CN106304170A - A kind of uplink resource optimization method of wireless mesh network - Google Patents

Abstract

The present invention is directed to wireless mesh network uplink resource and be difficult to efficient planning problem, by setting up uplink resource Optimized model, engineering equivalence optimization process and carrying out Business Stream measurement optimization, it is achieved the uplink resource optimization planning of wireless mesh network.

Description

A kind of uplink resource optimization method of wireless mesh network

Technical field

The present invention relates to communication network field, particularly relate to queueing theory, and optimum theory.

Background technology

The a moment being born from network, along with the development of network technology, the either experimental network of early stage, education network, office Territory net, wide area network the most now, network has become as the indispensable part of human social development.In recent years, the mankind Under the promotion of demand and research, radio network technique develops rapidly.Wireless Mesh netword is a kind of primary study therein Wireless networking mode.Wireless Mesh netword advantage in terms of merging heterogeneous network and improving wireless bandwidth resource utilization rate, makes It becomes the important component part of next generation wireless communication network.Wireless Mesh netword based on Ad Hoc network, be self-organizing, Self-configuring and the cancellated network from maintenance.Wireless Mesh netword is that one is multi-multipoint, has self-organizing and spontaneous recovery The broadband wireless network system of feature.Compared with Ad Hoc network, the mobility of wireless Mesh netword node and the weight to energy consumption Visual range degree is relatively low, and its network topology is more stable.Node is generally of multi-transceiver, can send simultaneously and receive data.Nothing In line Mesh network, node typically has a two types: routing node and user node, its interior joint can be simultaneously user node and Routing node.Wireless Mesh netword can be that part distributed network, i.e. network have converging information Centroid (such as route joint Point), it is also possible to be the communication between complete distributed network, i.e. network node be equality, each node can serve as Routing node.Wireless Mesh netword can be applied to military communication network, education network, public place of entertainment nerve of a covering, rescue and relief work are faced Time communication network, rapid deployment application network, communication recovery casual network, electric power station monitors network and should not destroy outside landform The wireless network etc. of the occasion of looks, the structure of wireless mesh network is as shown in Figure 1.

The Routing Protocol of wireless Mesh netword generally comprises generation route, chooses route and safeguards three processes of route, also By generating routing procedure and routing procedure can be chosen combine, referred to as find routing procedure.

Process according to network topological information and link quality state information generation route set is referred to as generation and is routed through Journey.Generating in routing procedure, network topological information and link quality state information are with the link information of node as main contents.Choosing Taking routing procedure is according to customer service demand information or other information, in generating the route set that routing procedure produces, choosing Select the process of most suitable route.

Safeguard that path process refers to the mistake safeguarding network topological information, link quality state information and selected path Journey.Network topological information and link quality state information can be along with the movement of wireless Mesh netword node or the changes of wireless channel And change, produced route also can change therewith.Route maintenance procedure cycle or irregularly send message, safeguard topology letter Breath, link quality state information and routing table.

Therefore, for realizing efficient path transmission ability, it is necessary to set up meet wireless mesh network transmission demand Uplink resource optimization mechanism.

Summary of the invention

The technical problem to be solved is: by setting up uplink resource Optimized model and engineering equivalence optimization Process, and carry out Business Stream measurement optimization, it is achieved the resource optimization management of wireless mesh network.

The present invention solves that the technical scheme that above-mentioned technical problem is used comprises the following steps, as shown in Figure 2:

A, set up uplink resource Optimized model；

B, engineering equivalence optimization process；

C, carry out Business Stream measure optimize.

In described step A, particularly as follows:

$\begin{array}{cc}\max & E\left\{\underset{b=1}{\overset{B}{\mathrm{\Σ}}}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2(1+\frac{Q_{k,n}^b\·x_{k,n}^b}{\underset{w=1,w\≠b}{\overset{B}{\mathrm{\Σ}}}\underset{v=1,v\≠k}{\overset{K}{\mathrm{\Σ}}}{Q}_{v,n}^w\·x_{v,n}^w+\left|{\mathrm{\σ}}_{k,n}^b\right|})\right\}\end{array}$

$\begin{array}{ccc}s.t.& P\{\underset{n=1}{\overset{N}{\Σ}}{p}_{k,n}^b\·x_{k,n}^b\≤P_k^b\}\≥(1-\mathrm{\α}),& \∀k\∈K,b\∈B\end{array}$

$\begin{array}{cc}\underset{k=1}{\overset{K}{\Σ}}{x}_{k,n}^b\≤1,& \∀n\∈N,b\∈B\end{array}$

$\begin{array}{cc}\underset{b=1}{\overset{n}{\Σ}}\underset{n=1}{\overset{N}{\Σ}}{x}_{k,n}^b\≥1,& \∀k\∈K\end{array}$

$\begin{array}{cc}{x}_{k,n}^b\∈\{0,1\},& \∀k,n,b\end{array}$

Subcarrier mark during wherein n is each base station, N is number of sub carrier wave, and k is the ID in each base station, K For number of users, b is Base Station Identification, and B is number of base stations,For decision variable, if subcarrier n being distributed to user in the b of base station K is thenOtherwise For distributing to the white Gaussian noise variance of the subcarrier n of user k in the b of base station,For The maximum available power of user k in the b of base station,For input power matrix,For in b sub-carriers n of base station The channel gain of user k, M is positive integer,For distributing to the power of the subcarrier n of user k in the b of base station, E () is upper Line link utilizes usefulness mean value function, P () to be probability function, w and b has same physical meaning, v and k has same physical Meaning, α is probability coefficent.

In described step B, particularly as follows:

$\begin{array}{ccc}st:& {\mathrm{\φ}}_{k,n}^b\≤a_m{t}_{k,n}^b+b_m,& \∀m,k,n,b\end{array}$

w≠b v≠k

$\begin{array}{cc}{x}_{k,n}^b\∈\{0,1\},& \∀k,n,b\end{array}$

$\begin{array}{cc}{\mathrm{\φ}}_{k,n}^b\≥0,& \∀k,n,b\end{array}.$

Wherein a_mWeight coefficient, b is used for link_mFor the estimation difference of link capacity,For user k quilt in the b of base station The uplink capacity used during distribution subcarrier n, N, k, w, b areLinear variable,For Uplink capacity when user k is allocated subcarrier n in the b of base station.

In described step B, through engineering approaches is processed the optimization solution that obtains and obtains by setting up uplink resource Optimized model The optimization solution obtained contrasts, and adaptive correction uplink resource Optimized model, particularly as follows: based on application service g, by work Journeyization is processed the solution that optimizes obtained and contrasts with the optimization solution obtained by uplink resource Optimized model, if the two is excellent Dissolve set completely the same, then using this optimization solution as the optimization solution of application-oriented service g；If the optimization solution set of the two differs Cause, then through engineering approaches is processed the optimization solution final optimization pass solution set as application service g of acquisition, and revises uplink resource The adjustable parameter of the wireless mesh network in Optimized model makes it obtain the optimization solution set that through engineering approaches processes, and by revised Uplink resource Optimized model is as the uplink resource Optimized model of next transmitting scene g+1, wireless mesh network In the application service collection with related application attribute be combined into G={1,2 ..., g, g+1}, the g+1 that will obtain in time U Optimize solution set and carry out statistical average process, and this statistical average is processed g+1 the optimization disaggregation optimizing solution set and obtaining Close and revise uplink resource Optimized model, and using this model as other related application in wireless mesh network system next time The priori uplink resource Optimized model of the application service set G+1 of attribute.

In described step C, particularly as follows: carry out from three planes, respectively network plane, sampling plane and network are put down Face, network plane comprise packet select rule and sample data bag select rule, sampling plane comprise Business Stream polymerized unit, Sampling arranges unit, sampling frame, and management plane comprises to be set up measuring point selection and information model, and by the request of measuring and Constraints Compromise programming measures point selection and information model is set up；Sampling frame comprise selection strategy, select trigger element and Granularity discrimination unit, selection strategy comprises systematicness and selects unit, randomness to select unit and adaptively selected unit, selects to touch Bill unit comprises counting driver element, event-driven unit and time driver element, and granularity discrimination unit comprises Business Stream level district Subdivision and packet-level discrimination unit, as shown in Figure 3.

Accompanying drawing explanation

The structural representation of Fig. 1 wireless mesh network

The uplink resource Optimizing Flow schematic diagram of Fig. 2 wireless mesh network

Fig. 3 Business Stream is measured and is optimized schematic diagram

Detailed description of the invention

For reaching above-mentioned purpose, technical scheme is as follows:

The first step, sets up uplink resource Optimized model, particularly as follows:

$\begin{array}{cc}\max & E\left\{\underset{b=1}{\overset{B}{\mathrm{\Σ}}}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\log }_2(1+\frac{Q_{k,n}^b\·x_{k,n}^b}{\underset{w=1,w\≠b}{\overset{B}{\mathrm{\Σ}}}\underset{v=1,v\≠k}{\overset{K}{\mathrm{\Σ}}}{Q}_{v,n}^w\·x_{v,n}^w+\left|{\mathrm{\σ}}_{k,n}^b\right|})\right\}\end{array}$

$\begin{array}{ccc}s.t.& P\{\underset{n=1}{\overset{N}{\Σ}}{p}_{k,n}^b\·x_{k,n}^b\≤P_k^b\}\≥(1-\mathrm{\α}),& \∀k\∈K,b\∈B\end{array}$

$\begin{array}{cc}\underset{k=1}{\overset{K}{\Σ}}{x}_{k,n}^b\≤1,& \∀n\∈N,b\∈B\end{array}$

$\begin{array}{cc}\underset{b=1}{\overset{n}{\Σ}}\underset{n=1}{\overset{N}{\Σ}}{x}_{k,n}^b\≥1,& \∀k\∈K\end{array}$

$\begin{array}{cc}{x}_{k,n}^b\∈\{0,1\},& \∀k,n,b\end{array}$

Subcarrier mark during wherein n is each base station, N is number of sub carrier wave, and k is the ID in each base station, K For number of users, b is Base Station Identification, and B is number of base stations,For decision variable, if subcarrier n being distributed to user in the b of base station K is thenOtherwise For distributing to the white Gaussian noise variance of the subcarrier n of user k in the b of base station,For The maximum available power of user k in the b of base station,For input power matrix,For in b sub-carriers n of base station The channel gain of user k, M is positive integer,For distributing to the power of the subcarrier n of user k in the b of base station, E () is upper Line link utilizes usefulness mean value function, P () to be probability function, w and b has same physical meaning, v and k has same physical Meaning, α is probability coefficent.

Second step, carries out engineering equivalence optimization and processes, particularly as follows:

$\begin{array}{ccc}st:& {\mathrm{\φ}}_{k,n}^b\≤a_m{t}_{k,n}^b+b_m,& \∀m,k,n,b\end{array}$

w≠b v≠k

$\begin{array}{cc}{x}_{k,n}^b\∈\{0,1\},& \∀k,n,b\end{array}$

$\begin{array}{cc}{\mathrm{\φ}}_{k,n}^b\≥0,& \∀k,n,b\end{array}.$

Wherein a_mWeight coefficient, b is used for link_mFor the estimation difference of link capacity,For user k quilt in the b of base station The uplink capacity used during distribution subcarrier n, N, k, w, b areLinearization,For Uplink capacity when user k is allocated subcarrier n in the b of base station.

3rd step, by through engineering approaches process obtain optimize solve with by set up uplink resource Optimized model acquisition excellent Dissolve and contrast, and adaptive correction uplink resource Optimized model, particularly as follows: based on application service g, at through engineering approaches The optimization that reason obtains is solved and contrasts with the optimization solution obtained by uplink resource Optimized model, if the optimization disaggregation of the two Close completely the same, then using this optimization solution as the optimization solution of application-oriented service g；If the optimization solution set of the two is inconsistent, then Through engineering approaches is processed the optimization solution final optimization pass solution set as application service g of acquisition, and revises uplink resource optimization The adjustable parameter of the wireless mesh network in model makes it obtain the optimization solution set that through engineering approaches processes, and by revised up Link circuit resource Optimized model is as the uplink resource Optimized model of next transmitting scene g+1, in wireless mesh network The application service collection with related application attribute is combined into G={1, and 2 ..., g, g+1}, g+1 the optimization that will obtain in time U Solution set carries out statistical average process, and g+1 the optimization solution set that this statistical average processes optimization solution set and acquisition is repaiied Positive uplink resource Optimized model, and using this model as other related application attribute in wireless mesh network system next time The priori uplink resource Optimized model of application service set G+1.

4th step, carries out Business Stream and measures optimization, particularly as follows: carry out from three planes, respectively network plane, take out Sample plane and network plane, network plane comprises packet and selects rule and sample data bag to select rule, sampling plane to comprise Business Stream polymerized unit, sampling arrange unit, sampling frame, and management plane comprises measuring point selection and information model foundation, And set up by the request of measuring and constraints Compromise programming measurement point selection and information model；Sampling frame comprises selection plan Slightly, selecting trigger element and granularity discrimination unit, selection strategy comprises systematicness and selects unit, randomness to select unit and adaptive Should select unit, select trigger element to comprise counting driver element, event-driven unit and time driver element, granularity is distinguished single Unit comprises Business Stream level discrimination unit and packet-level discrimination unit.

The present invention proposes the uplink resource optimization method of a kind of wireless mesh network, provides by setting up up-link Source optimization model, engineering equivalence optimization process and carry out Business Stream measures optimization, it is achieved the up-link money of wireless mesh network Source optimization is planned.

Claims (5)

1. a uplink resource optimization method for wireless mesh network, by setting up uplink resource Optimized model, work Journey equivalence optimization processes and carries out Business Stream measures optimization, it is achieved the uplink resource optimization planning of wireless mesh network, bag Include following steps:

A, set up uplink resource Optimized model；

B, engineering equivalence optimization process；

C, carry out Business Stream measure optimize.

Method the most according to claim 1, is characterized in that for described step A: particularly as follows:

$\begin{array}{cc}\max & E\left\{\underset{b=1}{\overset{B}{\Σ}}\underset{k=1}{\overset{K}{\Σ}}\underset{n=1}{\overset{N}{\Σ}}{\log }_2(1+\frac{Q_{k,n}^b\·x_{k,n}^b}{\underset{w=1,w\≠b}{\overset{B}{\Σ}}\underset{v=1,v\≠k}{\overset{K}{\Σ}}{Q}_{v,n}^w\·x_{v,n}^w+\left|{\mathrm{\σ}}_{k,n}^b\right|})\right\}\end{array}$

$\begin{array}{cc}s.t.& P\{\underset{n=1}{\overset{N}{\Σ}}{p}_{k,n}^b\·x_{k,n}^b\≤P_k^b\}\≥(1-\mathrm{\α}),\∀k\∈K,b\∈B\end{array}$

$\underset{k=1}{\overset{K}{\Σ}}{x}_{k,n}^b\≤1,\∀n\∈N,b\∈B$

$\underset{b=1}{\overset{n}{\Σ}}\underset{n=1}{\overset{N}{\Σ}}{x}_{k,n}^b\≥1,\∀k\∈K$

$x_{k,n}^b\∈\{0,1\},\∀k,n,b$

Subcarrier mark during wherein n is each base station, N is number of sub carrier wave, and k is the ID in each base station, and K is for using Amount, b is Base Station Identification, and B is number of base stations,For decision variable, if subcarrier n being distributed to user k in the b of base station,Otherwise For distributing to the white Gaussian noise variance of the subcarrier n of user k in the b of base station,For base station The maximum available power of user k in b,For input power matrix,For user in b sub-carriers n of base station The channel gain of k, M is positive integer,For distributing to the power of the subcarrier n of user k in the b of base station, E () is uplink Road utilizes usefulness mean value function, P () to be probability function, w and b has same physical meaning, v and k has same physical meaning, α is probability coefficent.

Method the most according to claim 1, is characterized in that for described step B: particularly as follows:

w≠b v≠k

$x_{k,n}^b\∈\{0,1\},\∀k,n,b$

${\mathrm{\φ}}_{k,n}^b\≥0,\∀k,n,b.$

Wherein a_mWeight coefficient, b is used for link_mFor the estimation difference of link capacity,It is allocated for user k in the b of base station The uplink capacity used during subcarrier n,ForLinear variable,For in base station Uplink capacity when user k is allocated subcarrier n in b.

Method the most according to claim 1, is characterized in that for described step B: by through engineering approaches process obtain optimization solution with The optimization solution obtained by setting up uplink resource Optimized model contrasts, and adaptive correction uplink resource optimizes Model, particularly as follows: based on application service g, process the optimization solution obtained and by uplink resource Optimized model by through engineering approaches The optimization solution obtained contrasts, if the optimization solution set of the two is completely the same, then using this optimization solution as application-oriented service g Optimization solution；If the optimization solution set of the two is inconsistent, then through engineering approaches is processed the optimization solution obtained as application service g Optimize eventually and solve set, and the adjustable parameter revising the wireless mesh network in uplink resource Optimized model makes it obtain engineering The optimization solution set that change processes, and using upper as next application service g+1 of revised uplink resource Optimized model Downlink resources Optimized model, the application service collection with related application attribute in wireless mesh network is combined into G={1, 2 ..., g, g+1}, g+1 obtained in time U optimization is solved set and carries out statistical average process, and by this statistical average Process optimize solve set with obtain g+1 optimization solve set correction uplink resource Optimized model, and using this model as In wireless mesh network system, the priori uplink resource of the application service set G+1 of other related application attribute is excellent next time Change model.

Method the most according to claim 1, is characterized in that for described step C: particularly as follows: carry out from three planes, Being respectively network plane, sampling plane and network plane, network plane comprises packet and selects rule and sample data bag to select Rule, sampling plane comprises Business Stream polymerized unit, sampling arranges unit, sampling frame, and management plane comprises and clicks measurement Select and information model foundation, and set up by the request of measuring and constraints Compromise programming measurement point selection and information model；Take out Sample rack structure comprises selection strategy, selects trigger element and granularity discrimination unit, and selection strategy comprises systematicness and selects unit, random Sexual behavior mode unit and adaptively selected unit, select trigger element to comprise counting driver element, event-driven unit and time drive Moving cell, granularity discrimination unit comprises Business Stream level discrimination unit and packet-level discrimination unit.