CN101895421A - Communication resource allocating method - Google Patents

Abstract

The invention relates to a communication resource allocating method, which comprises the following steps: establishing an ant colony algorithm node diagram for each cell in a communication system, and defining and initializing pheromones and heuristic information of each node and each path; determining an evaluation function of the communication system based on a preset optimization objective of the communication system; in each iteration, respectively allocating communication resources in parallel to users in each cell; after communication resources in all the cells are allocated, calculating an evaluation result based on the evaluation function and carrying out global pheromone update; and if a termination condition is met, stopping the allocating process and carrying out practical allocation on the communication resources based on the obtained allocating scheme, otherwise carrying out the next iteration. In the embodiment of the invention, the ant colony optimization algorithm is used to allocate the communication resources to the users in each cell in the communication system, and higher-quality solutions for optimizing problems are quickly obtained by probabilistic search of the ant colony optimization algorithm so as to guide the practical allocating process of the communication resources.

Description

Communication resource allocating method

Technical field

The present invention relates to the communications field, relate in particular to the scheduling of resource in a kind of communication system, the method for distribution.

Background technology

In the communications field, the use of how reasonably allocating the communication resource and optimizing the communication resource is unusual important problem, and communication resource optimization problem itself is owing to related to the optimization that a plurality of dimensional information realize the communication resource, for example information such as frequency, time slot, code word, power, antenna are therefore very difficult.

Existing common way is that the multidimensional resource optimization problem in some communications field is abstract in combinatorial optimization problem, attempts the optimal solution of the problem that obtains by protruding optimum theory, thus the direct communications Resource allocation and smoothing.But owing to changed the former restricted of problem, the optimal solution of acquisition also is not suitable for initial problem.Owing to adopted the power water-filling algorithm, in polynomial time, also be difficult to the optimal solution of the problem that obtains simultaneously.And, so be difficult to that more existing intelligent optimization algorithms are applied directly to the communications field and come the direct communications Resource allocation and smoothing because the application problem that the resource allocation of the communications field relates to and model and the resource allocation in other field have bigger differently.

Summary of the invention

The objective of the invention is to propose a kind of communication resource allocating method, can utilize ant colony optimization method to communicate the scheduling and the distribution of resource fast at the characteristics of the communications field.

For achieving the above object, the invention provides a kind of communication resource allocating method, comprising:

For setting up the ant group algorithm node diagram in each sub-district in the communication system, and the pheromones and the heuristic information in definition and each node of initialization and path;

Determine the evaluation function of communication system according to the optimization aim of default communication system;

In iteration each time, respectively to the parallel allotment of the user in each sub-district communication resource;

All finish the allotment of the communication resource when all sub-districts after, calculate evaluation result by described evaluation function, and carry out the renewal of global information element;

If reach end condition, then stop allocation process, and communicate the actual allotment of resource, otherwise carry out next iteration according to the allotment scheme that obtains.

Further, respectively to the parallel allotment of the user in each sub-district communication resource time, can also carry out the renewal of local message element simultaneously to each node.

Further, each node that is updated to the scheme of all previous iteration evaluation result optimum in the communication system of described global information element upgrades, and perhaps upgrades for each node to the scheme of evaluation result optimum in this iteration.

Further, each node that is updated to the more excellent a plurality of schemes of all previous iteration evaluation result in the communication system of described global information element upgrades, and perhaps upgrades for each node to the more excellent a plurality of schemes of evaluation result in this iteration.

Further, described heuristic information is one or more combinations and the function of user in channel quality information, momentary rate and the Mean Speed of allocate resource unit.

Further, described channel quality information comprises one or more the combination in signal to noise ratio, signal interference ratio, Signal to Interference plus Noise Ratio, channel gain interference ratio, channel condition information, the data rate that can transmit, the error rate, Block Error Rate and the interference strength.

Further, to be that described evaluation result meets pre-conditioned, ant group algorithm convergence and iterations reach one or more the combination in the preset times for described end condition.

Further, when the optimization aim of default communication system be under the limited situation of gross power during maximize throughput, it is big more that the evaluation function of communication system is defined as characterizing throughput, the function expression that pheromones release is many more.

Further, when the optimization aim of default communication system is the fairness of taking into account between throughput and user, the evaluation function of communication system adopts logarithmic function to act on the function expression that pheromones discharges, perhaps heuristic information adopts logarithmic function to act on the function expression of communication resource allocating process, perhaps regulates the allotment probability respectively to the parallel allotment of the user in each sub-district communication resource time.

Further, allotment ant group optimization model that the communication resource adopted be the ant system, the ant system that refines, based on the ant system of arranging, minimax ant system, ant group system or approximate uncertainty tree search system.

Further, to the parallel allotment of the user in each sub-district communication resource time, upgrade the selectable user set according to the difference of traffic performance and user's request.

Further, described communication resource allocating method adopts a plurality of parallel ant groups to carry out computing, in the calculating process between the ant group exchange of information in the hope of finding separating of problem quickly.

Based on technique scheme, the embodiment of the invention is allocated the communication resource by ant colony optimization algorithm to the user in each sub-district in the communication system, utilizes the probabilistic search of ant colony optimization algorithm to obtain the separating of better quality of optimization problem fast; The embodiment of the invention can also be according to the requirements definition heuristic information of communication system, determine evaluation function, to obtain the resource allocation proposal that compliance with system requires, instructs the allocation process of the actual communication resource.

Description of drawings

Accompanying drawing described herein is used to provide further understanding of the present invention, constitutes the application's a part, and illustrative examples of the present invention and explanation thereof are used to explain the present invention, do not constitute improper qualification of the present invention.In the accompanying drawings:

Fig. 1 is the schematic flow sheet of an embodiment of communication resource allocating method of the present invention.

Fig. 2 is the schematic diagram of the ant group algorithm node diagram of ant colony optimization algorithm of the present invention.

Fig. 3 is the schematic flow sheet of another embodiment of communication resource allocating method of the present invention.

Embodiment

Below by drawings and Examples, technical scheme of the present invention is described in further detail.

Among the present invention communication resource allocating based on ant colony optimization algorithm (Ant ColonyOptimization, be called for short ACO) be a kind of typical algorithm of first heuritic approach, the present invention utilizes the probabilistic search of ACO to obtain the suboptimal solution of communication resource optimization problem fast.

Ant colony optimization algorithm is one of typical algorithm of colony intelligence algorithm.Colony intelligence is meant that the main body of not having intelligence shows the characteristic of intelligent behavior by cooperation.Colony intelligence relies on the probabilistic search algorithm, and finding the solution of parallel distributed can make full use of multiprocessor, and the continuity of problem is not had specific (special) requirements, is a kind of new method that can effectively solve most of global optimization problems.

Ant colony optimization algorithm is proposed the nineties by Italian scholar Marco Dorigo the earliest, is referred to as ant system (Ant System is called for short AS).AS simulates the process of looking for food of true ant, can finish complicated work or reach the optimization of system problem by each the simple intermediary that distributes.The AS development has formed some improvement projects so far, has been applied to the research of traveling salesman problem, broad sense assignment problem etc.

The ant of ACO is represented a building process at random, thereby constructs complete separating by constantly separating the composition of separating that interpolation meets definition to part in building process.Adopting with the pheromones between the ant is the indirect communication form of media.Pheromones is a kind of media of ant according to the quality release of separating, and comprises discharging and evaporate two kinds of actions.The typical case of ACO is applied as traveling salesman problem, and (traveling salesman problem, TSP), the ant m that is positioned at city i selects city j as the probability in next one visit city to be

$p_{\mathrm{ij}}^m=\frac{{\left[{\mathrm{\τ}}_{\mathrm{ij}}\right]}^{\mathrm{\α}}{\left[{\mathrm{\η}}_{\mathrm{ij}}\right]}^{\mathrm{\β}}}{\underset{l\∈N_i^m}{\mathrm{\Σ}}{\left[{\mathrm{\τ}}_{\mathrm{il}}\right]}^{\mathrm{\α}}{\left[{\mathrm{\η}}_{\mathrm{il}}\right]}^{\mathrm{\β}}},\∀j\∈N_i^m---\left(1\right)$

Wherein, Represent the city set adjacent with city i, select formula as can be seen by above-mentioned probability, two factors are selected the city decisive role to ant: pheromones value τ _IjWith heuristic information η _Ij, the two has the effect of guiding the search skewed popularity to ant.Common η _IjBe made as 1/d _Ij, d _IjExpression city i, the distance between j, d _IjMore little, η _IjBig more, then ant selects the probability of j big more.τ _IjThen be to exchange a kind of media that uses between ant mutually, the path that ant makes up is short more, the τ of its release _IjBig more, thus attract other ant to select this paths.α, β have determined pheromones and the effect of heuristic information in the process that the ant search is separated.The ratio of the two has been controlled pheromones and the relative influence power of heuristic information when node is chosen.If α is bigger, then the relativity of pheromones is bigger, and algorithm may be stuck in a certain paths very soon, and this paths may not be to be optimal path; On the contrary, the relativity of pheromones is very little, and the influence that the renewal of pheromones is selected node is very little, and algorithm may not find optimal path for a long time yet.If α=0 then is equivalent to classical greedy algorithm.If β=0, the amplification coefficient that then has only pheromones in action, and the skewed popularity that brings without any heuristic information, this will make the performance of algorithm become no good.Next step that each ant is constantly selected to move according to formula (1) parallelly makes up separating of problem asynchronously.

After each ant is constructed complete separating, system according to different evaluation functions to separating the pheromones of assessing and upgrading each node or path.Release by pheromones reduces the scale of separating the search volume gradually, makes the hunting zone be retracted on the potential path of minority, thereby finds separating of problem fast.

In the resource allocation problem of communication system, its flow process comprises referring to Fig. 1 with the ACO algorithm application in the present invention:

Step 101, for setting up the ant group algorithm node diagram in each sub-district in the communication system, and the pheromones and the heuristic information in definition and each node of initialization and path;

The optimization aim of the communication system that step 102, basis are preset is determined the evaluation function of communication system;

Step 103, in this iteration, respectively to the parallel allotment of the user in each sub-district communication resource, the probability of foundation selects formula to select formula (1) for the probability among the AS, rely on the effect of heuristic information and pheromones merely.Probability in foundation AS is selected the formula (1), can also carry out probability according to the pseudorandom ratio rule among the ant group ACS of system and select., formula is as follows:

$\left\{\begin{array}{ccc}k={\arg \max }_{l\∈N_n^m}\left\{{\mathrm{\τ}}_{\mathrm{nl}}{\left[{\mathrm{\η}}_{\mathrm{nl}}\right]}^{\mathrm{\β}}\right\},& \mathrm{ifq}\≤q_0& (2-a)\\ p_{n,k}=\frac{{\left[{\mathrm{\τ}}_{n,k}\right]}^{\mathrm{\α}}{\left[{\mathrm{\η}}_{n,k}\right]}^{\mathrm{\β}}}{\underset{l\∈N_n^m}{\mathrm{\Σ}}{\left[{\mathrm{\τ}}_{n,l}\right]}^{\mathrm{\α}}{\left[{\mathrm{\η}}_{n,l}\right]}^{\mathrm{\β}}}& \mathrm{else}& (2-b)\end{array}\right.---\left(2\right)$

Compare the search experience that can develop ant better and accumulated with AS;

Step 104, judging whether all sub-districts all finish the allotment of the communication resource, is execution in step 105 then, continues the user who does not allocate the communication resource is carried out resource allocation otherwise return step 103;

Step 105, calculate evaluation result, and carry out the renewal of global information element by described evaluation function;

Step 106, judging whether to reach end condition, is execution in step 107 then, carries out next iteration otherwise return step 103;

Step 107, stop allocation process, and communicate the actual allotment of resource according to the allotment scheme that obtains.

In the present embodiment, heuristic information can be one or more combinations and the function of user in channel quality information, momentary rate and the Mean Speed of allocate resource unit, and channel quality information comprises one or more the combination in signal to noise ratio, signal interference ratio, Signal to Interference plus Noise Ratio, channel gain interference ratio, channel condition information, the data rate that can transmit, the error rate, Block Error Rate and the interference strength.

The present invention allocates ant group optimization model that the communication resource adopts can be for ant system, the ant system that refines, based on the ant system of arranging, minimax ant system, ant group system or approximate uncertainty tree search system.

In step 101, need set up the node diagram of ant group algorithm for each sub-district in the communication system, each node represents that ant makes up the state of process of separating in node diagram.Node diagram as shown in Figure 2, the user k in every line display sub-district of node diagram wherein, k ∈ K, wherein K represents the user's number in the sub-district, allocate resource n is shown in every tabulation of node diagram, n ∈ N, wherein N is a resource quantity adjustable in the communication system.In node diagram, be each node definition c _{N, k}Indicate the allotment situation of the communication resource, suppose that resource n is allocated k, then c to the user _{N, k}Be 1, otherwise be 0, promptly

Suppose that in the present embodiment there are 4 users in this node diagram corresponding district, 6 resources to be deployed, the arrow that points to by order between the node in Fig. 2 has been represented a schematic path, is used for representing the allotment situation of the communication resource in this sub-district, i.e. node c _1,3, c _2,2, c _3,1, c _4,4, c _5,3, c _6,2=1, all the other nodes are 0.

In step 101, communication system can be selected user's procedure definition for each allocate resource is a step-length, and ant is the user of current allocate resource according to certain probability selection allotment, in other words, allocates current resource to be deployed for the user exactly.Total N the resource to be deployed in every sub-district, then N step-length is an iteration, finishes until resource allocation.

Different resource can be stored among the matrix Φ the effectiveness of different user, and this matrix is K * N rank matrix, as the heuristic information of ACO algorithm.The effectiveness here promptly refers to heuristic information η _Nk, can require according to the difference of system to define.Can define different heuristic informations at different optimization aim.

The pheromones of different nodes is stored among the matrix Г, is K * N rank matrix, element τ _NkUser k is to the attraction degree of ant during expression allocate resource n.In step 101, the pheromones of each node can be set to empirical value or specific numerical value τ ₀

In step 102, evaluation function can be set according to the optimization aim of default communication system, two performances (throughput and fairness) of more often paying close attention to communication system are example, the optimization aim of communication system can be an optimization aim with the throughput-maximized of system, can with the fairness of resource allocation between the user optimization aim also, or compromise preferably as optimization aim with the throughput of system than the fairness of resource allocation between big and user.In order to realize different optimization aim, need to determine to realize the evaluation function of optimization aim.

Optimization aim with default communication system is an example for maximize throughput under the limited situation of gross power, and it is big more that the evaluation function of communication system can be defined as characterizing throughput, and pheromones discharges many more function expressions.In the embodiment of back, also can provide the example of more detailed evaluation function.

Optimization aim with default communication system is that the fairness of taking into account between throughput and user is an example, and the evaluation function of communication system can adopt logarithmic function to act on the function expression that pheromones discharges.In order to realize taking into account the fairness between throughput and user, it is also conceivable that and adopt function expression that logarithmic function acts on the communication resource allocating process, perhaps respectively to the parallel allotment of the user in each sub-district communication resource time, regulate the allotment probability parameter in the allotment formula as heuristic information.

In step 103, corresponding to the ant of each sub-district according to the allotment formula respectively to the parallel allotment of the user in each sub-district communication resource, the setting of allotment formula can be determined by heuristic information and/or pheromones.Respectively to the parallel allotment of the user in each sub-district communication resource time, can also select simultaneously each node to be carried out the renewal of local message element, make local pheromones dynamic change with this, increase the exploration of ant.

To the parallel allotment of the user in each sub-district communication resource time, can upgrade the selectable user set according to the difference of traffic performance and user's request.

In step 105, all finished first resource allotment corresponding to the ant of all sub-districts after, each sub-district has obtained a kind of scheme of resource allocation accordingly, the different user in the sub-district is formulated to the different communication resources.Calculate by predetermined evaluation function this moment, obtains evaluation result, and carry out the renewal of global information element.

The renewal of global information element can be for upgrading each node that iterates to the scheme of present evaluation result optimum from original execution in the communication system, also can upgrade for each node of the scheme of evaluation result optimum in this iteration.The renewal of global information element can select optimum scheme to upgrade, and also can select the scheme of more excellent some to carry out the renewal of pheromones.

In another embodiment, also can be that unit carries out that global information is plain to be upgraded with the sub-district, each sub-district is finished the result of this allotment of allotment post-evaluation and is carried out plain renewal of global information of this sub-district, will wait until that not necessarily all sub-districts all finish allotment and upgrade.

When obtaining to allocate scheme preferably, nodal information element after the renewal is also constantly being strengthened, make increasing ant when carrying out resource allocation, consider this node, and then construct more excellent allotment scheme by the continuous accumulation of pheromones with higher probability.

When algorithmic statement is all selected same allotment scheme to all ants, should the allotment scheme be optimum allotment scheme then.Certainly, not every resource allocation problem can both obtain optimum allotment scheme, perhaps in the limited time, can't obtain optimum allotment scheme, therefore need to set suitable end condition and come finishing iteration, and more excellent the separating that obtains also has positive effect on the direct communications Resource allocation and smoothing.

End condition can be restrained for ant group algorithm, it is pre-conditioned also can be that evaluation result meets, or iterations reaches preset times, end condition also can be that ant group algorithm convergence and iterations reach combinations two or more in the preset times, certainly end condition is not limited to these three kinds, and those skilled in the art can select other conditions of termination of iterations as required.

In addition, the present invention carries out resource allocation except adopting an ant group, can also realize the allotment of the communication resource by plural ant group, promptly set up many group ant group algorithm node diagrams for the resource allocation problem in the communication system, divide other to communicate the renewal of Resource allocation and smoothing and pheromones with parallel mode, in the process of algorithm operation, can be between the ant group exchange of information in the hope of finding separating of problem quickly.

As shown in Figure 3, be the schematic flow sheet of another embodiment of communication resource allocating method of the present invention.Compare with a last embodiment, the present embodiment position that adaptation step and pheromones updating steps are provided with in iterative process is different, and this resource allocation flow process specifically comprises:

Step 201, for setting up the ant group algorithm node diagram in each sub-district in the communication system, and the pheromones and the heuristic information in definition and each node of initialization and path;

The optimization aim of the communication system that step 202, basis are preset is determined the evaluation function of communication system;

Step 203, judging whether to reach end condition, is execution in step 204 then, otherwise execution in step 205;

Step 204, stop allocation process, and communicate the actual allotment of resource and end operation according to the allotment scheme that obtains;

Step 205, judging whether all sub-districts all finish the allotment of the communication resource, is execution in step 206 then, otherwise execution in step 207;

Step 206, calculate evaluation result, and carry out the renewal of global information element, and return the judgement whether step 203 reaches end condition by described evaluation function;

Step 207, current ant are that the communication resource to be allocated is selected the user, promptly to the parallel allotment of the user in each sub-district communication resource to be allocated, can also carry out local pheromones this moment and upgrade.

Present embodiment has provided another kind of realization flow of the present invention, can utilize the probabilistic search of ant colony optimization algorithm to obtain the separating of better quality of optimization problem fast equally.

The optimization aim that is directed to several communication systems below provides the example of several communication resource allocatings.

Example one

Consider that the multi-user OFDM downlink resource distributes the rate adaptation problem, the throughput of maximization system is promptly satisfying the distribution of carrying out adaptive subcarrier, bit and power under the constraints under the limited situation of gross power.Supposing the system can obtain each user's complete instantaneous channel information, and there is N Resource Block (RB) that consists of a plurality of sub-carriers every sub-district, and has K user, and for reducing the complexity of resource allocation, we suppose each RB mean allocation power.For satisfying user's error rate requirement, system adopts adaptive modulating-coding technology, definition r _{N, k}The transmission rate of expression user k on Resource Block RB-n.System selects suitable modulation coding mode according to the channel quality information of user feedback for the user.Mainly consider non-real-time service in this example.

So that one of model of ant group optimization---ACS is reference, and it is as follows that present embodiment comprises step:

1) initialization:

The pheromones initial value of each node is made as τ ₀, because for the first time in the iteration pheromones value of each node identical, have only heuristic information that ant is made up and separate the performance directive function.Therefore if expect the initial solution that quality is pretty good, the definition of heuristic information is very important.The definition heuristic information is as follows, and the transmission rate of user on this RB is big more, easy more this RB that is assigned to of user.It make ant algorithm tend at the very start construct separate.

${\mathrm{\η}}_{n,k}=\frac{r_{n,k}}{\underset{l\∈N_n^m}{\mathrm{\Σ}}{r}_{n,l}}---\left(2\right)$

Wherein,

Expression is positioned at the ant m optional user set of RB-n, and this example is mainly considered non-real-time service, so the selectable user of each ant set is gathered K for whole users.If real time business, the user who then satisfies the service rate requirement no longer is included in matrix

In, ant no longer is its Resources allocation in this iteration.

2) each ant begins to travel through each RB from first RB, is the user that current RB selects distribution according to formula (3).

It is q that ant is selected the probability of current possible optimum move mode ₀, q ₀Satisfy 0≤q ₀≤ 1, ant is with 1-q simultaneously ₀Probability have skewed popularity ground to explore each paths.Be in the n row if ant m is current, promptly will distribute RB-n, then it selects user k, and promptly user k is assigned to the probability of RB-n and is

$\left\{\begin{array}{ccc}k={\arg \max }_{l\∈N_n^m}\left\{{\mathrm{\τ}}_{\mathrm{nl}}{\left[{\mathrm{\η}}_{\mathrm{nl}}\right]}^{\mathrm{\β}}\right\},& \mathrm{ifq}\≤q_0& (3-a)\\ p_{n,k}=\frac{{\left[{\mathrm{\τ}}_{n,k}\right]}^{\mathrm{\α}}{\left[{\mathrm{\η}}_{n,k}\right]}^{\mathrm{\β}}}{\underset{l\∈N_n^m}{\mathrm{\Σ}}{\left[{\mathrm{\τ}}_{n,l}\right]}^{\mathrm{\α}}{\left[{\mathrm{\η}}_{n,l}\right]}^{\mathrm{\β}}}& \mathrm{else}& (3-b)\end{array}\right.---\left(3\right)$

Definition q is for being evenly distributed on a stochastic variable in the interval [0,1], if random number q≤q ₀, then ant m distributes to RB-n and satisfies the user k that formula (3-a) requires; Otherwise user k is assigned to the Probability p of RB-n _{N, k}(3-b) provides by formula.

3) when each ant be that current RB has selected a user, for example for after RB-n selected user k, promptly to pheromones τ _{N, k}Carry out the plain renewal of following local message:

τ _n，k←(1-ξ)τ _n,k+ξτ ₀ (4)

ξ represents the plain evaporation rate of local message, satisfies 0＜ξ＜1, τ ₀Be the pheromones initial value.Adopt that local message is plain to be upgraded, make the pheromones dynamic change, increase the exploration of ant.

4) traveled through all RB when each ant, promptly constructed after complete the separating, according to formula (5) node has been carried out promptly that global information is plain to be upgraded, pheromones evaporation and release movement are only carried out on the point of optimal path so far, make search efficient more direct.Definition matrix Ψ in the assigning process _BestStore the distribution condition of optimal resource allocation so far,, calculate optimum so far resource allocation value and write down distribution condition according to evaluation function, if Ψ whenever ant colony is finished iteration one time _BestMiddle RB-n distributes to user k, then pheromones τ _{N, k}Carry out following renewal:

${\mathrm{\τ}}_{n,k}\←(1-\mathrm{\ρ}){\mathrm{\τ}}_{n,k}+\mathrm{\ρ\Δ}{\mathrm{\τ}}_{n,k}^{\mathrm{bs}}---\left(5\right)$

Parameter ρ represents the plain evaporation rate of global information, satisfies 0＜ρ＜1,

Relevant with pheromones initial value, optimal solution.

Wherein evaluation function is defined as follows, and the throughput that the path that ant makes up reaches is big more, and pheromones discharges many more, and the probability that node is selected by other ants strengthens.

$f=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{c}_{n,k}{r}_{n,k}---\left(6\right)$

5) carry out next iteration, stagnate until reaching default iterations or algorithm

The embodiment initial value is as follows:

τ ₀Initial value for pheromones; α, the relative influence power of β control information element and heuristic information.ρ, ξ are the speed of pheromones evaporation, and evaporation can be avoided the unlimited accumulation of pheromones.q ₀Can regulate algorithm to the exploration degree of new route and the relative dynamics that prior information is utilized.

Example two

Consider that the multi-user OFDM downlink resource distributes the rate adaptation problem, take into account the throughput of system and the fairness between the user, heuristic information and evaluation function can be defined as follows, to reach different optimization aim.

Heuristic information: use logarithmic function as heuristic information, thereby the ant build path is provided the fairness that instructs between the assurance user.Logarithmic function its dependent variable y and independent variable x are non-linear relation, and the value of Δ y is not only closed with Δ x, and is also relevant with x, and when the x value was big, the Δ y that identical Δ x obtains was littler than the Δ y that x value hour obtains.Δ y represents the increment of y.

Therefore when the existing certain transmission rate of user, the transmission rate that new allocate resource obtains diminishes to its effectiveness, thus make this resource be more prone to allocate less or do not have the user of speed to speed, thereby the fairness between the assurance user.

${\mathrm{\η}}_{n,k}=\frac{\log (1+r_{n,k}+R_{n,k})-\log (1+R_{n,k})}{{\mathrm{\Σ}}_{l\∈N_n^m}(\log (1+r_{n,l}+R_{n,l})-\log (1+R_{n,l}))}---\left(7\right)$

R wherein _{N, k}Expression user's current transmission rate.

Evaluation function: guarantee fairness thereby logarithmic function acts on pheromones release as evaluation function.

$f=\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\log (1+c_{n,k}{r}_{n,k})---\left(8\right)$

Approximate with example one on concrete steps, just repeated no more here.

Example three

Consider that the multi-user OFDM downlink resource distributes the rate adaptation problem, take into account the throughput of system and the fairness between the user, wherein heuristic information definition, evaluation function, the embodiment flow process can be identical with embodiment one, just repeated no more here.

Consideration reaches the Different Optimization target of system by the different parameters of regulating deployment algorithm.Ant is selected the probability q of current possible optimum move mode ₀Can regulate the relative dynamics that original prior information and neighborhood information are explored as a parameter.As q ₀Be taken as bigger value, then ant is all selected optimum move mode with bigger probability in each iteration, be partial to utilize the original prior information of ant build path in early stage to make up the allotment scheme more, the original prior information of formula (3-a) is that a certainty is selected, and the definition of its heuristic information and evaluation function is all useful to throughput of system.Therefore its user fairness of scheme of ant structure is relatively poor.If q ₀Be taken as less value, then ant all is partial to utilize formula (3-b) to make up the allotment scheme in each iteration more, the neighborhood search of formula (3-b) is a probabilistic search, when being done to improve, throughput of system strengthened exploration to neighborhood, allotment scheme to other provides certain possibility, thereby aspect user fairness some improvement can be arranged.Therefore regulate q ₀Can reach trade off different of throughput of system with user fairness.

One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can be finished by the relevant hardware of program command, aforesaid program can be stored in the computer read/write memory medium, this program is carried out the step that comprises said method embodiment when carrying out; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CD.

Should be noted that at last: above embodiment is only in order to illustrate that technical scheme of the present invention is not intended to limit; Although with reference to preferred embodiment the present invention is had been described in detail, those of ordinary skill in the field are to be understood that: still can make amendment or the part technical characterictic is equal to replacement the specific embodiment of the present invention; And not breaking away from the spirit of technical solution of the present invention, it all should be encompassed in the middle of the technical scheme scope that the present invention asks for protection.

Claims (12)

1. communication resource allocating method comprises:

For setting up the ant group algorithm node diagram in each sub-district in the communication system, and the pheromones and the heuristic information in definition and each node of initialization and path;

Determine the evaluation function of communication system according to the optimization aim of default communication system;

In iteration each time, respectively to the parallel allotment of the user in each sub-district communication resource;

All finish the allotment of the communication resource when all sub-districts after, calculate evaluation result by described evaluation function, and carry out the renewal of global information element;

If reach end condition, then stop allocation process, and communicate the actual allotment of resource, otherwise carry out next iteration according to the allotment scheme that obtains.

2. communication resource allocating method according to claim 1 wherein respectively to the parallel allotment of the user in each sub-district communication resource time, also carries out the renewal of local message element simultaneously to each node.

3. communication resource allocating method according to claim 1, each node that is updated to the scheme of evaluation result optimum in all previous iteration in the communication system of wherein said global information element upgrades, and perhaps each node for the scheme of evaluation result optimum in this iteration upgrades.

4. communication resource allocating method according to claim 1, each node that is updated to the more excellent a plurality of schemes of evaluation result in all previous iteration in the communication system of wherein said global information element upgrades, and perhaps each node for the more excellent a plurality of schemes of evaluation result in this iteration upgrades.

5. communication resource allocating method according to claim 1, wherein said heuristic information are one or more combinations and the function of user in channel quality information, momentary rate and the Mean Speed of allocate resource unit.

6. communication resource allocating method according to claim 5, wherein said channel quality information comprise one or more the combination in signal to noise ratio, signal interference ratio, Signal to Interference plus Noise Ratio, channel gain interference ratio, channel condition information, the data rate that can transmit, the error rate, Block Error Rate and the interference strength.

7. that communication resource allocating method according to claim 1, wherein said end condition are that described evaluation result meets is pre-conditioned, ant group algorithm convergence and iterations reach one or more the combination in the preset times.

8. communication resource allocating method according to claim 1, wherein when the optimization aim of default communication system be under the limited situation of gross power during maximize throughput, it is big more that the evaluation function of communication system is defined as characterizing throughput, and pheromones discharges many more function expressions.

9. communication resource allocating method according to claim 1, wherein when the optimization aim of default communication system is the fairness of taking into account between throughput and user, the evaluation function of communication system adopts logarithmic function to act on the function expression that pheromones discharges, perhaps heuristic information adopts logarithmic function to act on the function expression of communication resource allocating process, perhaps regulates the allotment probability respectively to the parallel allotment of the user in each sub-district communication resource time.

10. communication resource allocating method according to claim 1, the ant group optimization model of wherein allocating the communication resource and being adopted is the ant system, the ant system that refines, based on the ant system of arranging, minimax ant system, ant group system or approximate uncertainty tree search system.

11., upgrade the selectable user set according to the difference of traffic performance and user's request according to the process of claim 1 wherein the time to the parallel allotment of the user in each sub-district communication resource.

12. according to the process of claim 1 wherein that described communication resource allocating method adopts a plurality of parallel ant groups to carry out computing, in the calculating process between the ant group exchange of information in the hope of finding separating of problem quickly.

Images