CN104955161A - Femtocell clustering and macro user scheduling-based interference coordination method in heterogeneous network - Google Patents

Abstract

The invention discloses a femtocell clustering and macro user scheduling-based interference coordination method in heterogeneous network, by utilizing the interference coordination of femtocell clustering and macro user scheduling in the heterogeneous network, a femtocell senses that whether the macro user exists around, and forms a femtocell interference cluster around each macro user, the interference cluster corresponding to the macro user asking for accessing a network reports feedback information to a macro base station through a feedback link, a macro cell selects a group K spare macro users for scheduling according to the feedback information, the macro cell selects (K-1) macro users to transmit data according to channel information of the macro user of the current time slot and informs the interference clusters corresponding to the rest macro users to communicate normally during the time slot, the macro user scheduling is repeated until the whole scheduling is finished. By utilizing the method, the coverage area of the macro cell is guaranteed; compared with the prior art, the method has the advantages that the more efficient frequency spectrum and power utilization rate can be realized.

Description

Based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling in heterogeneous network

Technical field

The invention belongs to cordless communication network technical field, particularly relate to the disturbance coordination method based on femtocell sub-clustering and grand user scheduling in a kind of heterogeneous network.

Background technology

Along with the explosive growth of wireless data amount and people are for the continuous lifting of data-rate requirements, new requirement is proposed to the deployment of wireless cellular network and design.Cell splitting can increase network capacity, and small cell occurs, especially femtocell (Home eNodeB) can make user side greatly reduce to the distance of base station end, thus reduces path loss.On the other hand, affix one's name to the spatial degrees of freedom that extensive antenna (massive MIMO) can increase antenna in macrocell end, increase link gain and reliability.Extensive antenna system can make the emittance of antenna focus on some specific region, thus strengthens the signal power of desired user, reduces the interference to undesired user.Therefore, dispose the heterogeneous network that the macrocell of extensive antenna and femtocell coexist and not only can provide reliable outdoor cover, the deficiency of macrocell to in-door covering can be made up simultaneously, increase the capacity of system and the satisfaction of user.But in this network configuration, macrocell and femtocell use identical frequency spectrum resource, cross-layer interference (interference between macrocell and femtocell) is the one of the main reasons threatening systematic function.

At present, have several main technology to cut down this kind of interference: one is interference cancellation techniques, this technology processes the mixed signal received, and therefrom remove interference component, this technology needs the priori knowing interference signal.Another kind is interference avoidance technique, and this technology makes each side of interference mutually use mutually orthogonal resource (such as: frequency, time, code word or space), thus plays the effect disturbed and avoid.Existing four kinds of interference reduction technology:

1. the interference eliminated in heterogeneous network is exactly the most simply carry out power control.If the through-put power of a femtocell base station can control completely, and can not be leaked to outdoor, the transfer of data of macrocell so effectively will be protected.Such as, power control techniques in indoor distributed system.But this control mode is very complicated and time-consuming.

2., in order to eliminate the interference of alternation of bed, there is a kind of decline link circuit resource allocation algorithm.Frequency spectrum resource can be divided into three parts by Macrocell, belongs to femtocell respectively and uses part and macrocell partly and common part.And the ratio of three is determined by the method optimized.

3. a cognitive method solves alternation of bed interference.By the wireless environment around perception, carry out frequency spectrum share by the mode of overlay and underlay, eliminate the interference of alternation of bed.

4. (Aliging guard zone coordination scheme) is a kind of has the algorithm eliminating interference in the heterogeneous network of the femtocell of cognitive ability.In this algorithm, femtocell needs to form some protection zones, and this can reduce the space availability ratio of communication network.

First three is planted the interference cancellation techniques mentioned and has respective limitation, and they only considered frequency domain and only there is the situation of unique user.4th kind of consultation loses the diversity gain of some users, and the femtocell in protection zone can not transmit data.

Summary of the invention

The object of the present invention is to provide the disturbance coordination method based on femtocell sub-clustering and grand user scheduling in a kind of heterogeneous network, the coverage being intended to solve the macrocell that existing interference avoids collaboration method to exist is less, the problem of frequency spectrum and power utilization complicated operation.

The present invention realizes like this, based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling in a kind of heterogeneous network, adopt the interference coordination of femtocell sub-clustering in the heterogeneous network coexisted at the macrocell and femtocell of disposing extensive antenna and grand user scheduling based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling in described heterogeneous network, whether each femtocell with cognitive ability perceives exists grand user around, a femtocell interference bunch is just defined around each grand user, the interference corresponding to grand user of request access network is bunch by feedback link to macro base station report feedback information, the grand user that macrocell selects one group of K alternative according to feedback information dispatches, macrocell is according to the channel information of the grand user of current time slots, K-1 grand user is selected to transmit data, and notify that interference corresponding to remaining grand user is bunch in time slot proper communication, repeat to dispatch grand user, and ensure that at each time slot be the different grand users that mourns in silence, until complete whole scheduling.

Further, specifically comprise the following steps based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling in described heterogeneous network:

The heterogeneous network that step one, the macrocell building the extensive antenna of deployment and femtocell coexist, femtocell and macrocell all adopt TDD and OFDM transmission pattern, FBS and UE all installs an antenna;

The existence of MUE around step 2, FBS perception and the frequency spectrum of macrocell, the femtocell spatially close from specific MUE is divided into the different IFC only disturbing MUE, dispatching cycle is T, T is divided into K time slot scheduling, and K is also the number of carrying out the MUE communicated in T with MBS;

Step 3, in time slot scheduling j, MBS only communicates with the MUE of K-1 activated state, the IFC relevant to activated state MUE is called state of mourning in silence, remaining next MUE is called state of mourning in silence, the IFC relevant to the state MUE that mourns in silence activates, normally communicate, in T time, each MUE has state of uniquely once mourning in silence;

Step 4, IFC chooser problem, first find the set { C of a mutually disjoint IFC by the method for exhaustion _jj=1,2 ..., | G|}, whether result is that an IFC only can disturb a MUE, obtains the set of MUE afterwards by iterative method, characterize optimum by class signal noise interference ratio;

Step 5, MUE user scheduling, planning step 4 basis to obtain bunch, discusses and makes information maximization in the time slot that certain is determined, obtain scheduling instruction vector, using iterative algorithm obtains dispatch matrix optimum in T time afterwards;

Step 6, method terminate.

Further, in described step one, the pattern of setting comprises: the transmission of channel model and down link.

Further, described channel model is in dispatching cycle T, and channel gain is constant;

Channel matrix: $G_{mj}=D_{mj}^{1/2}{H}_{mj};$

Wherein,

$D_{mj}^{1/2}=diag\{{\mathrm{\β}}_{m1j},{\mathrm{\β}}_{m2j},...,{\mathrm{\β}}_{mKj}\};$

${\mathrm{\β}}_{mkj}=r_{mkj}^{-a}\mathrm{\θ};$

$H_{mj}=\[h_{m1j}^T,...,h_{mKj}^T\];$

In formula be large scale fading matrix, contain path loss and shadow fading; r _mkjit is the distance in time slot j between MBS and a kth MUE; A is path loss index, and value is a ∈ [2,6]; θ is shadow fading variable, obeys distribution 10log ₁₀θ N (0, δ _shadow); Multipath fading matrix H _mjin each composition vector be independent same distribution h _mkjcN (0, I).

Further, described downlink transmission model:

The signal that in jth time slot, a kth MUE receives:

Wherein, p _fthe power that FBS launches, p _ithe power that MBS distributes to each MUE, p _kp _ione of them value, adopt constant power send mode, p is MBS total emission power, and A is all set carrying out the MUE communicated in T; x _mkjand x _mijit is the distribution that I.i.d. random variables obeys CN (0,1); g _mkjand g _fkjmBS channel gain and FBS channel gain respectively;

In like manner, the signal that FUE receives is obtained:

$y_{ffj}=\sqrt{p_F}{g}_{ffj}{x}_{ffj}+\underset{l\∈F,l\≠f}{\Σ}\sqrt{p_F}{g}_{lfj}{x}_{llj}+\underset{k\∈A}{\Σ}\sqrt{p_i}{w}_{mkj}{g}_{mfj}{x}_{mkj}+n;$

Wherein F is the set of the base station in IFC.

Further, the data rate in step one:

The Signal Interference and Noise Ratio of MUE is defined as:

The Signal Interference and Noise Ratio of FUE is defined as: ${\mathrm{\γ}}_{ffj}=\frac{p_F{||g_{ffj}||}^2}{\underset{l\∈F,l\≠f}{\Σ}{p}_F{||g_{lfj}||}^2+\underset{k\∈A}{\Σ}{p}_k{||w_{mkj}{g}_{mfj}||}^2+{\mathrm{\δ}}_n^2};$

Have according to Shannon's theorems:

In time slot j, a kth MUE transmitted data rates is: R _mkj=log ₂(1+ γ _mkj);

In IFC relevant to a kth MUE in time slot j, certain FUE transmitted data rates is: R _ffj=log ₂(1+ γ _ffj);

Transmission rate and time slot j have nothing to do, and obtain the data rate that in IFC relevant to a kth MUE in any one time slot, all FUE carry out to be

Further, described step 3 carries out sub-clustering selection and user scheduling method comprises after learning transmission model and downlink transmission state:

Wherein, Γ _kj=1 represents that a kth equipment is in activated state in a jth time slot, normally communicates, otherwise be in state of mourning in silence, target function is the maximization of message transmission rate total in T whole dispatching cycle;

The problem at the data rate optimum of whole time is replaced by the maximum transmission data rate in a time slot j:

Further, described step 4 concrete methods of realizing is:

Wherein A ⁽ⁱ⁾, be respectively candidate and dispatch grand user set and candidate's interference bunch set, R (i) for it before the class Signal to Interference plus Noise Ratio that defines, | G| is the number of the interference that obtains of the method for exhaustion bunch.

Further, described step 5 user scheduling problem method is as follows:

$\begin{array}{c}\underset{{\mathrm{\Γ}}_{kj}}{\max }\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}\left(R_{Fk}-R_{mkj}\right)\\ S_1:\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}=1\\ S_2:{\mathrm{\Γ}}_{kj}\∈\left\{0,1\right\},\∀k\∈\left\{1,2,...,\left|A\right|\right\}\end{array}.$

Based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling in heterogeneous network provided by the invention, not only ensure that the coverage of macrocell, and provide extra transmission opportunity for the FBS of interference MUE.This technology completes interference coordination between macrocell and femtocell by carrying out sub-clustering selection and user scheduling to the femtocell of the grand user of interference.Simultaneously, be easy in practice to make dispatching algorithm realize, the present invention have also been devised a kind of hierarchical solving algorithm of low complex degree, be two subproblems by former PROBLEM DECOMPOSITION: femtocell sub-clustering problem and grand user scheduling problem, grand user scheduling instruction vector is calculated by solving this two problems, by the grand user that mourns in silence in different time slot choose reasonable, thus provide communication opportunity for femtocell bunch, thus realize interference coordination, compared with prior art can realize frequency spectrum and power utilization more efficiently.

Accompanying drawing explanation

Fig. 1 is the disturbance coordination method flow chart based on femtocell sub-clustering and grand user scheduling in the heterogeneous network that provides of the embodiment of the present invention;

Fig. 2 is the antenna number N of the MBS that the embodiment of the present invention provides _tfor frequency spectrum and the power utilization simulation result schematic diagram of three kinds of agreements when independent variable;

The frequency spectrum of three kinds of agreements that Fig. 3 is the number K of the MUE that the embodiment of the present invention provides when being independent variable and power utilization simulation result schematic diagram;

Fig. 4 is the transmitting power p of the FBS that the embodiment of the present invention provides _ffor frequency spectrum and the power utilization simulation result schematic diagram of three kinds of agreements when independent variable.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with embodiment, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

The present invention proposes the disturbance coordination method based on femtocell sub-clustering and grand user scheduling in a kind of heterogeneous network, by carrying out sub-clustering to the femtocell of the grand user of interference, sub-clustering selection and user scheduling complete the interference coordination between macrocell and femtocell.Simultaneously, be easy in practice to make dispatching algorithm realize, the present invention have also been devised a kind of hierarchical solving algorithm of low complex degree, be two subproblems by former PROBLEM DECOMPOSITION: femtocell sub-clustering problem and grand user scheduling problem, grand user scheduling instruction vector is calculated by solving this two problems, by the grand user that mourns in silence in different time slot choose reasonable, thus provide communication opportunity for femtocell bunch, thus realize interference coordination.

Below in conjunction with accompanying drawing 1, application principle of the present invention is explained in detail.

As shown in Figure 1, the disturbance coordination method based on femtocell sub-clustering and grand user scheduling in the heterogeneous network of the embodiment of the present invention comprises the following steps:

Whether S101: under heterogeneous network environment, each femtocell with cognitive ability can perceive exists grand user around, just defines femtocell interference bunch (IFC) around each grand user;

S102: the interference corresponding to grand user of request access network is bunch by feedback link to macro base station report feedback information, and packets of information contains: in interference bunch, all femtocell send the identity (ID) of the grand user corresponding to the estimated value of data rate and interference bunch;

S103: macrocell selects one group (K) alternative grand user to dispatch according to feedback information, and the feature of this kind of grand user is, the interference bunch corresponding to grand user should comprise mutually different femtocell;

S104: macrocell, according to the channel information of grand user of current time slots, selects K-1 grand user to transmit data, and notify that interference corresponding to remaining grand user (mourn in silence grand user) is bunch in this time slot proper communication;

S105: the step repeating S104, and ensure that at each time slot be the different grand users that mourns in silence, until complete whole dispatching cycle.

Below in conjunction with specific embodiments of the invention, principle of the present invention is further described:

Specifically comprise the following steps based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling in the heterogeneous network of the embodiment of the present invention:

The heterogeneous network that step one, the macrocell building an extensive antenna of deployment and femtocell coexist, femtocell and macrocell all adopt TDD and OFDM transmission pattern, suppose that FBS (femtocell base station) and UE (user equipment) all installs an antenna, and understood channel condition information completely;

Step 2, FBS can the existence of MUE (grand user) around perception and the frequency spectrums of macrocell, the femtocell spatially close from specific MUE is divided into the different IFC only disturbing this MUE, dispatching cycle is T, T is divided into K time slot scheduling (identical for the correlated time of channel with MUE), K is also the number of carrying out the MUE communicated in T with MBS (macrocell base station);

Step 3, in time slot scheduling j, MBS only communicates with the MUE of (K-1) individual activated state, the IFC relevant to activated state MUE is called state of mourning in silence (not communicating in j), remaining next MUE is called state of mourning in silence, the IFC relevant to the state MUE that mourns in silence activates, normally communicate, in T time, each MUE can have state of uniquely once mourning in silence, and is divided into two subproblems because above-mentioned algorithm is too complicated;

Step 4, IFC (interference bunch) chooser problem, first find the set { C of a mutually disjoint IFC by the method for exhaustion _jj=1,2 ..., | G|}, whether result is that an IFC only can disturb a MUE, obtains the set of MUE afterwards by iterative method, characterize optimum by class signal noise interference ratio;

Step 5, MUE user scheduling, planning step 4 basis to obtain bunch, discusses and makes information maximization in the time slot that certain is determined, obtain scheduling instruction vector, using iterative algorithm obtains dispatch matrix optimum in T time afterwards;

Step 6, method terminate.

The present invention first sets up traffic model, on this basis, obtains optimum sub-clustering selection and user scheduling by iterative algorithm; In the heterogeneous network of macrocell mixing femtocell, by application said method, the higher availability of frequency spectrum and power efficiency can be obtained; By being two subproblems by this PROBLEM DECOMPOSITION, make this algorithm can physics realization.

Comprise further in the embodiment of the present invention:

Under the model of setting in step one, the transmission of channel model and down link:

1.1 channel models: think that channel gain is constant in dispatching cycle T, traditional only considers that the channel of multipath fading does not gear to actual circumstances, and is not also suitable for proposed method, so only consider the characteristic of large scale fading channel.

Channel matrix: $G_{mj}=D_{mj}^{1/2}{H}_{mj};$

Wherein,

$D_{mj}^{1/2}=diag\{{\mathrm{\β}}_{m1j},{\mathrm{\β}}_{m2j},...,{\mathrm{\β}}_{mKj}\};$

${\mathrm{\β}}_{mkj}=r_{mkj}^{-a}\mathrm{\θ};$

$H_{mj}=\[h_{m1j}^T,...,h_{mKj}^T\];$

In formula be large scale fading matrix, contain path loss and shadow fading.R _mkjit is the distance in time slot j between MBS and a kth MUE.A is path loss index, and value is a ∈ [2,6].θ is shadow fading variable, obeys distribution 10log ₁₀θ N (0, δ _shadow).Multipath fading matrix H _mjin each composition vector be independent same distribution h _mkjcN (0, I).

1.2 downlink transmission models:

The signal that in jth time slot, a kth MUE receives:

Wherein, p _fbe the power that FBS launches, suppose that it is a definite value.P _ithe power (p that MBS distributes to each MUE _kp _ione of them value), conveniently, adopt constant power send mode, (P is MBS total emission power, and A is all set carrying out the MUE communicated in T).X _mkjand x _mijit is the distribution that I.i.d. random variables obeys CN (0,1).G _mkjand g _fkjmBS channel gain and FBS channel gain respectively.

In like manner, the signal that FUE receives can also be obtained:

$y_{ffj}=\sqrt{p_F}{g}_{ffj}{x}_{ffj}+\underset{l\∈F,l\≠f}{\Σ}\sqrt{p_F}{g}_{lfj}{x}_{llj}+\underset{k\∈A}{\Σ}\sqrt{p_i}{w}_{mkj}{g}_{mfj}{x}_{mkj}+n;$

(wherein F is the set of the base station in IFC).

1.3 data rates:

The Signal Interference and Noise Ratio of MUE is defined as:

The Signal Interference and Noise Ratio of FUE is defined as: ${\mathrm{\γ}}_{ffj}=\frac{p_F{||g_{ffj}||}^2}{\underset{l\∈F,l\≠f}{\Σ}{p}_F{||g_{lfj}||}^2+\underset{k\∈A}{\Σ}{p}_k{||w_{mkj}{g}_{mfj}||}^2+{\mathrm{\δ}}_n^2};$

Have according to Shannon's theorems:

A) in time slot j, a kth MUE transmitted data rates is: R _mkj=log ₂(1+ γ _mkj);

B) in relevant to a kth MUE in time slot j IFC, certain FUE transmitted data rates is: R _ffj=log ₂(1+ γ _ffj);

C) suppose that b), transmission rate and time slot j have nothing to do, obtaining the data rate that in IFC relevant to a kth MUE in any one time slot, all FUE carry out is

Realization of the present invention is also: step 3 carries out sub-clustering selection and user scheduling method after knowing transmission model and downlink transmission state:

Be used in the optimization of message transmission rate total in T whole dispatching cycle to weigh the optimality of sub-clustering selection and user scheduling.S ₂constraints ensure that in each j time slot to be had and only has a MUE to mourn in silence.S ₃ensure that the slot length of mourning in silence that each MUE can be average in dispatching cycle T.

Consider by the problem of the maximum transmission data rate replacement in a time slot j at the data rate optimum of whole time:

Realization of the present invention is also: step 4 IFC chooser problem, and first IFC chooser problem finds the set { C of a mutually disjoint IFC by the method for exhaustion _jj=1,2 ..., | G|}, whether result is that an IFC only can disturb a MUE, obtains the set of MUE afterwards by iterative method, characterize optimum by class signal noise interference ratio.Concrete methods of realizing is:

Realization of the present invention is also: step 5 user scheduling problem, and algorithm is as follows:

$\begin{array}{c}\underset{{\mathrm{\Γ}}_{kj}}{\max }\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}\left(R_{Fk}-R_{mkj}\right)\\ S_1:\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}=1\\ S_2:{\mathrm{\Γ}}_{kj}\∈\left\{0,1\right\},\∀k\∈\left\{1,2,...,\left|A\right|\right\}\end{array};$

Find bunch selection optimum in a time slot by said method, in follow-up selection scheme, do not repeated to select by the IFC selected before only needing to ensure, then K time slot just completes the scheduling of one-period after terminating.

Simulation result of the present invention

The setting of parameter: hypothesis only has a FUE to access femtocell in time slot j; Assuming that the number of femtocell in IFC is (15); The channel gain of femtocell only considers that large scale declines; Carrier frequency is 2.6GHz; Macrocell covering radius is 1km; Power Spectrum of White Noise density-174dBm/Hz; Channel width 20MHz; The shade standard deviation 8dB of MBS; Path loss index a=4; Femtocell covering radius 10m; The shade standard deviation 6dB of FBS.

Fig. 2 illustrates the antenna number N of MBS _tfor frequency spectrum and the power utilization simulation result schematic diagram of three kinds of agreements when independent variable; MBS through-put power P=46dBm; FBS transmitting power is p _f=10dBm; Scheduling MUE number K=10; Can show that CSUS agreement (bunch selection and scheduling agreement) is eliminated interference agreement than AGZ agreement (Aliging guard zone coordination scheme) and tradition and had higher frequency spectrum and power utilization.Systematic function improves along with the increase of antenna number.

The frequency spectrum of three kinds of agreements and power utilization simulation result schematic diagram time the number K that Fig. 3 illustrates MUE is independent variable; MBS through-put power P=46dBm; FBS transmitting power is p _f=10dBm; The antenna number N of MBS _t=200; As can be seen from Figure 4 the performance of system improves along with the increase of the number of MUE; But the power efficiency hoisting velocity of MUE reduces along with the increase of MUE number; This is because the increase of MUE number can make increasing femtocell enter in IFC the chance losing normal transmission data; These femtocell can carry out interference cooperation, can not be higher than normal data transfer efficiency; And then whole system hydraulic performance decline.

Fig. 4 illustrates the transmitting power p of FBS _ffor frequency spectrum and the power utilization simulation result schematic diagram of three kinds of agreements when independent variable; MBS through-put power P=46dBm; The antenna number N of MBS _t=200; Scheduling MUE number K=10; As can be seen from the figure, when the transmitting power of femtocell increases, the systematic function that tradition eliminates interference agreement can not improve; And AGZ and CSUS consultation first increases rear reduction.The reason of this phenomenon is caused to have two 1) increase of the transmitting power of the femtocell signal power that not only can improve FBS also can improve the power of high people simultaneously; 2) increase of the transmitting power of femtocell can be the femtocell increase in IFC, makes this part femtocell lose a part of transmission opportunity.

The present invention disposes in the heterogeneous network that the macrocell of extensive antenna and femtocell coexist at one and proposes a kind of innovative interference collaboration method (name of this method is called CSUS); This method comprises sub-clustering and selects and grand user scheduling; First, MBS selects the set of the IFC comprising the different femtocell relevant to MUE, to meet the requirement of Signal Interference and Noise Ratio; MUE different afterwards mourns in silence in different time slots, and relative IFC activation of having an opportunity is communicated; Other IFCs relevant to activated state MUE are in state of mourning in silence.Become attainable to make agreement and be divided into two subproblems: sub-clustering selection and user scheduling.Simulation result shows, and CSUS method can improve frequency spectrum and the power usage efficiency of system to a great extent.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1. in heterogeneous network based on a disturbance coordination method for femtocell sub-clustering and grand user scheduling, it is characterized in that, specifically comprise the following steps based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling in described heterogeneous network:

The heterogeneous network that step one, the macrocell building the extensive antenna of deployment and femtocell coexist, femtocell and macrocell all adopt TDD and OFDM transmission pattern, FBS and UE all installs an antenna;

The existence of MUE around step 2, FBS perception and the frequency spectrum of macrocell, the femtocell spatially close from specific MUE is divided into the different interference femtocell bunch only disturbing MUE, dispatching cycle is T, T is divided into K time slot scheduling, and K is the number of carrying out the MUE communicated in T with MBS;

Step 3, in time slot scheduling j, the MUE of MBS and K-1 activated state communicates, the IFC relevant to activated state MUE is called state of mourning in silence, remaining next MUE is called state of mourning in silence, the IFC relevant to the state MUE that mourns in silence activates, normally communicate, in T time, each MUE has state of uniquely once mourning in silence;

Step 4, IFC chooser problem, first find the set { C of a mutually disjoint IFC by the method for exhaustion _jj=1,2 ..., | G|}, whether result is that an IFC only can disturb a MUE, obtains the set of MUE afterwards by iterative method, characterize optimum by class signal noise interference ratio;

Step 5, MUE user scheduling, planning step 4 basis to obtain bunch, discusses and makes information maximization in the time slot that certain is determined, obtain scheduling instruction vector, using iterative algorithm obtains dispatch matrix optimum in T time;

Terminate based on the interference coordination of femtocell sub-clustering and grand user scheduling in step 6, heterogeneous network.

2. in heterogeneous network as claimed in claim 1 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, whether described step 2 femtocell can perceive exists grand user around, around each grand user, form femtocell interference bunch, the interference corresponding to grand user of request access network is bunch by feedback link to macro base station report feedback information.

3. in heterogeneous network as claimed in claim 1 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, described step 3 carries out making in sub-clustering selection one of them grand user initiatively to mourn in silence, and the interference bunch for correspondence provides the chance of extra transmission information and whether optimum by the whether maximum characterization model of information rate.

4. in heterogeneous network as claimed in claim 1 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, described step 5 user scheduling problem method, under the prerequisite finding choose the best cluster in a time slot, do not repeated to select by the IFC selected, then K time slot just completes the scheduling of one-period after terminating.

5. in heterogeneous network as claimed in claim 1 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, in described step one, the pattern of setting comprises: the transmission of channel model and down link.

6. in heterogeneous network as claimed in claim 5 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, described channel model is in dispatching cycle T, and channel gain is constant;

Channel matrix: $G_{mj}=D_{mj}^{1/2}{H}_{mj},$

Wherein,

$D_{mj}^{1/2}=diag\{{\mathrm{\β}}_{m1j},{\mathrm{\β}}_{m2j},\mathrm{...},{\mathrm{\β}}_{mKj}\};$

${\mathrm{\β}}_{mkj}=r_{mkj}^{-a}\mathrm{\θ};$

$H_{mj}=\[h_{m1j}^T,\mathrm{...},h_{mKj}^T\];$

In formula $D_{mj}^{1/2}$ Be large scale fading matrix, contain path loss and shadow fading; r _mkjit is the distance in time slot j between MBS and a kth MUE; A is path loss index, and value is a ∈ [2,6]; θ is shadow fading variable, obeys distribution 10log ₁₀θ N (0, δ _shadow); Multipath fading matrix H _mjin each composition vector be independent same distribution h _mkjcN (0, I).

7. in heterogeneous network as claimed in claim 5 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, described downlink transmission model:

The signal that in jth time slot, a kth MUE receives:

Wherein, p _fthe power that FBS launches, p _ithe power that MBS distributes to each MUE, p _kp _ione of them value, adopt constant power send mode, p is MBS total emission power, and A is all set carrying out the MUE communicated in T; x _mkjand x _mijit is the distribution that I.i.d. random variables obeys CN (0,1); g _mkjand g _fkjmBS channel gain and FBS channel gain respectively;

In like manner, the signal that FUE receives is obtained:

$y_{ffj}=\sqrt{p_F}{g}_{ffj}{x}_{ffj}+\underset{1\∈F,l\≠f}{\Σ}\sqrt{p_F}{g}_{lfj}{x}_{llj}+\underset{k\∈A}{\Σ}\sqrt{p_i}{w}_{mkj}{g}_{mfj}{x}_{mkj}+n;$

Wherein F is the set of the base station in IFC.

8. in heterogeneous network as claimed in claim 1 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, the data rate in step one:

The Signal Interference and Noise Ratio of MUE is defined as:

The Signal Interference and Noise Ratio of FUE is defined as: ${\mathrm{\γ}}_{ffj}=\frac{p_F\left|\right|g_{ffj}|{|}^2}{\underset{l\∈F,l\≠f}{\Σ}{p}_F\left|\right|g_{lfj}|{|}^2+\underset{k\∈A}{\Σ}{p}_k\left|\right|w_{mkj}{g}_{mfj}|{|}^2+{\mathrm{\δ}}_n^2};$

Have according to Shannon's theorems:

In time slot j, a kth MUE transmitted data rates is: R _mkj=log ₂(1+ γ _mkj);

In IFC relevant to a kth MUE in time slot j, certain FUE transmitted data rates is: R _ffj=log ₂(1+ γ _ffj);

Transmission rate and time slot j have nothing to do, and obtain the data rate that in IFC relevant to a kth MUE in any one time slot, all FUE carry out to be

9. in heterogeneous network as claimed in claim 1 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, described step 3 carries out sub-clustering selection after learning transmission model and downlink transmission state and user scheduling method comprises:

$\underset{{\mathrm{\Γ}}_{kj}}{max}\underset{j=1}{\overset{K}{\Σ}}\underset{k=1}{\overset{K}{\Σ}}(1-{\mathrm{\Γ}}_{kj})R_{mkj}+\underset{j=1}{\overset{K}{\Σ}}\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}{R}_{Fk}$

$S_2:\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}=1$

$S_3:\underset{j=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}=1$

$S_4:{\mathrm{\Γ}}_{kj}\∈\{0,1\},\∀k,j\∈\{1,2,\mathrm{...},K\}$

Wherein, Γ _kj=1 represents that a kth equipment is in activated state in a jth time slot, normally communicates, otherwise be in state of mourning in silence, target function is the maximization of message transmission rate total in T whole dispatching cycle;

The problem at the data rate optimum of whole time is replaced by the maximum transmission data rate in a time slot j:

$\underset{{\mathrm{\Γ}}_{kj}}{max}\underset{k=1}{\overset{K}{\Σ}}(1-{\mathrm{\Γ}}_{kj})R_{mkj}+\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}{R}_{Fk}$

$S_3:{\mathrm{\Γ}}_{kj}\∈\{0,1\},\∀k\∈\{1,2,\mathrm{...},|A\left|\right\}$

10. in heterogeneous network as claimed in claim 1 based on the disturbance coordination method of femtocell sub-clustering and grand user scheduling, it is characterized in that, described step 4 concrete methods of realizing is:

${\mathrm{Step}}_1:M_C^{\left(0\right)}=\{C_i\∩C_{l}fori,j=1,\mathrm{...},|G|,andi\≠j\}$

Iteration terminates; Otherwise forward step3 to, perform circulation, until i=K.

${\mathrm{Step}}_3:M_C^{\left(i\right)}=G\backslash A^{(i-1)}$

Step ₄: class signal noise interference ratio $R^{\left(i\right)}\left(k\right)=\frac{p_i\left|\right|w_{mkj}{h}_{mk}|{|}^2}{\underset{g\∈G,g\≠i}{\Σ}\left|\right|w_{mgj}{g}_{mkj}|{|}^2+{\mathrm{\δ}}_n^2}$

Step ₅: selection criterion $A\left(i\right)=\arg \underset{k\∈M_C^{\left(i\right)}}{max}{R}^{\left(i\right)}\left(k\right),$ Return Step2

Wherein A ⁽ⁱ⁾, be respectively candidate and dispatch grand user set and candidate's interference bunch set, R (i) for it before the class Signal to Interference plus Noise Ratio that defines, | G| is the number of the interference that obtains of the method for exhaustion bunch;

Described step 5 user scheduling problem method is as follows:

$\underset{{\mathrm{\Γ}}_{kj}}{max}\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_{kj}(R_{Fk}-R_{mkj})$

$S_1:\underset{k=1}{\overset{K}{\Σ}}{\mathrm{\Γ}}_k=1.$

$S_2:{\mathrm{\Γ}}_{kj}\∈\{0,1\},\∀k\∈\{1,2,\mathrm{...},|A\left|\right\}$