CN101784119B - Distribution method of OFDMA (Orthogonal Frequency Division Multiple Access) distributed antenna network resources - Google Patents

Abstract

The invention discloses a distribution method of OFDMA (Orthogonal Frequency Division Multiple Access) distributed antenna network resources, which comprises the steps that: S1. a user terminal feedbacks channel state information of links by which the user terminal reaches all DAPs (Distributed Antenna Ports) on all subbands to a base station, which is used as a condition for the base station to use largest ratio transmission to carry out transmission; S2. the base station distributes the subbands to the user terminal according to the channel state information; S3. aiming at each subband distributed to the user terminal, the base station selects the DAPs the channel gain of which reaches a preset value for each subband as a DAP primary selection result so as to obtain a DAP set serving each subband; and S4. the power of each DAP is distributed to the subbadS served by the DAP by an iterated water-filling algorithm. The technical scheme of the invention can improve system throughput rate and has higher practicability.

Description

The distribution method of OFDMA spaced antenna Internet resources

Technical field

The present invention relates to the mobile communication technology field, relate in particular to a kind of distribution method of OFDMA spaced antenna Internet resources.

Background technology

Two Main Stage have been experienced in the development of mobile communications network so far, and 3G (Third Generation) Moblie (3G) has been at present the later stage of development, and the relevant criterion work of the 4th third-generation mobile communication is already in full swing and in the middle of carrying out.Along with the continuous increase of development of Communication Technique and user's request, the finiteness of frequency spectrum resource makes its use cost increase gradually.Therefore, the spectrum utilization efficiency that how to increase system has become people and has paid close attention to the focus with research.

(Multiple-Input Multiple-Out-put, the theoretical research that MIMO) technology is relevant is that more the wireless access of high power capacity provides new approaches to multiple-input and multiple-output.Under the support of the many antennas in base station, system has obtained higher spatial degrees of freedom.(DistributedAntenna System DAS) as a kind of MIMO mode of more broad sense, is dispersed in many antennas of base station end in its overlay area, and is connected to a central control unit through modes such as optical fiber distributing antenna system.In DAS, (Distributed AntennaPort DAP) to same user, has different path loss, shadow fading and declines soon different spaced antenna ports.This also is DAS and traditional centralized antenna system (centralized Antenna System, CAS) main distinction on network configuration.On the one hand; Compare among the CAS with the base station is that unit carries out the minizone base station collaboration and handles presence of intercell interference; DAS can coordinate all users in a plurality of antenna port service coverage areas through central control unit effectively; Utilize grand branch collection to bring more gain, be mainly reflected in aspects such as improving received signal quality, increase power system capacity and reduction power loss; In addition, the node number of DAS key-course rolls up than CAS, and these nodes have been for the user provides the chance that inserts backbone network more easily, also meet network configuration by verticalization to the flattening Development Trend.In view of this, DAS has incomparable advantage.

Along with the increase of user to the radio multimedium business demand; Wireless communication system need be strengthened the message transmission rate that the physical layer key technology improves system on the one hand, also needs RRM strategy efficiently to come to arrange effectively and reasonably the assignment problem of multiple radio resource on the other hand.(Orthogonal Frequency Division Multiplexing, OFDM) technology contrary frequency selectivity decline effectively occupies critical role to OFDM in the future wireless system.Insert (OrthogonalFrequency Division Multiple Access at an orthogonal frequency division multiplexing multiple access; OFDMA) in the system; Factors such as the dynamic assignment of subband, the loading of the self adaptation of power and multi-user diversity make that the resource allocation flexibility ratio of system is very high.

Brought new challenge and the OFDMA-DAS system is resource allocation, traditional resource distribution mode has been not suitable for the requirement of this new system.

On the one hand, the OFDMA-DAS technology has increased the selection of flexibility ratio a: DAP on the basis of dynamic subband, power division.Compare with traditional multiaerial system, different DAP have different path loss, shadow fading and decline soon to same user.In DAS, people have proposed the notion of " virtual subdistrict ", that is, do not having under the prerequisite of other restriction, and the user always inserts the own several nearer DAP links of distance, makes the user always be in the center of " virtual subdistrict " be made up of these several DAP links.Under the effect of " virtual subdistrict ", the user can accept better transmission service on the one hand, and the power efficiency of system also is improved on the other hand.

It is exactly that each spaced antenna port all receives the limited restriction of single port maximum power that DAS also has the another one important characteristic.When near the user density certain root DAP link is big; The transmitted power of this port can only satisfy the transmission requirement of certain customers; All the other users will have to insert DAP far away relatively, this means that these ports will expend the path loss that some power resources go to compensate these users far away relatively.Therefore, in an OFDMA-DAS system, the DAP selection under the port power limited prerequisite, the dynamic assignment of subband and the self adaptation of power load, and make that the resource allocation problem of OFDMA-DAS is more flexible.

Therefore, to the characteristics of OFDMA spaced antenna network, set up the framework of resource allocation problem; Take all factors into consideration aspects such as antenna port selection, allocation of subbands and power loading, design a kind of resource allocation methods of novel practical, fully excavate the dimension of OFDMA spaced antenna Resource Allocation in Networks; Improve effective utilization rate of resource; Increase user satisfaction, the enhanced system throughput has become the important topic that OFDMA spaced antenna network needs to be resolved hurrily.

Summary of the invention

The technical problem that (one) will solve

To defective that exists in the prior art and deficiency, the purpose of this invention is to provide a kind of distribution method of OFDMA spaced antenna Internet resources, the throughput that it can improve system has stronger practicality.

(2) technical scheme

For achieving the above object, the invention provides a kind of distribution method of OFDMA spaced antenna Internet resources, may further comprise the steps:

S1, according to the structure of OFDMA spaced antenna network, user terminal feeds back to the base station with its link channel condition information on each subband to each spaced antenna port DAP, adopts high specific to send the condition of transmission signals as this base station;

S2, said base station is the user terminal allocated subbands according to said channel condition information;

S3; The subband of getting to each user terminal; DAP or all DAP in the said OFDMA spaced antenna network that said base station reaches preset value for the gain of each subband selective channel obtain the DAP set into each subband service as the initial selected result of DAP;

S4, through the iteration water-filling algorithm respectively with the power division of each DAP to each subband of this DAP service.

Wherein, the water injection power distribution method to a DAP is found the solution based on following optimization problem among the said step S4:

1. the target of this optimization problem is a maximize system throughput;

2. the constraints of optimization problem is: the port Power Limitation of this DAP, and the sets of subbands of this DAP service;

3. when calculating the water injection power allocation result of a DAP; The to be sent power of other each DAP on its subband of serving separately keeps constant, and the power transmission strategy of this DAP on the subband of its service determined by following two factors: the channel situation of this DAP on subband; The transmitted power of other each DAP on each subband be to the contribution of subband received signal to noise ratio, wherein, above-mentioned first factor and this DAP power positive correlation to be sent on subband, second factor and this DAP sending out on subband are sent the power negative correlation;

4. finding the solution through Lagrangian method of this optimization problem asks extreme value to accomplish;

5. try to achieve after the extreme value, this DAP upgrades its power to be sent on the subband of service, if its power to be sent on a subband is zero, then this DAP is removed from the DAP set for this subband service, to accomplish the optimization that DAP selects.

Wherein, after the optimization through said DAP distribution, this DAP remains unchanged when other DAP carries out computing in this iteration at the power to be sent on the subband of its service.

Wherein, the iteration convergence condition of said iteration water-filling algorithm is: after the iteration, the variable quantity of the more preceding iteration of the throughput of system is during less than preset constant, iteration convergence.

Wherein, said DAP is for guaranteeing to cover antenna port or the relay station of setting up under the cellular cell framework.

Wherein, adopt high specific to send when transmitting, DAP for send each component in the weighing vector that meets design be independently carry out normalized.

Wherein, the initial value of the to be sent power of said each DAP on its subband of serving separately is through dividing equally acquisition with the power of DAP to the subband of its service.

Wherein, When said base station reaches the DAP of preset value for each subband selective channel gain; This preset value has determined to be the DAP quantity upper limit of subband service that the throughput positive correlation of the size of this quantity and system is with the operand negative correlation in the said iteration water-filling algorithm.

(3) beneficial effect

Compared with prior art; The present invention can produce following beneficial effect: through selective channel gain preferably some spaced antenna ports accomplished initial selected to antenna port, and accomplish further optimizing and revising of power division and antenna port through the iteration water-filling algorithm.The flexibility of resource allocation that this scheme has not only utilized OFDMA system dynamics port and power division to bring; And embodied the advantage of virtual subdistrict under the distributing antenna system; The degree of freedom to the user selects to insert antenna port has been carried out effective excavation; Finally can improve the throughput of system, and, therefore also have stronger practicality because algorithm complex can be adjusted through limiting the port number that inserts on each subband.

Description of drawings

Fig. 1 is the situation sketch map of the OFDMA spaced antenna Resource Allocation in Networks of the embodiment of the invention;

Fig. 2 is the schematic process flow diagram of the method for the embodiment of the invention;

Fig. 3 be the method for the embodiment of the invention under the situation of different virtual cell size with the comparative result figure of two kinds of rudimentary algorithms;

Fig. 4 be the method for the embodiment of the invention under different virtual cell size situation, the convergence situation sketch map of iterative power water filling.

Embodiment

Below in conjunction with accompanying drawing and embodiment, specific embodiments of the invention describes in further detail.Following examples are used to explain the present invention, but are not used for limiting scope of the present invention.

(Distributed Antenna Network DAN) is the macrocell that is made up of a plurality of antenna ports that are distributed in the overlay area to OFDMA spaced antenna network.At least comprise a base station BS (Base Station) in each macrocell, one or more DAP and one or more mobile terminal MS (Mobile Station) (being also referred to as " user terminal " in the present invention) as central control unit.MS can directly pass through BS or DAP access network.Among the present invention, the physical layer of system adopts the mode of OFDMA.

Fig. 1 shows the situation of OFDMA spaced antenna Resource Allocation in Networks of the present invention, and wherein, in general the resource allocation of system is mainly reflected in subband (band resource) distribution, power division and antenna port selection in the time slot scheduling.As shown in Figure 1, in a macrocell scope, the DAP (base station itself also possesses a DAP) of total M the responsible signal transmitting and receiving of system, each DAP is connected with the base station through ideal styles such as optical fiber.The overall process of resource allocation is accomplished as CCC in the base station.Distribution state among Fig. 1 shows that user 1 has obtained subband 1 and 2, and wherein subband 1 has inserted base station and No. 2 and No. 3 antenna ports, and subband 2 has inserted base station and No. 3 and No. 4 antenna ports.Similarly, user 2 has obtained subband 3, and has inserted base station and No. 1 antenna port.For the base station, it has participated in the transmission of 3 subbands simultaneously, and its port power will reasonably be assigned in the middle of these 3 subbands or other subband that relates to.And how each subband distributes to the user, and how each subband selects to insert antenna port, and how each antenna port to remove to distribute power in the sets of subbands of its service, is exactly the problem that the present invention will solve.

Fig. 2 shows the method flow diagram of OFDMA spaced antenna Resource Allocation in Networks of the present invention.Mainly comprise following steps: the user obtains to the channel condition information of the link between each DAP; The base station is user's allocated subbands; To the subband that each user gets, some DAP or all interior DAP of said OFDMA spaced antenna network of dynamically selecting its channel gain to reach preset value serve; With the maximize system throughput is that (maximize system throughput is meant target: calculate the normalization transmission rate according to each subband received signal to noise ratio through shannon formula; Then the normalization transmission rate of each subband summation and with its maximization), the gross power water filling of each DAP is assigned on its some subbands that relate to.Successively each spaced antenna port is carried out independently power water filling and distribute, accomplish one time iteration; Continue the result convergence of iteration until power division.

Specify the method for the embodiment of the invention below.

In Fig. 1, suppose that each macrocell comprises M spaced antenna port in the network, the terminal port antennae number is 1, and the system user number is K, and system bandwidth is divided into and is N subband.Use h _k[m, n, t] expression t moment spaced antenna port m is to the channel impulse response of link on subband n between the user k.Suppose the subband n that distributed to user k for one, system will be user's transmission diversity data through one or several DAP simultaneously.Use M _kThe DAP set of the subband n (having distributed to the subband of user k) that [n, t] expression t participates in serving constantly.The spaced antenna port is selected and power division because the embodiment of the invention is primarily aimed at, and supposes that further each subband can only give a user in a macrocell, and allocation of subbands is to be relatively independent of that DAP selects and the part of power division.Also be that allocation of subbands can realize through any one distribution method, like poll or equitable proportion etc.Similarly, make N _k[t] expression t distributes to the sets of subbands of user k constantly, makes N _kThe sets of subbands of [m, t] expression t DAPm service constantly.Make p _k[m, n, t] expression t is DAP constantly _mIn the transmitted power of distributing on the subband n of user k.In the spaced antenna network, each DAP need satisfy the port power upper limit, can be expressed as:

${\mathrm{\Σ}}_{n=1}^N{p}_k[m,n,t]\≤P_m,$ To any m

p _k[m, n, t]>=0, to any m, k and n

Wherein, P _mThe transmission power upper limit of expression DAPm, and for different port, P _mCan be different, this is to be determined by factors such as the appointed condition of this DAPm, position and peripheral user density and demands.

The flow process of the resource allocation that Fig. 2 shows is specific as follows:

The first step; Structure according to OFDMA spaced antenna network; User terminal feeds back to it with its channel condition information to the link between each spaced antenna port and inserts antenna port, and this port to the base station, adopts high specific to send the prerequisite of transmitting as the base station feedback information.

Suppose that the base station has obtained each user to the channel condition information of each DAP in each subband coidealization through certain feedback system.This mode can be the channel condition information through quantizing to feed back under the frequency division duplex system, also can be the idealized channel estimating of making through channel symmetry under the tdd systems.How explanation adopts the high specific transmission technology below.Suppose in the moment t base station subband n to be distributed to user k, then the user can be expressed as to the channel matrix of each DAP on this subband:

h _k[n，t]＝[h _k[1，n，t]，…h _k[M，n，t]]

In order to adopt high specific to send, be weighing vector of transmission Symbol Design at transmitting terminal:

ω _k[n，t]＝[ω _k[1，n，t]，…ω _k[M，n，t]] ^H

Wherein, ${\mathrm{\ω}}_k[m,n,t]=\left\{\begin{array}{cc}0,& \mathrm{Ifm}{\∉M}_k[n,t]\\ \frac{h_k[m,n,t]}{\left|h_k\right[m,n,t]|},& \mathrm{Ifm}\∈M_k[n,t]\end{array}\right.$

Notice that each element of weight vectors is independently to carry out normalizedly here, this also is that Power Limitation by DAP determines.Accordingly, the transmitted power of DAP can be expressed as:

p _k[n，t]＝[p _k[1，n，t]，…p _k[M，n，t]]，p _k[m，n，t]＝0，if

Can be at the reception signal on the subband n thus:

$y_k[n,t]={\mathrm{\Σ}}_{m=1}^M\left|h_k\right[m,n,t\left]\right|\·\sqrt{p_k[m,n,t]}\·x_k[n,t]+z_k\left[n\right]$

X wherein _k[n, t] is normalized transmission symbol, z _k[n] is additive white Gaussian noise.Wherein receiving signal strength signal intensity is:

${\mathrm{\γ}}_k[n,t]=\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{p}_k[m,n,t]\·{\left|h_k\right[m,n,t\left]\right|}^2$

Suppose that noise power spectral density is a constant, it is consistent that the noise power of different sub-band experience is then arranged, and for the ease of analyzing, in calculating, ignores this constant.According to Shannon (Shannon) formula, it is log at the normalization transmitted bit number that t can obtain constantly that subband n is distributed to user k ₂(1+ γ _k[n, t]).

Need to prove,, omitted the expression of t parameter in the definition to all variablees for the ease of following description.

In second step, the base station like the poll scheduling algorithm, is confirmed user's sets of subbands N according to giving the user with allocation of subbands someway _kIn algorithm of the present invention, the base station is user's allocated subbands, is independent of the joint antenna port and selects and power division.Therefore, allocated subbands can combine factors such as channel condition information and User Priority.At the channel condition information of getting on the subband, the base station is target with the maximize system throughput according to each user, for the user selects to insert antenna port and power division.This problem can be established as:

$\max \underset{n=1}{\overset{N}{\mathrm{\Σ}}}\log (1+\underset{m=1}{\overset{M}{\mathrm{\Σ}}}p[m,n]\·{\left|h_k\right[m,n\left]\right|}^2)$

$s.t.\underset{n=1}{\overset{N}{\mathrm{\Σ}}}p[m,n]\≤P_m,p[m,n]\≥0,$ To any n, m

Next system will adopt an alternative manner to accomplish above target.

In the 3rd step, to the subband that each user gets, the base station is according to situation such as cell deployment, user distribution and computational complexities, dynamically being limited in certain several channel gain for the DAP of its service preferably among the DAP.This is that an initialized antenna port is selected on the one hand, on the other hand, also ensuing iteration complexity has been carried out certain constraint.

To each subband n that has assigned, the base station selects some DAP to serve according to its channel gain.Concrete steps are following:

1) initialization N _k[m]=Φ, n=1;

2) make M _k[n]=Φ.Suppose that subband n has distributed to user k, also is n ∈ N in above-mentioned second step _kAccording to the channel condition information that obtains in the above-mentioned first step, this user to the channel of all M DAP on subband n is:

h _k[n]＝[h _k[1，n]，…h _k[M，n]]；

3) channel gain is sorted from big to small, supposes that ranking results is:

h _k[m ₁，n]≥h _k[m ₂，n]≥…≥h _k[m _M，n]

M wherein ₁, m ₂..., m _MBe 1,2 ..., the arrangement of M;

4) select preceding j DAPm ₁, m ₂..., m _jAs the serve port of this subband n, j≤M.Order

M _k[n]←M _k[n]∪{m ₁，m ₂，…，m _j}

N _k[m] ← N _k[m] ∪ { n}, m=m ₁, m ₂..., m _jWherein, " ← " expression assign operation.

5) if n=N accomplishes; Otherwise n=n+1 returns step 2).

Above step 1)～5) accomplished the function of selecting the initial DAP of access for each subband, this selection course is mainly according to the channel condition information of subband to different DAP.Different user to the path loss that link experienced, the shadow fading of DAP and to decline soon all be different and separate, this is also with relevant with its position that is assigned with the user.This selection course has embodied the advantage of virtual subdistrict under the spaced antenna network.

In the 4th step, systemic circulation ground to each DAP, is target with the maximize system throughput successively, and the gross power water filling of this DAP is assigned on its some subbands that relate to.

System accomplishes power division and corresponding antenna port adjustment algorithm through following iterative power water-filling algorithm:

1) initialization antenna port power distribution result: i=1, to antenna port m=1,2 ..., M participates in the subband N of transmission to it _k[m] divides equally its port power, also is

$p_k^{\left(i\right)}[m,n]=\left\{\begin{array}{cc}{P}_m/\left|N_k\right[m\left]\right|,& \mathrm{ifn}{\∈N}_k\left[m\right]\\ 0,& \mathrm{others}\end{array}\right.$

Wherein | N _k[m] | be set N _kThe element number of [m];

2) the i time iteration makes m=1;

3) upgrade p _k ⁽ⁱ⁾[m, n].Update method is following:

To DAPm ₀During renewal, the power distribution result of fixing other M-1 DAP, set up following optimization problem and find the solution (following formula has omitted subscript k):

$\max \underset{n=1}{\overset{N}{\mathrm{\Σ}}}\log (1+{\mathrm{\γ}}_k[n\left]\right)$

$s.t.p[m_0,n]\·{\left|h_k\right[m_0,n]|}^2+\underset{m=1,m\≠m_0}{\overset{M}{\mathrm{\Σ}}}p[m,n]\·{\left|h_k\right[m,n\left]\right|}^2={\mathrm{\γ}}_k\left[n\right],$ To any k, n

$\underset{n=1}{\overset{N}{\mathrm{\Σ}}}p[m_0,n]\≤P_{m_0},$ To any n

$p[m_0,n]\left\{\begin{array}{cc}\≥0,& \mathrm{Ifn}\∈N_k\left[m_0\right]\\ =0,& \mathrm{Others}\end{array}\right.,$ To any n

Wherein, p [m ₀, n] | h _k[m ₀, n] | ²Be DAPm ₀To the contribution of each subband received signal power, and

Then being the contribution of other DAP, is constant at this.Therefore, power division and corresponding D AP selection course had both been considered the instantaneous channel conditions of subband, had also considered the situation of other antenna port for this subband service, and the port that has embodied distributed network is selected the diversity gain that brings.

Finding the solution of above-mentioned optimization problem is a common protruding optimization problem, can be found the solution as follows by Lagrangian method:

$L(\mathrm{\μ}{,\mathrm{\λ}}_{m_0},\mathrm{\υ})=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\log (1+{\mathrm{\γ}}_k[n\left]\right)-{\mathrm{\λ}}_{m_0}\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}p\right[m_0,n]-P_{m_0})+\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\υ}}_{n}p[m_0,n]$

$+\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_n\left(p\right[m_0,n]\·{\left|h_k\right[m_0,n\left]\right|}^2+\underset{m=1,m\≠m_0}{\overset{M}{\mathrm{\Σ}}}p[m,n]\·{\left|h_k\right[m,n\left]\right|}^2-{\mathrm{\γ}}_k[n\left]\right)$

Wherein, μ,

υ is a Lagrange's multiplier.The Karush-Kuhn-Tucker of this optimization problem (KKT) condition is:

$\frac{\∂L}{\∂p[m_0,n]}={\mathrm{\μ}}_n^{*}\·{\left|h_k\right[m_0,n\left]\right|}^2-{\mathrm{\λ}}_{m_0}^{*}+{\mathrm{\υ}}_n^{*}=0,$ To any n,

$\frac{\∂L}{{\∂\mathrm{\γ}}_k\left[n\right]}=\frac{1}{1+{\mathrm{\γ}}_k^{*}\left[n\right]}-{\mathrm{\μ}}_n^{*}=0,{\mathrm{\μ}}_n^{*}\≥0,$ To any n,

${\mathrm{\μ}}_n^{*}\left(p^{*}\right[m_0,n]\·{\left|h_k\right[m_0,n\left]\right|}^2+G_n-{\mathrm{\γ}}_k^{*}[n\left]\right)=0,{\mathrm{\υ}}_n^{*}\≥0,$ To any n,

${\mathrm{\λ}}_{m_0}^{*}\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{p}^{*}\right[m_0,n]-P_{m_0})=0,{\mathrm{\υ}}_n^{*}{p}^{*}[m_0,n]=0,{\mathrm{\λ}}_{m_0}^{*}\≥0,$

$p^{*}[m_0,n]\·{\left|h_k\right[m_0,n\left]\right|}^2+G_n={\mathrm{\γ}}_k^{*}\left[n\right],$ $\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{p}^{*}[m_0,n]\≤P_{m_0},$ $p[m_0,n]\left\{\begin{array}{cc}\≥0,& \mathrm{ifn}\∈N_k\left[m_0\right]\\ =0,& \mathrm{others}\end{array}\right.$

Be prone to try to achieve separate into:

$p^{*}[m_0,n]={(\frac{1}{{\mathrm{\λ}}_{m_0}^{*}}-\frac{1+G_n}{{\left|h_k\right[m_0,n\left]\right|}^2})}^{+}$

Wherein, (x) ⁺=max (x, 0) asks parameter

Make ${\mathrm{\Σ}}_{n=1}^N{p}^{*}[m_0,n]=P_{m_0}.$

Order

N ∈ N _k[m ₀];

4) if m=M accomplishes one time iteration, execution in step 5); Otherwise m=m+1 returns step 3);

5) if

then finishing iteration; Otherwise i=i+1 returns step 2) the continuation iteration.

Above step 1)～5) accomplished each DAP on the one hand and served the process that subband carries out power division for it; On the other hand, also accomplished the adjustment of selected DAP.In fact, in the result that final iteration is accomplished, certain p _k[m ', n ']=0 means that port m ' will not be subband n ' service, thereby this DAP is removed from the DAP set for this subband service, with the optimization of completion DAP selection.

Distribute embodiment to verify that DAP of the present invention selects and the superiority of power division based on a concrete resource below.

In embodiments of the present invention, the macrocell of distributed network is made up of 1 center base station and 6 DAP, and distance between sites is 200m, and the user is 10m to the minimum range of any DAP.System's carrier frequency is 2GHz, and path loss PL adopts IMT-Advanced city microcellulor model to calculate: PL=36.7log ₁₀(d)+22.7+26log ₁₀(f _c), f wherein _cBe respectively GHz and m with the unit of d.It is the logarithm normal distribution of 8dB that shadow fading adopts variance.System bandwidth is 10MHz, and subcarrier bandwidth is 15kHz, and the granularity of resource allocation is unit with the subband, and being made up of 12 number of sub-carrier is 180kHz, and system has 50 subbands, and subband scheduler is a polling scheduler.The total transmitted power (also being the upper limit of the power summation of base station and other each DAP) of system is 38-43dBm, and the power upper limit of each DAP is 1/7 of gross power, and noise power is-174dBm/Hz.Number of users is 30/ port, also promptly adds up to 30*7=210, and MS evenly distributes in network, and business model is full buffer model (a full cache model).

The method that Fig. 3 has shown the embodiment of the invention under the situation of different virtual cell size with two kinds of comparisons that rudimentary algorithm carries out.Two kinds of rudimentary algorithms are respectively the whole day line and insert the power-sharing under power-sharing down and the single antenna access.Wherein, whole day line access power is divided equally be meant that each subband all transmits on all antenna port, and each port needles is divided equally its port power to whole subbands; The single antenna access power divide equally the system of being meant according to each subband at the channel condition information of its distributing user to the different antennae port, a best antenna port of selective channel gain transmits.As shown in Figure 3, the systematic function of method proposed by the invention always is superior to the performance of power-sharing.Compare the systematic function that the present invention combines the different virtual cell size again.On the one hand, virtual subdistrict bigger (j is big more), it is many more also promptly to allow the user to select to insert antenna number, and the performance of system is excellent more.But along with the increase of j, the systematic function increase is more and more slower.In fact, even the virtual subdistrict scope is bigger when initial selected, in the process that iterative power is distributed, channel status poor antenna port also can be closed by water-filling algorithm.Therefore, under the cell topology of inferior instance, the service antenna port that the method for using iterative power to distribute is finally confirmed also is the size of virtual subdistrict, and in the time of j=3 or j=4, performance is basically near the performance of whole day line virtual subdistrict.On the other hand, from the angle of amount of calculation, less j can reduce the complexity of iterative algorithm to a certain extent.Therefore in real network, use in this invention algorithm, can select suitable virtual subdistrict, reach more excellent performance faster, seek best compromise based on factors such as cell environment, user density.

Fig. 4 has shown the method for the embodiment of the invention under different virtual cell size situation, the convergence situation of iterative power water filling.4 convergence sample curve have been provided among Fig. 4.When as can be seen from Figure 4 virtual subdistrict is bigger (like j=7), the performance recruitment of former step convergences is less than the relative less situation (j=3) of virtual subdistrict.But restrain effect in order to reach final result, system needs similar iterations mostly.Like the result from this instance, system needs about 15 iteration can make the result stable.

It is to be noted; Though emulation is to specific OFDMA distributed network framework and the system parameters of a cover; But, and then have very strong generality through the method that iteration repeatedly is optimized adjustment and (be in particular in: on the one hand to the not restriction of port number of spaced antenna network because the present invention program adopts the base station original allocation; On the other hand, to virtual subdistrict size also not restriction), therefore, when method of the present invention was applied to have in distributed network and the system parameters of heterogeneous networks framework, the present invention program still can embody superiority.

Can find out by above embodiment; The antenna port of OFDMA spaced antenna network of the present invention is selected and power allocation scheme; The flexibility of resource allocation of not only having utilized dynamic subband of OFDMA and power division to bring; Also embodied the advantage of virtual subdistrict under the spaced antenna network, the degree of freedom that the user selects to insert antenna port has been carried out effective excavation.It has improved the throughput of system effectively, also can adjust through limiting the antenna port quantity that inserts on each subband on the algorithm complex, thereby seek compromise preferably, therefore also has stronger practicality.

The above only is an execution mode of the present invention; Should be pointed out that for those skilled in the art, under the prerequisite that does not break away from know-why of the present invention; Can also make some improvement and modification, these improve and modification also should be regarded as protection scope of the present invention.

Claims (7)

1. the distribution method of OFDMA spaced antenna Internet resources is characterized in that, may further comprise the steps:

S1, according to the structure of OFDMA spaced antenna network, user terminal feeds back to the base station with its link channel condition information on each subband to each spaced antenna port DAP, adopts high specific to send the condition of transmission signals as this base station;

S2, said base station is the user terminal allocated subbands according to said channel condition information;

S3; The subband of getting to each user terminal; DAP or all DAP in the said OFDMA spaced antenna network that said base station reaches preset value for the gain of each subband selective channel obtain the DAP set into each subband service as the initial selected result of DAP;

S4, through the iteration water-filling algorithm respectively with the power division of each DAP to each subband of this DAP service;

Water injection power distribution method to a DAP among the said step S4 is found the solution based on following optimization problem:

1. the target of this optimization problem is a maximize system throughput;

2. the constraints of optimization problem is: the port Power Limitation of this DAP, and the sets of subbands of this DAP service;

3. when calculating the water injection power allocation result of a DAP; The to be sent power of other each DAP on its subband of serving separately keeps constant, and the power transmission strategy of this DAP on the subband of its service determined by following two factors: the channel situation of this DAP on subband; The transmitted power of other each DAP on each subband be to the contribution of subband received signal to noise ratio, wherein, and above-mentioned first factor and this DAP power positive correlation to be sent on subband, second factor and this DAP power negative correlation to be sent on subband;

4. finding the solution through Lagrangian method of this optimization problem asks extreme value to accomplish;

5. try to achieve after the extreme value, this DAP upgrades its power to be sent on the subband of service, if its power to be sent on a subband is zero, then this DAP is removed from the DAP set for this subband service, to accomplish the optimization that DAP selects.

2. the distribution method of OFDMA spaced antenna Internet resources as claimed in claim 1; It is characterized in that; After the optimization through said DAP distribution, this DAP remains unchanged when other DAP carries out computing in this iteration at the power to be sent on the subband of its service.

3. according to claim 1 or claim 2 the distribution method of OFDMA spaced antenna Internet resources; It is characterized in that; The iteration convergence condition of said iteration water-filling algorithm is: after the iteration, the variable quantity of the more preceding iteration of the throughput of system is during less than preset constant, iteration convergence.

4. the distribution method of OFDMA spaced antenna Internet resources as claimed in claim 1 is characterized in that, said DAP is for guaranteeing to cover antenna port or the relay station of setting up under the cellular cell framework.

5. the distribution method of OFDMA spaced antenna Internet resources as claimed in claim 1 is characterized in that, adopts high specific to send when transmitting, DAP for each component in the weighing vector that sends Symbol Design be independently carry out normalized.

6. the distribution method of OFDMA spaced antenna Internet resources as claimed in claim 1 is characterized in that, the initial value of the to be sent power of said each DAP on its subband of serving separately is through dividing equally acquisition with the power of DAP to the subband of its service.

7. like the distribution method of claim 1 or 2 or 6 described OFDMA spaced antenna Internet resources; It is characterized in that; When said base station reaches the DAP of preset value for each subband selective channel gain; This preset value has determined to be the DAP quantity upper limit of subband service that the throughput positive correlation of the size of this quantity and system is with the operand negative correlation in the said iteration water-filling algorithm.

Images