CN104901732A - Pilot frequency multiplexing method in dense node configuration system - Google Patents

Abstract

The invention proposes a pilot frequency multiplexing method in a dense node configuration system. As the number of orthogonal pilot frequencies is limited, pilot frequency multiplexing is needed when the number K of orthogonal pilot frequencies is smaller than the number N of users. Previous pilot frequency multiplexing schemes adopt full-random multiplexing, and are applicable to a situation in which users are distributed evenly and randomly. However, due to the non-uniformity of user distribution in an actual system, full-random pilot frequency multiplexing may make users close to one another multiplex one pilot frequency, cause serious pilot frequency interference and reduce the capacity of the system. According to the pilot frequency multiplexing method provided by the invention, the position relationship between users is abstracted into an interference matrix, pilot frequency interference is quantified, and greater weight is given to users with greater interference to enable the users to use orthogonal pilot frequencies first. Therefore, pilot frequency interference is greatly reduced in the system level, the accuracy of channel state information estimation is improved, and the channel capacity is increased. The pilot frequency multiplexing method of the invention has the advantages of low pilot frequency distribution complexity, high feasibility and high extensibility.

Description

Pilot multiplex method in a kind of Dense nodes configuration-system

Technical field

The present invention relates to a kind of wireless communication system, particularly relate to the pilot frequency multiplexing scheme that a kind of multi-user space using dense distribution formula node to lay divides multi-address radio communication system.

Background technology

At frequency spectrum resource when day by day in short supply and wireless data service amount explosive growth, the wireless communication system after 4G needs to carry out new change.Dense distribution formula MIMO technology is not increasing under power and bandwidth resources and effectively can promote spectrum efficiency, and this technology has become the key technology of next generation mobile communication standard at present.The power system capacity of dense distribution formula MIMO depends on the precision of channel estimating.Utilize the reciprocity of channel, pilot tone can be sent by user to base station in up link and carry out channel estimating, obtain the channel condition information (CSI) that downlink transfer needs.But, in reality scene, because pilot resources is limited, make orthogonal guide frequency number can not meet the demand of intensive user.When multiple user uses identical pilot tone, pilot pollution will be caused.The existence of pilot pollution, the CSI that base station is obtained is inaccurate, thus capacity is restricted.So, need urgently to find the method weakening and even eliminate pilot pollution, and reduce the complexity of system realization.For this reason, the present invention provides a kind of pilot frequency multiplexing scheme of dense distribution formula wireless communication technology.

Summary of the invention

The object of this invention is to provide a kind of reasonable distribution and utilize limited pilot resources, support the technical method of dense distribution formula MIMO mobile communication.

The invention provides the pilot multiplex method in a kind of Dense nodes configuration-system, it is characterized in that:

(1) according to the user number K of access and the statistical channel state information of user, the interference matrix G between structuring user's, wherein G _i,jrepresent that i-th user (UE) is to the interference between a jth UE, the diagonal entry of G is set to 0.

(2) system designs one group of orthogonal guide frequency in advance, and available orthogonal guide frequency number is N, and pilot tone sequence number is expressed as 1,2 ..., N;

(3) set pilot frequency distribution vector as ξ=[ξ ₁... ξ _i... ξ _k], ξ _i=n (n=1...N) represents that i-th UE uses the n-th orthogonal guide frequency.If the UE number using the n-th orthogonal guide frequency is K _n, then have $\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{K}_n=K.$

(4) carry out pilot frequency distribution according to interference matrix to user, pilot frequency distribution is obtained by following iterative process:

Steps A, initialization pilot frequency distribution vector, ξ _i=1 (i=1 ..., K), all users use No. 1 pilot tone; If I _maxfor use same pilot UE between maximum interference, be initially n _tempfor the interim orthogonal guide frequency sequence number in iteration, be initially n _temp=1.If ξ '=(ξ ₁' ... ξ _i' ... ξ ' _k) for recording the pilot frequency distribution vector of current iteration process, be initially ξ '=ξ.

Step B, carries out iteration,

Successively pilot frequency distribution is carried out to user.When to user i (i=1,2 ..., K) when distributing pilot tone, traversal pilot tone 1 to pilot tone N, if so current pilot n is distributed to user i simultaneously, continue to perform ergodic process.

Step C, judges pilot frequency distribution result.If ξ ' is identical with ξ, then terminate pilot frequency distribution process, obtain pilot frequency distribution result; Otherwise, make ξ '=ξ, continue step B.

For the pilot multiplex method in above-mentioned a kind of Dense nodes configuration-system, the building method of a kind of interference matrix G is:

$G_{i,j}=\left\{\begin{array}{cc}{d}_{i,j}^{-2}& i\≠j\\ 0& i=j\end{array}\right.$

Wherein d _i,jrepresent the distance between i-th UE and a jth UE.

The present invention adopts technique scheme, has following beneficial effect:

1, when ensureing that CSI is accurately estimated in base station, can the good orthogonal pilot frequency sequence of designed in advance, and pilot number can be fixing, which greatly enhances uniformity and the operability of real system, the pilot-frequency expense of system can be reduced simultaneously, improve the information rate of system;

2, when orthogonal guide frequency number is given, can distribute simply and effectively and utilize existing pilot resources, reducing the pilot pollution between user, improving the accuracy that CSI is estimated in base station;

3, system is adaptive, change at intra-cell users number, intra-cell users change in location, during the change of intra-cell users clustered pattern, the pilot multiplex method that this invents provides still can be stablized and reasonably distribute pilot tone, minimized by pilot pollution between user, guarantee system normally works;

4, this to invent the pilot multiplex method complexity that provides low, but reliability and feasibility is good, system is always disturbed little, and total capacity is much larger than random pilot allocative decision;

5, this invents the pilot multiplex method that provides and has taken into account fairness, ensures that the Pilot Interference residing for each user minimizes, number of users and pilot number ratio comparatively rationally time, there is not the possibility that can not communicate under individual user is in strong jamming.

Accompanying drawing explanation

In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes only shows some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, the accompanying drawing of other embodiments can also be obtained according to these accompanying drawings.

The power system capacity of the pilot frequency multiplexing scheme that Fig. 1 provides for the embodiment of the present invention and Random assignment pilot schemes, with the simulation result schematic diagram of user uplink changed power;

The power system capacity of the pilot frequency multiplexing scheme that Fig. 2 provides for the embodiment of the present invention and Random assignment pilot schemes, with the simulation result schematic diagram of orthogonal guide frequency number change;

The flow chart of a kind of pilot multiplex method that Fig. 3 provides for the embodiment of the present invention.

Embodiment

The present invention program is understood better in order to make those skilled in the art person, below in conjunction with the accompanying drawing in the embodiment of the present invention, clear, complete description is carried out to the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, should belong to the scope of protection of the invention.

Suppose total K user in community, available orthogonal guide frequency number is N.

By pilot tone by user grouping, each group the multiplexing same pilot tone of user, be divided into N (1,2 ..., N) and group.Such as, the user of first group by pilot tone 1, i-th group use pilot tone i, N group pilot tone N.User's number symbol n of each group ₁, n ₂... n _n(often organizing number of users not necessarily identical) represents.

The pilot signal received is:

$y_{P,i}=\underset{j=1}{\overset{n_i}{\mathrm{\Σ}}}{g}_{i,j}+n_{P,i}$ [formula 1]

Represent and sue for peace to the pilot signal of the i-th group of user the received user of useful pilot tone i (namely), user's number is n _i.G _i,jbe the matrix of MN × 1, represent the channel matrix of i-th group of jth user to M RAUs

$g_{i,j}={\mathrm{\Λ}}_{i,j}^{\frac{1}{2}}{h}_{i,j}$

Wherein, Λ _i,jrepresent that corresponding large scale declines, h _i,jrepresent multipath fading, n _p,irepresent additive Gaussian noise, variance is γ _p.The MMSE channel estimating of [formula 1] can be expressed as

${\hat{g}}_{i,j}={\mathrm{\Λ}}_{i,j}{Q}_i^{\frac{1}{2}}{\hat{h}}_i$

$Q_i={(\underset{j=1}{\overset{n_i}{\mathrm{\Σ}}}{\mathrm{\Λ}}_{i,j}+{\mathrm{\γ}}_P{I}_{\mathrm{MN}})}^{-1}$

Definition channel estimation errors is,

${\tilde{g}}_{i,j}=g_{i,j}-{\hat{g}}_{i,j}$

Be defined as follows matrix:

G＝[G ₁… G _N]， $G_i=\left[\begin{array}{ccc}{g}_{i,1}& \·\·\·& g_{i,n_i}\end{array}\right]$

Represent ideal communication channel matrix.Correspondingly, the definition matrix that obtains of channel estimating and channel estimation errors matrix as follows,

${\hat{G}}_i=\left[\begin{array}{ccc}{\hat{g}}_{i,1}& \·\·\·& {\hat{g}}_{i,n_i}\end{array}\right],{\hat{G}}_i=\left[\begin{array}{ccc}{\hat{g}}_{i,1}& \·\·\·& {\hat{g}}_{i,n_i}\end{array}\right]$

$\hat{G}=\left[\begin{array}{ccc}{\hat{G}}_1& \·\·\·& {\hat{G}}_N\end{array}\right],\tilde{G}=\left[\begin{array}{ccc}{\tilde{G}}_1& \·\·\·& {\tilde{G}}_N\end{array}\right]$

Then ideal communication channel matrix can be expressed as

$G=\hat{G}+\tilde{G}$

So at BPU, utilize the channel matrix that channel estimating obtains, received signal vector and the relation between transmitting can be expressed as

$y=\tilde{G}s+\tilde{G}s+n$

Wherein y is the received signal vector of MN × 1, $s={\left[\begin{array}{ccc}{s}_1^T& \·\·\·& s_L^T\end{array}\right]}^T,$ S _lbe the transmission signal vector of l group user W × 1, n is covariance matrix is ε (nn ^h)=γ _uLi _mNthe multiple additive white Gaussian noise in MN × 1.Further, the covariance matrix of channel estimating interference plus noise can be calculated by following formula,

$\mathrm{\Σ}=\mathrm{cov}(\tilde{G}s+n,\tilde{G}s+n)=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}\underset{j=1}{\overset{n_i}{\mathrm{\Σ}}}({\mathrm{\Λ}}_{i,j}-{\mathrm{\Λ}}_{i,j}{Q}_i{\mathrm{\Λ}}_{i,j})+{\mathrm{\γ}}_{\mathrm{UL}}{I}_{\mathrm{MN}}$

The lower bound of the uplink channel capacity of known system can be determined by following formula

$C_{\mathrm{LB}}={\log }_2\det (I_M+{\mathrm{\Σ}}^{-1}\hat{G}{\hat{G}}^H)$

Due to det (I+A ^-1b)=det (A+B)-det (A), can obtain

Wherein, use law of great number, can obtain,

Wherein

Definition matrix so can obtain

Definition matrix for block diagonal matrix

Ξ＝diag([Ξ ₁…Ξ _n…Ξ _N])

Known

So the limiting value that can obtain lower bound capacity is

$C_{\mathrm{LB},\inf }={\log }_2\det (I_K+\mathrm{\Ξ})=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2\det (I_{K_n}+{\mathrm{\Ξ}}_n)$

The specific embodiments of pilot frequency multiplexing, as shown in Figure 3:

Step 1: system designs one group of orthogonal guide frequency in advance, available orthogonal guide frequency number is N, and pilot tone sequence number is expressed as 1,2 ..., N;

If pilot frequency distribution vector is ξ=[ξ ₁... ξ _i... ξ _k], ξ _i=n (n=1...N) represents that i-th UE uses the n-th orthogonal guide frequency.If the UE number using the n-th orthogonal guide frequency is K _n, then have

Step 2: according to the user number K of access and the location distribution of user, the interference matrix G between structuring user's, wherein G _i,jrepresent that i-th user (UE) is to the interference between a jth UE, the diagonal entry of G is set to 0,

$G_{i,j}=\left\{\begin{array}{cc}{d}_{i,j}^{-2}& i\≠j\\ 0& i=j\end{array}\right.$

Wherein d _i,jrepresent the distance between i-th UE and a jth UE.

Step 3: pilot frequency distribution is carried out to user according to interference matrix, pilot frequency distribution is obtained by following iterative process:

(1) initialization pilot frequency distribution vector, ξ _i=1 (i=1 ..., K), all users use No. 1 pilot tone; If I _maxfor use same pilot UE between maximum interference, be initially n _tempfor the interim orthogonal guide frequency sequence number in iteration, be initially n _temp=1.If ξ '=(ξ ₁' ... ξ _i' ... ξ ' _k) for recording the pilot frequency distribution vector of current iteration process, be initially ξ '=ξ.

(2) iteration is carried out

Successively pilot frequency distribution is carried out to user.When to user i (i=1,2 ..., K) and when distributing pilot tone, carry out the traversal of pilot tone 1 to pilot tone N, if so current pilot n is distributed to user i simultaneously, continue traversal.

(3) pilot frequency distribution result is judged.If ξ ' is identical with ξ, then terminate pilot frequency distribution process, obtain pilot frequency distribution result; Otherwise, make ξ '=ξ, continue the iterative process of (2).

At the end of iterative process, just obtain pilot frequency distribution result.At this moment, carry out performance evaluation, comparison system capacity, reference object is random pilot allocative decision.The capacity comparison of change user uplink power, as shown in Figure 1.As seen from Figure 1, the solution of the present invention is better than random pilot scheme when ascending power is greater than 5dBm, along with the increase of power, finally have an appointment 32% gain.The capacity comparison of change orthogonal guide frequency number, as shown in Figure 2.As seen from Figure 2, when the solution of the present invention changes pilot number at fixing ascending power, be better than random pilot scheme all the time, have an appointment at most 32% gain.

In the embodiment that the application provides, should be understood that, disclosed method, not exceeding in the spirit and scope of the application, can realize in other way.Current embodiment is a kind of exemplary example, should as restriction, and given particular content should in no way limit the object of the application.Such as, multiple unit or assembly can in conjunction with or another system can be integrated into, or some features can be ignored, or do not perform.

Claims (2)

1. the pilot multiplex method in Dense nodes configuration-system, is characterized in that:

(1) according to the user number K of access and the statistical channel state information of user, the interference matrix G between structuring user's, wherein G _i,jrepresent that i-th user (UE) is to the interference between a jth UE, the diagonal entry of G is set to 0;

(2) system designs one group of orthogonal guide frequency in advance, and available orthogonal guide frequency number is N, and pilot tone sequence number is expressed as 1,2 ..., N;

(3) set pilot frequency distribution vector as ξ=[ξ ₁... ξ _i... ξ _k], ξ _i=n (n=1...N) represents that i-th UE uses the n-th orthogonal guide frequency.If the UE number using the n-th orthogonal guide frequency is K _n, then have $\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{K}_n=K;$

(4) carry out pilot frequency distribution according to interference matrix to user, pilot frequency distribution is obtained by following iterative process:

Steps A, initialization pilot frequency distribution vector, ξ _i=1 (i=1 ..., K), all users use No. 1 pilot tone; If I _maxfor use same pilot UE between maximum interference, be initially n _tempfor the interim orthogonal guide frequency sequence number in iteration, be initially n _temp=1.If ξ '=(ξ ₁' ... ξ _i' ... ξ ' _k) for recording the pilot frequency distribution vector of current iteration process, be initially ξ '=ξ;

Step B, carries out iteration,

Successively pilot frequency distribution is carried out to user.When to user i (i=1,2 ..., K) when distributing pilot tone, traversal pilot tone 1 to pilot tone N, if so current pilot n is distributed to user i simultaneously, continue to perform ergodic process;

Step C, judges pilot frequency distribution result.If ξ ' is identical with ξ, then terminate pilot frequency distribution process, obtain pilot frequency distribution result; Otherwise, make ξ '=ξ, continue step B.

2. the pilot multiplex method in Dense nodes configuration-system according to claim 1, the building method of described interference matrix G is:

$G_{i,j}=\left\{\begin{array}{cc}{d}_{i,j}^{-2}& i\≠j\\ 0& i=j\end{array}\right.$

Wherein d _i,jrepresent the distance between i-th UE and a jth UE.