CN105978836A - Pilot frequency pollution elimination method for multiple cells in large-scale MIMO system - Google Patents

Abstract

The invention provides a pilot frequency pollution elimination method for multiple cells in a large-scale MIMO system. On a prior study basis, a new pilot frequency pollution alleviating method is proposed for reducing algorithm complexity and eliminating pilot frequency pollution in a compensation way. The pilot frequency eliminating scheme realizes pilot frequency pollution elimination through a method that corresponding cells keep silent in corresponding phases while sending pilot frequency signals repeatedly in other phases. The method can be applied to indoor environment and femto cell environments and is high in practicability.

Description

Multi-district pilots abatement of pollution method under extensive mimo system

Technical field:

The invention belongs to 5G radio communication physical-layer techniques field, particularly to multi-district pilots abatement of pollution method under the extensive mimo system of one.

Background technology:

Along with being skyrocketed through of mobile Internet and new business that cloud computing is representative, intelligent terminal becomes increasingly popular, the day by day growth of internet of things service, it is contemplated that following 10 years, mobile data services will develop at a speed unheard of before.The big advantages such as extensive MIMO is as the key technology for PHY of following 5G mobile communication, and with its safety, energy consumption is low, handling capacity is big, and efficiency is high, by world research, personnel are paid close attention to.Extensive mimo system, by increasing antenna for base station number extremely hundreds of the most thousands of, obtains huge channel capacity and spectrum efficiency.Meanwhile, the impact of channel uncorrelated noise and rapid fading will disappear, and will tend to optimum during uniform enconding method.

Along with continuing to increase of antenna for base station number, base station needs the channel condition information that accurate acquisition is current, ensures the reliability of channel.Under time division duplex (TDD) pattern, channel reciprocity principle can be passed through, obtain channel condition information.But, this will produce new problem: pilot pollution.The pilot signal Spatial Dimension sent due to terminal is limited, so the user terminal that inevitably there is different districts uses same pilot to launch simultaneously, thus causes base station cannot be distinguished by, then forms the pilot pollution problem.

The information being disclosed in this background section is merely intended to increase the understanding of the general background to the present invention, and is not construed as recognizing or imply in any form this information structure prior art well known to persons skilled in the art.

Summary of the invention:

It is an object of the invention to provide a kind of multi-district pilots abatement of pollution method under extensive mimo system, thus overcome above-mentioned defect of the prior art.

For achieving the above object, the invention provides

Multi-district pilots abatement of pollution method under extensive mimo system, the steps include:

(1) setting a cellular network with L time synchronized, wherein, the radius of single femtocell is R, and the number of this cell-site antenna is M, and user terminal number is K；The dissemination channel factor between m root antenna and the kth user of jth community of l community can be expressed as{β_jlkIt is nonnegative constant, { h_jlkmIt is normalized to unit 1, namelySo, the signal received by m root antenna of l community can be expressed as:

$y_{lm}=\underset{j=1}{\overset{L}{\Σ}}\underset{k=1}{\overset{K}{\Σ}}\sqrt{p_r{\mathrm{\β}}_{jlk}}{h}_{jlkm}{\mathrm{\ζ}}_{jk}+{\mathrm{\ω}}_{lm}---\left(1\right)$

Wherein, p_rFor user's pilot signal power, ζ_jkThe pilot frequency sequence sent for the kth user of jth community；ω_lmFor independent identically distributed average be zero, variance beWhite Gaussian noise；

(2) Y is made_l=[y_l1y_l2…y_lM], W_l=[ω_l1ω_l2…ω_lM], D_jl=diag{ [β_jl1β_jl2…β_jlK], h_jlk=[h_jlk1h_jlk2…h_jlkM], H_jl=[h_jl1h_jl2…h_jlK], then the channel matrix of two minizones can be expressed as:

$G_{jl}=D_{jl}^{1/2}{H}_{jl}---\left(2\right)$

The pilot signal that the l community receives can be expressed as:

$Y_l=\sqrt{p_r}\underset{j=1}{\overset{L}{\Σ}}{\mathrm{\Λ}}_l{G}_{jl}+W_l---\left(3\right)$

Wherein Λ_l=[ζ_l1ζ_l2…ζ_lK], use least mean-square estimate (LS), formula (3) can be expressed as:

${\hat{G}}_{ll}^{LS}=G_{ll}+\underset{j=1,j\≠l}{\overset{L}{\Σ}}{G}_{jl}+{\mathrm{\Λ}}_l^H{\hat{W}}_l---\left(4\right)$

Wherein,From formula (4), G_llFor expecting pilot signal,For the pilot signal from other cellular user terminal, it is pilot pollution；

(3) pilot pollution is eliminated:

A, channel estimation time is divided into L+1；

B, the 0th stage, named Phase 0, the Base Transmitter pilot signal that the terminal use of all honeycombs was correlated with to it within the same time in this stage, then the signal that base station, community 1 receives can be expressed as:

$Y\left(0\right)=\sqrt{p_r}{\mathrm{\ΛG}}_{11}+\sqrt{p_r}\underset{l=2}{\overset{L}{\Σ}}{\mathrm{\ΛG}}_{l1}+W\left(0\right)---\left(5\right)$

C, the 1st stage, named Phase 1, the user in community 1 keeps mourning in silence, and the user in remaining community is simultaneously to respective community pilot signal transmitted, then the signal that base station, community 1 receives can be expressed as:

$Y\left(1\right)=\sqrt{p_r}\underset{l=2}{\overset{L}{\Σ}}{\mathrm{\ΛG}}_{l1}+W\left(1\right)---\left(6\right)$

D, L-1 stage of residue, it is divided into:

D1, single community keep mourning in silence:

In remaining L-1 the stage, each stage only one of which community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted, then the signal that base station, community 1 receives can be expressed as:

$Y\left(l\right)=\sqrt{p_r}{\mathrm{\ΛG}}_{11}+\sqrt{p_r}\underset{j=2,j\≠l}{\overset{L}{\Σ}}{\mathrm{\ΛG}}_{j1}+W\left(l\right)---\left(7\right)$

In 2≤l≤L stage, Y (l) does not comprise the interference signal from l honeycomb, calculates the sum of all Y (l), then has:

$\begin{array}{c}{\tilde{Y}}_1=Y\left(0\right)+\underset{l=2}{\overset{L}{\Σ}}Y\left(l\right)-(L-1)Y\left(1\right)\\ =\widetilde{\mathrm{\Λ}}{G}_{11}+\tilde{W}\end{array}---\left(8\right)$

Wherein,

Using least square (LS) to estimate, its normal equation is

$\widehat{\mathrm{\θ}}={\left(H^{T}H\right)}^{-1}{H}^{T}x---\left(9\right)$

Thus obtain

$\begin{array}{c}{\tilde{G}}_{CE-1}={\left({\widetilde{\mathrm{\Λ}}}^H\widetilde{\mathrm{\Λ}}\right)}^{-1}{\widetilde{\mathrm{\Λ}}}^H{\tilde{Y}}_1\\ =G_{11}+\frac{1}{L^{2}K}{\widetilde{\mathrm{\Λ}}}^H\tilde{W}\end{array}---\left(10\right)$

Its estimation difference can be obtained according to formula (10)

${\mathrm{MSE}}_{1,LS}=\frac{(L^2-L+1)}{L^3{\mathrm{Kp}}_r{\mathrm{\β}}_{11}^2}{\mathrm{\σ}}_Z^2---\left(11\right)$

D2, two communities keep mourning in silence:

The two community situations of mourning in silence are divided into asymmetric two communities to mourn in silence and mourn in silence in symmetrical two communities；

D2-1, two communities are asymmetric mourns in silence, and in remaining L-1 the stage, in each stage, adjacent Liang Ge community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted；

D2-2, two community symmetries are mourned in silence, and symmetric case is divided into relative cell and is separated by community.In remaining L-1 the stage, in each stage, Liang Ge community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted；

The system estimation (LS) mourned in silence in two communities is

$\begin{array}{c}{\tilde{G}}_{CE-2}={\left({\widetilde{\mathrm{\Λ}}}^H\widetilde{\mathrm{\Λ}}\right)}^{-1}{\widetilde{\mathrm{\Λ}}}^H{\tilde{Y}}_1\\ =G_{11}+\frac{1}{L^{2}K}{\widetilde{\mathrm{\Λ}}}^H\tilde{W}\end{array}---\left(12\right)$

Wherein,

Its estimation difference is:

${\mathrm{MSE}}_{2,LS}=\frac{(L^2-3L+4)}{L^3{\mathrm{Kp}}_r{\mathrm{\β}}_{11}^2}{\mathrm{\σ}}_Z^2---\left(13\right)$

Compared with prior art, there is advantages that

Keep mourning in silence at stages respectively by being separated by symmetrical community, then realized the elimination of pilot pollution by least mean-square estimate.Simulation result shows that this method achieves preferable effect in terms of complexity, actual application, systematic function.

Accompanying drawing illustrates:

Fig. 1 is pilot pollution schematic diagram；

Fig. 2 is the present invention up phases of training schematic diagram；

Fig. 3 is to mourn in silence situation schematic diagram in different districts of the present invention；

Fig. 4 is the present invention channel estimation errors simulation result coordinate diagram with or without pilot pollution；

Fig. 5 is the system under different situations of the present invention and rate simulating result coordinate diagram；

Fig. 6 is system and rate simulating result coordinate diagram under difference path-loss factor of the present invention；

Fig. 7 is signal to noise ratio and channel estimation errors simulation result coordinate diagram under different situations of the present invention.

Detailed description of the invention:

Below the detailed description of the invention of the present invention is described in detail, it is to be understood that protection scope of the present invention is not limited by detailed description of the invention.

Explicitly indicate that unless otherwise other, otherwise in entire disclosure and claims, term " includes " or its conversion such as " comprising " or " including " etc. will be understood to comprise stated element or ingredient, and does not get rid of other element or other ingredient.

As shown in Figure 1-2, a kind of multi-district pilots abatement of pollution method under extensive mimo system, the steps include:

(1) setting a cellular network with L time synchronized, wherein, the radius of single femtocell is R, and the number of this cell-site antenna is M, and user terminal number is K；The dissemination channel factor between m root antenna and the kth user of jth community of l community can be expressed as{β_jlkIt is nonnegative constant, { h_jlkmIt is normalized to unit 1, namelySo, the signal received by m root antenna of l community can be expressed as:

$y_{lm}=\underset{j=1}{\overset{L}{\Σ}}\underset{k=1}{\overset{K}{\Σ}}\sqrt{p_r{\mathrm{\β}}_{jlk}}{h}_{jlkm}{\mathrm{\ζ}}_{jk}+{\mathrm{\ω}}_{lm}---\left(1\right)$

Wherein, p_rFor user's pilot signal power, ζ_jkThe pilot frequency sequence sent for the kth user of jth community；ω_lmFor independent identically distributed average be zero, variance beWhite Gaussian noise；

(2) Y is made_l=[y_l1y_l2…y_lM], W_l=[ω_l1ω_l2…ω_lM], D_jl=diag{ [β_jl1β_jl2…β_jlK], h_jlk=[h_jlk1h_jlk2…h_jlkM], H_jl=[h_jl1h_jl2…h_jlK], then the channel matrix of two minizones can be expressed as:

$G_{jl}=D_{jl}^{1/2}{H}_{jl}---\left(2\right)$

The pilot signal that the l community receives can be expressed as:

$Y_l=\sqrt{p_r}\underset{j=1}{\overset{L}{\Σ}}{\mathrm{\Λ}}_l{G}_{jl}+W_l---\left(3\right)$

Wherein Λ_l=[ζ_l1ζ_l2…ζ_lK], use least mean-square estimate (LS), formula (3) can be expressed as:

${\hat{G}}_{ll}^{LS}=G_{ll}+\underset{j=1,j\≠l}{\overset{L}{\Σ}}{G}_{jl}+{\mathrm{\Λ}}_l^H{\hat{W}}_l---\left(4\right)$

Wherein,From formula (4), G_llFor expecting pilot signal,For the pilot signal from other cellular user terminal, it is pilot pollution；

(3) pilot pollution is eliminated:

A, channel estimation time is divided into L+1；

B, the 0th stage, named Phase 0, the Base Transmitter pilot signal that the terminal use of all honeycombs was correlated with to it within the same time in this stage, then the signal that base station, community 1 receives can be expressed as:

$Y\left(0\right)=\sqrt{p_r}{\mathrm{\ΛG}}_{11}+\sqrt{p_r}\underset{l=2}{\overset{L}{\Σ}}{\mathrm{\ΛG}}_{l1}+W\left(0\right)---\left(5\right)$

C, the 1st stage, named Phase 1, the user in community 1 keeps mourning in silence, and the user in remaining community is simultaneously to respective community pilot signal transmitted, then the signal that base station, community 1 receives can be expressed as:

$Y\left(1\right)=\sqrt{p_r}\underset{l=2}{\overset{L}{\Σ}}{\mathrm{\ΛG}}_{l1}+W\left(1\right)---\left(6\right)$

D, L-1 stage of residue, it is divided into:

D1, single community keep mourning in silence:

In remaining L-1 the stage, each stage only one of which community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted, then the signal that base station, community 1 receives can be expressed as:

$Y\left(l\right)=\sqrt{p_r}{\mathrm{\ΛG}}_{11}+\sqrt{p_r}\underset{j=2,j\≠l}{\overset{L}{\Σ}}{\mathrm{\ΛG}}_{j1}+W\left(l\right)---\left(7\right)$

In 2≤l≤L stage, Y (l) does not comprise the interference signal from l honeycomb, calculates the sum of all Y (l), then has:

$\begin{array}{c}{\tilde{Y}}_1=Y\left(0\right)+\underset{l=2}{\overset{L}{\Σ}}Y\left(l\right)-(L-1)Y\left(1\right)\\ =\widetilde{\mathrm{\Λ}}{G}_{11}+\tilde{W}\end{array}---\left(8\right)$

Wherein,

Using least square (LS) to estimate, its normal equation is

$\widehat{\mathrm{\θ}}={\left(H^{T}H\right)}^{-1}{H}^{T}x---\left(9\right)$

Thus obtain

$\begin{array}{c}{\tilde{G}}_{CE-1}={\left({\widetilde{\mathrm{\Λ}}}^H\widetilde{\mathrm{\Λ}}\right)}^{-1}{\widetilde{\mathrm{\Λ}}}^H{\tilde{Y}}_1\\ =G_{11}+\frac{1}{L^{2}K}{\widetilde{\mathrm{\Λ}}}^H\tilde{W}\end{array}---\left(10\right)$

Its estimation difference can be obtained according to formula (10)

${\mathrm{MSE}}_{1,LS}=\frac{(L^2-L+1)}{L^3{\mathrm{Kp}}_r{\mathrm{\β}}_{11}^2}{\mathrm{\σ}}_Z^2---\left(11\right)$

D2, two communities keep mourning in silence:

As it is shown on figure 3, the two community situations of mourning in silence are divided into asymmetric two communities to mourn in silence and mourn in silence in symmetrical two communities；

D2-1, two communities are asymmetric mourns in silence, and in remaining L-1 the stage, in each stage, adjacent Liang Ge community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted；

D2-2, two community symmetries are mourned in silence, and symmetric case is divided into relative cell and is separated by community.In remaining L-1 the stage, in each stage, Liang Ge community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted；

The system estimation (LS) mourned in silence in two communities is

$\begin{array}{c}{\tilde{G}}_{CE-2}={\left({\widetilde{\mathrm{\Λ}}}^H\widetilde{\mathrm{\Λ}}\right)}^{-1}{\widetilde{\mathrm{\Λ}}}^H{\tilde{Y}}_1\\ =G_{11}+\frac{1}{L^{2}K}{\widetilde{\mathrm{\Λ}}}^H\tilde{W}\end{array}---\left(12\right)$

Wherein,

Its estimation difference is:

${\mathrm{MSE}}_{2,LS}=\frac{(L^2-3L+4)}{L^3{\mathrm{Kp}}_r{\mathrm{\β}}_{11}^2}{\mathrm{\σ}}_Z^2---\left(13\right).$

For the systematic function of check algorithm, utilize MATLAB R2014a that algorithm is emulated.The parameter of extensive MIMO TDD system is as follows:

Number of cells is L=7, and single subdistrict number of users is K=8, and single user pilot transmission power is p_r=10dB, the variance of additive noisePath-loss factor β in community_ll=1, interval path-loss factor is β_jl=a.

As shown in Figure 4, emulation is simulated for the channel estimation errors under different interval path-loss factor.In the presence of having pilot pollution, the channel estimation errors of system is continuously increased with the increase of interval path-loss factor；In the presence of not having pilot pollution, system only exists noise, and channel estimation errors does not changes with the change of path-loss factor.The existence of pilot pollution has bigger inhibition for extensive MIMO TDD system performance.

As it is shown in figure 5, for the certain condition of interval path-loss factor, the cancellation schemes under different situations, in the case of antenna for base station number is continuously increased, system and speed be ever-increasing on the whole.Liang Ge community is separated by symmetric case and is better than other communities and mourns in silence scheme.The single subdistrict scheme of mourning in silence is better than other Liang Ge communities and mourns in silence scheme simultaneously, essentially consists in single subdistrict and mourns in silence so that more community carries out downlink data transmission so that system and speed increase.

As shown in Figure 6, for the condition that antenna for base station number is certain, when the path-loss factor in community is continuously increased, system is constantly to reduce with speed.When path-loss factor is less than 0.28, the single subdistrict situation of mourning in silence is better than Liang Ge community and mourns in silence situation；When path-loss factor is more than 0.28, it is separated by the symmetrical community situation of mourning in silence and is better than other several situations of mourning in silence.

As it is shown in fig. 7, certain for antenna for base station number, the interval condition certain with path-loss factor in community, when the signal to noise ratio of system is continuously increased, channel estimation errors tapers into.The mourn in silence community channel estimation errors of situation of Liang Ge is better than single subdistrict and mourns in silence the channel estimation errors of situation, and it about improves 1.3dB.

The aforementioned description to the specific illustrative embodiment of the present invention illustrates that and the purpose of illustration.These descriptions are not wishing to limit the invention to disclosed precise forms, and it will be apparent that according to above-mentioned teaching, can much change and change.The purpose selected exemplary embodiment and describe is to explain the certain principles of the present invention and actual application thereof, so that those skilled in the art is capable of and utilizes the various different exemplary of the present invention and various different selection and change.The scope of the present invention is intended to be limited by claims and equivalents thereof.

Claims (1)

1. multi-district pilots abatement of pollution method under an extensive mimo system, it is characterised in that: the steps include:

(1) setting a cellular network with L time synchronized, wherein, the radius of single femtocell is R, and the number of this cell-site antenna is M, and user terminal number is K；The dissemination channel factor between m root antenna and the kth user of jth community of l community can be expressed as{β_jlkIt is nonnegative constant, { h_jlkmIt is normalized to unit 1, namelySo, the signal received by m root antenna of l community can be expressed as:

Wherein, p_rFor user's pilot signal power, ζ_jkThe pilot frequency sequence sent for the kth user of jth community；ω_lmFor independent identically distributed average be zero, variance beWhite Gaussian noise；

(2) Y is made_l=[y_l1y_l2…y_lM], W_l=[ω_l1ω_l2…ω_lM], D_jl=diag{ [β_jl1β_jl2…β_jlK], h_jlk=[h_jlk1h_jlk2…h_jlkM], H_jl=[h_jl1h_jl2…h_jlK], then the channel matrix of two minizones can be expressed as:

The pilot signal that the l community receives can be expressed as:

Wherein Λ_l=[ζ_l1ζ_l2…ζ_lK], use least mean-square estimate (LS), formula (3) can be expressed as:

Wherein,From formula (4), G_llFor expecting pilot signal,For the pilot signal from other cellular user terminal, it is pilot pollution；

(3) pilot pollution is eliminated:

A, channel estimation time is divided into L+1；

B, the 0th stage, named Phase 0, the Base Transmitter pilot signal that the terminal use of all honeycombs was correlated with to it within the same time in this stage, then the signal that base station, community 1 receives can be expressed as:

C, the 1st stage, named Phase 1, the user in community 1 keeps mourning in silence, and the user in remaining community is simultaneously to respective community pilot signal transmitted, then the signal that base station, community 1 receives can be expressed as:

D, L-1 stage of residue, it is divided into:

D1, single community keep mourning in silence:

In remaining L-1 the stage, each stage only one of which community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted, then the signal that base station, community 1 receives can be expressed as:

In 2≤l≤L stage, Y (l) does not comprise the interference signal from l honeycomb, calculates the sum of all Y (l), then has:

Wherein,

Using least square (LS) to estimate, its normal equation is

Thus obtain

Its estimation difference can be obtained according to formula (10)

D2, two communities keep mourning in silence:

The two community situations of mourning in silence are divided into asymmetric two communities to mourn in silence and mourn in silence in symmetrical two communities；

D2-1, two communities are asymmetric mourns in silence, and in remaining L-1 the stage, in each stage, adjacent Liang Ge community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted；

D2-2, two community symmetries are mourned in silence, and symmetric case is divided into relative cell and is separated by community.In remaining L-1 the stage, in each stage, Liang Ge community keeps mourning in silence, and remaining community is simultaneously to respective base station pilot signal transmitted；

The system estimation (LS) mourned in silence in two communities is

Wherein,

Its estimation difference is: