CN102938666A - Forming method of downlink multi-cell combined beams - Google Patents

Abstract

The invention provides a forming method of downlink multi-cell combined beams. The method comprises that all users estimate own channel state information in accordance with received signals, the channel state information is fed back to a central processor of a cell, the channel state information comprises the instant channel state information from all distributed antenna elements of the cell to users and the large-scale channel state information from all distributed antenna elements of other cells to users, central processors of all cells transmit the received channel state information to centralized processing nodes, and according to selected optimization targets, the centralized processing nodes calculate all user transmission beam forming vectors in cells in an optimization mode based on the received channel state information. By the aid of the method, the system performance can be improved, and the pilot overhead, the channel estimation overhead and the feedback overhead are reduced effectively.

Description

A kind of descending multi-plot joint beam-forming method

Technical field

The present invention relates to the invention belongs to the wireless data transmission technology field, particularly a kind of descending multi-plot joint beam-forming method based on impaction of partial channel state information.

Background technology

In the second generation and the third generation wireless communication system, cellular technology is widely adopted.In order to improve the performance of cellular system small area edge customer, distributing antenna system (DistributedAntenna System, DAS), base station collaboration system, DWCS (Distributed Wireless Communication System, DWCS) etc. new network architecture technology and multipoint cooperative (Coordinated Multi-Point, CoMP) technology are proposed in succession.These technology have embodied the key character of future broadband wireless communication systems: " spaced antenna placement " and " collaborative signal processing ".

Because complexity is too high, system-wide overall allied signal is processed and is difficult in practice realize.A kind of more feasible framework is that the antenna element with distributed placement in the system is divided into generalized cell one by one, signal Combined Treatment in the center processor of this residential quarter of inner each antenna in each residential quarter, and adopt relatively simple cooperation mode between different districts, as working in coordination with resource distribution etc.For down link, the multi-plot joint beam forming can effectively reduce presence of intercell interference, improves systematic function.In the design of multi-plot joint beam forming scheme, the channel condition information of transmitting terminal (Channel State Information, CSI) plays very important effect, and the difference of transmitting terminal CSI quantity has very large impact to the performance of system.

Generally speaking, we can think that the decline of wireless channel comprises two parts: the one, and the large scale decline comprises path loss and shadow fading; The 2nd, multipath fading, i.e. Rayleigh fading.Multipath fading changes fast, is difficult to real-time tracking, and the large scale decline changes slowly, obtains than being easier to.In real system, when particularly system scale is larger, obtains all instant CSI(at transmitting terminal and comprise large scale CSI and small scale CSI) can bring sizable pilot-frequency expense, channel estimating expense and feedback overhead to system.

Summary of the invention

(1) technical problem to be solved

The object of the invention provides a kind of descending multi-plot joint beam-forming method, resolution system pilot-frequency expense, channel estimating expense and the sizable problem of feedback overhead.

(2) technical scheme

The invention provides a kind of descending multi-plot joint beam-forming method, the method comprises:

S1, each user be according to the channel condition information of the Signal estimation that receives self, and information feedback is given the center processor of this residential quarter; Described channel condition information comprises that the antenna element of each distributed placement in this residential quarter arrives user's self large scale channel condition information to user's self instant channel state information and the antenna element of other distributed placements in residential quarter;

The center processor of S2, each residential quarter is transmitted to the channel condition information of receiving and focuses on node;

S3, based on the channel condition information that receives, focus on node according to predetermined optimization aim, optimize the transmission beamforming vectors that calculates each user in the residential quarter.

Optionally, the method comprises that also focusing on node is distributed to the center processor of each residential quarter with the beamforming vectors that calculates, and the center processor of each residential quarter carries out pretreated step to the beamforming vectors of relative users.

(3) beneficial effect

The object of the invention is to propose a kind of descending multi-plot joint beam-forming method based on impaction of partial channel state information, being applicable to every residential quarter has the far-end antenna element of a plurality of distributed placements and a plurality of users' multi-cell wireless communication system, can effectively reduce pilot-frequency expense, channel estimating expense and feedback overhead when improving systematic function.

Description of drawings

Fig. 1 is flow chart of steps of the present invention;

Fig. 2 is the DWCS system schematic that 3 residential quarters are arranged.

Embodiment

The object of the present invention is to provide a kind of based on the inner instant CSI in transmitting terminal part CSI-each residential quarter and minizone large scale CSI-descending multi-plot joint beam-forming method, being applicable to every residential quarter has the far-end antenna element of a plurality of distributed placements and the multi-cell wireless communication system that there are a plurality of users every residential quarter.The method comprises as shown in Figure 1:

S1, each user be according to the channel condition information of the Signal estimation that receives self, and information feedback is given the center processor of this residential quarter; Described channel condition information comprises that the antenna element of each distributed placement in this residential quarter arrives user's self large scale channel condition information to user's self instant channel state information and the antenna element of other distributed placements in residential quarter;

The center processor of S2, each residential quarter is transmitted to the channel condition information of receiving and focuses on node;

S3, based on the channel condition information that receives, focus on node according to selected optimization aim, optimize the transmission beamforming vectors that calculates each user in the residential quarter.

Wherein channel condition information is estimated to be divided into:

1) short corresponding to pilots insertion cycle and the channel estimation period of instant channel state information in the residential quarter, be enough to follow the tracks of the quick variation of instant channel state information;

2) pilots insertion cycle and the channel estimation period corresponding to minizone large scale channel condition information is longer, as long as can follow the tracks of the variation of large scale channel condition information.

Wherein information feedback is divided into:

1) instant channel shape body information feedback period is short in the residential quarter, be enough to follow the tracks of the quick variation of instant channel state information;

2) the minizone large scale channel state feedback cycle longer, as long as can follow the tracks of the variation of large scale channel condition information

Because it is slower that large scale CSI changes, the pilots insertion cycle that is used for estimation large scale CSI can prolong, and channel estimation period and feedback cycle also can proper extensions.Thereby pilot-frequency expense, channel estimating expense and feedback overhead can effectively be reduced.

Wherein, the method comprises that also focusing on node is distributed to the center processor of each residential quarter with the beamforming vectors that calculates, and the center processor of each residential quarter carries out pretreated step to the beamforming vectors of relative users.

Concrete:

Consider the DWCS system of a generalized cell, K residential quarter arranged in the system, wherein in k residential quarter N is arranged _kIndividual distributed antenna unit, U _kThe user of individual single antenna and a center processor, the center processor of each residential quarter is connected to one by wire link and focuses on node.Distributed antenna unit in each residential quarter is connected to the separately center processor of residential quarter by optical fiber, the sending and receiving signal of each distributed antenna unit Combined Treatment in corresponding center processor all in the same residential quarter.As shown in Figure 2, be a DWCS system that 3 residential quarters are arranged.

S1, the signal indication that u user receives in k residential quarter in down link are

$y_u^{\left(k\right)}=h_u^{(k,k)}{w}_u^{\left(k\right)}{x}_u^{\left(k\right)}+\underset{l=1}{\overset{K}{\mathrm{\Σ}}}\underset{j=1,(l,j)\≠(k,u)}{\overset{U_l}{\mathrm{\Σ}}}{h}_u^{(k,l)}{w}_j^{\left(l\right)}{x}_j^{\left(l\right)}+n_u^{\left(k\right)},---\left(1\right)$

Wherein

A 1 * N _lComplex vector, each spaced antenna that represents l residential quarter u user's in k the residential quarter channel,

The beamforming vectors of j user in l the residential quarter, The transmitted signal of j user in l the residential quarter,

Be white complex gaussian noise, obey multiple Gaussian Profile

According to expression formula 1,

Can be expressed as

$h_u^{(k,l)}=g_u^{(k,l)}{L}_u^{(k,l)}---\left(2\right)$

Wherein

A 1 * N _lComplex vector, its each element is obeyed CN (0,1) and is distributed, expression multipath fading, i.e. Rayleigh fading;

Be a diagonal matrix, the decline of expression large scale comprises path loss and shadow fading.

U the user's that be n the distributed antenna unit in l residential quarter in k the residential quarter distance, Be shadow fading between the two, γ is the path loss factor.In this system, the inner instant CSI in each residential quarter is:

$\{h_u^{(k,k)}=g_u^{(k,k)}{L}_u^{(k,k)}|k=1,...,K,u=1,...,U_k\}---\left(3\right)$

Minizone large scale CSI is:

$\{h_u^{(k,l)}=g_u^{(k,l)}{L}_u^{(k,l)}|k,l=1,...,K,k\≠l,u=1,...,U_k\}---\left(4\right)$

All information feedbacks are given the center processor of this residential quarter; If U _k=U, N _k=N, then the inner CSI in each residential quarter is contained NUK bar single-shot list collection of letters road, and minizone CSI is contained (K-1) NUK bar single-shot list collection of letters road, and CSI accounts for whole CSI's between visible cells

To count K larger when the residential quarter, and the user in the system only estimates and feeds back the inner instant CSI in residential quarter and minizone large scale CSI, can greatly reduce pilot tone, channel estimating and feedback overhead.

The center processor of S2, each residential quarter is received behind the CSI of each user feedback it is transmitted to and is focused on node

S3, based on the CSI that receives, focus on node adopts each user in certain optimization method computing system according to selected optimization aim transmission beamforming vectors

S4, focus on node the beamforming vectors that calculates is distributed to the center processor of each residential quarter, make each center processor carry out preliminary treatment to the data of relative users.

If the business of each user in the DWCS system is insensitive to time delay, and different User Priorities is different, and weighted sum rate that then can selecting system is optimization aim.Owing to only know large scale CSI for the channel transmitting terminal of minizone, the hypothesis of the interference plus noise Gaussian distributed that receives based on the user, the power of the interference plus noise that u user receives in k residential quarter

$I_{k,u}^2\left(w\right)=\underset{l=1}{\overset{K}{\mathrm{\Σ}}}\underset{j=1,(l,j)\≠(k,u)}{\overset{U_l}{\mathrm{\Σ}}}{\left|h_u^{(k,l)}{w}_j^{\left(l\right)}\right|}^2+{\mathrm{\σ}}_{k,u}^2---\left(5\right)$

Can approximate representation be

$J_{k,u}^2\left(w\right)=E_{\left\{g_u^{(k,l)}\right|l=1,...,K,l\≠k\}}\left(I_{k,u}^2\left(w\right)\right)=\underset{i=1,i\≠u}{\overset{U_k}{\mathrm{\Σ}}}{\left|h_u^{(k,k)}{w}_i^{\left(k\right)}\right|}^2+\underset{l=1,l\≠k}{\overset{K}{\mathrm{\Σ}}}\underset{j=1}{\overset{U_l}{\mathrm{\Σ}}}{\left|\right|L_u^{(k,l)}{w}_j^{\left(l\right)}\left|\right|}_2^2+{\mathrm{\σ}}_{k,u}^2---\left(6\right)$

The weight coefficient of supposing u the user in k residential quarter is

Total transmitted power of all distributed antenna units of k residential quarter is constrained to P ^(k), then multi-plot joint beam forming problem can be written as

$\underset{\left\{w_u^{\left(k\right)}\right|k=1,...,K,u=1,...,U_k\}}{\max }\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\underset{u=1}{\overset{U_k}{\mathrm{\Σ}}}{\mathrm{\α}}_u^{\left(k\right)}{\log }_2(1+\frac{{\left|h_u^{(k,k)}{w}_u^{\left(k\right)}\right|}^2}{J_{k,u}^2\left(w\right)})---\left(7\right)$

s.t. $\underset{u=1}{\overset{U_k}{\mathrm{\Σ}}}{\left|\right|w_u^{\left(k\right)}\left|\right|}_2^2\≤P^{\left(k\right)},k=1,...,K.$

Calculate each user's transmission beamforming vectors, focus on the solution that node only need to adopt certain algorithm to obtain this problem and get final product.

The above only is preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the technology of the present invention principle; can also make some improvement and replacement, these improvement and replacement also should be considered as protection scope of the present invention.

Claims (2)

1. descending multi-plot joint beam-forming method is characterized in that the method comprises:

S1, each user be according to the channel condition information of the Signal estimation that receives self, and information feedback is given the center processor of this residential quarter; Described channel condition information comprises that the antenna element of each distributed placement in this residential quarter arrives user's self large scale channel condition information to user's self instant channel state information and the antenna element of other distributed placements in residential quarter;

The center processor of S2, each residential quarter is transmitted to the channel condition information of receiving and focuses on node;

S3, based on the channel condition information that receives, focus on node according to predetermined optimization aim, optimize the transmission beamforming vectors that calculates each user in the residential quarter.

2. method as claimed in claim 1, it is characterized in that, the method comprises that also focusing on node is distributed to the center processor of each residential quarter with the beamforming vectors that calculates, and the center processor of each residential quarter carries out pretreated step to the beamforming vectors of relative users.

Images