CN108064077B - The power distribution method of full duplex D2D in cellular network - Google Patents

Abstract

The invention discloses a full-duplex D2D power distribution method in a cellular network, which solves the problem of low spectral efficiency of a D2D communication system in the prior art. On the basis of satisfying the minimum communication rate constraints of cellular users and the individual power constraints of cellular users and D2D users, the present invention establishes a cellular network system model that maximizes system spectrum efficiency, and uses traditional optimization methods for the obtained nonlinear and non-convex optimization model. The power allocation method combining the method and the intelligent algorithm obtains the transmitting power of cellular users and D2D users. The present invention considers the total spectrum efficiency of full-duplex D2D users and cellular users, and the proposed power allocation method has the characteristics of high precision and low complexity, and improves the reliability and spectrum efficiency of the D2D communication system.

Description

Power Allocation Method for Full-duplex D2D in Cellular Networks

technical field

The present invention belongs to the technical field of communication, and further relates to a full-duplex terminal direct D2D (Device-to-Device) power allocation method in a cellular network in the technical field of terminal direct communication. The present invention can be used in a cellular network of wireless mobile communication. Cellular users use traditional cellular communication methods to communicate with base stations, and D2D users use full-duplex mode to multiplex the spectrum resources of cellular users for D2D communication, and optimize cellular users and D2D users. Power allocation maximizes the spectrum efficiency of the system and improves the spectrum utilization and reliability of the communication system.

Background technique

D2D communication can bring channel gain, hop count gain and multiplexing gain. It has the advantages of improving system performance, improving user experience, and expanding cellular communication applications. It is widely regarded as one of the key technologies in the next generation of cellular networks. The so-called D2D communication means that D2D users can communicate directly without going through the relay of the base station. D2D communication can use dedicated spectrum resources, and can also reuse spectrum resources of traditional cellular users. When D2D multiplexes spectrum resources of cellular users, spectrum utilization can be improved, but it will cause interference to cellular users, and at the same time, interference from cellular users will also be received. The working modes of D2D users are divided into half-duplex mode and full-duplex mode. When D2D users use the full-duplex mode to communicate, the spectrum efficiency can be further improved, but it will bring its own loop interference, making the interference situation more complicated. Therefore, it is necessary to control the interference through an effective resource allocation scheme, so as to maximize the spectrum efficiency of the communication system on the basis of ensuring the communication quality of cellular users.

In their paper "Power Allocation Optimization for Full-Duplex D2D Communications Underlaying Cellular Networks" (Networking and Network Applications (NaNA), 2016 International Conference, pp.103-108), Boqun Zuo et al. A spectrally efficient power allocation method for full-duplex D2D users. This method is used in the terminal direct system. Under the constraints of ensuring the communication rate of cellular users and the individual power of each user, a mathematical model for maximizing the spectral efficiency of D2D users is constructed, and the non-convex problem is approximated as a convex problem by a convex approximation method. And the power allocation is carried out through the method of multi-layer iteration convergence to the optimal solution of the convex problem. The disadvantage of this method is that the objective function only considers the spectral efficiency of the D2D link, but does not consider the spectral efficiency of cellular users, and does not fully utilize the performance of the full-duplex system.

Southeast University discloses a D2D user pair and cellular user sharing spectrum power allocation method in its patent document "A D2D user pair shares spectrum with cellular users" (application number: 201710346637.X publication number: CN 107172574 A). Spectrum power allocation method. The implementation steps of the method are as follows: first, initialize the number of iterations, user allocated power and Lagrangian multiplier; second, calculate the allocated power of cellular users; third, perform a first-order convex approximation on the data rate of the D2D user pair After simplification; fourth, use the subgradient descent method to update the user's allocated power and Lagrangian multipliers; fifth, judge whether the Lagrangian multipliers converge, and if so, the currently calculated allocated power matrix is the final Allocate power, if not, add one to the number of cycles, and return to step 2. The disadvantage of this method is that this method is suitable for D2D working in a half-duplex mode, the objective function only considers the sum rate of the D2D link, and the spectrum utilization rate of the system is low.

Xidian University disclosed a method of power allocation based on SINR incremental iteration in embedded D2D cellular network (application number: 201310606828.7 publication number: CN 103582104 A) which was applied for by Xidian University. A Power Allocation Method Based on SINR Incremental Iteration. The method first builds a model about the power allocation problem, and uses the increase of the utility function to represent the benefit of each iteration. The iterative algorithm includes: calculating the utility function, changing the SINR of the receiving end of the cellular communication and the SINR of the receiving end of the D2D communication, calculating the related income function, and comparing the value of the income function to determine which type of user's SINR to increase. The disadvantage of this method is that this method is suitable for D2D working in a half-duplex mode, and the spectrum utilization rate of the whole system is not fully utilized.

Contents of the invention

The purpose of the present invention is to propose a full-duplex D2D power allocation method in a cellular network to address the shortcomings of the above-mentioned existing technologies. The method of the present invention can maximize the total spectrum efficiency of cellular users and full-duplex D2D users. Improve the spectrum utilization and reliability of the communication system.

The idea of realizing the purpose of the present invention is: the cellular network includes a base station, a cellular user and two D2D users, the cellular user communicates with the base station in half-duplex mode, and the two D2D users use the full-duplex mode to multiplex the spectrum of the cellular user For resource communication, on the basis of satisfying the minimum communication rate constraints of cellular users and the individual power constraints of cellular users and D2D users, a cellular network system model that maximizes the system spectrum efficiency is established. For the obtained nonlinear and non-convex optimization model, traditional The power allocation method combining optimization method and intelligent algorithm obtains the transmit power of cellular users and D2D users.

The concrete steps that realize the object of the present invention are as follows:

(1) According to the following formula, calculate the maximum information rate from the cellular user to the base station when no D2D user reuses the spectrum resource of the cellular user:

Among them, R _cu ′ represents the maximum information rate from the cellular user cu to the base station when no D2D user reuses the spectrum resources of the cellular user cu, B represents the bandwidth of the cellular network system, log ₂ represents the logarithmic operation with base 2, Indicates the maximum transmission power of the cellular user cu, h _cu indicates the channel gain of the communication link between the cellular user cu and the base station, and n _c indicates the background noise power received by the base station;

(2) Determine whether the maximum information rate R _cu ' from the cellular user cu to the base station is greater than the minimum information rate of the cellular user cu itself in the cellular network system when there is no D2D user reusing the spectrum resources of the cellular user cu, and if so, perform step (3 ), otherwise, perform step (11);

(3) According to the following formula, the base station establishes the power allocation model of the cellular network system that maximizes the spectrum efficiency:

objective function Restrictions

Among them, max{ } represents the operation of finding the maximum value, p ₁ represents the transmit power of the first D2D user, h ₁ represents the channel gain of the interference link from the first D2D user to the base station, and p ₂ represents the second D2D user The transmit power of the user, h ₂ represents the channel gain of the interference link from the second D2D user to the base station, h ₁₂ represents the channel gain of the communication link from the first D2D user to the second D2D user, h ₂ ′ represents the The channel gain of the interference link from the user to the second D2D user, η represents the residual loop self-interference coefficient of the first D2D user after self-interference cancellation, and the residual loop self-interference coefficient between the second D2D user and the first D2D user The self-interference coefficient is equal, n _d represents the background noise power received by the first D2D user, the background noise received by the second D2D user is equal to that of the first D2D user, h ₂₁ represents the second D2D user to the first D2D user The channel gain of the communication link of , h ₁ ′ represents the channel gain of the interference link from the cellular user to the first D2D user, Indicates the minimum information rate of the cellular user cu set by the cellular network system, Indicates the maximum transmit power of the first D2D user, Indicates the maximum transmission power of the second D2D user;

(4) When the transmit power of at least one D2D user is 0, the transmit power values of the cellular user and two D2D users:

(4a) Let the transmit power of the first D2D user be 0, use positive constraints to calculate the transmit power values of the cellular user and the second D2D user, and denote the system spectrum efficiency value under this power combination as e1;

(4b) Let the transmit power of the second D2D user be 0, use positive constraints to calculate the transmit power values of the cellular user and the first D2D user, and denote the system spectrum efficiency value under this power combination as e2;

(4c) If e1 is greater than e2, record the power of the cellular user, the first D2D user, and the second D2D user obtained in step (4a) as power combination 1, otherwise, record the power of the cellular user obtained in step (4b) , the power of the first D2D user and the second D2D user is recorded as power combination 1;

(5) According to the following formula, calculate the maximum information rate of the cellular user when two D2D users simultaneously multiplex the spectrum resources of the cellular user:

Among them, R _cu ″ represents the maximum information rate of the cellular user cu when two D2D users simultaneously multiplex the spectrum resources of the cellular user cu to the base station, Indicates the minimum transmission power of the first D2D user and the second D2D user;

(6) Determine whether the maximum information rate R _cu ″ of the cellular user cu is not less than the minimum information rate of the cellular user cu when two D2D users simultaneously multiplex the spectrum resources of the cellular user; if so, perform step (7); otherwise , execute step (9);

(7) When the transmission power of the two D2D users is not 0, the transmission power values of the cellular user and the two D2D users:

According to the minimum rate constraints of cellular users, calculate the more accurate value range of the transmitting power of cellular users and two D2D users, use the adaptive particle swarm optimization algorithm to calculate the transmitting power of cellular users and two D2D users, and obtain the cellular users , the power of the first D2D user and the second D2D user is recorded as power combination 2;

(8) Judging whether the system spectral efficiency of power combination 1 is greater than the system spectral efficiency of power combination 2, if so, then perform step (9), otherwise, perform step (10);

(9) Take the transmit powers of the cellular user, the first D2D user and the second D2D user respectively, and then perform step (12) after taking the three values in the power combination 1;

(10) Take the transmit power of the cellular user, the first D2D user, and the second D2D user, respectively, and then perform step (12) after taking the three values in the power combination 2;

(11) Take the maximum value of the transmission power of the cellular user cu, take the transmission power of the first D2D user and the second D2D user as 0, and perform step (12);

(12) The base station notifies the user of the obtained transmission power corresponding to each user, and the user transmits a signal using the transmission power notified by the base station.

Compared with the prior art, the present invention has the following advantages:

First, on the basis of satisfying the minimum communication rate constraints of cellular users and the individual power constraints of users, the present invention makes D2D work in full-duplex mode, maximizes the total spectrum efficiency of cellular users and D2D users, and overcomes the existing technology Only considering the D2D user spectrum efficiency or the low spectrum efficiency of the D2D system operating in the half-duplex mode, the present invention improves spectrum utilization.

Second, the present invention aims at the spectral efficiency optimization problem in the nonlinear non-convex power distribution model, utilizes the traditional optimization method in the step (4) and the method that the self-adaptive particle swarm optimization algorithm combines in the step (7) to solve the user power, overcome The prior art only adopts the traditional optimization method to solve the problem that the user power affects the precision, so that the present invention has the advantages of high precision and low complexity.

Description of drawings

FIG. 1 is a schematic diagram of a full-duplex D2D communication scenario in a cell according to the present invention;

Fig. 2 is a flowchart of the present invention;

Fig. 3 is a simulation diagram of the present invention.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings.

Referring to Fig. 1, the scene used in the present invention is further described.

In Figure 1, the cellular network includes a base station bs, a cellular user cu, the first D2D user du1, and the second D2D user du2. The cellular user communicates with the base station in half-duplex mode, and the two D2D users use full-duplex communication. Mode multiplexes communication of spectrum resources of cellular users. In Figure 1, h _cu represents the channel gain of the communication link from the cellular user cu to the base station, h ₁ represents the channel gain of the interference link from the first D2D user to the base station, and h ₂ represents the interference link from the second D2D user to the base station h ₁ ′ represents the channel gain of the interference link from the cellular user to the first D2D user, h ₂ ′ represents the channel gain of the interference link from the cellular user to the second D2D user, and h ₁₂ represents the channel gain of the interference link from the cellular user to the second D2D user. The channel gain of the communication link from one D2D user to the second D2D user, h ₂₁ represents the channel gain of the communication link from the second D2D user to the first D2D user, η represents the two D2D users after self-interference cancellation The residual loop self-interference coefficient of .

Referring to FIG. 2 , the specific steps of the present invention for completing the full-duplex D2D power allocation method in the cellular network will be further described in detail.

Step 1, according to the following formula, calculate the maximum information rate from the cellular user to the base station when no D2D user reuses the spectrum resource of the cellular user:

Among them, R _cu ′ represents the maximum information rate from the cellular user cu to the base station when no D2D user reuses the spectrum resources of the cellular user cu, B represents the bandwidth of the cellular network system, log ₂ represents the logarithmic operation with base 2, Indicates the maximum transmission power of the cellular user cu, h _cu indicates the channel gain of the communication link between the cellular user cu and the base station, and n _c indicates the background noise power received by the base station.

Step 2, judging whether the maximum information rate R _cu ' from the cellular user cu to the base station is greater than the minimum information rate of the cellular user cu itself in the cellular network system when no D2D user reuses the spectrum resources of the cellular user cu, and if so, perform step 3, Otherwise, go to step 11.

Step 3, according to the following formula, the base station establishes the power allocation model of the cellular network system that maximizes the spectrum efficiency: the objective function Restrictions

Among them, max{ } represents the operation of finding the maximum value, p ₁ represents the transmit power of the first D2D user, h ₁ represents the channel gain of the interference link from the first D2D user to the base station, and p ₂ represents the second D2D user The transmit power of the user, h ₂ represents the channel gain of the interference link from the second D2D user to the base station, h ₁₂ represents the channel gain of the communication link from the first D2D user to the second D2D user, h ₂ ′ represents the The channel gain of the interference link from the user to the second D2D user, η represents the residual loop self-interference coefficient of the first D2D user after self-interference cancellation, and the residual loop self-interference coefficient between the second D2D user and the first D2D user The self-interference coefficient is equal, n _d represents the background noise power received by the first D2D user, the background noise received by the second D2D user is equal to that of the first D2D user, h ₂₁ represents the second D2D user to the first D2D user The channel gain of the communication link of , h ₁ ′ represents the channel gain of the interference link from the cellular user to the first D2D user, Indicates the minimum information rate of the cellular user cu set by the cellular network system, Indicates the maximum transmit power of the first D2D user, Indicates the maximum transmit power of the second D2D user.

Step 4, when the transmit power of at least one D2D user is 0, transmit power values of the cellular user and two D2D users are obtained.

Let the transmit power of the first D2D user be 0, use positive constraints to calculate the transmit power values of the cellular user and the second D2D user, and denote the system spectrum efficiency value under this power combination as e1.

The active constraints mean that at least one of the constraints in the power allocation model is an active constraint, and the transmit power of the cellular user and the D2D user is located at the boundary point of the definition domain determined by the constraints.

The steps for calculating the transmit power values of the cellular user and the second D2D user are as follows:

In the first step, according to the following formula, use the minimum rate constraint condition of cellular users to calculate the judgment condition value p ₂ ′:

In the second step, according to the following formula, calculate the maximum power value of the cellular user that satisfies the minimum rate constraint

The third step is to take the transmit power of the cellular user as The transmit power of two D2D users is taken as 0, which is recorded as power combination 3, and the transmit power of cellular users is taken as The transmit power of the first D2D user is 0, and the transmit power of the second D2D user is p ₂ ′, which is recorded as power combination 4, and the transmit power of the cellular user is taken as The transmit power of the first D2D user is set to 0, and the transmit power of the second D2D user is set to It is recorded as power combination 5, and the transmit power of cellular users is taken as The transmit power of the first D2D user is set to 0, and the transmit power of the second D2D user is set to Record it as power combination 6.

The fourth step, if Then the transmit power of the cellular user and the two D2D users adopts the value in power combination 3; if Calculate the system spectral efficiencies e3 and e4 corresponding to power combination 3 and power combination 4. If e3>e4, the transmit power of the cellular user and the two D2D users adopts the value in power combination 3; otherwise, the cellular user and the two D2D users The transmit power of adopts the value in power combination 4; if Calculate the system spectral efficiencies e3, e5, and e6 corresponding to power combination 3, power combination 5, and power combination 6. If e3≥e5 and e3≥e6, the transmit power of the cellular user and the two D2D users adopts the value in power combination 3, If e5>e3 and e5>e6, the transmit power of the cellular user and the two D2D users adopts the value in power combination 5; if e6>e3 and e6>e5, the transmit power of the cellular user and the two D2D users adopts the value in power combination 6 value in .

Let the transmit power of the second D2D user be 0, and use positive constraints to calculate the transmit power values of the cellular user and the first D2D user, and denote the system spectral efficiency value under this power combination as e2.

If e1 is greater than e2, record the power of the cellular user, the first D2D user, and the second D2D user obtained in step (4a) as power combination 1; otherwise, record the power of the cellular user, the first D2D user obtained in step (4b) The power of one D2D user and the second D2D user is recorded as power combination 1.

Step 5, according to the following formula, calculate the maximum information rate of the cellular user when two D2D users simultaneously multiplex the spectrum resources of the cellular user:

Among them, R _cu ″ represents the maximum information rate of the cellular user cu when two D2D users simultaneously multiplex the spectrum resources of the cellular user cu to the base station, Indicates the minimum transmit power of the first D2D user and the second D2D user.

Step 6: Determine whether the maximum information rate R _cu ″ of the cellular user cu is not less than the minimum information rate of the cellular user cu when two D2D users simultaneously multiplex the spectrum resources of the cellular user; if so, execute step 7; otherwise, execute Step 9.

Step 7, when the transmission power of the two D2D users is not 0, the transmission power values of the cellular user and the two D2D users are obtained.

According to the minimum rate constraints of cellular users, calculate the more accurate value range of the transmitting power of cellular users and two D2D users, use the adaptive particle swarm optimization algorithm to calculate the transmitting power of cellular users and two D2D users, and obtain the cellular users , the power of the first D2D user, and the second D2D user are recorded as power combination 2.

The specific steps for calculating the more accurate value range of the transmit power of the cellular user and the two D2D users are as follows:

In the first step, calculate the upper limit value of the transmit power of the first D2D user according to the following formula

Among them, min means to take the minimum value operation.

In the second step, calculate the upper limit value of the transmit power of the second D2D user according to the following formula

The third step is to calculate the lower limit of the transmit power of the cellular user cu according to the following formula

In the fourth step, according to the individual power constraints of the cellular user and the D2D user, the more accurate value ranges of the transmission power of the cellular user, the first D2D user, and the second D2D user are respectively obtained as

Step 8, judging whether the system spectral efficiency of power combination 1 is greater than the system spectral efficiency of power combination 2, if yes, perform step 9, otherwise, perform step 10.

In step 9, the transmit powers of the cellular user, the first D2D user and the second D2D user are respectively taken from three values in power combination 1, and then step 12 is executed.

In step 10, the transmit powers of the cellular user, the first D2D user and the second D2D user are respectively taken from three values in the power combination 2, and then step 12 is executed.

Step 11, take the maximum value of the transmit power of the cellular user cu, take the transmit power of the first D2D user and the second D2D user as 0, and execute step 12.

In step 12, the base station notifies the user of the obtained transmit power corresponding to each user, and the user transmits a signal using the transmit power notified by the base station.

The effects of the present invention will be further described in detail in combination with simulation experiments.

1. Simulation conditions:

The simulation experiment of the present invention is carried out under MATLAB 7.11 software. In the simulation experiment of the present invention, there is a base station in the center of the cellular network, and the coverage area is a circle with a radius of 500m. Cellular users and D2D users are randomly distributed in the cell, the distance to the base station is not less than 20m, and the maximum distance between two D2D users is 50m. The upper limit of the transmit power of the cellular user cu is 24dBm, and the upper limit of the transmit power of the two D2D users is 24dBm. The channel gain model of the link in the system is g _i,j =d _i,j ^-n , where i is the transmitting user, j is the receiving user, d is the distance between user i and user j, n is the path loss index, The value is 4. The system bandwidth is 1Hz, the minimum rate requirement of cellular users is 3bit/s/Hz, the noise power received by the base station and D2D users is 10 ^-13 W, and the minimum transmit power of D2D users is 0.001W. The simulation residual self-interference coefficient ranges from (-110) to (-60) dB, and the number of simulations is 1000 times.

2. Simulation content and result analysis:

The simulation experiment of the present invention is to use the present invention and two existing technologies to maximize the spectral efficiency of the communication network system for user power allocation (the prior art 1 is to maximize the power of the total spectral efficiency of cellular users and half-duplex D2D users Allocation method, prior art 2 is a power allocation method that maximizes the spectrum efficiency of two full-duplex D2D users). The average spectral efficiency of the cellular communication system is simulated with respect to the D2D user residual loop interference coefficient.

The simulation experiment results of the present invention are shown in FIG. 3 . The horizontal axis in FIG. 3 represents the D2D user residual loop interference coefficient, and the vertical axis represents the system spectrum efficiency. In Fig. 3, the solid line marked with a right-facing triangle represents the curve of the system spectral efficiency obtained by prior art 1, the solid line marked with squares represents the curve of the system spectral efficiency obtained by prior art 2, and the curve marked with a circle The solid line represents the curve of the spectral efficiency of the system obtained by the present invention.

As can be seen from Fig. 3, with the reduction of the remaining loop interference coefficient, the spectral efficiency of the present invention is improved compared with the spectral efficiency of the prior art 1, and the spectral efficiency of the present invention is higher than that of the prior art 1. The efficiency is up to 8bit/s/Hz higher. Within the value range of the remaining loop interference coefficient, the spectral efficiency of the present invention is higher than that of the prior art 2, and the spectral efficiency of the present invention is at most 1.8 bit/s/Hz higher than that of the prior art 2. The simulation results show that the full-duplex D2D communication system adopted by the present invention has a higher spectrum utilization rate and can alleviate the requirement of spectrum shortage.

Claims (2)

1. A power allocation method for full-duplex D2D in a cellular network, characterized in that the cellular network includes a base station, a cellular user and two D2D users, the cellular user communicates with the base station in a half-duplex mode, and the two D2D The user uses the full-duplex mode to multiplex the spectrum resource communication of the cellular user, and optimizes the power allocation between the cellular user and the two D2D users to maximize the spectral efficiency of the cellular network system. The specific steps include the following: (1) According to the following formula, calculate the maximum information rate from the cellular user to the base station when no D2D user reuses the spectrum resource of the cellular user: Among them, R _cu ′ represents the maximum information rate from the cellular user cu to the base station when no D2D user reuses the spectrum resources of the cellular user cu, B represents the bandwidth of the cellular network system, log ₂ represents the logarithmic operation with base 2, Indicates the maximum transmission power of the cellular user cu, h _cu indicates the channel gain of the communication link between the cellular user cu and the base station, and n _c indicates the background noise power received by the base station; (2) Determine whether the maximum information rate R _cu ' from the cellular user cu to the base station is greater than the minimum information rate of the cellular user cu itself in the cellular network system when there is no D2D user reusing the spectrum resources of the cellular user cu, and if so, perform step (3 ), otherwise, perform step (11); (3) According to the following formula, the base station establishes the power allocation model of the cellular network system that maximizes the spectral efficiency: the objective function Restrictions 0≤p ₁ ≤P ₁ ^max ; Among them, max{ } represents the operation of finding the maximum value, p ₁ represents the transmit power of the first D2D user, h ₁ represents the channel gain of the interference link from the first D2D user to the base station, and p ₂ represents the second D2D user The transmit power of the user, h ₂ represents the channel gain of the interference link from the second D2D user to the base station, p _cu represents the transmit power of the cellular user cu, and h ₁₂ represents the communication link from the first D2D user to the second D2D user h ₂ ′ represents the channel gain of the interference link from the cellular user to the second D2D user, η represents the remaining loop self-interference coefficient after self-interference cancellation of the first D2D user, and the second D2D user It is equal to the residual loop self-interference coefficient of the first D2D user, n _d represents the background noise power received by the first D2D user, and the background noise received by the second D2D user is equal to that of the first D2D user, h ₂₁ represents the background noise power received by the first D2D user The channel gain of the communication link from two D2D users to the first D2D user, h ₁ ′ represents the channel gain of the interference link from the cellular user to the first D2D user, Indicates the minimum information rate of the cellular user cu set by the cellular network system, P ₁ ^max indicates the maximum transmission power of the first D2D user, Indicates the maximum transmission power of the second D2D user; (4) When the transmit power of at least one D2D user is 0, the transmit power values of the cellular user and two D2D users: (4a) Let the transmit power of the first D2D user be 0, use positive constraints to calculate the transmit power values of the cellular user and the second D2D user, and denote the system spectrum efficiency value under this power combination as e1; The steps to calculate the transmit power values of the cellular user and the second D2D user are as follows: In the first step, according to the following formula, use the minimum rate constraint condition of cellular users to calculate the judgment condition value p ₂ ′: In the second step, according to the following formula, calculate the maximum power value of the cellular user that satisfies the minimum rate constraint The third step is to take the transmit power of the cellular user as The transmit power of two D2D users is taken as 0, which is recorded as power combination 3, and the transmit power of cellular users is taken as The transmit power of the first D2D user is 0, and the transmit power of the second D2D user is p ₂ ′, which is recorded as power combination 4, and the transmit power of the cellular user is taken as The transmit power of the first D2D user is set to 0, and the transmit power of the second D2D user is set to It is recorded as power combination 5, and the transmit power of cellular users is taken as The transmit power of the first D2D user is set to 0, and the transmit power of the second D2D user is set to Recorded as power combination 6; The fourth step, if Then the transmit power of the cellular user and the two D2D users adopts the value in power combination 3; if Calculate the system spectral efficiencies e3 and e4 corresponding to power combination 3 and power combination 4. If e3>e4, the transmit power of the cellular user and the two D2D users adopts the value in power combination 3; otherwise, the cellular user and the two D2D users The transmit power of adopts the value in power combination 4; if Calculate the system spectral efficiencies e3, e5, and e6 corresponding to power combination 3, power combination 5, and power combination 6. If e3≥e5 and e3≥e6, the transmit power of the cellular user and the two D2D users adopts the value in power combination 3, If e5>e3 and e5>e6, the transmit power of the cellular user and the two D2D users adopts the value in power combination 5; if e6>e3 and e6>e5, the transmit power of the cellular user and the two D2D users adopts the value in power combination 6 the value in; (4b) Let the transmission power of the second D2D user be 0, use the positive constraint condition, adopt the same method as step (4a), calculate the transmission power value of the cellular user and the first D2D user, and combine the power The system spectrum efficiency value is denoted as e2; (4c) If e1 is greater than e2, record the power of the cellular user, the first D2D user, and the second D2D user obtained in step (4a) as power combination 1, otherwise, record the power of the cellular user obtained in step (4b) , the power of the first D2D user and the second D2D user is recorded as power combination 1; (5) According to the following formula, calculate the maximum information rate of the cellular user when two D2D users simultaneously multiplex the spectrum resources of the cellular user: Among them, R _cu ″ represents the maximum information rate of the cellular user cu when two D2D users simultaneously multiplex the spectrum resources of the cellular user cu to the base station, Indicates the minimum transmission power of the first D2D user and the second D2D user; (6) Determine whether the maximum information rate R _cu ″ of the cellular user cu is not less than the minimum information rate of the cellular user cu when two D2D users simultaneously multiplex the spectrum resources of the cellular user; if so, perform step (7); otherwise , execute step (9); (7) When the transmission power of the two D2D users is not 0, the transmission power values of the cellular user and the two D2D users: According to the minimum rate constraints of cellular users, calculate the more accurate value range of the transmitting power of cellular users and two D2D users, use the adaptive particle swarm optimization algorithm to calculate the transmitting power of cellular users and two D2D users, and obtain the cellular users , the power of the first D2D user and the second D2D user is recorded as power combination 2; According to the minimum rate constraint of the cellular user, the specific steps of calculating the more accurate value range of the transmitting power of the cellular user and the two D2D users are as follows: In the first step, calculate the upper limit value of the transmit power of the first D2D user according to the following formula Among them, min means to take the minimum value operation; In the second step, calculate the upper limit value of the transmit power of the second D2D user according to the following formula The third step is to calculate the lower limit of the transmit power of the cellular user cu according to the following formula In the fourth step, according to the individual power constraints of the cellular user and the D2D user, the more accurate value ranges of the transmission power of the cellular user, the first D2D user, and the second D2D user are respectively obtained as (8) Judging whether the system spectral efficiency of power combination 1 is greater than the system spectral efficiency of power combination 2, if so, then perform step (9), otherwise, perform step (10); (9) Take the transmit powers of the cellular user, the first D2D user and the second D2D user respectively, and then perform step (12) after taking the three values in the power combination 1; (10) Take the transmit power of the cellular user, the first D2D user, and the second D2D user, respectively, and then perform step (12) after taking the three values in the power combination 2; (11) Take the maximum value of the transmission power of the cellular user cu, take the transmission power of the first D2D user and the second D2D user as 0, and perform step (12); (12) The base station notifies the user of the obtained transmission power corresponding to each user, and the user transmits a signal using the transmission power notified by the base station. 2. The power allocation method for full-duplex D2D in a cellular network according to claim 1, wherein the positive constraints described in step (4a) refer to at least one of the constraints in the power allocation model being Active constraints, the transmit power of cellular users and D2D users are located at the boundary points of the defined domain determined by the constraints.