}1.0000& 0.0625& 0.0002& 0.0000& 0.0002& 0.0625& 0.0625\\ 0.0625& 1.0000& 0.0625& 0.0002& 0.0000& 0.0002& 0.0625\\ 0.0002& 0.0625& 1.0000& 0.0625& 0.0002& 0.0000& 0.0625\\ 0.0000& 0.0002& 0.0625& 1.0000& 0.0625& 0.0002& 0.0625\\ 0.0002& 0.0000& 0.0002& 0.0625& 1.0000& 0.0625& 0.0625\\ 0.0625& 0.0002& 0.0000& 0.0002& 0.0625& 1.0000& 0.0625\\ 0.0625& 0.0625& 0.0625& 0.0625& 0.0625& 0.0625& 1.0000\end{array}\right]$

As shown in Figure 5, the radiation gain matrix of collaboration user can be modeled as the third user location distribution:

$G=\left[\begin{array}{ccccccc}0.4058& 0.4622& 0.0229& 0.0010& 0.0009& 0.0176& 0.3212\\ 0.0176& 0.4058& 0.4622& 0.0229& 0.0010& 0.0009& 0.3212\\ 0.0009& 0.0176& 0.4058& 0.4622& 0.0229& 0.0010& 0.3212\\ 0.0010& 0.0009& 0.0176& 0.4058& 0.4622& 0.0229& 0.3212\\ 0.0229& 0.0010& 0.0009& 0.0176& 0.4058& 0.4622& 0.3212\\ 0.4622& 0.0229& 0.0010& 0.0009& 0.0176& 0.4058& 0.3212\\ 0.0625& 0.0625& 0.0625& 0.0625& 0.0625& 0.0625& 1.0000\end{array}\right]$

C. optimal group pre-coding matrix is calculated.Based on the partial channel knowledge of user, system power constraint and the rate request of user, the optimization problem of structure precoding and power division, the target of this optimization problem is: between guarantee user under fairness condition, the speed of user is maximized, is specifically expressed as follows:

$\begin{array}{c}\underset{\{w_k,p_k\}}{\min }\max (1-\frac{p_k}{F_k})\\ s.t.C_1:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left|w_k\right|}^2{p}_k\&le;P_{\max },C_2:R_k\&le;F_k\end{array}---\left(11\right)$

P in formula (11) _maxfor the maximum power of system retrains, F _kit is the rate request of user k;

Adopt ZF pre-coding power optimized algorithm, order and by ZF pre-coding matrix be applied in (11), formula (11) is rewritten as

$\begin{array}{c}\underset{\{t,p_k\}}{\min }\mathrm{t\; s}.t.C_1:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left[{\left({\mathrm{GG}}^T\right)}^{-1}\right]}_{k,k}{p}_k\&le;P_{\max },\\ C_2:\frac{b_{k,\max }^2{u}_k^{*}\left(P_{\mathrm{out}}\right)p_k}{(1/{2b}_0-c_0)}\&GreaterEqual;\left(2^{\frac{(1-t)F_{}k}{B_w}}\right),C_3:t\&le;1\end{array}---\left(12\right)$

Solve formula (12), obtain optimal power allocation { p _k, k=1,2 ..., K}, wherein the distribution power of user k is [(GG ^t) ^-1] _{k, k}p _k;

Adopt precoding optimized algorithm, formula (11) is rewritten as

$\begin{array}{c}\underset{\{w_k,t\}}{\min }\mathrm{t\; s}.t.C_1:t\&le;1,C_2:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left|w_k\right|}^2\&le;P_{\max }\\ C_3:\frac{{\left|g_K^T{w}_k\right|}^2}{\frac{(1/{2b}_0-c_0)}{b_{k,\max }^2{u}_k^{*}\left(P_{\mathrm{out}}\right)}+\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2}\&GreaterEqual;(2^{\frac{(1-t)F_k}{B_W}}-1)\end{array}---\left(\text{13}\right)$

C in formula (13) ₂a convex function, C ₃be a convex Second-order cone programming constraint, adopt the dichotomy of iteration to solve and obtain optimum precoding vectors { w _k, k=1,2 ..., K}.

D. according to the optimum pre-coding matrix calculated, precoding is carried out to transmitting of cooperative beam.According to the precoding vectors of the optimum obtained cooperative system transmit for:

$x=\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{w}_k^{*}{s}_k,k=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,K---\left(14\right).$

In this example, supposing the system is operated in L-band, frequency f _c=2GHz, the distribution bandwidth of each frequency range is B _w=7 × 10 ⁴hz, satellite antenna in the emission maximum antenna gain of beam center is the Q factor of mobile subscriber's receiver is the covering diameter of terrestrial beam is

Fig. 3 is a kind of random user distribution environment, and Fig. 6 compares the rate capability of algorithm and the traditional channeling joint Power allocation algorithm proposed in this environment; To channeling algorithm, the distribution bandwidth of each user is B _w/ f _r, wherein, f _rbe frequency reuse, the present embodiment selects f _r=4.Fig. 6 shows, two kinds of cooperation transmission methods of proposition achieve larger rate gain than legacy frequencies multiplexing joint Power optimization method, and, the increase of the maximum power that can provide along with system, this gain further increases; Such as, when system emission power is 8dBW, precoding optimization method realizes gain 87.7%, and when transmitting power is 20dBW, this gain reaches 165.5%.We can also notice, system works is when low transmitting power, and zero forcing joint power optimization method performance is starkly lower than precoding optimization method.

Fig. 7 compares the performance of method and the multiplexing joint Power optimization method of legacy frequencies proposed in Fig. 4 position distribution environment; The user location distribution feature of this environment is, the direction vector { g of collaboration user _j, j=1,2 ..., K} has best orthogonality; The Performance Ratio embodiment 1 that Fig. 7 also show cooperation transmission method realization is better, and such as, when transmitting power is 20dBW, the gain that cooperation transmission method realizes than conventional method reaches 200%; Further, two kinds of cooperation transmission method performances of proposition are all very good.

Fig. 8 compares the performance of method and the multiplexing joint Power optimization method of legacy frequencies proposed in Fig. 5 position distribution environment; Because the position distribution of user has suitable symmetry relative to the center of cooperative beam, the direction vector { g of collaboration user _j, j=1,2 ..., K} has very large correlation, and the distribution of this customer location is one very hostile environment; Fig. 8 shows, the performance of collaboration method has very large decline, especially ZF precoding joint Power optimized algorithm; But precoding optimization method still achieves very large performance gain, such as, when transmitting power is 20dBW, precoding optimization method still achieves the performance boost of 76% than traditional method.Indicate the robustness of precoding optimization method to user location distribution.

Specific embodiment described herein is only to the explanation for example of the present invention's spirit.Those skilled in the art can make various amendment or supplement or adopt similar mode to substitute to described specific embodiment, but can't depart from spirit of the present invention or surmount the scope that appended claims defines.

Claims (5)

1., based on a satellite multi-beam cooperation transmission method for partial channel knowledge, it is characterized in that: comprise the following steps:

A. carry out random packet to user in the cooperative beam area of coverage, the user in each collaboration user group, respectively from the different beams area of coverage, often organizes user and distributes a frequency range;

B. the link effective transmission gain of satellite to user is calculated according to known satellite subscriber channel partial information: adopt this shade channel model of Lay to derive the outage probability expression formula of satellite user link, according to User Part channel information and the requirement of user link outage probability, try to achieve the effective transmission gain of subscriber channel;

C. calculate optimal group pre-coding matrix: the optimization problem structure obtaining optimum precoding and power division target function according to partial channel knowledge, system power constraints and user rate request, Solve problems structure obtains optimum pre-coding matrix;

D. according to the optimum pre-coding matrix calculated, precoding is carried out to transmitting of cooperative beam.

2. a kind of satellite multi-beam cooperation transmission method based on partial channel knowledge according to claim 1, it is characterized in that step a detailed process is: multi-beam cooperation transmission method is realized by juncture station, juncture station manages K wave beam, N number of user is uniform-distribution with in K footprint of a beam, adopt the method for Stochastic choice by user grouping, often organize K user respectively from K the different beams area of coverage.

3. a kind of satellite multi-beam cooperation transmission method based on partial channel knowledge according to claim 2, is characterized in that step b detailed process is:

The signal often organizing user adopts multi-beam cooperation to send, then the Received signal strength of terrestrial user k can be expressed as:

$y_k=a_k{b}_{k,\max }{g}_k^T\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{w}_j{s}_j+n_k---\left(1\right)$

Wherein k is a kth user, a _kit is the fading channel coefficients of user k; w _jfor the transmitting precoding weight vector of user k; S _jit is the predetermined data-signal sending to user; be the direction vector of user, K represents K the wave beam that juncture station manages, or is G=[g by radiation gain matrix notation ₁g ₂g _k], it reflects the radiation gain characteristic of satellite antenna to terrestrial user; n _kit is the Gaussian noise in Received signal strength; for the beam gain of a kth beam center, it contains the impact of transmitter antenna gain (dBi), path loss, receiving antenna gain and noise power, is defined as

$b_{k,\max }^2=\frac{G_R^2{G}_T^2\left(\mathrm{\&alpha;}\right)}{{\left(4\mathrm{\&pi;}\frac{L_k}{\mathrm{\&lambda;}}\right)}^2{k}_B{T}_R{B}_W}---\left(2\right)$

L in formula (2) _kfor the distance between user k and satellite, λ is carrier frequency, k _bfor Boltzmann constant, user's receiving antenna gain, be maximum satellite transmitting antenna gain, depend on the mode parameter α of transmitting antenna, T _rand B _wbe respectively the bandwidth of receiver noise temperature and transmission link;

Then the Received signal strength Signal to Interference plus Noise Ratio of cooperative transmission system downlink user k can be expressed as

${\mathrm{\&Gamma;}}_k=\frac{{\left|a_k\right|}^2{b}_{k,\max }^2{\left|g_k^T{w}_k\right|}^2}{{\left|a_k\right|}^2{b}_{k,\max }^2\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2+1},---\left(3\right)$

Therefore, the channel capacity of user k is C _k=log ₂(1+ Γ _k);

Calculate satellite to user link outage probability:

$P_{\mathrm{out}}=\mathrm{Pr}(C_k<C_{\mathrm{out},k})=\mathrm{Pr}({\mathrm{\&gamma;}}_k<{\mathrm{\&gamma;}}_k^{\mathrm{thr}})---\left(4\right)$

Wherein C _{out, k}be the target capacity meeting outage probability, be outage capacity;

Adopt this shadow fading channel model of Lay, the probability density function of fading channel is:

$f_{{\mathrm{\&gamma;}}_k}\left({\mathrm{\&gamma;}}_k\right)=K_0\exp (-\frac{{\mathrm{\&gamma;}}_k}{2b_0})F_1^1(m,1;c_0{\mathrm{\&gamma;}}_k)---\left(5\right)$

Wherein channel parameter $K_0=\frac{1}{2b_0}{\left(\frac{2b_{0}m}{2b_{0}m+\mathrm{\&Omega;}}\right)}^m,c_0=\frac{1}{2b_0}\frac{\mathrm{\&Omega;}}{2b_{0}m+\mathrm{\&Omega;}},$ Ω is the average power of sight line component, and m is Nakagami distributed constant, 2b ₀the average power of scattering component, ₁f ₁(m, 1; c ₀γ _k) be confluent hypergeometric function, thresholding for:

${\mathrm{\&gamma;}}_k^{\mathrm{thr}}=\frac{(2^{C_{\mathrm{out},k}}-1)/b_{k,\max }^2}{({\left|g_k^T{w}_k\right|}^2-(2^{C_{t\arg \mathrm{et},k}}-1)\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2)}---\left(6\right)$

Solve formula (4) to obtain

P _out＝Φ(u _k)＝1-exp(-u _k)Q(u _k) (7)

In formula (7) q (u _k) be a unlimited item value of series of restraining,

$Q\left(u_k\right)=\frac{K_0}{(1/{2b}_0-c_0)}\underset{n=0}{\overset{\&infin;}{\mathrm{\&Sigma;}}}(\underset{j=0}{\overset{n-1}{\mathrm{\&Pi;}}}(m-1-j)/n!){\left(\frac{c_0}{1/{2b}_0-c_0}\right)}^n\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{u}_k^i/k!---\left(8\right)$

Apply limited item number with approximant (8) in value of series, obtain

$P_{\mathrm{out}}\&ap;1-(1+u(1-\frac{K_0}{(1/{2b}_0-c_0)})+\frac{u^2}{2}(1-\frac{K_0}{(1/{2b}_0-c_0)}-\frac{(m-1)K_0{c}_0}{{(1/{2b}_0-c_0)}^2}))\exp (-u)---\left(9\right)$ Satellite user partial channel knowledge comprises the directional information { g of subscriber channel _j, j=1,2 ..., K} sum test statistics information under User Part channel information known case, based on the user terminal probability P determined _outrequirement, the dichotomy calculating formula (9) of using iterative, obtains the effective transmission gain of subscriber channel

According to u _kdefinition, user's outage capacity can be expressed as

$C_{\mathrm{out},k}\&ap;{\log }_2(1+\frac{u_k^{*}\left(P_{\mathrm{out}}\right){\left|g_k^T{w}_k\right|}^2}{(1/{2b}_0-c_0)/b_{k,\max }^2+u_k^{*}\left(P_{\mathrm{out}}\right)\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2})---\left(10\right)$

Suppose the band bandwidth B often organizing user's distribution _w, then the signal transmission rate of user k can be expressed as R _k=B _wc _{out, k}.

4. a kind of satellite multi-beam cooperation transmission method based on partial channel knowledge according to claim 3, is characterized in that step c detailed process is:

Based on User Part channel information, system power constraint and the rate request of user, the optimization problem of structure precoding and power division, problem is:

$\begin{array}{c}\underset{\{w_k,p_k\}}{\min }\max (1-\frac{p_k}{F_k})\\ s.t.C_1:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left|w_k\right|}^2{p}_k\&le;P_{\max },C_2:R_k\&le;F_k\end{array}---\left(11\right)$

P in formula (11) _maxfor the constraint of system maximum power, F _kit is the rate request of user k;

Adopt ZF pre-coding power optimized algorithm, order and by ZF pre-coding matrix be applied in (11), formula (11) is rewritten as

$\begin{array}{c}\underset{\{t,p_k\}}{\min }\mathrm{t\; s}.t.C_1:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left[{\left({\mathrm{GG}}^T\right)}^{-1}\right]}_{k,k}{p}_k\&le;P_{\max },\\ C_2:\frac{b_{k,\max }^2{u}_k^{*}\left(P_{\mathrm{out}}\right)p_k}{(1/{2b}_0-c_0)}\&GreaterEqual;\left(2^{\frac{(1-t)F_{}k}{B_w}}\right),C_3:t\&le;1\end{array}---\left(12\right)$

Solve formula (12), obtain optimal power allocation { p _k, k=1,2 ..., K}, wherein the distribution power of user k is [(GG ^t) ^-1] _{k, k}p _k;

Adopt precoding optimized algorithm, formula (11) is rewritten as

$\begin{array}{c}\underset{\{w_k,t\}}{\min }\mathrm{t\; s}.t.C_1:t\&le;1,C_2:\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{\left|w_k\right|}^2\&le;P_{\max }\\ C_3:\frac{{\left|g_K^T{w}_k\right|}^2}{\frac{(1/{2b}_0-c_0)}{b_{k,\max }^2{u}_k^{*}\left(P_{\mathrm{out}}\right)}+\underset{j=1,j\&NotEqual;k}{\overset{K}{\mathrm{\&Sigma;}}}{\left|g_k^T{w}_j\right|}^2}\&GreaterEqual;(2^{\frac{(1-t)F_k}{B_W}}-1)\end{array}---\left(\text{13}\right)$

C in formula (13) ₂a convex function, C ₂be a convex Second-order cone programming constraint, adopt the dichotomy of iteration to solve and obtain optimum precoding vectors { w _k, k=1,2 ..., K}.

5. a kind of satellite multi-beam cooperation transmission method based on partial channel knowledge according to claim 3, is characterized in that the detailed process of steps d is: according to the precoding vectors of the optimum obtained cooperative system transmit for:

$x=\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}{w}_k^{*}{s}_k,k=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,K---\left(14\right).$