CN103986558B - Adaptive cooperation transmission method in a kind of cellular mobile communication D2D systems - Google Patents

Abstract

The invention discloses adaptive cooperation transmission method in a kind of cellular mobile communication D2D systems, carry out as follows：The Doppler shift of the user that cooperation is participated in cell is estimated first, obtains channel time delay coefficient correlation；Then according to dispatching algorithm, it is determined that participating in community user and the D2D user of cooperation transmission, base station determines the transmission mode that cooperation will be used；Then user estimates respective channel information, and by these feedback of the information to base station；Finally carry out data transmission under the pre-coding scheme for choosing pattern.The present invention can be effectively applied in actual scene, under the control of core net, in cell mobile communication systems the frequency resource of community user share to Cell Edge User carry out device-to-device between data transfer, to reduce the cost of the cell transmission capacity of cellular communication system on a small quantity, greatly improve the reachable transmission rate of Cell Edge User so that average frequency spectrum utilization rate is improved in cell.

Description

Adaptive cooperation transmission method in a kind of cellular mobile communication D2D systems

Technical field

The invention belongs to the cooperation transmission field in mobile communication multiple-input and multiple-output (MIMO) system, it is related to a kind of honeycomb Cooperation transmission method in GSM.

Background technology

For in cell mobile communication systems, the available signal power that Cell Edge User faces is low and interfering signal power High, it is impossible to effectively to obtain the difficult point of high data rate, constantly having scholar to propose in recent years can be by equipment and equipment Direct communication link is set up between (Device-to-Device, D2D) to effectively improve the data rate of Cell Edge User. This D2D technologies are in existing cellular mobile communication environment so that positioned at cell edge user can by shared cell its The frequency resource of his user, directly sets up link and carries out data transmission with neighbouring terminal.Due to D2D communication two users it Between distance it is very near, it is possible to use less transmission power obtains message transmission rate higher, is filled using multiplexing technology Divide excavated space resource, the availability of frequency spectrum and power efficiency are improved to greatest extent.But inevitably bring significant Co-channel interference (ICI), constrains lifting of the system up to transmission rate.

The interference produced to D2D user 1 when being communicated with community user 0 for base station, selects suitable communication pattern so that The total performance of system is best.There are two kinds of conventional precoding processing patterns base station in sending signal, maximizes speed (maximum rate, MR) precoding and interference eliminate (interference cancellation, IC) precoding mode： Under IC patterns, AF panel is done in base station using the channel information that D2D user 1 feeds back to it；Under MR patterns, base station needs to obtain The channel information of the feedback of community user 0, maximization speed precoding processing is done to cell communication link.Two kinds of collaboration modes are not With being that under IC patterns, base station not only needs the channel information that community user 0 feeds back, while needing the letter that D2D user 1 feeds back Road information is, it is necessary to extraly feedback channel information is come disturbing of eliminating that D2D user is subject to；And under MR patterns, base station only needs The channel condition information of the feedback of community user 0, thus feedback link load is obviously reduced, system delay can also be decreased.Two Plant collaboration mode and different system achievable rates are obtained in that in different application scene, thus a real system can be two Switched in the pattern of kind, to realize that maximum rate is transmitted.

Existing MR and IC mode evaluations scheme is all built upon assuming the basis that base station can know ideal communication channel information On, but in practical application scene, either time division duplex (TDD) or FDD (FDD) system, this hypothesis is all very Difficulty is met.If system uses time division duplex (TDD), transmitting terminal can be by reciprocity principle, in base station estimation uplink Path channels information and as the channel information of downlink, but up-link and downlink are not full symmetric , thus just unavoidably there is evaluated error in the channel information that obtains by reciprocity of base station.If system uses frequency division Duplexing (FDD), is generally estimated downlink channel information by user terminal, and then user passes through feedback link by downlink Feedback of channel information to base station, but user can be limited with feedack amount in real system.User is in order to meet The capacity requirement of feedback link, is often first quantified channel information, and the information after quantization then is passed through into limited ratio Spy feeds back to base station, and last base station recovers channel information as much as possible according to this feedback information.This scheme may result in base There is a certain amount error between the system channel information and real channel information for obtaining of standing, the size of this quantization error by The capacity of feedback link is determined.There is the information delay problem that feedback link brings when in addition using feedback mechanism, i.e. base station is obtained The feedback information for obtaining not is the channel information at current time, but the channel information before one section of time delay.Above-mentioned many Under the influence of factor, the downlink information that base station obtains not fully is matched with real channel information, is caused based on ideal The adaptive transmission scheme performance of channel information is remarkably decreased, therefore we need to study and can only know non-ideal communication channel in base station Under conditions of information, the self adaptation switching problem of MR and IC collaboration modes.

The content of the invention

Technical problem：The present invention provides one kind and can be effectively applied in actual scene, can only be prolonged in base station end When Limited Feedback channel information, the channel information that obtains there is the non-ideal factors such as quantization error, delay of feedback and path loss In the case of, it would be preferable to support device-to-device direct communication, effectively improve the reachable transmission rate and the availability of frequency spectrum of system Adaptive cooperation transmission method in cellular mobile communication D2D systems.

Technical scheme：Adaptive cooperation transmission method in cellular mobile communication D2D systems of the invention, including following step Suddenly：

(1) user's average Doppler frequency deviation is estimated：The average movement of the user terminal of cooperation is participated in base station investigation cell Speed, calculates the Doppler shift f of community user 0₀=f_cv₀/ c, then calculates channel time delay correlation coefficient ρ₀=J₀(2π f₀T_s), and then computing device receives the Doppler shift f of end subscriber 1 to device link₁=f_cv₁/ c, then calculates channel and prolongs When correlation coefficient ρ₁=J₀(2πf₁T_s), wherein, c represents the light velocity, f_cRepresent carrier frequency, v₀Represent the average shifting of community user 0 Dynamic speed, v₁Represent that device-to-device link receives the average rate travel of end subscriber 1, π represents pi, T_sRepresent that each is accorded with The length in number cycle, J₀() represents first kind zero Bessel function；

(2) channel information that the activation user that is scheduled quantifies to base station feedback：

Base station scheduling chooses a community user, its band resource is shared between edge customer carries out device-to-device and is led to Letter, the user carries out channel estimation to obtain channel information and path-loss factor first, further according to the known code in itself and base station This quantifies to channel information, and quantized channel information and path-loss factor are fed back into base station；

Equipment carries out channel estimation to obtain channel information and path-loss factor first with equipment communication user simultaneously, then Channel information is quantified according to the known code book in itself and base station, and quantized channel information and path-loss factor are fed back to Base station；

(3) base station determines the cooperation transmission mode of nearly base station user：Base station according in step (2) obtain path loss because The channel time delay coefficient correlation obtained in son, the quantizing bit number of channel information, and step (1), feeds back real in feedback link When channel condition information before, respectively calculate maximize speed precoding mode under system up to transmission rate R^MRClosed solutions and System is up to transmission rate R under interference elimination precoding mode^ICClosed solutions, and based on the two closed solutions, to maximizing speed Rate precoding mode and interference eliminate the performance gain obtained under precoding mode both transmission modes and are compared, selectivity The transmission mode that a kind of transmission mode of energy high gain is used as nearly base station user；

(4) data transfer is realized：Base station according to the transmission mode selected in step (3), using the amount of feedback in step (2) Change channel information and calculate precoding vector, the data that then will be sent to nearly base station user are multiplied by precoding vector, carry out data Transmission, edge customer is utilized and carried out data transmission with frequency resource.

Calculated according to below scheme in the preferred scheme of the inventive method, in step (3) and maximize speed precoding mode Under, the closed solutions R of the reachable transmission rate of system^MR：

The enclosed of reachable transmission rate of the community user 0 in the case where speed precoding mode is maximized is calculated according to following formula first Solution

Reachable biography of the end subscriber 1 in the case where speed precoding mode is maximized is received according to following formula computing device to device link The closed solutions of defeated speed

In formula, subscript b represents cell base station, and subscript 0 represents the community user of scheduling, and subscript s represents the transmission of D2D links End subscriber, subscript 1 represents the reception end subscriber of D2D links, L_b0Cell base station b to the path-loss factor of community user 0 is represented, L_b1Represent that cell base station b to D2D links receive the path-loss factor of end subscriber 1, L_s0Represent D2D link transmitting terminals s to cell The path-loss factor of user 0, L_s1Represent D2D link transmitting terminals s to the path-loss factor for receiving end subscriber 1, P_bRepresent small The transmission power of area base station b, P_sRepresent the transmission power of D2D link transmitting terminals s, ρ₀Represent that the channel time delay of community user 0 is related Coefficient, B₀The channel condition information bit number of the feedback of community user 0 is represented, M represents the quantity of Base Transmitter antenna, function F₁ (c₀,c₁,c₂, M) definition be：

Wherein, c₀, c₁, c₂, M is argument of function, j in functional expression！Represent the factorial of j, log₂E it with 2 is bottom that () is The logarithm of natural number e, the natural logrithm with 2 is reciprocal each other, i.e. log₂E ()=1/ln2, works as c₁When=0, function F₁(c₀,c₁,c₂, M) it is reduced to：

F₁(c₀,0,c₂, 1) and=log₂(e)I₀₁(c₀,c₂)；

Function I (a, b, m, n) is defined as follows：

Wherein, a, b, m, n are argument of function, in functional expressionIt is to take out s element from m different elements Number of combinations,It is incomplete gamma functions,For gamma function its In, u, x are the independent variable of gamma function；

I₀₁(a, b) is special cases of the function I (a, b, m, n) in m=0, n=1：

In formulaIt is the first rank exponential integral function, wherein k is argument of function；

Then calculated according to following formula and maximized under speed precoding mode, the closed solutions R of the reachable transmission rate of system^MR：

Interference is calculated in step (3) according to below scheme to eliminate under precoding mode, the reachable transmission rate of system is closed Formula solution R^IC：

The closed solutions that community user 0 eliminates the reachable transmission rate under precoding mode in interference are calculated according to following formula first

End subscriber 1 is received according to following formula computing device to device link and eliminates the reachable transmission under precoding mode in interference The closed solutions of speed

In formula, ρ₁Represent that device-to-device link receives the channel time delay coefficient correlation of end subscriber 1, B₁Expression equipment is to setting Channel condition information bit number from standby terminating subscriber 1 to base station feedback；Function F₂(c₀,c₁,c₂) definition be：

Wherein c₀, c₁, c₂It is argument of function；

Then calculate interference according to following formula to eliminate under precoding mode, the closed solutions R of the reachable transmission rate of system^IC：

In the preferred scheme of the inventive method, in step (3), maximization speed precoding mode is calculated according to following formula and is prelisted The performance gain R that the lower system of code is obtained^MR,↑：

The performance gain R that interference eliminates system acquisition under precoding mode is calculated according to following formula^IC,↑：

Wherein, R₀It is achievable rate of the community user 0 in the case where device-to-device communication condition is not introduced, is calculated according to following formula Obtain：

Then the performance gain that system is obtained under the latter two transmission mode for introducing D2D communications is compared：If meeting R^{MR ,↑}≥R^IC,↑, then select to maximize the transmission mode that speed precoding mode is used as nearly base station user, otherwise select interference to disappear Except the transmission mode that precoding mode is used as nearly base station user.

Beneficial effect：Compared with prior art, the present invention has advantages below：

1) in systems in practice, signal sending end estimates channel status to there is path loss, receiving terminal between receiving terminal When feeding back to transmitting terminal, there is the quantization error of feedback, the channel condition information that signal sending end is obtained is real-time with present channel There is delay time error between state, base station can accurately obtain channel condition information if assuming, these non-ideal factors are such as made Become second nature loss of energy, and the present invention can effectively resist path loss present in real system, quantization error and feedback and prolong The penalty that the non-ideal factors such as time error are caused, so as to obtain system transfer rate higher；

2) when the invention is applied with cellular mobile communication D2D systems, it is possible to use the closed solutions of achievable rate, in feedback Before Link Feedback instantaneous channel conditions information, system when calculating between base station and community user using which kind of pre-coding scheme Performance is more excellent, when system environments can obtain preferable performance under MR precoding modes, using local MR pre-coding schemes, only needs Wanting the feedback channel information of community user 0 is carried out as precoding vector, when system environments can be obtained under IC precoding modes compared with During good performance, using ZF pre-coding scheme, it is necessary to community user 0 and D2D links receive the feedback channel information simultaneously of end subscriber 1 The calculating of precoding vector is carried out, therefore, different pre-coding schemes are selected in different system environments, being capable of abundant land productivity With feedback channel, implementation complexity is reduced, obtain preferably systematic function.

Brief description of the drawings

Fig. 1 is the system block diagram of the inventive method；

Fig. 2 be when the distance between each user terminal and base station, when distance each other determines, different feedback bits and In the case of different delay of feedback, system achievable rate is with the graph of a relation for sending signal to noise ratio change under two kinds of collaboration modes；

Fig. 3 be when the distance between each user terminal and base station, when distance each other determines, different feedback bits and In the case of different delay of feedback, the achievable rate that the loss of community user achievable rate and D2D user obtain under two kinds of collaboration modes With the graph of a relation for sending signal to noise ratio change；

Fig. 4 be when the distance between each user terminal and base station, when distance each other determines, different feedback bits and In the case of different delay of feedback, system performance gain is with the graph of a relation for sending signal to noise ratio change under two kinds of collaboration modes.

Specific embodiment

The present invention is further described below in conjunction with the accompanying drawings.As shown in figure 1, holding equipment of the invention is to equipment A kind of adaptive cooperation transmission method based on Time-delayed Feedback is implemented as follows in the cell mobile communication systems of communication：

(1) system investigates the average rate travel of intra-cell users terminal, to calculate Doppler shift, then according to carat Gram (Clarke) scattering model calculates channel time delay coefficient correlation.

With the Doppler shift f of the community user 0 of base station communication₀=f_cv₀/ c, channel time delay coefficient correlation is ρ₀=J₀(2π f₀T_s), device-to-device communication link receives the Doppler shift f of end subscriber 1₁=f_cv₁/ c, channel time delay coefficient correlation is ρ₁ =J₀(2πf₁T_s), wherein, c represents the light velocity, f_cRepresent carrier frequency, v₀Represent the average shifting with base station communication user 0 in cell Dynamic speed, v₁Represent that device-to-device link receives the average translational speed of end subscriber 1, π represents pi, T_sRepresent that each is accorded with The length in number cycle, J₀() represents the first kind zero-order Bessel (Bessel) function；

(2) channel information that the activation user that is scheduled quantifies to base station feedback：

A community user wait data transfer is dispatched in base station using dispatching algorithm is commonly used in existing communication system, passes every time During defeated beginning, the user and D2D links receive end subscriber 1 and carry out channel estimation first to obtain channel information and path loss The factor, further according to base station and the known code book of user 0, quantized channel information, and by quantized channel information and path-loss factor Feed back to base station；

Equipment carries out channel estimation to obtain channel information and path-loss factor first with equipment communication user simultaneously, then Channel information is quantified according to the known code book in itself and base station, and quantized channel information and path-loss factor are fed back to Base station；

(3) base station determines the cooperation transmission mode of nearly base station user：

The channel time delay coefficient correlation that base station obtains according to the path-loss factor and step (1) that are obtained in step (2), Before feedback link feedback instantaneous channel conditions information, calculate maximize the reachable transmission of system under speed precoding mode respectively Speed R^MRClosed solutions and interference eliminate precoding mode under system up to transmission rate R^ICClosed solutions, and based on the two Closed solutions, eliminate the performance obtained under precoding mode both transmission modes and increase to maximizing speed precoding mode and interference Benefit is compared, the transmission mode for selecting a kind of transmission mode of performance high gain to be used as nearly base station user；

(4) base station is fed back according to the transmission mode selected in step (3) using the feedback channel of step (1) (2) acquisition The quantized channel information for having time delay calculates precoding vector ω, and data to be sent then are multiplied by into precoding vector ω, enters line number According to transmission.

Various non-ideal factors such as path loss, quantization error, delay of feedback error present in analysis real system On the basis of, system is up to transmission rate R under maximization speed precoding mode is calculated respectively^MRClosed solutions and interference eliminate pre- System is up to transmission rate R under coding mode^ICClosed solutions, be then based on the two closed solutions, to maximize speed precoding Pattern and interference eliminate the performance gain obtained under precoding mode both transmission modes and are compared, being adaptive selected property A kind of transmission mode of energy high gain, the transmission mode used as nearly base station user carries out data transmission,

Calculated according to below scheme in step (3) and maximized under speed precoding mode, the reachable transmission rate of system Closed solutions R^MR：

The enclosed of reachable transmission rate of the community user 0 in the case where speed precoding mode is maximized is calculated according to following formula first Solution

Reachable biography of the end subscriber 1 in the case where speed precoding mode is maximized is received according to following formula computing device to device link The closed solutions of defeated speed

In formula, subscript b represents cell base station, and subscript 0 represents the user of scheduling, and subscript s represents that the transmitting terminal of D2D links is used Family, subscript 1 represents the reception end subscriber of D2D links, L_b0Represent cell base station b to the path-loss factor of community user 0, L_b1 Represent that cell base station b to D2D links receive the path-loss factor of end subscriber 1, L_s0Represent that D2D link transmitting terminals s is used to cell The path-loss factor at family 0, L_s1Represent D2D link transmitting terminals s to the path-loss factor for receiving end subscriber 1, P_bRepresent cell The transmission power of base station b, P_sRepresent the transmission power of D2D link transmitting terminals s, ρ₀Represent the channel time delay phase relation of community user 0 Number, B₀The channel condition information bit number of the feedback of community user 0 is represented, M represents the quantity of Base Transmitter antenna, function F₁(c₀, c₁,c₂, M) definition be：

Wherein, c₀, c₁, c₂, M is argument of function, j in functional expression！Represent the factorial of j, log₂E it with 2 is bottom that () is The logarithm of natural number e, the natural logrithm with 2 is reciprocal each other, i.e. log₂E ()=1/ln2, works as c₁When=0, function F₁(c₀,c₁,c₂, M) it is reduced to：

F₁(c₀,0,c₂, 1) and=log₂(e)I₀₁(c₀,c₂)

Function I (a, b, m, n) is defined as follows：

Wherein, a, b, m, n are argument of function, in functional expressionIt is to take out s element from m different elements Number of combinations,It is incomplete gamma functions,For gamma function its In, u, x are the independent variable of gamma function；

I₀₁(a, b) is special cases of the function I (a, b, m, n) in m=0, n=1：

In formulaIt is the first rank exponential integral function, wherein k is argument of function；

Then calculated according to following formula and maximized under speed precoding mode, the closed solutions R of the reachable transmission rate of system^MR：

Interference is calculated in step (3) according to below scheme to eliminate under precoding mode, the reachable transmission rate of system is closed Formula solution R^IC：

The closed solutions that community user 0 eliminates the reachable transmission rate under precoding mode in interference are calculated according to following formula first

End subscriber 1 is received according to following formula computing device to device link and eliminates the reachable transmission under precoding mode in interference The closed solutions of speed

In formula, ρ₁Represent that device-to-device link receives the channel time delay coefficient correlation of end subscriber 1, B₁Expression equipment is to setting Channel condition information bit number from standby terminating subscriber 1 to base station feedback；Function F₂(c₀,c₁,c₂) definition be：

Wherein c₀, c₁, c₂It is argument of function；

Then calculate interference according to following formula to eliminate under precoding mode, the closed solutions R of the reachable transmission rate of system^IC：

In cellular mobile communication D2D systems of the invention in adaptive cooperation transmission method, in step (3), according to following formula Calculating maximizes the performance gain R of system acquisition under speed precoding mode precoding^MR,↑：

The performance gain R that interference eliminates system acquisition under precoding mode is calculated according to following formula^IC,↑：

Wherein, R₀It is achievable rate of the community user 0 in the case where device-to-device communication condition is not introduced, is calculated according to following formula Obtain：

Then the performance gain that system is obtained under the latter two transmission mode for introducing device-to-device communication is compared：If full Sufficient R^MR,↑≥R^IC,↑, then selection maximizes the transmission mode that speed precoding mode is used as nearly base station user, under the pattern, The precoding vector of base station isWhereinRepresent that community user 0 is believed to the feedback channel that base station is submitted to Breath；If meetingThen selection interference eliminates the transmission mode that precoding mode is used as nearly base station user, the pattern Under, base station would leak into the interference ZF of D2D user, and as far as possible maximize the useful signal for being sent to community user, base In such design requirement, the precoding vector ω of base station is selectedDirection existsKernel on throwing Shadow；

After the precoding vector of base station determines, will treat to be multiplied by precoding vector ω with base station user communication data, carry out Carrier modulation simultaneously sends.

Base station according to channel information calculate precoding vector when, it is necessary to the channel status at current time [n], real system In, what base station was obtained from feedback channel is the channel condition information that has time delay, so we assume that receiving previous moment [n-1] Channel condition information, then by the coefficient correlation of channel time delay, estimate time delay produce error so that as far as possible reduce prolong When the penalty that brings.

Fig. 2 be when the distance between each user terminal and base station, when distance each other determines, different feedback bits and In the case of different delay of feedback, system achievable rate is with the graph of a relation for sending signal to noise ratio change under two kinds of collaboration modes.It is bent in figure Line assumes that base station transmission antenna number is M=4, and information feedback bit number is B₀=B₁=B, the skew of unit slot frequency It is f₀T_s=f₁T_s=fT_s, base station and D2D send the proportional relation (P of power of equipment_s=P_b/ 20), as can be seen from the figure： When transmit power is smaller, two kinds of precoding mode performances are relative, from from the point of view of system implementation complexity, this appropriate to the occasion use Realize simple, it is necessary to the less MR precoding modes of channel information amount.When transmit power is larger, when the letter of feedback channel feedback When the application condition of channel state information is small, when quantization error and all smaller delay time error, the performance of IC precoding modes It is substantially better than MR precoding modes.When quantization error and delay time error all than it is larger when, be adapted to use MR precoding modes.

Fig. 3 be when the distance between each user terminal and base station, when distance each other determines, different feedback bits and In the case of different delay of feedback, the achievable rate that the loss of community user achievable rate and D2D user obtain under two kinds of collaboration modes With the graph of a relation for sending signal to noise ratio change.Curve assumes that base station transmission antenna number is 4, information feedback bit number in figure It is B₀=B₁=8, the skew of unit slot frequency is f₀T_s=f₁T_s=0.08, base station and D2D send the proportional pass of power of equipment System (P_s=P_b/ 20), when base-station transmitting-power increases, D2D transmitting terminal power also increases, and is obtaining D2D message capacities higher While, bigger interference is also generated to community user, now, the loss of the achievable rate of community user is also increased.

Fig. 4 be when the distance between each user terminal and base station, when distance each other determines, different feedback bits and In the case of different delay of feedback, system performance gain is with the graph of a relation for sending signal to noise ratio change under two kinds of collaboration modes.It is bent in figure Line assumes that base station transmission antenna number is 4, and information feedback bit number is B₀=B₁=8, unit slot frequency offsets and is f₀T_s=f₁T_s=0.08, base station and D2D send the proportional relation (P of power of equipment_s=P_b/ 20), emulation is obtained, when base station hair When sending power increase to a certain extent, because the interference that D2D links bring so that the gain trend of system slows down, continue to increase Transmit power, can produce the rollback of performance, and curve can be used as the reference for sending precoding mode selection.In systems in practice, remove Selection is suitable to send precoding mode, can also be without keeping base station and D2D to send the power proportional relation of equipment, reduces The transimission power of D2D links, because D2D communications are typical short-range communications, transmitting terminal can send letter with relatively low power Number transmit data.

It should be understood that above-described embodiment is merely to illustrate the specific embodiment of technical solution of the present invention, rather than limitation is originally The scope of invention.After the present invention has been read, those skilled in the art are to the modification of various equivalents of the invention and replace Change and fall within the protection domain that the application claim is limited.

Claims (2)

1. adaptive cooperation transmission method in a kind of cellular mobile communication D2D systems, it is characterised in that：The transmission method include with Lower step：

(1) user's average Doppler frequency deviation is estimated：The average rate travel of the user terminal of cooperation is participated in base station investigation cell, Calculate the Doppler shift f of community user 0₀=f_cv₀/ c, then calculates channel time delay correlation coefficient ρ₀=J₀(2πf₀T_s), enter And computing device receives the Doppler shift f of end subscriber 1 to device link₁=f_cv₁/ c, then calculates channel time delay phase relation Number ρ₁=J₀(2πf₁T_s), wherein, c represents the light velocity, f_cRepresent carrier frequency, v₀Represent the average rate travel of community user 0, v₁ Represent that device-to-device link receives the average rate travel of end subscriber 1, π represents pi, T_sRepresent each symbol period Length, J₀() represents first kind zero Bessel function；

(2) channel information that the activation user that is scheduled quantifies to base station feedback：

Base station scheduling chooses a community user, its band resource is shared between edge customer carries out device-to-device and is communicated, The user carries out channel estimation to obtain channel information and path-loss factor first, further according to the known code book pair in itself and base station Channel information is quantified, and quantized channel information and path-loss factor are fed back into base station；

Equipment carries out channel estimation to obtain channel information and path-loss factor first with equipment communication user simultaneously, further according to It quantifies with the known code book in base station to channel information, and quantized channel information and path-loss factor are fed back into base Stand；

(3) base station determines the cooperation transmission mode of nearly base station user：Base station according in the step (2) obtain path loss because The channel time delay coefficient correlation obtained in son, the quantizing bit number of channel information, and step (1), feeds back real in feedback link When channel condition information before, respectively calculate maximize speed precoding mode under system up to transmission rate closed solutions R^MRWith Closed solutions R of the system up to transmission rate under interference elimination precoding mode^IC, and based on the two closed solutions, to maximizing speed Rate precoding mode and interference eliminate the performance gain obtained under precoding mode both transmission modes and are compared, selectivity The transmission mode that a kind of transmission mode of energy high gain is used as nearly base station user；Under the maximization speed precoding mode Closed solutions R of the system up to transmission rate^MRCalculated according to below scheme：

The closed solutions of reachable transmission rate of the community user 0 in the case where speed precoding mode is maximized are calculated according to following formula first

$R_0^{MR}=F_1\left(L_{b0}{P}_b{\mathrm{\ρ}}_0^2\right(1-2^{-\frac{B_0}{M-1}}),L_{b0}{P}_b(1-{\mathrm{\ρ}}_0^2),L_{s0}{P}_s,M)$

Reachable transmission speed of the end subscriber 1 in the case where speed precoding mode is maximized is received according to following formula computing device to device link The closed solutions of rate

$R_1^{MR}=F_1(L_{s1}{P}_s,0,L_{b1}{P}_b,1)$

In formula, subscript b represents cell base station, and subscript 0 represents the community user of scheduling, and subscript s represents that the transmitting terminal of D2D links is used Family, subscript 1 represents the reception end subscriber of D2D links, L_b0Represent cell base station b to the path-loss factor of community user 0, L_b1 Represent that cell base station b to D2D links receive the path-loss factor of end subscriber 1, L_s0Represent that D2D link transmitting terminals s is used to cell The path-loss factor at family 0, L_s1Represent D2D link transmitting terminals s to the path-loss factor for receiving end subscriber 1, P_bRepresent cell The transmission power of base station b, P_sRepresent the transmission power of D2D link transmitting terminals s, ρ₀Represent the channel time delay phase relation of community user 0 Number, B₀The channel condition information bit number of the feedback of community user 0 is represented, M represents the quantity of Base Transmitter antenna, function F₁(c₀, c₁,c₂, M) definition be：

$F_1(c_0,c_1,c_2,M)={\log }_2\left(e\right){\left(\frac{c_1}{c_1-c_0}\right)}^M\left(I_{01}\right(c_1,c_2)-\frac{c_0}{c_1}\underset{i=0}{\overset{M-1}{\Σ}}\underset{l=0}{\overset{i}{\Σ}}\underset{j=0}{\overset{i-l}{\Σ}}\frac{{(c_1-c_0)}^i}{{c_0}^j{c_1}^{i}j!}I(c_0,c_2,i-l,i-l-j+1\left)\right)$

Wherein, c₀, c₁, c₂, M is argument of function, j in functional expression！Represent the factorial of j, log₂E () is the nature bottom of for 2 The logarithm of number e, the natural logrithm with 2 is reciprocal each other, i.e. log₂E ()=1/ln2, works as c₁When=0, function F₁(c₀,c₁,c₂, M) and letter Turn to：

F₁(c₀,0,c₂, 1) and=log₂(e)I₀₁(c₀,c₂)；

Function I (a, b, m, n) is defined as follows：

$\begin{array}{c}I(a,b,m,n)\\ =\left\{\begin{array}{cc}\frac{a}{b}{(\frac{a}{b}-1)}^{-n}\left(e^{\frac{1}{a}}\mathrm{\Γ}\left(m+1\right)\mathrm{\Γ}\left(-m,\frac{1}{a}\right)-\underset{s=0}{\overset{m}{\Σ}}\underset{k=1}{\overset{n}{\Σ}}{\left(-1\right)}^s\left(\begin{array}{c}m\\ s\end{array}\right){\left(\frac{1}{b}\right)}^s{\left(\frac{1}{a}\right)}^{k-m-1}{\left(\frac{a}{b}-1\right)}^{k-1}{e}^{\frac{1}{b}}\mathrm{\Γ}\left(m-s-k+1,\frac{1}{b}\right)\right)& a\≠b\\ \underset{s=0}{\overset{m}{\Σ}}{(-1)}^s\left(\begin{array}{c}m\\ s\end{array}\right){\left(\frac{1}{a}\right)}^{n+s-m}{e}^{\frac{1}{a}}\mathrm{\Γ}(m-s-n,\frac{1}{a})& a=b\end{array}\right.\end{array}$

Wherein, a, b, m, n are argument of function, in functional expressionIt is to take out the s combination of element from m different elements Number,It is incomplete gamma functions,For gamma function wherein, u, x It is the independent variable of gamma function；

I₀₁(a, b) is special cases of the function I (a, b, m, n) in m=0, n=1：

$I_{01}(a,b)=\left\{\begin{array}{cc}\frac{a}{a-b}\left(e^{\frac{1}{a}}{E}_1\right(\frac{1}{a})-e^{\frac{1}{b}}{E}_1(\frac{1}{b}\left)\right)& a\≠b\\ \frac{1}{a}{e}^{\frac{1}{a}}\mathrm{\Γ}(-1,\frac{1}{a})& a=b\end{array}\right.$

In formulaIt is the first rank exponential integral function, wherein k is argument of function；

Then calculated according to following formula and maximized under speed precoding mode, the closed solutions R of the reachable transmission rate of system^MR：

$R^{MR}=R_0^{MR}+R_1^{MR};$

Closed solutions R of the system up to transmission rate under the interference elimination precoding mode^ICCalculated according to below scheme：

The closed solutions that community user 0 eliminates the reachable transmission rate under precoding mode in interference are calculated according to following formula first

$R_0^{IC}=F_1\left(L_{b0}{P}_b{\mathrm{\ρ}}_0^2\right(1-2^{-\frac{B_0}{M-1}}),0,L_{s0}{P}_s,M-1),$

End subscriber 1 is received according to following formula computing device to device link and eliminates the reachable transmission rate under precoding mode in interference Closed solutions

$R_1^{IC}=F_2(L_{s1}{P}_s,L_{b1}{P}_b{\mathrm{\ρ}}_1^2{2}^{-\frac{B_1}{M-1}},L_{b1}{P}_b(1-{\mathrm{\ρ}}_1^2\left)\right),$

In formula, ρ₁Represent that device-to-device link receives the channel time delay coefficient correlation of end subscriber 1, B₁Represent that device-to-device is received Channel condition information bit number from end subscriber 1 to base station feedback；Function F₂(c₀,c₁,c₂) definition be：

Wherein c₀, c₁, c₂It is argument of function；

Then calculate interference according to following formula to eliminate under precoding mode, the closed solutions R of the reachable transmission rate of system^IC：

$R^{\mathrm{IC}}=R_0^{\mathrm{IC}}+R_1^{\mathrm{IC}};$

(4) data transfer is realized：Base station according to the transmission mode selected in the step (3), using the amount of feedback in step (2) Change channel information and calculate precoding vector, the data that then will be sent to nearly base station user are multiplied by precoding vector, carry out data Transmission, edge customer is utilized and carried out data transmission with frequency resource.

2. adaptive cooperation transmission method in cellular mobile communication D2D systems according to claim 1, it is characterised in that In the step (3), the performance gain R of system acquisition under maximization speed precoding mode is calculated according to following formula^MR,↑：

$R^{MR,\↑}=R_1^{MR}+R_0^{MR}-R_0;$

The performance gain R that interference eliminates system acquisition under precoding mode is calculated according to following formula^IC,↑：

$R^{IC,\↑}=R_1^{IC}+R_0^{IC}-R_0;$

Wherein, R₀It is achievable rate of the community user 0 in the case where device-to-device communication condition is not introduced, is calculated according to following formula：

$R_0=F_1\left(L_{b0}{P}_b{\mathrm{\ρ}}_0^2\right(1-2^{-\frac{B_0}{M-1}}),L_{b0}{P}_b(1-{\mathrm{\ρ}}_0^2),0,M);$

Then the performance gain that system is obtained under the latter two transmission mode for introducing D2D communications is compared：If meeting R^MR,↑≥ R^IC,↑, then the transmission mode that selection maximization speed precoding mode is used as nearly base station user, otherwise selection interference elimination The transmission mode that precoding mode is used as nearly base station user.