CN106454906A - Method for reducing time delay of relay network of LTE terminal - Google Patents

Abstract

The invention discloses a method for reducing time delay of a relay network of an LTE terminal. The method comprises the following steps of step 1.searching users requiring relay service; step 2.deciding a relay service provider; step 3.solving a state change probability of the user as a relay service requestor and a state change probability of the user as the relay service provider; and step 4.designing a buffer strip for each user to obtain average delay of all users. According to the method designed by the invention, the variables, such as Poisson strength, total electronic money and user motion state, are likely to change under different actual scenes. Through adoption of a system model, the appropriate variables can be designed for different scenes to realize an effect of minimum system average delay.

Description

Method for reducing relay network time delay of LTE terminal

Technical Field

The invention belongs to the technical field of networks, and particularly relates to a method for reducing the time delay of an LTE terminal relay network.

Background

At present, in order to increase the capacity of an LTE (Long Term Evolution) network system, a relay technology for a terminal is proposed, and electronic money is used to buy and sell a relay service. In such a cooperative relay system, not only each relay server receiving the relay service will obtain a benefit, but also the whole network system will obtain a certain increase in network capacity.

Existing system models that study this technology only address the efficiency and capacity of the system. However, the average transmission delay of the system user side is not studied sufficiently, and now in the LTE network, more and more multimedia data are used by the user, and the data are often sensitive to the delay.

[1]N.Mastronarde,V.Patel,J.Xu,and M.van der Schaar,“To relay or notto relay:learning relaying strategies in cellular device-to-device networks”.

[2]N.Salodkar,A.Bhorkar,A.Karandikar,V.S.Borkar,“An on-line learningalgorithm for energy efficient delay constrained scheduling over a fadingchannel,”IEEE Journal on Selected Areas in Communications,vol.26,no.4,pp.732-742,Apr.2008.

[3]J.Xu and M.van der Schaar,“Token system design for autonomicwireless relay networks,”

IEEE Trans on Communications,vol.61,no.7,pp.2924-2935,July 2013..

Disclosure of Invention

In the LTE network, besides capacity, the requirement of a user on transmission delay is higher and higher, and the invention provides a system model for observing the delay performance of the system in various states, so that proper parameters are designed for the system to reduce delay and improve the delay perception of a user terminal.

The method comprises the following steps:

step 1, searching a user needing relay service; the location of the user may change over time, but at time node t, the location of a user is fixed. When the data rate received by the user is less than the target data rate, the user will request the relay service to other surrounding users.

Step 2, determining a relay service provider;

step 3, solving the state change probability of the user as a relay service demander and the state change probability of the user as a relay service provider;

and 4, designing a buffer zone for each user to obtain the average time delay of all users.

Supposing that a user i needs to receive data from a base station, namely a user 0, at a time node t, i ∈ {1,2, …, N }, wherein N is the total number of the users, the value of N is a natural number, and tau is a target Signal-to-interference plus Noise Ratio (SINR)^i,targetIn the case where user i has a target acceptance rate γ^i,targetThis can be achieved by noting that the target acceptance rate is time-varying, and calculating the external relay demand index for user i at time node t by the following formula

Wherein, γ^i,targetWhich is indicative of the target acceptance rate,indicating that user i is connected from the base stationReceiving rate of data, external relay demand indexThat is, at time node t, the probability that the receiving rate of the user i receiving data from the base station is smaller than the target receiving rate is also represented, and the probability that the signal-to-noise ratio between the user i and the base station is smaller than the target signal-to-noise ratio is also represented.

The step 2 of the invention comprises:

setting an optimal energy consumption thresholdAnd an optimal electronic money amount threshold value K_th(Pⁱ) When the energy of the relay service provider is less than the optimal energy consumption threshold value and the electronic money amount of the relay service provider is less than the optimal electronic money amount threshold value, the relay service occurs, and the formula is as follows:

wherein k isⁱNumber of electronic money, p, for user iⁱFor the energy state of user i, σⁱ＝1∈σⁱ(kⁱ,pⁱ) Time indicates that user i can provide relay service, σⁱ＝0∈σⁱ(kⁱ,pⁱ) Time indicates that user i cannot provide the relay service.

In step 2 of the invention, the optimal energy threshold value is set by a user according to different terminal users; the optimal threshold value for the amount of electronic money is determined by three factors: the method comprises the steps that an external relay demand index, a discount coefficient beta and energy consumption C are set by a network operator, the beta value of each user is the same after the setting is finished, and the corresponding optimal electronic currency quantity threshold value is obtained according to different external relay demands ORDR.

In step 2 of the present invention, if the user uses sigmaⁱ(kⁱ,pⁱ) After 1, set Vⁱ(kⁱ,pⁱ) For optimal revenue expectation (when the user acts as a relay service provider, c is accompanied by the optimal cooperation strategyⁱEnergy consumption of, expected value minus cⁱThe net benefit is obtained, and this value can be used to measure the system performance):

wherein,for the benefit obtained by user i at time node t,is composed ofExpected value of (1), discount coefficients β [0,1), β^tIs the discount coefficient at the time t,indicating the expected value of the revenue obtained by the user i after t time nodes.

The step 3 of the invention comprises:

user state starts to be Sⁱ＝(kⁱ,pⁱ)，σⁱWhen 1 indicates that the user can provide the relay service, σⁱWhen the value is 0, it means that the user cannot provide the relay service, and if the user is a relay service demander, the user status changes when the user obtains the relay service (k)ⁱ,pⁱ)→(kⁱ-1,pⁱ)，cⁱThe energy consumed when providing relay service to other users for user i,corresponding probability is λⁱeⁱThe calculation formula is as follows:

λⁱeⁱ＝Pⁱ([kⁱ-1,pⁱ]|[kⁱ,pⁱ],σⁱ)；

wherein, cⁱEnergy consumed when providing relay service for user i to other users;

if the user acts as a relay service provider, when the user provides relay service for other users, the user state changes (k)ⁱ,pⁱ)→(kⁱ+1,pⁱ-cⁱ) Corresponding probability of yⁱσⁱThe calculation formula is as follows:

yⁱσⁱ＝Pⁱ([kⁱ+1,pⁱ-cⁱ]|[kⁱ,pⁱ],σⁱ)。

the step 4 of the invention comprises: the average time delay D is calculated by the average queue length Q and the average arrival rate a:

D＝Q/a；

at time t, Q is set_tFor queue length, the dynamic queue length is calculated by the following formula:

Q_t+1＝max[Q_t-U_t,0]+W_t+1，

by Q_t+1Obtaining an average queue length Q, wherein W_tFor the number of data sets arriving in the buffer zone, U_tIn order to complete the data set for the transmission process,

at present, the LTE network structure consists of two layers of structure, core network (EPC) and access network (E-UTRAN), the access network including access part (eNodeB) and terminal (UE). The terminal can be a mobile phone, a tablet computer and the like, and can be used as a data receiver and a relay node to transmit data. The invention provides a system model for observing the time delay performance of a system in various states, thereby designing a proper use state for the system to achieve the purpose of reducing the time delay.

Therefore, the invention provides a method for researching the time delay performance of the system in various states, and the system parameters are designed through analysis to achieve the effect of optimal time delay.

Has the advantages that: in the LTE network, more and more multimedia data are transmitted in the system, and the conventional system usually pays more attention to the capacity of the system, but the delay analysis of the system is insufficient. The method designed by the invention can be used for visually observing the time delay performance of the system in various states, so that a proper system state is designed to achieve the optimal system average time delay.

In the method designed by the invention, variables such as Poisson strength, total electronic money amount and user motion state can be changed under different actual scenes. The system model of the invention can be adopted to design suitable variables for different scenes so as to achieve the effect of minimum average time delay of the system.

Drawings

The foregoing and other advantages of the invention will become more apparent from the following detailed description of the invention when taken in conjunction with the accompanying drawings.

FIG. 1 is a flow chart of the system of the present invention.

Fig. 2 is a diagram illustrating a relationship between an optimal threshold value of the amount of electronic money and external relay demand and discount coefficient β.

Fig. 3 is a diagram illustrating the influence of different amounts of electronic money on the delay.

Fig. 4 is a diagram showing the number of electronic money 9000 and the poisson strength of 20.

Detailed Description

The invention is further explained below with reference to the drawings and the embodiments.

In the LTE network, the cooperative relaying of the terminal user can improve the performance of the whole system, but when the terminal is normally used in the LTE network, they are interested in the data resources occupied by themselves and often do not help other users to transfer data by themselves when there is no reasonable incentive strategy. In this case, an electronic money is proposed for buying and selling the relay service. When a user needs the relay service, the user pays an electronic money to buy the relay service, and the user receiving the service request decides not to provide the relay service for the user according to objective conditions, if the service is provided, the total amount of the electronic money of a relay service provider is increased by 1, and if the service is not needed, no relay service occurs. Next, some factors affecting the occurrence of this relay service will be described.

Firstly, the method comprises the following steps: when a relay service occurs, the user needs to consume a certain amount of energy when he is acting as a relay service provider. The invention can set an energy threshold value for the user in the system, and can provide relay service for other users only when the energy state of the user is smaller than the threshold value. Users often stay in the network for a long time and therefore need to propose an automatic strategy to decide not to accept these ongoing relay service requests.

Secondly, the method comprises the following steps: the motion profile of the user is another factor. For example, when all users are stationary or they move very slowly, the probability that the user farthest from the base station needs relay service is much greater than the user closer to the base station.

Thirdly, the method comprises the following steps: another factor to be considered is the total amount of electronic money, and how much this amount affects the delayed performance of the system. If the amount of electronic money is small, when the user needs the relay service, he is likely not to have a relay to make a request for the relay service; if the amount of electronic money is large, it is highly likely that the user cannot provide the relay service to other users at this time.

The model of the method of the invention: it is assumed that there are N users in the LTE network, time is divided into discrete time points, and at each time point t, some users need to contact the base station, and some users do not. The invention considers that at each time point, the number of users needing to carry out data transmission with the base station obeys the poisson process, and the strength is theta. Meaning that at each point in time, there is an average of θ users randomly assigned to the system that need to provide communication service to the base station. However, some users may communicate directly with the base station, but others need relay service because of some objective environmental factors. It is first necessary to provide the system N users with a certain amount of electronic money that is used to buy and sell the relay service. When a user needs relay service, the user takes out an electronic currency to send a relay service request to other users, the other users determine that the relay service is not required to be provided according to the electric quantity state and the quantity of the electronic currency, if the user agrees, an ACK (Acknowledgement) is sent to a relay service requester, and a relay service provider increases the electronic currency and consumes certain energy; if not, a NACK (not granted) is sent to the relay service requester, and there is no exchange of electronic money and no energy loss. Meanwhile, each user has a buffer zone, and at each time point, the data volume required to be transmitted by each user determines the length of the buffer zone, so that each buffer delay is determined, and the performance condition of the system is reflected by the buffer delay. Next, important parameters in this method will be described.

First, an external relay demand index (ORDR) assumes that, at time node t, user i ∈ {1,2, …, N } wants to receive data from the base station (user 0), and that τ is a target signal-to-Noise Ratio (SINR)^i,targetIn the case of (1), user i has a target acceptance rate τ^i,targetIs achievable. It should be noted that the target acceptance rate may vary with time, and is obtained hereTime node t, external relay demand index (ORDR):

the formula means that the external relay demand index is the probability that the receiving rate of the data received by the user i from the base station is less than the target receiving rate at the time node t, and the probability that the signal-to-noise ratio between the user i and the base station is less than the target signal-to-noise ratio.

Second, an internal relay demand Index (IRDR): suppose that at time node t, the probability y that user i is required to provide relay serviceⁱ。

Third, Relay acquisition efficiency (Relay received rate abbreviated RRE):

eⁱpr (ACK received)kⁱ>0，pⁱ>0) (2)

This formula is the probability of getting relay service in case the user needs it. k is a radical ofⁱIs the electronic money number, p, of the userⁱIs the energy state of the user.

Fourthly, energy budget: is provided withThe highest energy threshold for the user. When the energy state of the user is less than the value, the user can decide to provide the relay service, and each time the user provides the relay service, the user consumes a certain energy valueWhen the user's electronic energy is 0, the user can neither make a relay request nor provide a relay service.

Fifthly, cooperation strategy: when the user receives the relay request, it must decide whether to provide the relay service. Setting sigmaⁱ(kⁱ,pⁱ) Collaborating policies for users

When sigma isⁱ＝1∈σⁱ(kⁱ,pⁱ) The representative subscriber can provide the relay service when the value is sigmaⁱ＝0∈σⁱ(kⁱ,pⁱ) The representative user cannot provide the relay service. The objective condition of the cooperation strategy is that the user has K electronic currencies and the power state is P, and when the power of the user is slowly consumed, the user wants to save the energy of the user and then uses a lower threshold strategy.

User i optimal cooperation strategy obeys Markov process and setsFor the benefit to be gained at time node t,is composed ofSet a discount coefficient β [0,1), and then obtain the expected value of the user i getting the profit after t time nodes are passed as

And setting the discount coefficients of all the users to be the same, wherein the coefficient is used for measuring the endurance degree of the users in the network, if beta → 1, the users are very tolerant, and if beta → 0, the users only consider the current income, so the users cannot serve as the relay service provider. In the system simulation diagram below, it will be shown how different discount coefficients affect the user cooperation strategy.

At each time node, if the user adopts the optimal strategy sigmaⁱ(kⁱ,pⁱ) After 1, set Vⁱ(kⁱ,pⁱ) For optimal revenue expectation:

sixthly, state transition: setting the state of the user as sⁱ＝(kⁱ,pⁱ)∈S,kⁱ∈ K is the amount of electronic money of the user {1,2, …, K },is the energy state of the user. When a user provides a relay service to another user, it gets an electronic money, loss amount cⁱEnergy (k) ofⁱ,pⁱ)→(kⁱ+1,pⁱ-cⁱ) (ii) a When a user requests a relay service from another user, it pays an electronic money, and the electronic energy remains unchanged (k)ⁱ,pⁱ)→(kⁱ-1,pⁱ). Setting Pⁱ([k^j,p^j]|[kⁱ,pⁱ],σⁱ) For joint probabilities of state transitions, from a state s for a userⁱ＝(kⁱ,pⁱ) Transition to another state s^j＝(k^j,p^j) The two probabilities are the basis of the user state change in the system, the user state in the system can change in real time, and the following 6 probability formulas are also the basis of the user state change. The formula for this joint probability is:

Pⁱ([0,pⁱ]|[0,pⁱ],σⁱ)＝1-yⁱσⁱ(5)

Pⁱ([1,pⁱ-cⁱ]|[0,pⁱ],σⁱ)＝yⁱσⁱ(6)

Pⁱ([kⁱ-1,pⁱ]|[kⁱ,pⁱ],σⁱ)＝λⁱeⁱ(7)

Pⁱ([kⁱ,pⁱ]|[kⁱ,pⁱ],σⁱ)＝1-λⁱeⁱ-yⁱσⁱ(8)

Pⁱ([kⁱ+1,pⁱ-cⁱ]|[kⁱ,pⁱ],σⁱ)＝yⁱσⁱ(9)

Pⁱ([k^j,0]|[kⁱ,0],σⁱ)＝1 (10)

seventhly, buffering the belt: the time is divided into time points with the same length, and the data to be transmitted can be arranged in a queue after reaching the buffer zone until the data is transmitted. Each user has an identical buffer zone. At time t, set W_tThe data group arriving at the buffer zone is {1,2, …, W }, and the data transmission process follows poisson distribution. The present invention considers the size of each data group transmitted to be L bits. At time t, Q is set_tFor queue length, then set U_tTo complete the data set for the transmission process, a dynamic queue length is then obtained as:

Q_t+1＝max[Q_t-U_t,0]+W_t+1，(11)

the working process of the invention is as follows:

having detailed some of the important parameters of the process of the present invention, we now need a brief description of how the process of the present invention needs to be done: first, all users needing relay service need to be traversed, then according to their locations, candidate lists of relay service providers need to be found, and when a candidate service provider receives a request for relay service, they can determine whether to provide the service according to their own electronic money amount and electric quantity state. Whether or not the candidate relay service provider decides to answer the request, at time node t, this information, together with the data set to be transferred arriving at the buffer zone, may determine the queue length of the buffer zone, and then the corresponding transmission delay may be known.

Step 1, searching users needing relay service: the location of the user may change over time, but at time node t, the location of a user is fixed. When the data rate received by the user is less than the target data rate, the user will request the relay service to other surrounding users, see formula (1).

Step 2, determining a relay service provider: the relay service provider candidates are a list that can provide the relay service, but they may decide not to provide the relay service according to their own cooperation policy. Firstly, setting an optimal energy consumption threshold value, wherein the relay service provider provides the service only when the energy of the relay service provider is less than the threshold value; and secondly, setting an optimal threshold value of the electronic money amount, wherein the relay service provider considers to provide the service to obtain the electronic money only when the electronic money amount is less than the value. Only when the above two conditions are simultaneously satisfied, relay service occurs, see equation (3).

Step 3, further, the optimal energy threshold value is set by a user according to different terminal users; the optimal threshold value for the amount of electronic money is determined by three factors: external relay demand, ORDR; a discount coefficient β; energy consumption C. See fig. 2, where count Factor (β): a discount coefficient β; ORDR (λ): an external relay demand index λ; optimalthreshold: an optimal electronic money amount threshold value; policy (cost ═ 0.5): collaborative strategy (energy consumption 0.5); setting the energy consumption C to 0.5, it can be seen that the optimal threshold value of the amount of electronic money increases as the external relay demand ORDR and the discount coefficient β increase. Meanwhile, the value of the discount coefficient beta is set by a network operator, the beta value of each user is the same after the setting is finished, and the corresponding optimal threshold value of the electronic currency quantity can be obtained according to different external relay requirements ORDR.

Step 4, further, the user state is started to be Sⁱ＝(kⁱ,pⁱ)，σⁱWhen 1 represents the relay service, σⁱWhen 0, it represents that the relay service cannot be provided. If the user is a relay service demander, the user state changes when the user obtains the relay service (k)ⁱ,pⁱ)→(kⁱ-1,pⁱ) Corresponding probability is λⁱeⁱThe product of the external relay demand index and the relay acquisition efficiency, see equation (7); if the user acts as a relay service provider, when the user provides relay service for other users, the user state changes (k)ⁱ，pⁱ)→(kⁱ+1,pⁱ-cⁱ) Corresponding probability of yⁱσⁱInner relay demand index and σⁱSee formula (9).

Step 5, further, designing a buffer zone for each user, wherein the corresponding buffer delay is as follows: according to the ktle's Law, the average delay D is determined by the average queue length Q and the average arrival rate a, i.e., Q/a. In this formula, the average arrival time a can be obtained by treating the arrival process as a poisson process, and the queue length can also be obtained by the above formula (11), where in (11) only U_tThis parameter can be derived from the relay system, assuming that the radio channels are orthogonal, so that the amount of data transmitted at time node t can be calculated. At the time node t, the size of the data packet is set to be L bits, so that the quantity U of the transmission data packets can be obtained_tThus, the average queue length and the average arrival rate can be obtained, and the average delay can be obtained through the two values.

The working flow of the method of the invention is shown in figure 1;

by simulating the system model, the following graphs were obtained in MATLAB: a graph in which the threshold value of the amount of electronic money, which is respectively the optimum, changes according to different factors, a graph of the influence of the amount of electronic money on the delay, and a graph of the influence of the movement state of the user on the delay.

Attention is paid to: the simulation area size is 1km multiplied by 1km, 100 users, the poisson strength is the number of data packets arriving at each user at a time node t, the interval between each time node is one minute, and the 600 m/time point is 10 m/s.

The optimal e-currency threshold is determined by three factors: ORDR, discount factor, energy consumption. As shown in fig. 2, the energy consumption is set to 0.5, so that the relationship between the threshold value of the electronic money and the discount coefficient and ORDR can be seen. It can be observed from the graph that the optimal threshold K increases with increasing discount coefficient and ORDR. Meaning that if more and more users seek relay services, the users are increasingly able to patiently wait for the relay services, and the e-currency threshold increases accordingly.

The invention compares the influence of different electronic money amounts on the delay. As shown in fig. 3, in the figure, Tokensupply: an electronic money amount; average delay: averaging the time delays; relay mode: a relay mode; poisson intensity of 20, average velocity of 700 m/time point (11.7 m/s),

in the LTE with Relay service system (Relay Mode), it can be seen that the average delay of the system is relatively minimum when the total number of electronic money is 11000.

The invention compares the influence of the moving state of the user on the system. Relay service-less LTE system (directttransmission Mode). As shown in FIG. 4, Speed (meters/time slot): speed (meters/time point); average delay: averaging the time delays; relay mode: a relay mode; the number of relay mode electronic money is 9000, and the poisson strength is 20. It can be seen that if the rate of movement of the user is relatively fast, the average delay of the system with relay service will be relatively small. When the user is close to the base station, it has a high possibility of directly connecting with the base station to transmit data without a relay service, and when the user is far from the base station, it has a high possibility of requiring a relay service. When the user moves faster, the user frequently changes the position, and the user alternately changes the roles of the relay service demander and the relay service provider, so that the whole system can perform better. When the moving speed of the user is 900 m/time point (15 m/sec), the average delay is relatively minimum.

Examples

The method is realized by writing a program in MATLAB language, and the LTE terminal wireless relay channel model is modeled in the program, so that the time delay performance of the LTE terminal network system in each state is analyzed. When the system is in a state that the poisson strength is 20 and the user moving state is 700 m/time point (11.7 m/s), 11000 electronic money is designed for the system, and the minimum average transmission time delay of the system can be observed. The transmission time delay of different use scenes, such as short messages and picture receiving and sending, online video watching and the like, of the user is different, and the time delay can be reduced by designing appropriate parameters for the different use scenes by using the method disclosed by the invention.

The present invention provides a method for reducing the delay of the relay network of the LTE terminal, and a number of methods and approaches for implementing the technical solution are provided, the above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, a number of improvements and refinements may be made without departing from the principle of the present invention, and these improvements and refinements should also be regarded as the protection scope of the present invention. All the components not specified in the present embodiment can be realized by the prior art.

Claims (6)

1. A method for reducing the time delay of an LTE terminal relay network is characterized by comprising the following steps:

step 1, searching a user needing relay service;

step 2, determining a relay service provider;

step 3, solving the state change probability of the user as a relay service demander and the state change probability of the user as a relay service provider;

and 4, designing a buffer zone for each user to obtain the average time delay of all users.

2. The method of claim 1 wherein step 1 comprises assuming that at time node t, user i is to receive data from the base station, i ∈ {1,2, …, N }, N being the total number of users, N being a natural number, and τ being the target SNR^i,targetIn the case of (1), the external relay demand index of the user i at the time node t is calculated by the following formula

${\mathrm{\&lambda;}}_t^i=({\mathrm{\&gamma;}}_t^{0i}<{\mathrm{\&gamma;}}^{i,t\arg et})=({\mathrm{\&tau;}}_t^{0i}<{\mathrm{\&tau;}}^{i,t\arg et}),$

Wherein, γ^i,targetWhich is indicative of the target acceptance rate,indicating the rate at which user i receives data from the base station, external relay demand indexThat is, at time node t, the probability that the receiving rate of the user i receiving data from the base station is smaller than the target receiving rate is also represented, and the probability that the signal-to-noise ratio between the user i and the base station is smaller than the target signal-to-noise ratio is also represented.

3. The method of claim 2, wherein step 2 comprises:

setting an optimal energy consumption thresholdAnd an optimal electronic money amount threshold value K_th(Pⁱ) When the energy of the relay service provider is less than the optimal energy consumption threshold value and the electronic money amount of the relay service provider is less than the optimal electronic money amount threshold value, the relay service occurs, and the formula is as follows:

wherein k isⁱNumber of electronic money, p, for user iⁱFor the energy state of user i, σⁱ＝1∈σⁱ(kⁱ,pⁱ) Indicates that user i is able to provide a relay service, σⁱ＝0∈σⁱ(kⁱ,pⁱ) Indicating that user i cannot provide the relay service.

4. Method according to claim 3, characterized in that in step 2, if the user employs the optimal policy σⁱ(kⁱ,pⁱ) After 1, set Vⁱ(kⁱ,pⁱ) For optimal revenue expectation:

$V^i(k^i,p^i)={\max }_{{\mathrm{\&sigma;}}^i}E\&lsqb;{\mathrm{\&Sigma;}}_{t=0}^{\mathrm{\&infin;}}{\mathrm{\&beta;}}^t{u}_t^i\&rsqb;,$

wherein,for the benefit obtained by user i at time node t,is composed ofExpected value of (1), discount coefficients β∈ [0,1), β^tIs the discount coefficient at the time t,indicating the expected value of the revenue obtained by the user i after t time nodes.

5. The method of claim 4, wherein step 3 comprises:

user i state starts to be Sⁱ＝(kⁱ,pⁱ)，σⁱWhen 1 indicates that the user can provide the relay service, σⁱWhen the value is 0, it means that the user cannot provide the relay service, and if the user is a relay service providerWhen it provides relay service to other users, the user status changes (k)ⁱ,pⁱ)→(kⁱ+1,pⁱ-cⁱ) Corresponding probability of yⁱσⁱThe calculation formula is as follows:

yⁱσⁱ＝Pⁱ([kⁱ+1,pⁱ-cⁱ]|[kⁱ,pⁱ],σⁱ)，

wherein, cⁱEnergy consumed when providing relay service for user i to other users;

if the user i is a relay service demander, the user state changes when the user i obtains the relay service (k)ⁱ,pⁱ)→(kⁱ-1,pⁱ) Corresponding probability is λⁱeⁱThe calculation formula is as follows:

λⁱeⁱ＝Pⁱ([kⁱ-1,pⁱ]|[kⁱ,pⁱ],σⁱ)。

6. the method of claim 5, wherein step 4 comprises: the average time delay D is calculated by the average queue length Q and the average arrival rate a:

D＝Q/a；

at time t, Q is set_tFor queue length, the dynamic queue length is calculated by the following formula:

Q_t+1＝max[Q_t-U_t,0]+W_t+1，

by Q_t+1Obtaining an average queue length Q, wherein W_tFor data sets arriving in buffer zone, U_tTo complete the data set of the transmission process.