CN107277819B - A Spectrum Resource Division Method Based on Heterogeneous Networks - Google Patents

Abstract

The invention belongs to communicate 5G and heterogeneous network technologies field, specifically a kind of frequency spectrum resource division methods based on heterogeneous network.The invention proposes the heterogeneous network models of a kind of the 5G cellular system for having Fairness Guarantee and D2D mixture of networks, it is intended to the communication requirement to guarantee two kinds of users of heterogeneous network is distributed by reasonable frequency spectrum resource.Basic ideas of the invention are to solve the convex optimization problem of frequency spectrum resource distribution by introducing a fairness constraint condition, and by adjusting the weight of the adjustable two kinds of user's communication requirements of parameter in fairness constraint condition, achieve the purpose that flexible configuration.

Description

A Spectrum Resource Division Method Based on Heterogeneous Networks

technical field

The invention belongs to the technical field of communication 5G and heterogeneous network (Heterogeneous Network), specifically a heterogeneous network model in which a fairness-guaranteed 5G cellular system and a D2D network are mixed. The present invention relates to convex optimization (Convex Optimal), 5G, resource allocation (Resource Allocation) and other technologies.

Background technique

D2D communication does not require the base station to forward data, and it is directly sent from the D2D sending end to the D2D receiving end. Without considering the signaling overhead, it can save half of the system spectrum resources compared with traditional cellular communication, and has the characteristics of low data transmission delay, good link quality, low transmission power, and the ability to reuse cellular spectrum resources. Therefore, carrying the local services with a large amount of data in the current network through D2D transmission can effectively reduce the load of the cellular network, and the architecture of a hybrid network composed of cellular and D2D may be applied in the future. In this hybrid network, the cellular network is the primary system using the licensed frequency band, and the D2D network, as a secondary unlicensed network, can share the licensed frequency band of the cellular network. Therefore, how to allocate spectrum resources to D2D and cellular systems will be the key to system performance.

Contents of the invention

In view of the above problems, the present invention proposes a heterogeneous network model in which a 5G cellular system and a D2D network are mixed with fairness, aiming to ensure the communication needs of two users of the heterogeneous network through reasonable spectrum resource allocation. The basic idea of the present invention is to solve the convex optimization problem of frequency spectrum resource allocation by introducing a fairness constraint condition. Obtain the ratio of two types of users allocated to subcarriers. And by adjusting the parameters in the fairness constraints, the weights of the two kinds of user communication requirements can be adjusted. To achieve the purpose of flexible configuration.

First introduce the system model:

As shown in FIG. 1 , considering a heterogeneous network in which a single-cell cellular and D2D networks are mixed, there is one base station (BS), several cellular users, and several D2D users. Cellular users and D2D users obey Poisson Point Process distribution (Poisson Point Process, PPP) with different parameters. D2D users and cellular users use a complete frequency band through a dedicated mode (Dedicated Mode). There is a scale factor λ∈[0,1], and the subcarriers with a proportion of λ are allocated to cellular users; while the subcarriers with a proportion of 1-λ are allocated to D2D users. Since there is no spectrum reuse and it is a single-cell system, co-channel interference is not considered here.

The system takes the spectrum utilization rate per unit area (ASE bits/(Hz·m ² ·s)) as the performance index, and considers that the signal-to-noise ratio of the received signal is higher than the threshold value before it can communicate. The channel gain follows an exponential distribution with parameter 1. The parameter table of the system is as follows

system variable meaning L The total number of subcarriers in the cell λ_ue Parameters of the PPP distribution of cellular users λ_D2D Parameters of D2D user PPP distribution d_ue Average distance from user to base station d_D2D The average distance between the sending and receiving ends of D2D users σ² thermal noise power P_ue Cellular user transmit power P_D2D D2D user transmit power τ_ue Signal-to-Noise Ratio Threshold for Cellular Users τ_D2D D2D user signal-to-noise ratio threshold gamma large-scale fading coefficient h channel gain λ₁,λ₂ The allocation ratio of subcarriers, λ₁+λ₂=1, the variable to be solved beta Fairness Constraint Parameters alpha optimize the parameters of the expression

Table 1

Technical scheme of the present invention is divided into three steps:

S1, establish the mathematical model that obtains solution λ ₁ , λ ₂ :

Firstly, the optimization index of this problem is given: the spectrum utilization rate per unit area of the system ASE and the ASE expressions of two kinds of users:

Among them, P{SNR _ue ≥τ _ue } and P{SNR _D2D ≥τ _D2D } represent the probability of successful communication between two users.

Among them, q _u and q _D represent the probability that cellular users and D2D users are assigned to subcarriers respectively:

Convert this infinite series to integral form:

Substituting formula (1.4) instead of formula (1.3) into formula (1.1) to get A' _u (λ ₁ ), A' _D (λ ₁ ).

Next define a function with parameters λ ₁ , λ ₂ :

Then the mathematical model to obtain the solution λ ₁ , λ ₂ is as follows

And the solution λ ₁ , λ ₂ of this convex optimization problem is a function of parameter α.

S2. The solution process of the optimization problem

In this step, the curves of the solution λ ₁ and λ ₂ with respect to the parameter α are obtained by solving the mathematical model (1.5). The detailed steps are as follows.

In order to simplify the solution of the problem, two constraint conditions A _u =A _u (λ ₁ ) and A _D =A _D (λ ₂ ) are introduced first. Then the mathematical model (1.5) for solving λ ₁ and λ ₂ becomes:

Then construct the Lagrangian function λ ₁ ,λ ₂ of the objective function

L(u,v,w)＝ln(A _u )+ln(A _D )+u[A _u -A _u (λ ₁ )]+v[A _D -A _D (λ ₂ )]+w(1 -λ ₁ -λ ₂ ) (1.7)

according to A system of equations is obtained, and when the value of α is determined, the numerical solutions of λ ₁ and λ ₂ can be obtained through the dichotomy method. According to the solutions of λ ₁ and λ ₂ obtained under different α values, a curve can be drawn. That is to say, through this step we get λ ₁ , λ ₂ is a function of parameter α as an independent variable. In the next step, we will determine the value of the parameter α to determine the specific λ ₁ , λ ₂ .

S3. Introduce KSBS constraint conditions to determine parameter α

As mentioned above, λ ₁ and λ ₂ are functions of parameter α, and in this step we will introduce KSBS fairness constraints to determine the value of parameter α to obtain specific λ ₁ , λ ₂ .

KSBS fairness constraints are:

The meaning of this formula is to constrain the ASE growth of cellular users and the ASE growth of D2D users to be a fixed ratio β. For example, when β=1, it indicates that the ASE growth of the cellular user is the same as that of the D2D user. Other cases can be deduced by analogy. And this expression is actually a linear function, the focus of its image and λ ₁ , λ ₂ on the α function image is the solution of the whole problem. as shown in picture 2.

The beneficial effect of the present invention is that a fairness constraint condition is introduced to solve the convex optimization problem of frequency spectrum resource allocation, and by adjusting the parameters in the fairness constraint condition, the weights of two kinds of user communication requirements can be adjusted to achieve the purpose of flexible configuration.

Description of drawings

Figure 1 is a schematic diagram of the model;

Figure 2 is a schematic diagram of the function image.

Detailed ways

The technical solution of the present invention has been described in detail in the content of the invention, and will not be repeated here.

Claims (1)

1. A method for dividing spectrum resources based on a heterogeneous network. The heterogeneous network is a heterogeneous network composed of a mixture of a 5G cellular system and a D2D network. It is assumed that there is a base station BS and several cellular users in the heterogeneous network. Several D2D users; cellular users and D2D users obey the Poisson point process distribution with different parameters, D2D users and cellular users use a complete frequency band through a dedicated mode, with a proportional coefficient λ∈[0,1], the ratio is λ ₁ subcarriers are allocated to cellular users; and subcarriers with a proportion of λ ₂ are allocated to D2D users, λ ₁ + λ ₂ = 1, the method uses the spectrum utilization rate per unit area as the performance index, and considers that the received signal Communication can only be carried out when the signal-to-noise ratio is higher than the threshold value, and the channel gain obeys an exponential distribution with a parameter of 1; it is characterized in that the method comprises the following steps: S1, establish the model that obtains λ ₁ and λ ₂ : Firstly, the spectrum utilization ratio expressions of two kinds of users are given as formula 1: Among them, A _u is a cellular user, A _D is a D2D user, λ ₁ and λ ₂ are the allocation ratios of subcarriers, satisfying λ ₁ +λ ₂ =1, q _u , q _D respectively indicate that the cellular user and the D2D user are allocated to The probability of the subcarrier, λ _u is the parameter of the PPP distribution of the cellular user, P _u is the transmit power of the cellular user, τ _ue is the SNR threshold of the cellular user, λ _D is the parameter of the PPP distribution of the D2D user, P _D is the D2D User transmit power, τ _D2D is the signal-to-noise ratio threshold of D2D users; P{SNR _ue ≥τ _ue }, P{SNR _D2D ≥τ _D2D } represents the probability of successful communication between two users: q _u , q _D represent the probability that cellular users and D2D users are assigned to subcarriers respectively: Convert the infinite series of Equation 3 into integral form: Substitute Formula 4 into Formula 1 to get A' _u (λ ₁ ), A' _D (λ ₁ ); (Formula: 5) The function whose parameters are defined as λ ₁ and λ ₂ is: The mathematical model to obtain the solutions λ ₁ and λ ₂ is: The solutions λ ₁ and λ ₂ of formula 5 are functions about the parameter α; S2. Solving the optimization problem: Introduce two constraints A _u =A _u (λ ₁ ), A _D =A _D (λ ₂ ), and transform formula 5 into: Construct the Lagrange function: L(u,v,w)＝ln(A _u )+ln(A _D )+u[A _u -A _u (λ ₁ )]+v[A _D -A _D (λ ₂ )]+w(1 -λ ₁ -λ ₂ ) (Equation 7) according to Obtain a system of equations, in the case of determining the value of α, the solutions λ ₁ , λ ₂ can be obtained through the dichotomy method; S3. Introduce the KSBS constraint condition to determine the parameter α: According to step S2, λ ₁ and λ ₂ can be obtained as a function of parameter α as an independent variable, and KSBS fairness constraints are introduced to determine the value of parameter α to obtain specific λ ₁ , λ ₂ ; The fairness constraints are: The ASE growth of cellular users and the ASE growth of D2D users are constrained to a fixed ratio β by formula 8.