CN103260192A - Home base station and macro base station heterogeneous double-layer network power distribution method based on double utilities - Google Patents

Abstract

The invention relates to a method for solving the problem of power distribution in down links in a home base station and macro base station heterogeneous double-layer network. According to a home base station and macro base station heterogeneous double-layer network power distribution method based on double utilities, regarding the heterogeneously allocated network environment for home base stations and a macro base station, the differences of responsibility and capacity of the home base stations and the macro base station are taken into account, the home base stations and the macro base station respectively carry out power distribution according to different goals, the goal of the home base stations is the cell capacity of a maximized home base station, and the goal of the macro base station is that the energy efficiency of the macro base station is improved on the premise that the lowest signal interference noise ratio of each link is guaranteed. The optimization problem for the double utilities is modeled as a stackelberg game model, the macro base station can forecast the power distribution strategy of each home base station, and selects the optimum power distribution strategy based on the power distribution strategy of each home base station, the home base station can not forecast the power distribution strategy of the macro base station, and carries out power distribution by adopting an iteration water-filling algorithm. According to the home base station and macro base station heterogeneous double-layer network power distribution method based on the double utilities, due to the fact that the stackelberg game model is introduced, the essence of the problem of the power distribution of the home base stations and macro base station heterogeneous double-layer network is reflected more truly, and therefore the performance of power control can be largely improved.

Description

A kind of Home eNodeB and macro base station isomery double-layer network power distribution method of using based on economic benefits and social benefits

Technical field

The present invention relates to the mobile communication technology field, especially, the present invention is be used to providing a kind of under Home eNodeB and macro base station isomery double-layer network deployment scenario, Home eNodeB and macro base station carry out the power control mechanism of power division respectively with different targets, can be used in Long Term Evolution of future generation (LTE) mobile communication system.

Background technology

The next generation mobile communication system network will be operated in the 2GHz frequency range, the loss aggravation through walls of its wireless signal, the more blind area of indoor generation, simultaneously, because occurring in indoor data traffic constantly increases, indoor covering problem becomes problem very important in the next generation mobile communication network design.

Home eNodeB is that a kind of user disposes, the small-power femto cell equipment of plug and play, being mainly used to user for indoor user and hot spot region provides more the signal of high-quality to cover and macroreticular capacity more, it is the important means that solves indoor covering problem and shunt for the hot spot region mobile communications network, have boundless application prospect, can predict Home eNodeB and macro base station isomery double-layer network will become the typical deployment scenario of next generation mobile communication network.

Along with the continuous aggravation of global warming trend, energy-saving and emission-reduction cause the great attention of society day by day, the CO of the annual discharging of ICT industry ₂Accounted for the whole world 2%, and consumed the electric energy in the whole world 0.5%, wherein quite a few energy consumes in base station side, and for energy-saving and emission-reduction and minimizing operator operation expenses, the energy efficiency that promotes the base station has become one of hot issue of whole moving communicating field concern.

In LTE-Advanced the 4th third generation mobile communication network, different, equipment the capacity variances with the function of macro base station of the Home eNodeB of Home eNodeB and macro base station isomery double-layer network are huge, therefore the two responsibility of bearing is also different, and this is the problem that must consider in the design of RRM mechanism.For Home eNodeB, because the fast development of mobile Internet, mobile data traffic presents explosive growth, simultaneously, because the data service more than 70% occurs in indoor, therefore, the major function of Home eNodeB is the high-speed radio access to be provided, to be the hot spot region shunting of mobile communications network for indoor environment.And because the own low-power consumption of Home eNodeB and disposed and the feature of management by the user, its energy-conservation needs are not urgent.For macro base station, the wide area covering that provides the QoS guarantee for the user will be its main responsibility, the great burden that the while brings for environment and operator owing to its huge energy consumption, and therefore, the efficiency of raising macro base station is more even more important than Home eNodeB.

The document of research Home eNodeB and macro base station isomery double-layer network has much at present, but study at the energy efficiency simple target of maximization network capacity or optimization network mostly, at two kinds of characteristics that the base station is different in the heterogeneous network, carry out economic benefits and social benefits and lack relatively with Study on optimized.Patent of the present invention is made as the maximization efficiency with the optimization aim of macro base station, and the optimization aim of Home eNodeB is made as the maximization cell capacity, and the mode of distributing by realtime power is optimized the two effectiveness of Home eNodeB and macro base station, improves the performance of network integral body.

Summary of the invention

Under the double-layer network sight that the present invention is intended to dispose at Home eNodeB and macro base station isomery, how effectively to improve the capacity of home base station network, and the energy efficiency problem that promotes macro base station when guaranteeing macro base station user QoS, adopt the modeling of stackelberg betting model, proposed a kind of comparatively effectively Home eNodeB and macro base station downlink power allocation mechanism.

The present invention mainly adopts following technical method to solve the problems of the technologies described above.

In the Home eNodeB of using based on economic benefits and social benefits and macro base station isomery double-layer network power distribution method, described family base station system comprises a plurality of Home eNodeB, be deployed in the macrocell of some macro base station coverings, each Home eNodeB is by the network of cable broadband mode access carriers such as ADSL, wire cable, optical cable, and can carry out information interaction with macro base station, Home eNodeB can be gathered the information in this base station network, and passes to macro base station in real time.

As shown in Figure 1, consider that random distribution has K Home eNodeB in a macrocell, macro base station and Home eNodeB common spectrum resource, therefore, on each subchannel, Home eNodeB and macro base station can produce mutually and disturb.

The present invention considers the two performance of Home eNodeB and macro base station respectively, has set different utility functions accordingly.Set up the stackelberg betting model simultaneously, macro base station is as the leader of game, and each base station is as the follower of game, macro base station can be that the policy selection of Home eNodeB is made a prediction to the follower, and based on predicting the power division of at first making the best, and Home eNodeB does not have predictive ability, can only carry out policy selection after macro base station makes decisions again.Macro base station and Home eNodeB constantly carry out policy update, and after the iteration of certain number of times, Home eNodeB and macro base station can reach poised state, i.e. stackelberg equilibrium.

For Home eNodeB k, its optimization aim function can be expressed as

$\max :R_k=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2(1+\frac{{P_k}^n}{{\mathrm{\σ}}_k^n+\underset{j\≠k}{\mathrm{\Σ}}{P}_j^n{\mathrm{\α}}_{\mathrm{jk}}^n+p_m^n{\mathrm{\α}}_{\mathrm{mk}}^n})$

$s.t:\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{P_k}^n\≤P_k^{\max },$

Wherein, R _kThe cell capacity of expression Home eNodeB k, N represents the quantity of Traffic Channel in the network, P _k ⁿRepresent Home eNodeB k in the transmitting power of channel n,

Represent Home eNodeB j in the transmitting power of channel n, Represent macro base station in the transmitting power of channel n,

Be the power of the user's of busy channel n additive white Gaussian noise in the Home eNodeB

Channel gain with channel n

Ratio,

Be normalized interference channel gain, wherein

It is the gain of the interference channel of the Home eNodeB k of Home eNodeB j on channel n.Restrictive condition

Be to say that the total transmitting power of down link of Home eNodeB is less than the maximum transmit power limit of Home eNodeB

Owing to do not predict the ability of other adversary's competitive strategies, Home eNodeB suffered noise and disturb sum on each channel Can obtain by measuring, can think a constant in the game iteration each time.Therefore, can adopt classical power water-filling algorithm to calculate transmitting power on subchannel n for Home eNodeB k, for $p_k^n={\mathrm{\μ}}_k-({\mathrm{\σ}}_k^n+\underset{j\≠k}{\mathrm{\Σ}}{P}_j^n{\mathrm{\α}}_{\mathrm{jk}}^n+p_m^n{\mathrm{\α}}_{\mathrm{mk}}^n),$ μ wherein _kBe " horizontal plane " of power water filling, try to achieve each with Lagrangian Y-factor method Y After, bring into

Middle summation can obtain μ _k

Home eNodeB can be in real time reported information in this base station to macro base station, and macro base station can be selected make a prediction based on this power policy to home base station cells.Therefore, the optimization aim of Home eNodeB can be expressed as

$\max :\mathrm{\η}=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2(1+\frac{p_m^n}{{\mathrm{\σ}}_k^n+\underset{k=1}{\overset{k}{\mathrm{\Σ}}}{p}_k^n{\mathrm{\α}}_{\mathrm{km}}^n})}{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{p}_m^n+p_c}$

$s.t.\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{p}_m^n\≤P_m^{\max }---\left(1\right)$

$p_m^n{H}_{\mathrm{mm}}^n\≥{\mathrm{\γ}}_m^n,\∀n\∈N---\left(2\right)$

$p_k^n=\underset{p_k^n}{\arg \max }\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2(1+\frac{p_k^n}{{\mathrm{\σ}}_k^n+\underset{j\≠k}{\mathrm{\Σ}}{p}_j^n{\mathrm{\α}}_{\mathrm{jk}}^n+p_m{\mathrm{\α}}_{\mathrm{mk}}^n}),\∀k\∈K---\left(3\right)$

Restrictive condition (2) refers to that the channel of each macro base station will satisfy certain qos requirement.What restrictive condition (3) reacted is the predictive ability of macro base station, when the transmitting power of macro base station on channel n is

The time, according to water-filling algorithm and non-cooperative game theory, the transmitting power of Home eNodeB k on channel n is

Therefore, when macro base station is at first made after power policy selects, the power distribution strategies that each Home eNodeB is determined by non-cooperative game subsequently and the power distribution strategies of macro base station are linear, and namely restrictive condition (3) can be used

Equivalent representation.Wherein,

Refer to that in the macro base station power distribution strategies be P _mAfter, when Home eNodeB reaches Nash Equilibrium by non-cooperative game, the transmitting power of each Home eNodeB on channel n; E is unit matrix, $g^n=\left[\begin{array}{c}{\mathrm{\α}}_{m1}^n\\ {\mathrm{\α}}_{m2}^n\\ .\\ .\\ .\\ {\mathrm{\α}}_m^n\end{array}\right],$ $\mathrm{\μ}=\left[\begin{array}{c}{\mathrm{\μ}}_1\\ {\mathrm{\μ}}_2\\ .\\ .\\ .\\ {\mathrm{\μ}}_K\end{array}\right],$ ? ${\mathrm{\σ}}^n=\left[\begin{array}{c}{\mathrm{\σ}}_1^n\\ {\mathrm{\σ}}_2\\ .\\ .\\ .\\ {\mathrm{\σ}}_K^n\end{array}\right].$ At this moment, macro base station suffered interference on channel n $\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{p}_k^n{\mathrm{\α}}_{\mathrm{km}}^n=h^n\×\mathrm{NE}\left(P_m\right)=h^n(E+G)(\mathrm{\μ}-\mathrm{\σ})-h^n{(E+G)}^{-1}{g}^n{P}_m^n,$ Wherein, $h^n=\left[\begin{array}{cccc}{\mathrm{\α}}_{1m}^n& {\mathrm{\α}}_{2m}^n& ...& {\mathrm{\α}}_{\mathrm{Km}}^n\end{array}\right],$ Therefore in the target function macro base station suffered interference on channel n can be expressed as with

Linear function for independent variable.For finding the solution of optimal solution, we can adopt the method for fractional programming, and target function is changed into $\max :\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2(1+\frac{p_m^n}{{\mathrm{\σ}}_k^n+\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{p}_k^n{\mathrm{\α}}_{\mathrm{km}}^n})-q(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{p}_m^n+p_c),$ Wherein q is a variable, for each q value, all to there being one group Make

Get maximum, and for satisfying restrictive condition, making maximum simultaneously is that group of 0

Be the best power allocation strategy, the q of correspondence is the maximum of macro base station efficiency simultaneously.Can adopt the mode of iteration to calculate best power distribution strategies, elder generation is q fixedly, finds the solution $\max :\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2(1+\frac{p_m^n}{{\mathrm{\σ}}_k^n+\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{p}_k^n{\mathrm{\α}}_{\mathrm{km}}^n})-q(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{p}_m^n+p_c),$ Judge whether its maximum is 0; Upgrade q by dichotomy, continue to find the solution $\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2(1+\frac{p_m^n}{{\mathrm{\σ}}_k^n+\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{p}_k^n{\mathrm{\α}}_{\mathrm{km}}^n})-q(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{p}_m^n+p_c)$ Maximum, be 0 q up to finding corresponding maximum.

For whole isomery double-layer network power distribution method based on the stackelberg game, its key step is

Step 1: before Home eNodeB carried out power division, macro base station was target with the maximization efficiency, at first carries out power division one time;

Step 2: Home eNodeB carries out interferometry, and carries out the power division of non-cooperative game, reaches Nash Equilibrium up to all Home eNodeB;

Step 3: Home eNodeB comprises the channel gain of its each down channel to the macro base station reporting information $H_k=\left[\begin{array}{cccc}{H}_{\mathrm{kk}}^1& H_{\mathrm{kk}}^2& ...& H_{\mathrm{kk}}^N\end{array}\right],$ Additive white Gaussian noise power ${\mathrm{\σ}}_k=\left[\begin{array}{cccc}{\mathrm{\σ}}_k^1& {\mathrm{\σ}}_k^2& ...& {\mathrm{\σ}}_k^N\end{array}\right],$ The numerical value of " horizontal plane " of water-filling algorithm $\mathrm{\μ}={\left[\begin{array}{cccc}{\mathrm{\μ}}_1& {\mathrm{\μ}}_2& ...& {\mathrm{\μ}}_K\end{array}\right]}^T,$ Macro base station on each channel to the channel gain of the interference channel of this base station $H_k^M=\left[\begin{array}{cccc}{H}_{\mathrm{Mk}}^1& H_{\mathrm{Mk}}^2& ...& H_{\mathrm{Mk}}^N\end{array}\right],$ Other Home eNodeB on each channel to the channel gain of the interference channel of this base station

Step 4: macro base station reports and to next step the information of forecasting of power distribution strategies of Home eNodeB, is that target is carried out best power and distributed with the maximization efficiency based on Home eNodeB.

Step 5: judge whether this power division of macro base station and last time be consistent, if, then jump to step 8, otherwise, jump to step 6;

Step 6: Home eNodeB carries out interferometry, and carries out the power division of non-cooperative game, reaches Nash Equilibrium up to all Home eNodeB;

Step 7: judge whether this is balanced identical with equilibrium last time, if, then jump to step 8, otherwise, jump to step 3;

Step 8: the power division of the isomery double-layer network of Home eNodeB and macro base station has reached stackelberg game equilibrium state, and this power division finishes, and waits for scheduling next time.

Description of drawings

In order more clearly to set forth embodiments of the invention and existing technical method, below the explanation accompanying drawing of using in technical method explanation accompanying drawing of the present invention and the description of the Prior Art is done simple introduction, obviously, under the prerequisite of not paying creative work, those of ordinary skills can obtain other accompanying drawing by this accompanying drawing.

Fig. 1 is the system's scene graph that comprises Home eNodeB, macro base station and mobile subscriber in the embodiment of the invention.

Fig. 2 is power division flow chart in the embodiment of the invention.

Embodiment

Clearer for what technical method advantage of the present invention was described, below in conjunction with accompanying drawing the specific embodiment of the present invention is described in further detail, obvious described embodiment is part embodiment of the present invention, rather than whole embodiment.Embodiments of the invention can be expanded on this basis, under the situation of overall architecture unanimity, be obtained more optimization methods.According to embodiments of the invention, those of ordinary skill in the art can realize every other embodiment of the present invention on without the basis of creative work, all belong to protection scope of the present invention.

Main thought of the present invention is: in Home eNodeB and macro base station isomery double-layer network deployment scenario, the power division problem of the down link of Home eNodeB and macro base station is modeled as a stackelberg game, macro base station is target with the maximization efficiency, and can predict the power distribution strategies of Home eNodeB, and Home eNodeB can not be predicted other adversarys' strategy, Home eNodeB maximizes self capacity by the mode of non-cooperative game, after the final game iteration by limited number of times, Home eNodeB and macro base station will reach the stackelberg equilibrium.

Its concrete steps are described below:

Step 201: before Home eNodeB carried out power division, macro base station was target with the maximization efficiency, at first carries out power division one time;

Step 202: Home eNodeB carries out interferometry, and carries out the power division of non-cooperative game, reaches Nash Equilibrium up to all Home eNodeB;

Step 203: Home eNodeB comprises the channel gain of its each down channel to the macro base station reporting information $H_k=\left[\begin{array}{cccc}{H}_{\mathrm{kk}}^1& H_{\mathrm{kk}}^2& ...& H_{\mathrm{kk}}^N\end{array}\right],$ Additive white Gaussian noise power ${\mathrm{\σ}}_k=\left[\begin{array}{cccc}{\mathrm{\σ}}_k^1& {\mathrm{\σ}}_k^2& ...& {\mathrm{\σ}}_k^N\end{array}\right],$ The numerical value of " horizontal plane " of water-filling algorithm $\mathrm{\μ}={\left[\begin{array}{cccc}{\mathrm{\μ}}_1& {\mathrm{\μ}}_2& ...& {\mathrm{\μ}}_K\end{array}\right]}^T,$ Macro base station and other Home eNodeB on each channel to the channel gain of the interference channel of this base station $H_k^M=\left[\begin{array}{cccc}{H}_{\mathrm{Mk}}^1& H_{\mathrm{Mk}}^2& ...& H_{\mathrm{Mk}}^N\end{array}\right],$

Step 204: macro base station reports and next step the information of forecasting of power distribution strategies of Home eNodeB is carried out best power distribute based on Home eNodeB.

Step 205: judge whether this power division of macro base station and last time be consistent, if, then jump to step 208, otherwise, jump to step 206;

Step 206: Home eNodeB carries out interferometry, and carries out the power division of non-cooperative game, reaches Nash Equilibrium up to all Home eNodeB;

Step 207: judge whether this is balanced identical with equilibrium last time, if, then jump to step 208, otherwise, jump to step 203;

Step 208: the power division of Home eNodeB and macro base station isomery double-layer network has reached stackelberg game equilibrium state, and this power division finishes, and waits for scheduling next time.

Claims (6)

1. Home eNodeB and macro base station isomery double-layer network power distribution method of using based on economic benefits and social benefits is characterized in that may further comprise the steps:

Step 1: before Home eNodeB carried out power division, macro base station was target with the maximization efficiency at first, carries out power division one time;

Step 2: Home eNodeB carries out interferometry, and carries out the power division of non-cooperative game, reaches Nash Equilibrium up to all Home eNodeB;

Step 3: Home eNodeB comprises the channel gain of its each down channel to the macro base station reporting information

Additive white Gaussian noise power

The numerical value of " horizontal plane " of water-filling algorithm Macro base station on each channel to the channel gain of the interference channel of this base station

Other Home eNodeB on each channel to the channel gain of the interference channel of this base station

Step 4: macro base station reports and to next step the information of forecasting of power distribution strategies of Home eNodeB, is that target is carried out best power and distributed with the maximization efficiency based on Home eNodeB.

Step 5: judge whether this power division of macro base station and last time be consistent, if, then jump to step 8, otherwise, jump to step 6;

Step 6: Home eNodeB carries out interferometry, and carries out the power division of non-cooperative game, reaches Nash Equilibrium up to all Home eNodeB;

Step 7: judge whether this is balanced identical with equilibrium last time, if, then jump to step 8, otherwise, jump to step 3;

Step 8: the isomery double-layer network power division of Home eNodeB and macro base station has reached stackelberg game equilibrium state, and this power division finishes, and waits for scheduling next time.

2. a kind of Home eNodeB and macro base station isomery double-layer network power distribution method of using based on economic benefits and social benefits according to claim 1 is characterized in that:

In the described step 1, macro base station at first carries out power division one time, because there is not the interference of Home eNodeB, simultaneously because macro base station lacks the relevant information that the Home eNodeB power division is predicted, therefore, the power division of macro base station is equivalent to optimize following problem at this moment:

。

3. a kind of Home eNodeB and macro base station isomery double-layer network power distribution method of using based on economic benefits and social benefits according to claim 1 is characterized in that:

In the described step 2, the interference of Home eNodeB perception macro base station and other Home eNodeB, and carry out power division according to the power water-filling algorithm, through after the iteration of certain number of times, whole home base station network reaches the Nash Equilibrium state.

4. a kind of Home eNodeB and macro base station isomery double-layer network power distribution method of using based on economic benefits and social benefits according to claim 1 is characterized in that:

In described step 3, Home eNodeB is for the power division prediction of macro base station to Home eNodeB to the macro base station reporting information, namely provides calculating to macro base station

Relevant information.

5. a kind of Home eNodeB and macro base station isomery double-layer network power distribution method of using based on economic benefits and social benefits according to claim 1 is characterized in that:

In described step 4, macro base station carries out power division according to the solution of following optimization problem,

Wherein, restrictive condition (3) can pass through

Equivalent representation, above optimization problem can adopt the method for fractional programming, obtains optimal solution by the iteration of certain number of times.

6. a kind of Home eNodeB and macro base station isomery double-layer network power distribution method of using based on economic benefits and social benefits according to claim 1 is characterized in that:

In described step 8, the criterion that reaches the stackelberg game equilibrium refers to that power division that the power division of Home eNodeB reaches Nash Equilibrium or macro base station equates with last calculating.

Images