CN105554808A

CN105554808A - D2D (Device-to-Device) resource distribution method and device in cellular network

Info

Publication number: CN105554808A
Application number: CN201610096832.7A
Authority: CN
Inventors: 喻鹏; 李文璟; 赵攀; 丰雷; 周凡钦
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2016-02-22
Filing date: 2016-02-22
Publication date: 2016-05-04
Anticipated expiration: 2036-02-22
Also published as: CN105554808B

Abstract

The embodiment of the invention discloses a D2D resource distribution method and device in cellular network. The method is: collecting channel state information of all cellular users and D2D pairs in the coverage of a base station, determining every channel gain, respectively calculating throughput capacity gains of every D2D pair respectively multiplexing the uplink and downlink channels of every cellular user; aiming at every D2D pair, determining the multiplexing mode of the D2D pair basing on the comparison of the throughput capacity gains multiplexing the uplink and downlink channels of every cellular user, determining multiple channels corresponding to the D2D pairs, finally determining the transmitting power of every D2D pair on corresponding multiple channels, and distributing the channels and the transmitting power to every D2D pair. According to the method and the device of the invention, the multiplexing mode is selected through analyzing and comparing the throughput capacity gains of the D2D user pairs; the multiplexing resources in the D2D pair system are greatly increased; and the system throughput capacity is greatly increased.

Description

Method and device for allocating resources to D2D in cellular network

Technical Field

The invention particularly relates to a method and a device for allocating resources by D2D in a cellular network.

Background

In recent years, with the rapid development of mobile multimedia services, the requirement of users for transmission rate is higher and higher, and the limited frequency band resource has not met the development requirement of wireless communication systems, so that D2D (Device-to-Device) communication technology has come into play and has attracted strong attention in the industry and academia at home and abroad. D2D communication is a new technology that allows nearby end users to communicate directly by reusing cell resources under the control of a cellular system. The D2D communication under the coverage of the cellular network can improve the frequency spectrum efficiency of the network, increase the system throughput, enlarge the cell coverage, reduce the burden of a base station, reduce the transmission delay and the equipment power consumption, and meet the user experience requirements.

However, on the other hand, because the D2D communication technology multiplexes spectrum resources of a cell, when a D2D user pair (referred to as a D2D pair for short) and a cellular user communicate on the same channel, co-channel interference may occur between the two users, and when the interference is serious, the communication quality of the two users may be seriously reduced, and the performance of the system may be reduced. Therefore, a reasonably effective channel resource allocation scheme for the D2D user pairs is a key issue for D2D communication research in cellular networks.

In the prior art, there are two main resource reuse modes of D2D communication in cellular networks: uplink channel multiplexing mode, such as chinese patent CN 103079262A; and a downlink channel multiplexing mode. Obviously, in both of the foregoing multiplexing modes, all D2D within the coverage area of the base station multiplex all downlink channels or all uplink channels, and the number of the D2D pairs of reusable resources is small, resulting in a small throughput of the D2D to the system.

Disclosure of Invention

The embodiment of the invention discloses a method and a device for allocating resources to D2D in a cellular network, which are used for solving the problems that in the existing resource allocation scheme, the number of reusable resources to D2D is small, and the throughput of the D2D to a system is small. The technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a method for allocating resources by D2D in a cellular network, including the steps of:

collecting channel state information of all cellular users and D2D pairs in the coverage area of the base station, and determining various channel gains, wherein the various channel gains comprise the channel gain between each cellular user and the base station, the channel gain between each D2D pair and the base station, the channel gain between each D2D pair and each cellular user respectively, and the channel gain inside each D2D pair;

respectively calculating the throughput gain of each D2D pair of the uplink channel and the downlink channel which respectively multiplex each cellular user according to the channel gains;

for each D2D pair, based on the comparison of the throughput gain of the uplink channel multiplexing each cellular user and the throughput gain of the downlink channel multiplexing each cellular user, determining the multiplexing mode of the D2D pair as: independently multiplexing an uplink channel for communication or independently multiplexing a downlink channel for communication;

determining a plurality of channels corresponding to each D2D pair according to the throughput gain of the uplink channel and the downlink channel of each cellular user respectively multiplexed by each D2D pair and the multiplexing mode of all D2D pairs;

and determining the transmission power of each D2D pair on the plurality of channels corresponding to the D2D pair and the plurality of channels corresponding to each D2D pair according to the multiplexing mode of the D2D pair and the plurality of channels corresponding to the D2D pair, and allocating the plurality of channels corresponding to the D2D pair and the corresponding transmission power on the plurality of channels corresponding to the D2D pair.

Preferably, the formula for separately calculating the throughput gain of each D2D on the uplink channel and the downlink channel that are respectively multiplexed by each cellular user includes:

T_{j i}^{u} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{C} g_{i j} + N_{0}}) + \log_{2} (1 + \frac{P_{C} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{C} g_{B i}}{N_{0}}),

T_{j i}^{d} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{B} g_{B j} + N_{0}}) + \log_{2} (1 + \frac{P_{B} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{B} g_{B i}}{N_{0}});

wherein,is D2D to D_jMultiplexing cellular user C_iThe throughput gain of the uplink channel of (1),is D2D to D_jMultiplexing cellular user C_iThe throughput gain of the downlink channel of (a),representing the maximum transmit power, P, of the D2D pair_BRepresenting the transmission power, P, of the base station_CRepresenting the transmission power, g, of a cellular user_BiIndicating cellular user C_iChannel gain with base station, g_BjRepresenting D2D vs D_jChannel gain with base station, g_ijRepresenting D2D vs D_jAnd cellular user C_iChannel gain between, g_jjRepresenting D2D vs D_jInternal channel gain G-G β d^-αWherein G is a path loss constant, β represents a channel fading constant, α is a path loss exponent, d is a distance from a transmitting end to a receiving end, and N is₀Representing the noise value.

Preferably, for each D2D pair, based on the comparison between the respective throughput gains of the multiplexed uplink channels and the respective throughput gains of the multiplexed downlink channels, the multiplexing mode of the D2D pair is determined as follows: the method for separately multiplexing the uplink channel for communication or separately multiplexing the downlink channel for communication comprises the following steps:

for each D2D pair, the average throughput gain of its multiplexed uplink channel is determined from the individual throughput gains of its multiplexed uplink channel, and further,

comparing each throughput gain of the multiplexing downlink channel with the average throughput gain respectively, counting the number of the throughput gains of the multiplexing downlink channels with values larger than the average throughput gain, if the number exceeds a preset integer, determining that the D2D communicates with the single multiplexing downlink channel, otherwise, determining that the D2D communicates with the single multiplexing uplink channel;

wherein the predetermined integer is 2M/N of rounding up or rounding down, M is the number of cellular users in the coverage area of the base station, and N is the number of pairs of D2D in the coverage area of the base station.

Preferably, the determining, according to the throughput gain of the uplink channel and the downlink channel of each D2D pair respectively multiplexing each cellular user and the multiplexing mode of all D2D pairs, a plurality of channels corresponding to each D2D pair includes:

a1: the uplink throughput gain matrix omega based on the multiplexing pattern of all D2D pairs^uIn the method, elements corresponding to D2D pairs for communicating separately multiplexed downlink channelsUpdating to any value less than or equal to zero; in the downlink throughput gain matrix omega^dIn the method, elements corresponding to D2D pairs for communicating by independently multiplexing uplink channelsUpdating to any value less than or equal to zero;

a2: initialization matrix X ═ X_ji]_N×MDefine phi 0_i＝{1,2,…,M}，W_j＝{1,2,…,N}；

A3 finding the row j ∈ W in the throughput gain matrix omega that satisfies_jColumn i ∈ Φ_iDetermines the element X of the matrix X corresponding to the row and to the column in which said maximum element is located_ji＝1；

A4: updating phi_i＝Ф_i\i，W_j＝W_j\j，

If it is notAnd isAnd in the throughput gain matrix omega, the corresponding i ∈ phi exists_iAnd j ∈ W_jElement T of_jiIf greater than 0, returning to step A3;

if it is notAnd in the throughput gain matrix omega, the corresponding i ∈ phi exists_iElement T of_jiGreater than 0, update W_jReturning to step A3;

if it is notOr throughput gain matrix omega, all corresponding i ∈ phi_iElement T of_jiIf the sum of the channel numbers is less than or equal to 0, determining an output matrix X, and determining a plurality of channels corresponding to each D2D pair according to the matrix X;

wherein, the cellular user C in the coverage area of the base station_iIs M, i ∈ {1,2, …, M }, D2D to D in the coverage area of the base station_jN, j ∈ {1,2, …, N }, M > N, throughput gain matrix Ω ═ T_ji]_N×MThe matrix X represents the uplink channel multiplexing condition matrix X^uOr downlink channel multiplexing situation matrix X^dWhen the matrix X represents the uplink channel multiplexing condition matrix X^uWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the uplink channel of the multiplexing cellular user^uElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of uplink channelWhen the matrix X represents the downlink channel multiplexing condition matrix X^dWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the downlink channel of the multiplexing cellular user^dElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of downlink channelCharacterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j，Characterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j。

Preferably, the determining the transmission power of each D2D pair on the plurality of channels corresponding to each D2D pair according to the multiplexing mode of the D2D pair and the plurality of channels corresponding to each D2D pair includes:

based on the multiplexing mode of the D2D pair, determining the signal-to-interference-and-noise ratio calculation formulas of the cellular users and the D2D pair as follows:

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}},

{SINR}_{j i}^{u} = \frac{P_{j i}^{u} g_{j j}}{x_{j i}^{u} P_{C} g_{i j} + N_{0}}, {SINR}_{j i}^{d} = \frac{P_{j i}^{d} g_{j j}}{x_{j i}^{d} P_{B} g_{B j} + N_{0}},

meanwhile, the formula for determining the maximum total rate and the constraint condition of the D2D system is as follows:

\begin{matrix} \max & R_{D 2 D} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} [x_{j i}^{u} \log_{2} (1 + {SINR}_{j i}^{u}) + x_{j i}^{d} \log_{2} (1 + {SINR}_{j i}^{d})] \end{matrix}

s.t.

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}} &GreaterEqual; γ^{u}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}} &GreaterEqual; γ^{d};

0 \leq Σ_{i = 1}^{M} x_{j i}^{u} P_{j i}^{u} \leq P_{j}^{D}; 0 \leq Σ_{i = 1}^{M} x_{j i}^{d} P_{j i}^{d} \leq P_{j}^{D};

0 \leq Σ_{j = 1}^{N} x_{j i}^{u} \leq 1; 0 \leq Σ_{j = 1}^{N} x_{j i}^{d} \leq 1;

(Σ_{i = 1}^{M} x_{j i}^{u}) (Σ_{i = 1}^{M} x_{j i}^{d}) = 0;

if it is

x_{j i}^{u} > 0, P_{j i}^{u} > 0,

Otherwise

P_{j i}^{u} = 0;

If it is

x_{j i}^{d} > 0, P_{j i}^{d} > 0,

Otherwise

P_{j i}^{d} = 0;

Based on the plurality of channels corresponding to each D2D pair, reducing the corresponding formula of the maximum total rate and the constraint condition of the D2D system to:

\begin{matrix} \max & R_{D 2 D}^{u} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{u} g_{j j}}{P_{C} g_{i j} + N_{0}}), \end{matrix}

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{u} g_{B j} \leq \frac{P_{C} g_{B i}}{γ^{u}} - N_{0}, \end{matrix}

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

\begin{matrix} \max & R_{D 2 D}^{d} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{d} g_{j j}}{P_{B} g_{B j} + N_{0}}), \end{matrix}

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{d} g_{i j} \leq \frac{P_{B} g_{B i}}{γ^{d}} - N_{0}, \end{matrix}

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

the function of the maximum total rate of the D2D system and the corresponding formula of the constraint condition are converted into Lagrangian functions:

its dual function is:

the operation formula for obtaining the optimal power solution according to the KKT condition is as follows:

solving each power optimal solution P based on an operation formula of the power optimal solution_ji ^*Based on the obtained optimal solution P for each power_ji ^*Allocating transmit power for each D2D pair on its corresponding plurality of channels; wherein, P_ji ^*Is D2D to D_jIn cellular user C_iTransmit power on an uplink or downlink channel;

wherein,

cellular user C in the coverage area of the base station_iIs M, i ∈ {1,2, …, M }, D2D to D in the coverage area of the base station_jThe number of (c) is N, j ∈ {1,2, …, N },for cellular user C_iThe signal-to-interference-and-noise ratio of the uplink channel,for cellular user C_iThe signal-to-interference-and-noise ratio of the downlink channel,is D2D to D_jMultiplexing cellular user C_iThe signal-to-interference-and-noise ratio of the uplink channel,is D2D to D_jMultiplexing cellular user C_iThe signal-to-interference-and-noise ratio of the downlink channel,representing D2D vs D_jFor cellular user C_iIn the upstream ofChannel reuse case, if cellular user C_iIs D2D to D_jMultiplex thenOtherwiseRepresenting D2D vs D_jFor cellular user C_iIf cellular user C is the downlink channel multiplexing situation_iIs D2D to D_jMultiplex thenOtherwiseChannel gain G-G β d^-αWherein G is a path loss constant, β represents a channel fading constant, α is a path loss exponent, d is a distance from a transmitting end to a receiving end, and the parameterRepresenting D2D vs D_jIn cellular user C_iThe transmit power on the uplink channel is,representing D2D vs D_jIn cellular user C_iTransmission power, P, on a downlink channel_BRepresenting the transmission power, P, of the base station_CIndicating cellular user C_iTransmit power of g_BiIndicating cellular user C_iChannel gain with base station, g_BjRepresenting D2D vs D_jChannel gain with base station, g_ijRepresenting D2D vs D_jAnd cellular user C_iChannel gain between, g_jjRepresenting D2D vs D_jInternal channel gain, N₀Representing a noise value; MaxR_D2DRepresents the maximum rate, γ, of a system of all D2D pairs^uIndicating uplink minimum for preset cellular usersSignal to noise ratio, gamma^dIndicating a preset downlink minimum signal-to-noise ratio of the cellular subscriber,representing the maximum transmit power of the D2D pair.

P_ji ^*Characterization of D2D vs D_jIn cellular user C_iPower optimum solution on uplink channelOr power optimum solution on downlink channelWhen P is present_ji ^*Characterizing power optimal solutions on uplink channelsWhen is, P_jiCharacterization of D2D vs D_jAt C_iTransmit power on uplink channelGamma represents the preset uplink minimum signal-to-noise ratio gamma of the cellular user^u(ii) a When P is present_ji ^*Characterizing power optimal solutions on downlink channelsWhen is, P_jiCharacterization of D2D vs D_jAt C_iTransmit power on downlink channelGamma represents the preset downlink minimum signal-to-noise ratio gamma of the cellular user^d；Andrespectively, D2D to D_jIn cellular user C_iTransmit power on an uplink channel and a downlink channel; mu.s_jAndfor two multiplicative factors, mu, in the Lagrangian function_j ^*Andis the optimal solution of the dual function obtained by the gradient search algorithm; maxAnd maxThe maximum total rate of the system formed by all D2D pairs in the uplink channel and the downlink channel respectively.

Preferably, the gradient search algorithm comprises the steps of:

b1: initializing multiplication factorsμ_j(0) And a step-size parameter,

where i ∈ {1,2, …, M }, j ∈ {1,2, …, N }, and the corresponding element x_ji≠0；

B2: updating

B3: updating t to t +1, and returning to the step B2; up toAnd (mu)_i(0),μ_i(1),…,μ_i(t +1)) all converge, so thatμ_j ^*＝μ_i(t+1)。

Wherein t represents the number of iterations of the gradient search algorithm,and mu_j(t) denotes multiplication factors at the t-th iteration, respectivelyμ_jElement x_jiCharacterizing elementsOr elementWhen P is present_ji ^*Characterizing power optimal solutions on uplink channelsWhen, the element x_jiCharacterizing elementsWhen P is present_ji ^*Characterizing power optimal solutions on downlink channelsWhen, the element x_jiCharacterizing elements

In another aspect, an embodiment of the present invention further provides an apparatus for allocating resources by D2D in a cellular network, including:

the channel gain determining module is used for collecting channel state information of all cellular users and D2D pairs in the coverage area of the base station and determining each channel gain, wherein each channel gain comprises the channel gain between each cellular user and the base station, the channel gain between each D2D pair and the base station, the channel gain between each D2D pair and each cellular user respectively and the channel gain inside each D2D pair;

a throughput gain calculation module, configured to calculate, according to the channel gains, throughput gains of each D2D pair of an uplink channel and a downlink channel that respectively multiplex each cellular user;

a multiplexing mode determining module, configured to determine, for each D2D pair, based on a comparison between a throughput gain of an uplink channel multiplexing each cellular user and a throughput gain of a downlink channel multiplexing each cellular user, a multiplexing mode of the D2D pair as follows: independently multiplexing an uplink channel for communication or independently multiplexing a downlink channel for communication;

a channel determining module, configured to determine, according to the throughput gain of each D2D pair for multiplexing the uplink channel and the downlink channel of each cellular user respectively and the multiplexing mode of all D2D pairs, a plurality of channels corresponding to each D2D pair;

and the resource allocation module is used for determining the transmitting power of each D2D pair on the plurality of channels corresponding to the D2D pair according to the multiplexing mode of the D2D pair and the plurality of channels corresponding to each D2D pair, and allocating the plurality of channels corresponding to each D2D pair and the corresponding transmitting power on the plurality of channels corresponding to the plurality of channels.

Preferably, the formula adopted by the throughput gain calculation module to calculate the throughput gain of each D2D for the uplink channel and the downlink channel that respectively multiplex each cellular user includes:

T_{j i}^{u} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{C} g_{i j} + N_{0}}) + \log_{2} (1 + \frac{P_{C} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{C} g_{B i}}{N_{0}}),

T_{j i}^{d} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{B} g_{B j} + N_{0}}) + \log_{2} (1 + \frac{P_{B} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{B} g_{B i}}{N_{0}});

wherein,is D2D to D_jMultiplexing cellular user C_iThe throughput gain of the uplink channel of (1),is D2D to D_jMultiplexing cellular user C_iThe throughput gain of the downlink channel of (a),representing the maximum transmit power, P, of the D2D pair_BRepresenting the transmission power, P, of the base station_CRepresenting the transmission power, g, of a cellular user_BiIndicating cellular user C_iChannel gain with base station, g_BjRepresenting D2D vs D_jWith base stationsChannel gain, g_ijRepresenting D2D vs D_jAnd cellular user C_iChannel gain between, g_jjRepresenting D2D vs D_jInternal channel gain G-G β d^-αWherein G is a path loss constant, β represents a channel fading constant, α is a path loss exponent, d is a distance from a transmitting end to a receiving end, and N is₀Representing the noise value.

Preferably, the multiplexing mode determining module is specifically configured to:

Preferably, the channel determination module includes:

a throughput gain updating unit for updating the throughput gain matrix omega in the uplink based on the multiplexing mode of all D2D pairs^uIn the method, elements corresponding to D2D pairs for communicating separately multiplexed downlink channelsUpdating to any value less than or equal to zero; in the downlink throughput gain matrix omega^dIn the method, elements corresponding to D2D pairs for communicating by independently multiplexing uplink channelsUpdating to any value less than or equal to zero;

an initialization unit for initializing a matrix X ═ X_ji]_N×MDefine phi 0_i＝{1,2,…,M}，W_j＝{1,2,…,N}；

Element x_jiA determination unit for finding out the row j ∈ W in the throughput gain matrix omega where it is satisfied_jColumn i ∈ Φ_iDetermines the element X of the matrix X corresponding to the row and to the column in which said maximum element is located_ji＝1；

Channel determination unit for updating phi_i＝Ф_i\i，W_j＝W_j\j，

If it is notAnd isAnd in the throughput gain matrix omega, the corresponding i ∈ phi exists_iAnd j ∈ W_jElement T of_jiGreater than 0, return element x_jiA determination unit;

if it is notAnd in the throughput gain matrix omega, the corresponding i ∈ phi exists_iElement T of_jiGreater than 0, update W_j1,2, …, N, returning element x_jiA determination unit;

wherein, the cellular user C in the coverage area of the base station_iIs M, i ∈ {1,2, …, M }, D2D to D in the coverage area of the base station_jN, j ∈ {1,2, …, N }, M > N, throughput gain matrix Ω ═ T_ji]_N×MThe matrix X represents the uplink channel multiplexing condition matrix X^uOr downlink channel multiplexing situation matrix X^dWhen the matrix X represents the uplink channel multiplexing condition matrix X^uWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the uplink channel of the multiplexing cellular user^uElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of uplink channelWhen the matrix X represents the downlink channel multiplexing condition matrix X^dWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the downlink channel of the multiplexing cellular user^dElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of downlink channelCharacterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j，Characterization ofCellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j。

Preferably, the resource allocation module includes:

a transmit power determination unit, configured to determine, based on the multiplexing mode of the D2D pair, the sir calculation formulas of the cellular users and the D2D pair to be:

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}},

{SINR}_{j i}^{u} = \frac{P_{j i}^{u} g_{j j}}{x_{j i}^{u} P_{C} g_{i j} + N_{0}}, {SINR}_{j i}^{d} = \frac{P_{j i}^{d} g_{j j}}{x_{j i}^{d} P_{B} g_{B j} + N_{0}},

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}} &GreaterEqual; γ^{u}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}} &GreaterEqual; γ^{d};

s.t.

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}} &GreaterEqual; γ^{u}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}} &GreaterEqual; γ^{d};

0 \leq Σ_{i = 1}^{M} x_{j i}^{u} P_{j i}^{u} \leq P_{j}^{D}; 0 \leq Σ_{i = 1}^{M} x_{j i}^{d} P_{j i}^{d} \leq P_{j}^{D};

0 \leq Σ_{j = 1}^{N} x_{j i}^{u} \leq 1; 0 \leq Σ_{j = 1}^{N} x_{j i}^{d} \leq 1;

(Σ_{i = 1}^{M} x_{j i}^{u}) (Σ_{i = 1}^{M} x_{j i}^{d}) = 0;

if it is

x_{j i}^{u} > 0, P_{j i}^{u} > 0,

Otherwise

P_{j i}^{u} = 0;

If it is

x_{j i}^{d} > 0, P_{j i}^{d} > 0,

Otherwise

P_{j i}^{d} = 0;

\begin{matrix} \max & R_{D 2 D}^{u} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{u} g_{j j}}{P_{C} g_{i j} + N_{0}}), \end{matrix}

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{u} g_{B j} \leq \frac{P_{C} g_{B i}}{γ^{u}} - N_{0}, \end{matrix}

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

\begin{matrix} \max & R_{D 2 D}^{d} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{d} g_{j j}}{P_{B} g_{B j} + N_{0}}), \end{matrix}

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{d} g_{i j} \leq \frac{P_{B} g_{B i}}{γ^{d}} - N_{0}, \end{matrix}

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

its dual function is:

wherein,

cellular user C in the coverage area of the base station_iIs M, i ∈ {1,2, …, M }, D2D to D in the coverage area of the base station_jThe number of (c) is N, j ∈ {1,2, …, N },for cellular user C_iThe signal-to-interference-and-noise ratio of the uplink channel,for cellular user C_iThe signal-to-interference-and-noise ratio of the downlink channel,is D2D to D_jMultiplexing cellular user C_iThe signal-to-interference-and-noise ratio of the uplink channel,is D2D to D_jMultiplexing cellular user C_iThe signal-to-interference-and-noise ratio of the downlink channel,representing D2D vs D_jFor cellular user C_iIf cellular user C is the uplink channel reuse situation_iIs D2D to D_jMultiplex thenOtherwiseRepresenting D2D vs D_jFor cellular user C_iIf cellular user C is the downlink channel multiplexing situation_iIs D2D to D_jMultiplex thenOtherwiseChannel gain G-G β d^-αWherein G is a path loss constant, β represents a channel fading constant, α is a path loss exponent, d is a distance from a transmitting end to a receiving end, and the parameterRepresenting D2D vs D_jIn cellular user C_iThe transmit power on the uplink channel is,representing D2D vs D_jIn cellular user C_iTransmission power, P, on a downlink channel_BRepresenting the transmission power, P, of the base station_CFor indicating cellsHousehold C_iTransmit power of g_BiIndicating cellular user C_iChannel gain with base station, g_BjRepresenting D2D vs D_jChannel gain with base station, g_ijRepresenting D2D vs D_jAnd cellular user C_iChannel gain between, g_jjRepresenting D2D vs D_jInternal channel gain, N₀Representing a noise value; MaxR_D2DRepresents the maximum rate, γ, of a system of all D2D pairs^uRepresenting a predetermined minimum signal-to-noise ratio, gamma, of the uplink of the cellular subscriber^dIndicating a preset downlink minimum signal-to-noise ratio of the cellular subscriber,represents the maximum transmit power of the D2D pair;

P_ji ^*characterization of D2D vs D_jIn cellular user C_iPower optimum solution on uplink channelOr power optimum solution on downlink channelWhen P is present_ji ^*Characterizing power optimal solutions on uplink channelsWhen is, P_jiCharacterization of D2D vs D_jAt C_iTransmit power on uplink channelGamma represents the preset uplink minimum signal-to-noise ratio gamma of the cellular user^u(ii) a When P is present_ji ^*Characterizing power optimal solutions on downlink channelsWhen is, P_jiCharacterization of D2D vs D_jAt C_iDownlink channelTransmit power ofGamma represents the preset downlink minimum signal-to-noise ratio gamma of the cellular user^d；Andrespectively, D2D to D_jIn cellular user C_iTransmit power on an uplink channel and a downlink channel; mu.s_jAndfor two multiplicative factors, mu, in the Lagrangian function_j ^*Andis the optimal solution of the dual function obtained by the gradient search algorithm; maxAnd maxThe maximum total rate of a system formed by all D2D pairs in the uplink channel and the downlink channel respectively;

and the channel and power allocation unit is used for allocating the plurality of corresponding channels and the corresponding transmitting powers on the plurality of corresponding channels for each pair of D2D.

Preferably, the apparatus for implementing the gradient search algorithm comprises:

initial submodule: for initializing multiplication factorsμ_j(0) And a step-size parameter,

Updating the submodule: for updating

Determining a submodule: the updating submodule is used for updating t to t +1 and returning the updated t to the updating submodule; up toAnd (mu)_i(0),μ_i(1),…,μ_i(t +1)) all converge, so thatμ_j ^*＝μ_i(t+1)；

The method and the device for allocating the D2D pair resources in the cellular network provided by the embodiment of the invention adopt a mode that one D2D user pair multiplexes a plurality of cellular user resources, and simultaneously analyze and compare throughput gains generated by specific space diversity of the D2D user pair to select an optimal multiplexing mode, so that the reusable resources of the D2D pair system are greatly increased, the throughput of the D2D pair system is greatly increased, the problem of serious interference between the D2D pair and the cellular users is further avoided, more D2D user pairs are accessed into the network, and the overall performance of the system formed by the D2D user pairs is improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a diagram of a suitable scenario for an embodiment of the present invention;

fig. 2 is a flowchart illustrating a method for allocating resources by D2D in a cellular network according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of a D2D pair resource allocation apparatus in a cellular network according to an embodiment of the present invention;

FIG. 4 is a bar graph of various system throughputs provided by embodiments of the present invention;

FIG. 5 is a line graph of D2D throughput versus number of cellular users provided by an embodiment of the present invention;

FIG. 6 is a line graph of D2D throughput versus D2D log number provided by an embodiment of the present invention;

fig. 7 is a line graph of the relationship between D2D throughput and the distance of two users in each D2D pair provided by an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that there are three channel allocation manners for D2D communication in the cellular network: firstly, one D2D user multiplexes one resource block, for example, in Chinese patent CN103686743A, the problem of low spectrum utilization rate exists in the channel allocation mode; secondly, a plurality of D2D users reuse one resource block, and the channel allocation mode can generate serious interference; third, one D2D user multiplexes multiple resource blocks, and the third channel multiplexing method described above is adopted in the embodiment of the present invention. A resource block as referred to herein refers to a channel, including an uplink channel and a downlink channel of a cellular user.

The method of the embodiment of the invention can be applied to a server of the base station, and of course, can also be applied to an equipment server associated with the base station equipment. The application scenario of the embodiment of the present invention is shown in fig. 1, and within the coverage area of the base station BS, there are three cellular users C₁、C₁、C₃And two D2D pairs D₁、D₂(ii) a In the figure, a straight solid line with an arrow indicates an uplink data signal, a straight dotted line with an arrow indicates a downlink data signal, a wavy solid line with an arrow indicates an uplink interference signal, and a wavy dotted line with an arrow indicates a downlink interference signal.

The method for allocating resources by D2D in the cellular network provided by the embodiment of the present invention may include the following steps:

s1: collecting channel state information of all cellular users and D2D pairs in the coverage area of the base station, and determining various channel gains, wherein the various channel gains comprise the channel gain between each cellular user and the base station, the channel gain between each D2D pair and the base station, the channel gain between each D2D pair and each cellular user respectively, and the channel gain inside each D2D pair;

s2: respectively calculating the throughput gain of each D2D pair of the uplink channel and the downlink channel which respectively multiplex each cellular user according to the channel gains;

in the embodiment of the present invention, the formula for calculating the throughput gain of each D2D for multiplexing the uplink channel and the downlink channel of each cellular user respectively may be:

T_{j i}^{u} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{C} g_{i j} + N_{0}}) + \log_{2} (1 + \frac{P_{C} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{C} g_{B i}}{N_{0}}),

T_{j i}^{d} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{B} g_{B j} + N_{0}}) + \log_{2} (1 + \frac{P_{B} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{B} g_{B i}}{N_{0}});

wherein,is D2D to D_jMultiplexing cellular user C_iThe throughput gain of the uplink channel of (1),is D2D to D_jMultiplexing cellular user C_iIt should be noted that D is described herein_jAnd C_iThe D2D pair and the cellular user number, respectively;

representing the maximum transmit power, P, of the D2D pair_BRepresenting the transmission power, P, of the base station_CThe transmission power of the cellular user is shown, and it should be noted that the maximum transmission power of the D2D pair, the transmission power of the base station, and the transmission power of the cellular user are all determined by searching in the communication standard, and can be used directly without going throughPerforming additional calculation;

g in the formula_BiIndicating cellular user C_iChannel gain with base station, g_BjRepresenting D2D vs D_jChannel gain with base station, g_ijRepresenting D2D vs D_jAnd cellular user C_iChannel gain between, g_jjRepresenting D2D vs D_jThe internal channel gain is calculated according to the formula G-G β d^-αWherein G is a path loss constant, β represents a channel fading constant, α is a path loss exponent, d is a distance from a transmitting end to a receiving end, and the unit of d is km, and N₀Represents the noise value, N₀The four parameters G, β, α and d belong to the parameters collected in step S1.

S3: for each D2D pair, based on the comparison of the throughput gain of the uplink channel multiplexing each cellular user and the throughput gain of the downlink channel multiplexing each cellular user, determining the multiplexing mode of the D2D pair as: independently multiplexing an uplink channel for communication or independently multiplexing a downlink channel for communication;

it should be noted that some D2D pairs reuse channels of any cellular user due to their own location diversity, and their throughput gain is greater than that of other D2D pairs, which results in that the base station allocates more channels to the same D2D pair, and other D2D pairs cannot access the network because their throughput gain is smaller and cannot be divided into channels.

In order to enable more D2D pairs to access the network, in the embodiment of the present invention, for each D2D pair, based on the comparison between the throughput gain of the uplink channel multiplexing each cellular user and the throughput gain of the downlink channel multiplexing each cellular user, the multiplexing mode of the D2D pair is determined as follows: the multiplexing uplink channels separately for communication or multiplexing downlink channels separately for communication may include:

for each D2D pair, determining the average throughput gain of the multiplexing uplink channel according to the throughput gain of the multiplexing uplink channel, further comparing the throughput gain of the multiplexing downlink channel with the average throughput gain, and counting the number of the throughput gains of the multiplexing downlink channels with the value larger than the average throughput gain, if the number exceeds a predetermined integer, determining that the D2D communicates with the individual multiplexing downlink channel, otherwise, the D2D communicates with the individual multiplexing uplink channel;

For example, in the embodiment of the present invention, it is assumed that there are M cellular users and N D2D pairs, and it should be noted that the base station allocates one uplink channel and one downlink channel to each cellular user, so for each D2D pair, the sum of M throughput gains of its multiplexed uplink channel may be calculated first, then divided by M to obtain an average throughput gain of its multiplexed uplink channel, then the M throughput gains of its multiplexed downlink channel are compared with the average throughput gain, and the statistical value is greater than the number of throughput gains of the multiplexed downlink channel of the average throughput gain;

if the number exceeds a predetermined integer, it is determined that the D2D communicates with the individual multiplexed downlink channel, otherwise, the D2D communicates with the individual multiplexed uplink channel, for example, M is 20, N is 15, in the embodiment of the present invention, 2M/N is 2.667, if rounded up, the value is 3, and if rounded down, the value is 2, assuming that rounded up is adopted, the number of throughput gains of the corresponding D2D to the multiplexed downlink channel, whose value should be larger than the average throughput gain, is counted, if the number exceeds 3, it is determined that the D2D communicates with the individual multiplexed downlink channel, otherwise, the D2D communicates with the individual multiplexed uplink channel.

The 2M/N is selected mainly in consideration of fairness of channel allocation and multiplexing of uplink or downlink channels of cellular users by only one D2D pair at most. In the embodiment of the invention, when M uplink channels and M downlink channels are provided in the coverage area of the base station, the total number of the D2D pairs of reusable channels is 2M, and the number of the channels allocated to each D2D pair is as same as possible in consideration of the fairness among the D2D pairs, so that the number of the resources allocated to each D2D pair is 2M/N.

S4: determining a plurality of channels corresponding to each D2D pair according to the throughput gain of the uplink channel and the downlink channel of each cellular user respectively multiplexed by each D2D pair and the multiplexing mode of all D2D pairs;

the determining, according to the throughput gain of the uplink channel and the downlink channel of each D2D pair respectively multiplexing each cellular user and the multiplexing mode of all D2D pairs, a plurality of channels corresponding to each D2D pair may include:

any value less than or equal to zero in the above steps indicates that D2D corresponds to the corresponding elementOrAny value less than or equal to zero is updated, and it should be noted that any value less than or equal to zero corresponding to updating may be equal or different for different pairs of D2D; of course, in an actual computer algorithm, the elements corresponding to the pairs of D2D that communicate with the individual multiplexed downlink channels may be directly mapped toUpdating to zero value, and corresponding elements of D2D pair for separately multiplexing uplink channels for communicationAnd also updated to a value of zero.

A2: initialization matrix X ═ X_ji]_N×MSetting all elements in matrix X with N rows and M columns to zero to make matrix X be zero matrix, and defining two arrays phi_i＝{1,2,…,M}，W_j＝{1,2,…,N}。

For example, assume that the throughput gain matrix Ω at D2D for the uplink channel of the multiplexed cellular users^uIn row 5, column 7For the current matrix omega^uBased on the maximum element in (i), the maximum element X in matrix X is set to 5 in row j and 7 in column i, where j is 5 in row and i is 7 in column₅₇The value is 1, it is noted that j-5 and i-7 as described herein should both satisfy 5 ∈ W_jAnd 7 ∈ Φ_i。

A4: updating phi_i＝Ф_i\i，W_j＝W_j\ j, i.e. the updated array phi_iTo remove element i from the original array; updated array W_jTo remove element j from the original array, e.g. original array W_j＝{1,2,3,4,5,6}，Ф_i1,2,3,4,5,6,7,8, in step a3, the element in row 5, column 7For the current matrix omega^uThe largest element in (1), then from array W_jRemoving element 5 from the array phi_iElement 7 is removed, so the two updated arrays are: w_j＝{1,2,3,4,6}，Ф_i＝{1,2,3,4,5,6,8}。

if it is notOr throughput gain matrix omega, all corresponding i ∈ phi_iElement T of_jiAnd if the sum is less than or equal to 0, determining an output matrix X, and determining a plurality of channels corresponding to each D2D pair according to the matrix X.

In addition, it should be noted that in the embodiment of the present invention, one channel can be multiplexed by only one D2D pair, so that when the channel is multiplexed, the channel can be multiplexed by only one D2D pairThat is, M uplink channels or M downlink channels of the cellular user have been all multiplexed by the D2D pairs, and there is no channel that can be multiplexed within the coverage area of the base station, at this time, the algorithm should be stopped, and the output matrix X is determined;

and when the throughput gain matrix omega is in, all corresponding i ∈ phi_iAnd j ∈ W_jElement T of_jiWhen the number of channels is less than or equal to 0, no matter whether channels which can be multiplexed exist in the coverage area of the base station, the throughput gain of the D2D to the remaining channels for multiplexing is zero or even negative, and is not beneficial or not beneficial to the increase of the throughput of the whole D2D system, so the algorithm needs to be stopped in this case, and the output matrix X needs to be determined.

In the above steps a 1-a 4, the number of cellular users in the coverage area of the base station is M, i ∈ {1,2, …, M }, the number of pairs D2D is N, j ∈ {1,2, …, N }, where M > N, and the throughput gain matrix Ω ═ T ═ N_ji]_N×M；

Matrix X represents uplink channel multiplexing condition matrix X^uOr downlink channel multiplexing situation matrix X^dWhen the matrix X represents the uplink channel multiplexing condition matrix X^uWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the uplink channel of the multiplexing cellular user^uElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of uplink channelWhen the matrix X represents the downlink channel multiplexing condition matrix X^dWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the downlink channel of the multiplexing cellular user^dElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of downlink channel

In the matrix X that is finally output,characterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j，Characterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j。

It should be emphasized that, the steps a1 to a4 are more effective to avoid that the base station allocates too many channels to the D2D pair with better throughput gain, and the D2D pair with worse throughput gain cannot allocate channels; steps a1 to a4 allow all D2D pairs to be allocated to channels for communication, and thus, embodiments of the present invention allow more D2D pairs to access the network.

S5: and determining the transmission power of each D2D pair on the plurality of channels corresponding to the D2D pair and the plurality of channels corresponding to each D2D pair according to the multiplexing mode of the D2D pair and the plurality of channels corresponding to the D2D pair, and allocating the plurality of channels corresponding to the D2D pair and the corresponding transmission power on the plurality of channels corresponding to the D2D pair.

The determining the transmission power of each D2D pair on the plurality of channels corresponding to each D2D pair according to the multiplexing mode of the D2D pair and the plurality of channels corresponding to each D2D pair may be:

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}},

{SINR}_{j i}^{u} = \frac{P_{j i}^{u} g_{j j}}{x_{j i}^{u} P_{C} g_{i j} + N_{0}}, {SINR}_{j i}^{d} = \frac{P_{j i}^{d} g_{j j}}{x_{j i}^{d} P_{B} g_{B j} + N_{0}},

maximum total rate calculation formula for the D2D system:

\begin{matrix} \max & R_{D 2 D} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} [x_{j i}^{u} \log_{2} (1 + {SINR}_{j i}^{u}) + x_{j i}^{d} \log_{2} (1 + {SINR}_{j i}^{d})] \end{matrix}

s.t. the constraint is:

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}} &GreaterEqual; γ^{u}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}} &GreaterEqual; γ^{d},

this constraint represents each cellular user C_iMinimum channel quality requirement of;

0 \leq Σ_{i = 1}^{M} x_{j i}^{u} P_{j i}^{u} \leq P_{j}^{D}; 0 \leq Σ_{i = 1}^{M} x_{j i}^{d} P_{j i}^{d} \leq P_{j}^{D},

this constraint represents D2D versus D_jPower constraints of (d);

0 \leq Σ_{j = 1}^{N} x_{j i}^{u} \leq 1; 0 \leq Σ_{j = 1}^{N} x_{j i}^{d} \leq 1,

in the above formulaRepresentation for cellular user C_iAll ofThe sum of (a) and (b),representation for cellular user C_iAll ofDue to the fact thatAndis 0 or 1, so the constraint indicates that each cellular user's uplink and downlink channels can be multiplexed by only one D2D pair at most;

(Σ_{i = 1}^{M} x_{j i}^{u}) (Σ_{i = 1}^{M} x_{j i}^{d}) = 0,

in view of the embodiment of the present invention, the multiplexing mode of the D2D pair is: multiplexing uplink channels alone for communication or multiplexing downlink channels alone for communication, for D2D for D_jIt is not possible to multiplex the uplink and downlink channels simultaneously, so when D2D is paired with D_jWhen the uplink channel is multiplexed individually for communication, D2D is associated with D_jAll ofAre all 0, i.e.Similarly, when D2D is paired with D_jWhen the downlink channel is multiplexed individually for communication, D2D is associated with D_jAll ofAre all 0, i.e.

If it is

x_{j i}^{u} > 0, P_{j i}^{u} > 0,

Otherwise

P_{j i}^{u} = 0;

If it is

x_{j i}^{d} > 0, P_{j i}^{d} > 0,

Otherwise

P_{j i}^{d} = 0,

This constraint represents: D2D pair D_jMultiplexing cellular user C_iIn the uplink channel of the wireless communication system,thenHas a value; D2D pair D_jMultiplexing cellular user C_iIn the case of the downlink channel of (1),thenThere is a value. Obviously, D2D vs. D_jMultiplexing cellular user C_iIn the uplink channel of the wireless communication system,D2D pair D_jMultiplexing cellular user C_iIn the case of the downlink channel of (1),

after each D2D determines the multiplexed channels, the D2D method for calculating the transmission power on the corresponding channel can be implemented by the prior art.

In order to optimize the power calculation method, in the embodiment of the present invention, first, based on a plurality of channels corresponding to each pair of D2D, the corresponding formulas of the maximum total rate and the constraint condition of the D2D system are simplified as follows:

\begin{matrix} \max & R_{D 2 D}^{u} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{u} g_{j j}}{P_{C} g_{i j} + N_{0}}), \end{matrix}

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{u} g_{B j} \leq \frac{P_{C} g_{B i}}{γ^{u}} - N_{0}, \end{matrix}

\begin{matrix} \max & R_{D 2 D}^{d} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{d} g_{j j}}{P_{B} g_{B j} + N_{0}}), \end{matrix}

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{d} g_{i j} \leq \frac{P_{B} g_{B i}}{γ^{d}} - N_{0}, \end{matrix}

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

its dual function is:

wherein,

cellular user C in the coverage area of the base station_iIs M, i ∈ {1,2, …, M }, D2D to D in the coverage area of the base station_jThe number of (c) is N, j ∈ {1,2, …, N },for cellular user C_iThe signal-to-interference-and-noise ratio of the uplink channel,for cellular user C_iThe signal-to-interference-and-noise ratio of the downlink channel,is D2D to D_jMultiplexing cellular user C_iThe signal-to-interference-and-noise ratio of the uplink channel,is D2D to D_jMultiplexing cellular user C_iThe signal-to-interference-and-noise ratio of a downlink channel;

representing D2D vs D_jFor cellular user C_iIf cellular user C is the uplink channel reuse situation_iIs D2D to D_jMultiplex thenOtherwiseRepresenting D2D vs D_jFor cellular user C_iIf cellular user C is the downlink channel multiplexing situation_iOf the downlink channelIs paired with D2D_jMultiplex thenOtherwise

x_{j i}^{d} = 0;

Channel gain G-G β d^-αWherein G is a path loss constant, β represents a channel fading constant, α is a path loss exponent, d is a distance from a transmitting end to a receiving end, and the parameterRepresenting D2D vs D_jIn cellular user C_iThe transmit power on the uplink channel is,representing D2D vs D_jIn cellular user C_iTransmission power, P, on a downlink channel_BRepresenting the transmission power, P, of the base station_CIndicating cellular user C_iTransmit power of g_BiIndicating cellular user C_iChannel gain with base station, g_BjRepresenting D2D vs D_jChannel gain with base station, g_ijRepresenting D2D vs D_jAnd cellular user C_iChannel gain between, g_jjRepresenting D2D vs D_jInternal channel gain, N₀Representing a noise value;

maxR_D2Drepresents the maximum rate, γ, of a system of all D2D pairs^uRepresenting a predetermined minimum signal-to-noise ratio, gamma, of the uplink of the cellular subscriber^dIndicating a preset downlink minimum signal-to-noise ratio of the cellular subscriber,represents the maximum transmit power of the D2D pair;

P_ji ^*characterization of D2D vs D_jIn cellular user C_iPower optimum solution on uplink channelOr power optimum solution on downlink channelWhen P is present_ji ^*Characterizing power optimal solutions on uplink channelsWhen is, P_jiCharacterization of D2D vs D_jAt C_iTransmit power on uplink channelGamma represents the preset uplink minimum signal-to-noise ratio gamma of the cellular user^u(ii) a When P is present_ji ^*Characterizing power optimal solutions on downlink channelsWhen is, P_jiCharacterization of D2D vs D_jAt C_iTransmit power on downlink channelGamma represents the preset downlink minimum signal-to-noise ratio gamma of the cellular user^d；Andrespectively, D2D to D_jIn cellular user C_iTransmit power on an uplink channel and a downlink channel;

μ_jandfor two multiplicative factors, mu, in the Lagrangian function_j ^*Andis the optimal solution of the dual function obtained by the gradient search algorithm; maxAnd maxThe maximum total rate of the system formed by all D2D pairs in the uplink channel and the downlink channel respectively.

In addition, the gradient search algorithm includes the steps of:

b1: initializing multiplication factorsμ_j(0) And step size parameter, in the embodiment of the invention, assignment can be initializedμ_j(0)＝0、＝0.01；

B2: updating

It should be noted that when x_jiWhen equal to 0, corresponding P_ji ^*The value of (a) is inevitably zero, so when using the gradient search algorithm, the x corresponding to the corresponding i and j should be judged first_jiWhether or not the value of (A) is zero, only x_jiThe gradient search algorithm is only needed if not equal to 0.

The embodiment of the invention carries out simulation experiments in a single base station cell with the radius of 500M, wherein cellular users and D2D users are uniformly distributed, the transmitting power of the base station is 46dB, the transmitting power of the cellular users is 23dB, the maximum transmitting power of D2D is 23dB, and the interference threshold values of uplink and downlink channels of the cellular users are 7dB and 13dB respectively.

In fig. 4 to 7, the PCUOD, the WU, and the WD characterize three channel allocation modes, the PCUOD characterizes a channel allocation manner adopted in the embodiment of the present invention, the WU characterizes an uplink channel of D2D to all multiplexing cellular users, and the WD characterizes a downlink channel of D2D to all multiplexing cellular users. The CUs in fig. 3 and 4 represent cellular users. The D2D throughputs illustrated in FIGS. 5-7 and described below represent the throughputs of the system of all D2D pairs.

When the number of cellular users is 60, the number of D2D pairs is 20, and the distance between two users in each D2D pair is 20 meters, the situation that the three channel allocation manners correspond to different system throughputs is shown in fig. 4, and it is obvious that, by using the channel allocation manner adopted in the embodiment of the present invention, the throughputs of all D2D pairs and the throughputs of all D2D pairs and all cellular users are the largest.

In fig. 4, "CUup" represents a system formed by uplink channels of all cellular subscribers, "CUdown" represents a system formed by downlink channels of all cellular subscribers, "CUall" represents a system formed by all cellular subscribers, "D2D" represents a system formed by all D2D pairs, and "Total" represents a system formed by all D2D pairs and all cellular subscribers.

When the number of D2D pairs is 20 and the distance between two users in each D2D pair is 20 meters, the throughput of the system formed by all D2D pairs varies as the number of cellular users varies as shown in fig. 5. As can be seen from the figure, the throughput of the system formed by all D2D pairs increases with the number of cellular users, regardless of the channel allocation method; obviously, as can be seen from fig. 5, the channel allocation method adopted in the embodiment of the present invention has a significantly better improvement on the D2D throughput than other channel allocation methods.

As shown in fig. 5, when the number of cellular users is 100, the channel allocation method according to the embodiment of the present invention has a corresponding D2D throughput that is 52% higher than the D2D throughput when the WU channel allocation method is used, and 225% higher than the D2D throughput when the WD channel allocation method is used.

When the number of cellular users is 60 and the distance between two users in each D2D pair is 20 meters, the throughput of the system formed by all D2D pairs varies as the number of D2D pairs varies as shown in fig. 6. As can be seen from the figure, in any channel allocation method, the throughput of the system formed by all D2D pairs increases with the number of D2D pairs, and since the total number of D2D pairs of reusable channels is not changed, the throughput of D2D increases more slowly.

As can be seen from fig. 6, when the number of cellular users and the distance between two users in each D2D pair are constant, as the number of D2D pairs changes, the corresponding D2D throughput is always significantly higher than the corresponding D2D throughput of other channel allocation methods when the channel allocation method of the embodiment of the present invention is adopted. When the number of the D2D pairs is 30, the channel allocation method of the embodiment of the invention has the corresponding D2D throughput which is 45% higher than the D2D throughput of the WU channel allocation method and 190% higher than the D2D throughput of the WD channel allocation method.

When the number of cellular users is 60 and the number of pairs of D2D is 20, the throughput of the system formed by all pairs of D2D varies as the distance between the two users in each pair of D2D varies as shown in fig. 7. As can be seen from the figure, all D2D throughputs decrease with increasing distance between two users in the D2D pair, regardless of the channel allocation method. The distance between the abscissas "D2D" in fig. 7 is the distance between the two users in each of the D2D pairs described above.

As can be seen from fig. 7, in the process of changing the distance between two users in the D2D pair, the throughput of D2D corresponding to the channel allocation method according to the embodiment of the present invention is always significantly higher than the throughput of D2D corresponding to the other channel allocation methods. When the distance between two users in each D2D pair is 35 meters, the corresponding D2D throughput is 65% higher than the D2D throughput when the WU channel allocation method is adopted and 186% higher than the D2D throughput when the WD channel allocation method is adopted by adopting the channel allocation method of the embodiment of the invention.

The method for allocating the resources by the D2D pair in the cellular network provided by the embodiment of the invention adopts a mode that one D2D user pair multiplexes a plurality of cellular user resources, and simultaneously analyzes and compares the throughput gain generated by the specific space diversity of the D2D user pair to select the optimal multiplexing mode, so that the resources which can be multiplexed by the D2D pair in the system are greatly increased, the throughput of the D2D pair in the system is greatly increased, the problem of serious interference between the D2D pair and the cellular users is further avoided, more D2D user pairs are accessed into the network, and the overall performance of the system formed by the D2D user pairs is improved.

Corresponding to the foregoing method embodiment, as shown in fig. 3, an embodiment of the present invention further provides an apparatus for allocating resources by D2D pair in a cellular network, where the apparatus includes:

a channel gain determining module 210, configured to collect channel state information of all cellular users and D2D pairs within the coverage area of the base station, and determine channel gains, where the channel gains include a channel gain between each cellular user and the base station, a channel gain between each D2D pair and the base station, a channel gain between each D2D pair and each cellular user, and a channel gain inside each D2D pair;

a throughput gain calculation module 220, configured to calculate, according to the channel gains, throughput gains of each D2D pair of an uplink channel and a downlink channel that respectively multiplex each cellular user;

a multiplexing mode determining module 230, configured to determine, for each D2D pair, a multiplexing mode of the D2D pair based on a comparison between a throughput gain of an uplink channel multiplexing each cellular user and a throughput gain of a downlink channel multiplexing each cellular user: independently multiplexing an uplink channel for communication or independently multiplexing a downlink channel for communication;

a channel determining module 240, configured to determine, according to the throughput gain of each D2D pair for multiplexing the uplink channel and the downlink channel of each cellular user respectively and the multiplexing mode of all D2D pairs, a plurality of channels corresponding to each D2D pair;

and the resource allocation module 250 is configured to determine, according to the multiplexing mode of the D2D pairs and the multiple channels corresponding to each D2D pair, the transmit power of each D2D pair on the multiple channels corresponding to the pair, and allocate, for each D2D pair, the multiple channels corresponding to the pair and the corresponding transmit powers on the multiple channels corresponding to the pair.

Specifically, the formula adopted by the throughput gain calculation module 220 to calculate the throughput gain of each D2D for multiplexing the uplink channel and the downlink channel of each cellular user respectively includes:

T_{j i}^{u} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{C} g_{i j} + N_{0}}) + \log_{2} (1 + \frac{P_{C} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{C} g_{B i}}{N_{0}}),

T_{j i}^{d} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{B} g_{B j} + N_{0}}) + \log_{2} (1 + \frac{P_{B} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{B} g_{B i}}{N_{0}});

Specifically, the multiplexing mode determining module 230 is specifically configured to:

Specifically, the channel determining module 240 includes:

Channel determination unit for updating phi_i＝Ф_i\i，W_j＝W_j\j，

wherein, the cellular user C in the coverage area of the base station_iIs M, i ∈ {1,2, …, M }, D2D to D in the coverage area of the base station_jN, j ∈ {1,2, …, N }, M > N, throughput gain matrix Ω ═ T_ji]_N×M；

Matrix X represents uplink channel multiplexing condition matrix X^uOr downlink channel multiplexing situation matrix X^dWhen the matrix X represents the uplink channel multiplexing condition matrix X^uWhen, the element x_jiCharacterizing elementsThroughput gain matrix omega characterizes D2D pair multiplexing cellular user uplink channelsOf the throughput gain matrix omega^uElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of uplink channelWhen the matrix X represents the downlink channel multiplexing condition matrix X^dWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the downlink channel of the multiplexing cellular user^dElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of downlink channel

Characterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j，Characterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j。

Specifically, the resource allocation module 250 includes a transmission power determination unit and a channel and power determination unit.

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}},

{SINR}_{j i}^{u} = \frac{P_{j i}^{u} g_{j j}}{x_{j i}^{u} P_{C} g_{i j} + N_{0}}, {SINR}_{j i}^{d} = \frac{P_{j i}^{d} g_{j j}}{x_{j i}^{d} P_{B} g_{B j} + N_{0}},

\begin{matrix} \max & R_{D 2 D} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} [x_{j i}^{u} \log_{2} (1 + {SINR}_{j i}^{u}) + x_{j i}^{d} \log_{2} (1 + {SINR}_{j i}^{d})] \end{matrix}

s.t.

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}} &GreaterEqual; γ^{u}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}} &GreaterEqual; γ^{d};

0 \leq Σ_{i = 1}^{M} x_{j i}^{u} P_{j i}^{u} \leq P_{j}^{D}; 0 \leq Σ_{i = 1}^{M} x_{j i}^{d} P_{j i}^{d} \leq P_{j}^{D};

0 \leq Σ_{j = 1}^{N} x_{j i}^{u} \leq 1; 0 \leq Σ_{j = 1}^{N} x_{j i}^{d} \leq 1;

(Σ_{i = 1}^{M} x_{j i}^{u}) (Σ_{i = 1}^{M} x_{j i}^{d}) = 0;

if it is

x_{j i}^{u} > 0, P_{j i}^{u} > 0,

Otherwise

P_{j i}^{u} = 0;

If it is

x_{j i}^{d} > 0, P_{j i}^{d} > 0,

Otherwise

P_{j i}^{d} = 0;

\begin{matrix} \max & R_{D 2 D}^{u} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{u} g_{j j}}{P_{C} g_{i j} + N_{0}}), \end{matrix}

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{u} g_{B j} \leq \frac{P_{C} g_{B i}}{γ^{u}} - N_{0}, \end{matrix}

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

\begin{matrix} \max & R_{D 2 D}^{d} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{d} g_{j j}}{P_{B} g_{B j} + N_{0}}), \end{matrix}

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{d} g_{i j} \leq \frac{P_{B} g_{B i}}{γ^{d}} - N_{0}, \end{matrix}

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

its dual function is:

wherein,

representing D2D vs D_jFor cellular user C_iIf cellular user C is the uplink channel reuse situation_iIs D2D to D_jMultiplex thenOtherwiseRepresenting D2D vs D_jFor cellular user C_iIf cellular user C is the downlink channel multiplexing situation_iIs D2D to D_jMultiplex thenOtherwise

x_{j i}^{d} = 0;

Specifically, the device for implementing the gradient search algorithm includes:

Updating the submodule: for updating

Determining a submodule: the updating submodule is used for updating t to t +1 and returning the updated t to the updating submodule; up toAndall converge and makeμ_j ^*＝μ_i(t+1)；

The resource allocation device of the D2D pair in the cellular network provided by the embodiment of the invention adopts a mode that one D2D user pair multiplexes a plurality of cellular user resources, and simultaneously, the optimal multiplexing mode is selected according to the analysis and comparison of throughput gains generated by specific space diversity of the D2D user pair, so that the reusable resources of the D2D pair in the system are greatly increased, the throughput of the D2D pair in the system is greatly increased, the problem of serious interference between the D2D pair and the cellular users is further avoided, more D2D user pairs are accessed into the network, and the overall performance of the system formed by the D2D user pairs is improved.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. The term "comprising", without further limitation, means that the element so defined is not excluded from the group consisting of additional identical elements in the process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

Those skilled in the art will appreciate that all or part of the steps in the above method embodiments may be implemented by a program to instruct relevant hardware to perform the steps, and the program may be stored in a computer-readable storage medium, which is referred to herein as a storage medium, such as: ROM/RAM, magnetic disk, optical disk, etc.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for D2D pair resource allocation in a cellular network, the method comprising the steps of:

2. The method of claim 1, wherein the formula for separately calculating the throughput gain of each D2D for the uplink channel and the downlink channel multiplexing each cellular user comprises:

T_{j i}^{u} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{C} g_{i j} + N_{0}}) + \log_{2} (1 + \frac{P_{C} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{C} g_{B i}}{N_{0}}),

T_{j i}^{d} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{B} g_{B j} + N_{0}}) + \log_{2} (1 + \frac{P_{B} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{B} g_{B i}}{N_{0}});

3. The method of claim 1, wherein for each D2D pair, based on the comparison of the respective throughput gain of its multiplexed uplink channel and the respective throughput gain of the multiplexed downlink channel, the multiplexing mode of the D2D pair is determined as: the method for separately multiplexing the uplink channel for communication or separately multiplexing the downlink channel for communication comprises the following steps:

4. The method of claim 1, wherein the determining the plurality of channels corresponding to each D2D pair according to the throughput gain of each D2D pair for multiplexing the uplink channel and the downlink channel of each cellular user respectively and the multiplexing mode of all D2D pairs comprises:

A4: updating phi_i＝Ф_i\i，W_j＝W_j\j，

wherein, the cellular user C in the coverage area of the base station_iIs M, i ∈ {1,2, …, M }, D2D to D in the coverage area of the base station_jThe number of (c) is N, j ∈ {1,2, …, N }, M>N, throughput gain matrix Ω ═ T_ji]_N×MThe matrix X represents the uplink channel multiplexing condition matrix X^uOr downlink channel multiplexing situation matrix X^dWhen the matrix X represents the uplink channel multiplexing condition matrix X^uWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the uplink channel of the multiplexing cellular user^uElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of uplink channelWhen the matrix X represents the downlink channel multiplexing condition matrix X^dWhen, the element x_jiCharacterizing elementsThe throughput gain matrix omega characterizes the throughput gain matrix omega of the D2D to the downlink channel of the multiplexing cellular user^dElement T_jiCharacterization of D2D vs D_jMultiplexing cellular user C_iThroughput gain of downlink channel Characterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j，Characterizing cellular user C_iIs allocated to D2D pair D_j，Characterizing cellular user C_iIs not allocated to D2D pair D_j。

5. The method of claim 1, wherein the determining the transmit power of each D2D pair on the plurality of channels corresponding thereto according to the multiplexing pattern of the D2D pairs and the plurality of channels corresponding to each D2D pair comprises:

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}}, {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}},

{SINR}_{j i}^{u} = \frac{P_{j i}^{u} g_{j j}}{x_{j i}^{u} P_{C} g_{i j} + N_{0}}, {SINR}_{j i}^{d} = \frac{P_{j i}^{d} g_{j j}}{x_{j i}^{d} P_{B} g_{B j} + N_{0}},

\begin{matrix} \max & R_{D 2 D} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} [x_{j i}^{u} \log_{2} (1 + {SINR}_{j i}^{u}) + x_{j i}^{d} \log_{2} (1 + {SINR}_{j i}^{d})] \end{matrix}

s.t.

{SINR}_{i}^{u} = \frac{P_{C} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{u} P_{j i}^{u} g_{B j} + N_{0}} &GreaterEqual; γ^{u}; {SINR}_{i}^{d} = \frac{P_{B} g_{B i}}{Σ_{j = 1}^{N} x_{j i}^{d} P_{j i}^{d} g_{i j} + N_{0}} &GreaterEqual; γ^{d};

0 \leq Σ_{i = 1}^{M} x_{j i}^{u} P_{j i}^{u} \leq P_{j}^{D}; 0 \leq Σ_{i = 1}^{M} x_{j i}^{d} P_{j i}^{d} \leq P_{j}^{D};

0 \leq Σ_{j = 1}^{N} x_{j i}^{u} \leq 1; 0 \leq Σ_{j = 1}^{N} x_{j i}^{d} \leq 1;

(Σ_{i = 1}^{M} x_{j i}^{u}) (Σ_{i = 1}^{M} x_{j i}^{d}) = 0;

if it isOtherwiseIf it isOtherwise

\begin{matrix} \max & R_{D 2 D}^{u} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{u} g_{j j}}{P_{C} g_{i j} + N_{0}}) \end{matrix},

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{u} g_{B j} \leq \frac{P_{C} g_{B i}}{γ^{u}} - N_{0} \end{matrix},

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

\begin{matrix} \max & R_{D 2 D}^{d} = Σ_{j = 1}^{N} Σ_{i = 1}^{M} \log_{2} (1 + \frac{P_{j i}^{d} g_{j j}}{P_{B} g_{B j} + N_{0}}) \end{matrix},

\begin{matrix} s . t . & Σ_{j = 1}^{N} P_{j i}^{d} g_{i j} \leq \frac{P_{B} g_{B i}}{γ^{d}} - N_{0} \end{matrix},

0 \leq Σ_{i = 1}^{M} P_{j i}^{u} \leq P_{j}^{D};

its dual function is:

wherein,

cellular user C in the coverage area of the base station_iIs M, i ∈ {1,2, …, M }, D2D to D in the coverage area of the base station_jThe number of (c) is N, j ∈ {1,2, …, N },for cellular user C_iThe signal-to-interference-and-noise ratio of the uplink channel,for cellular user C_iThe signal-to-interference-and-noise ratio of the downlink channel,is D2D to D_jMultiplexing cellular user C_iThe signal-to-interference-and-noise ratio of the uplink channel,is D2D to D_jMultiplexing cellular user C_iThe signal-to-interference-and-noise ratio of the downlink channel,representing D2D vs D_jFor cellular user C_iIf cellular user C is the uplink channel reuse situation_iIs D2D to D_jMultiplex thenOtherwise Representing D2D vs D_jFor cellular user C_iIf cellular user C is the downlink channel multiplexing situation_iIs D2D to D_jMultiplex thenOtherwiseChannel gain G-G β d^-αWherein G is a path loss constant, β represents a channel fading constant, α is a path loss exponent, d is a distance from a transmitting end to a receiving end, and the parameterRepresenting D2D vs D_jIn cellular user C_iThe transmit power on the uplink channel is,representing D2D vs D_jIn cellular user C_iTransmission power, P, on a downlink channel_BRepresenting the transmission power, P, of the base station_CIndicating cellular user C_iTransmit power of g_BiIndicating cellular user C_iChannel gain with base station, g_BjRepresenting D2D vs D_jChannel gain with base station, g_ijRepresenting D2D vs D_jAnd cellular user C_iChannel gain between, g_jjRepresenting D2D vs D_jInternal channel gain, N₀Representing a noise value; MaxR_D2DRepresents the maximum rate, γ, of a system of all D2D pairs^uRepresenting a predetermined minimum signal-to-noise ratio, gamma, of the uplink of the cellular subscriber^dIndicating a preset downlink minimum signal-to-noise ratio of the cellular subscriber,represents the maximum transmit power of the D2D pair;

P_ji ^*characterization of D2D vs D_jIn cellular user C_iPower optimum solution on uplink channelOr power optimum solution on downlink channelWhen P is present_ji ^*Characterizing power optimal solutions on uplink channelsWhen is, P_jiCharacterization of D2D vs D_jAt C_iTransmit power on uplink channelGamma represents the preset uplink minimum signal-to-noise ratio gamma of the cellular user^u(ii) a When P is present_ji ^*Characterizing power optimal solutions on downlink channelsWhen is, P_jiCharacterization of D2D vs D_jAt C_iTransmit power on downlink channelGamma represents the preset downlink minimum signal-to-noise ratio gamma of the cellular user^d；Andrespectively, D2D to D_jIn cellular user C_iTransmit power on an uplink channel and a downlink channel; mu.s_jAndfor two multiplicative factors, mu, in the Lagrangian function_j ^*Andis the optimal solution of the dual function obtained by the gradient search algorithm;andthe maximum total rate of the system formed by all D2D pairs in the uplink channel and the downlink channel respectively.

6. The method of claim 5, wherein the gradient search algorithm comprises the steps of:

b1: initializing multiplication factorsμ_j(0) And a step-size parameter,

B2: updating

B3: updating t to t +1, and returning to the step B2; up toAnd (mu)_i(0),μ_i(1),…,μ_i(t +1)) all converge, so thatμ_j ^*＝μ_i(t+1)；

7. An apparatus for D2D pair resource allocation in a cellular network, the apparatus comprising:

8. The apparatus of claim 7, wherein the formula for the throughput gain calculation module to calculate the throughput gain of each D2D for the uplink channel and the downlink channel multiplexing each cellular user respectively comprises:

T_{j i}^{u} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{C} g_{i j} + N_{0}}) + \log_{2} (1 + \frac{P_{C} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{C} g_{B i}}{N_{0}}),

T_{j i}^{d} = \log_{2} (1 + \frac{P_{j}^{D} g_{j j}}{P_{B} g_{B j} + N_{0}}) + \log_{2} (1 + \frac{P_{B} g_{B j}}{P_{j}^{D} g_{j j} + N_{0}}) - \log_{2} (1 + \frac{P_{B} g_{B i}}{N_{0}});

9. The apparatus of claim 7, wherein the multiplexing mode determination module is specifically configured to:

10. The apparatus of claim 7, wherein the channel determining module comprises:

a throughput gain updating unit, usingIn the multiplexing mode based on all D2D pairs, the throughput gain matrix omega in the uplink^uIn the method, elements corresponding to D2D pairs for communicating separately multiplexed downlink channelsUpdating to any value less than or equal to zero; in the downlink throughput gain matrix omega^dIn the method, elements corresponding to D2D pairs for communicating by independently multiplexing uplink channelsUpdating to any value less than or equal to zero;

Channel determination unit for updating phi_i＝Ф_i\i，W_j＝W_j\j，