CN106507388A - A kind of wireless mesh network link optimizing method based on dynamic application scene - Google Patents

Abstract

The present invention is difficult to cooperate with optimum management problem for the wireless mesh network energy under dynamic application scene, by setting up wireless mesh network energy optimizing model and carrying out engineering equivalent process, and the optimization solution for obtaining through engineering approaches process is contrasted with the optimization solution obtained by wireless mesh network energy optimizing model, and adaptive correction wireless mesh network energy optimizing model, realize the energy-optimised utilization of wireless mesh network.

Description

A kind of wireless mesh network link optimizing method based on dynamic application scene

Technical field

The present invention relates to intelligent grid field, more particularly to communication network, and optimum theory.

Background technology

Wireless Mesh netword is a kind of emerging radio technology emerged for specific business application, is the Internet Service provider (ISP) and other-end user, set up strong, reliable Access mode and provide selection.With traditional Wireless network is different, and wireless Mesh netword is a dynamic self-organizing, each node in self-configuring network, i.e. Mesh network Automatically set up and maintain the connection of network.Various advanced radio-frequency techniques are employed in wireless Mesh netword so that Wireless Mesh The capacity of network can be increased dramatically, and wireless mesh network structure is as shown in Figure 1.

Due to the static characteristic of Mesh routers in wireless Mesh netword so that one strong, effective link pipe of design Reason mechanism is possibly realized, and already present single path and multipath flow protocol are for single source and destination in wireless network On the premise of node pair propose, wireless Mesh netword is different, it be using multiple gateways node with Internet is coupled together.In the flow towards Internet, each node can be selected independently the gateway node of optimum.Institute The reason for traditional wireless network is different from wireless Mesh netword with balancing link load technology has lower two：

(1) node static；Node is all in order at for the load-balancing mechanism for having pointed out next in Wireless Ad hoc network Dynamic is considering.And the load-balancing mechanism for being applied to wireless sensor network is then to consider node energy and calculate energy The restriction of power.Therefore, these load-balancing mechanisms for having existed are unsuitable for wireless Mesh netword, wireless Mesh netword to be Design a special load-balancing algorithm to realize network optimal performance.

(2) multiple available gateways:According to this characteristic, load balancing can be made to carry out between multiple gateways, that is, observe net The congestion situation of pass, if it find that gateway congestion, then reselect other gateways in non-congested state, by such side Formula, realizes the load balancing of network.Autonomous Adhoc networks are different from, wireless Mesh netword is mainly wired for providing Basic network the Internet connective and design.Therefore, in order to improve the utilization rate of backbone bandwidth, gateway with Internet is connected.In wireless Mesh netword, the unbalanced of offered load can bring down several subject matters:Network capacity Decline, gateway node overload, near network center area jamming, easily form bottleneck node and bottleneck node overload occur.Cause It is the side for providing a quick, inexpensive access Internet for fixing user and mobile subscriber for wireless Mesh netword Formula, lacks effective link optimizing strategy.

Therefore, in wireless Mesh netword, it is the service quality for improving QoS of customer and network, carries out link optimizing Design is very important, and so as to extend the life cycle of network, increases network throughput.

Content of the invention

The technical problem to be solved is：Equivalent by setting up wireless mesh network link optimizing model and engineering Process, and the optimization solution that through engineering approaches process is obtained and the optimization by setting up the acquisition of wireless mesh network link optimizing model Solution is contrasted, and adaptive correction wireless mesh network energy optimizing model, and set up through-put power and time delay joint excellent Change model, realize the adaptive adjustment capability of wireless mesh network link circuit resource.

The present invention is comprised the following steps by solving the technical scheme that above-mentioned technical problem is adopted, as shown in Figure 2：

A, set up wireless mesh network link optimizing model；

B, carry out engineering equivalent process；

C, by through engineering approaches process obtain optimization solution with by set up wireless mesh network link optimizing model obtain excellent Neutralizing is contrasted, and adaptive correction wireless mesh network link optimizing model, and sets up through-put power and time delay joint Optimized model.

In step A, specially：Optimized model is specially：

x_ij,z_j,y_jl,w_jN∈{0,1}

Wherein a_ijFor decision variable, if j-th BS covers i-th TP, a_ij=1, on the contrary then a_ij=0, b_jlFor decision-making Variable, if can realize data transfer, b by wireless link between j-th BS and l-th BS_jl=1, on the contrary then b_jl=0, d_i For the service request of i-th TP, x_ijFor decision variable, x if i-th TP to be distributed to j-th BS_ij=1, on the contrary then x_ij= 0, z_j>1 for j-th CS weight coefficient, w_jNFor decision variable, w if j-th BS is comprising mesh gateways_jN=1, on the contrary then w_jN =0, y_jlFor decision variable, y if it there is wireless link between j-th BS and l-th BS_jl=1, on the contrary then y_jl=0, f_jlFor Business Stream in wireless link (j, l), f_jNFor the Business Stream between the mesh gateways and backbone network in j-th BS, I is test node Set, T are time interval set, and S is the collection of base stations comprising mesh routers and mesh gateway devices,For mesh nets Close set,For covering the collection of base stations of i-th TP that there is reduction to cover receiving power request, h is covered with reduction for covering The Base Station Identification of i-th TP of lid receiving power request, c_jFor the installation cost of base station j, p_jFor the company between gateway and core net Meet cost, d_itIt is the business demand in i-th TP of time t, u_jlFor the link capacity between base station j and base station l, v_jFor base station j In distribute to the access capacity of TP, L_iFor covering the number of base stations of i-th TP, M is the backbone network incoming end towards mesh gateways Capacity, BS is base station, and TP is test node, l₁Higher limit for h.

In step B, it is to reduce room and time complexity, carries out engineering equivalent process, specially：

Wherein h_itFor decision variable, if sending service request d in i-th TP of moment t_itThen h_it=1, on the contrary then h_it=0, x_ijtFor decision variable, x if i-th TP to be distributed to j-th BS in moment t_ijt=1, on the contrary then x_ijt=0, z_jtBecome for decision-making Amount, z if base station j is in running order at moment t_jt=1, on the contrary then z_jt=0, f_jltIt is in moment t wireless link (j, l) Business Stream, f_j0tIt is the Business Stream between j-th base station of moment t and the 0th node, ξ_jThe power consumed by base station j ∈ S.

In step C, specially：In transmitting scene g, by the optimization solution of through engineering approaches process acquisition and by wireless The optimization solution that mesh network links Optimized model is obtained is contrasted, if the optimization solution set of the two is completely the same, this is excellent Neutralizing is used as the optimization solution in transmitting scene g；If the optimization solution set of the two is inconsistent, by the optimization of through engineering approaches process acquisition Solution is used as the final optimization pass solution set in transmitting scene g, and revises the wireless mesh in wireless mesh network link optimizing model The adjustable parameter of network makes which obtain the optimization solution set that through engineering approaches are processed, and by revised wireless mesh network link optimizing Wireless mesh network link optimizing model of the model as next transmitting scene g+1；The business transmission field of wireless mesh network The scape time is G={ 1,2 ..., g, g+1 }, and optimizing solution set by g+1 obtained in time G carries out statistical average process, And the g+1 optimization solution set correction wireless mesh network link optimizing mould that the statistical average is processed the set of optimization solution and obtained Type, and the priori wireless mesh network link optimizing mould that the model is transmitted as business in wireless mesh network system next time Type.

In step C, specially：Through-put power and time delay combined optimization model is set up, specially：

(s_r+θ(η_r)-s_q)(s_q+θ(η_q)-s_r)H(η_q·η_r) >=0, q ≠ r, q, r=0,1 ..., ω

η_q∈ W, q=0,1 ..., ω

θ(η_q)>0, q=0,1 ..., ω

Wherein θ (η_q) it is the node v when network adopts multicast mechanism_qTransmission business flow to the propagation delay time of its child node, η_q For v_qThrough-put power, I (η_q) it is by node v_qThe node set of interference, H (η_q,η_r) it is decision variable, if { π_e∈C_q}∪{π_e ∈I(η_q) and { π_e∈C_r}∪{π_e∈I(η_r) while setting up, then H (η_q,η_r)=1, otherwise then H (η_q,η_r)=0, C_rFor v_r's Child node set, π_rFor v_rChild node, η_qFor v_qThrough-put power, s_qFor v_qBusiness Stream send initial time, W be network in The transmit power set of node,For v_qChild node through-put power, ω be network in node number.

In step C, specially：By the optimization solution of through engineering approaches process acquisition and by setting up wireless mesh network link The optimization solution that Optimized model is obtained is contrasted, and adaptive correction through-put power and time delay combined optimization model.

Description of the drawings

Fig. 1 wireless mesh network structural representations

Wireless mesh network link optimizing schematic flow sheets of the Fig. 2 based on dynamic application scene

Specific embodiment

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

The first step, sets up wireless mesh network link optimizing model, specially：

x_ij,z_j,y_jl,w_jN∈{0,1}

Wherein a_ijFor decision variable, if j-th BS covers i-th TP, a_ij=1, on the contrary then a_ij=0, b_jlFor decision-making Variable, if can realize data transfer, b by wireless link between j-th BS and l-th BS_jl=1, on the contrary then b_jl=0, d_i For the service request of i-th TP, x_ijFor decision variable, x if i-th TP to be distributed to j-th BS_ij=1, on the contrary then x_ij= 0, z_j>1 for j-th CS weight coefficient, w_jNFor decision variable, w if j-th BS is comprising mesh gateways_jN=1, on the contrary then w_jN =0, y_jlFor decision variable, y if it there is wireless link between j-th BS and l-th BS_jl=1, on the contrary then y_jl=0, f_jlFor Business Stream in wireless link (j, l), f_jNFor the Business Stream between the mesh gateways and backbone network in j-th BS, I is test node Set, T are time interval set, and S is the collection of base stations comprising mesh routers and mesh gateway devices,For mesh nets Close set,For covering the collection of base stations of i-th TP that there is reduction to cover receiving power request, h is covered with reduction for covering The Base Station Identification of i-th TP of lid receiving power request, c_jFor the installation cost of base station j, p_jFor the company between gateway and core net Meet cost, d_itIt is the business demand in i-th TP of time t, u_jlFor the link capacity between base station j and base station l, v_jFor base station j In distribute to the access capacity of TP, L_iFor covering the number of base stations of i-th TP, M is the backbone network incoming end towards mesh gateways Capacity, BS is base station, and TP is test node, l₁Higher limit for h.

Second step, is to reduce room and time complexity, carries out engineering equivalent process, specially：

Wherein h_itFor decision variable, if sending service request d in i-th TP of moment t_itThen h_it=1, on the contrary then h_it=0, x_ijtFor decision variable, x if i-th TP to be distributed to j-th BS in moment t_ijt=1, on the contrary then x_ijt=0, z_jtBecome for decision-making Amount, z if base station j is in running order at moment t_jt=1, on the contrary then z_jt=0, f_jltIt is in moment t wireless link (j, l) Business Stream, f_j0tIt is the Business Stream between j-th base station of moment t and the 0th node, ξ_jThe power consumed by base station j ∈ S.

3rd step, by the optimization solution of through engineering approaches process acquisition and by setting up the acquisition of wireless mesh network link optimizing model Optimization solution contrasted, and adaptive correction wireless mesh network energy optimizing model and through-put power and time delay joint excellent Change model, in transmitting scene g, through engineering approaches are processed the optimization solution of acquisition and obtained by wireless mesh network link optimizing model Optimization solution contrasted, if the optimization solution set of the two is completely the same, using the optimization solution as excellent in transmitting scene g Neutralizing；If the optimization solution set of the two is inconsistent, the optimization solution that through engineering approaches process is obtained is used as final in transmitting scene g Optimize solution set, and revising the adjustable parameter of the wireless mesh network in wireless mesh network link optimizing model makes which obtain work The optimization solution set of journeyization process, and using revised wireless mesh network link optimizing model as next transmitting scene g+ 1 wireless mesh network link optimizing model；The business transmitting scene time of wireless mesh network be G=1,2 ..., g, g+ 1 }, optimizing solution set by g+1 obtained in time G carries out statistical average process, and the statistical average is processed optimization solution Set optimizes solution set with g+1 for obtaining and revises wireless mesh network link optimizing model, and using the model as next time The priori wireless mesh network link optimizing model of business transmission in wireless mesh network system.

4th step, sets up through-put power and time delay combined optimization model, specially：

(s_r+θ(η_r)-s_q)(s_q+θ(η_q)-s_r)H(η_q·η_r) >=0, q ≠ r, q, r=0,1 ..., ω

η_q∈ W, q=0,1 ..., ω

θ(η_q)>0, q=0,1 ..., ω

Wherein θ (η_q) it is the node v when network adopts multicast mechanism_qTransmission business flow to the propagation delay time of its child node, η_q For v_qThrough-put power, I (η_q) it is by node v_qThe node set of interference, H (η_q,η_r) it is decision variable, if { π_e∈C_q}∪{π_e ∈I(η_q) and { π_e∈C_r}∪{π_e∈I(η_r) while setting up, then H (η_q,η_r)=1, otherwise then H (η_q,η_r)=0, C_rFor v_r's Child node set, π_rFor v_rChild node, η_qFor v_qThrough-put power, s_qFor v_qBusiness Stream send initial time, W be network in The transmit power set of node,For v_qChild node through-put power, ω be network in node number.

5th step, by the optimization solution of through engineering approaches process acquisition and by setting up the acquisition of wireless mesh network link optimizing model Optimization solution contrasted, and adaptive correction through-put power and time delay combined optimization model.

The present invention proposes a kind of wireless mesh network link optimizing method based on dynamic application scene, by setting up nothing Line mesh network links Optimized model and engineering equivalent process, and by the optimization solution of through engineering approaches process acquisition and by setting up nothing The optimization solution that line mesh network links Optimized model is obtained is contrasted, and the energy-optimised mould of adaptive correction wireless mesh network Type and through-put power and time delay combined optimization model, realize the adaptive adjustment capability of wireless mesh network link circuit resource.

Claims (6)

1. a kind of wireless mesh network link optimizing method based on dynamic application scene, by setting up wireless mesh network link Optimized model and carry out engineering equivalent process, and the optimization solution that through engineering approaches process is obtained with pass through wireless mesh network link The optimization solution that Optimized model is obtained is contrasted, and adaptive correction wireless mesh network link optimizing model, is realized wireless The link optimizing of mesh networks is utilized, and is comprised the steps：

A, set up wireless mesh network link optimizing model；

B, carry out engineering equivalent process；

C, the optimization solution for obtaining through engineering approaches process are carried out with the optimization solution obtained by wireless mesh network link optimizing model Contrast, and adaptive correction wireless mesh network link optimizing model and through-put power and time delay combined optimization model.

2. method according to claim 1, for step A it is characterized in that：Optimized model is specially：

$\begin{array}{cc}min& \underset{j\∈S}{\Σ}{c}_j{z}_{jt}+p_j{w}_{jN}\end{array}$

$\begin{array}{ccc}s.t.& \underset{j\∈S}{\Σ}{x}_{ij}=1,& \∀i\∈I\end{array},$

$\begin{array}{cc}{x}_{ij}\≤z_j{a}_{ij},& \∀i\∈I,j\∈S\end{array},$

$\begin{array}{cc}\underset{l\∈S}{\Σ}(f_{lj}-f_{jl})+\underset{i\∈I}{\Σ}{d}_i{x}_{ij}=f_{jN},& \∀j\∈S\end{array},$

$\begin{array}{cc}{f}_{lj}-f_{jl}\≤u_{jl}{y}_{jl},& \∀j,l\∈S\end{array},$

$\begin{array}{cc}\underset{i\∈I}{\Σ}{d}_i{x}_{ij}\≤v_j,& \∀j\∈S\end{array},$

$\begin{array}{cc}{f}_{jN}\≤{\mathrm{Mw}}_{jN},& \∀j\∈S\end{array},$

$\begin{array}{ccc}{y}_{jl}\≤z_j,& y_{jl}\≤z_l,& \∀j,l\∈S\end{array},$

$\begin{array}{cc}{y}_{jl}\≤b_{jl},& \∀j,l\∈S\end{array},$

$\begin{array}{cc}\underset{h=l+1}{\overset{l_1}{\Σ}}{x}_{ijh}+z_{jl}\≤1,& \∀l=1,...,L_i-1,i\∈I\end{array},$

x_ij,z_j,y_jl,w_jN∈{0,1}

Wherein a_ijFor decision variable, if j-th BS covers i-th TP, a_ij=1, on the contrary then a_ij=0, b_jlBecome for decision-making Amount, if can realize data transfer, b by wireless link between j-th BS and l-th BS_jl=1, on the contrary then b_jl=0, d_iFor The service request of i-th TP, x_ijFor decision variable, x if i-th TP to be distributed to j-th BS_ij=1, on the contrary then x_ij=0, z_j>1 for j-th CS weight coefficient, w_jNFor decision variable, w if j-th BS is comprising mesh gateways_jN=1, on the contrary then w_jN= 0, y_jlFor decision variable, y if it there is wireless link between j-th BS and l-th BS_jl=1, on the contrary then y_jl=0, f_jlIt is nothing Business Stream in wired link (j, l), f_jNFor the Business Stream between the mesh gateways and backbone network in j-th BS, I is test node collection Close, T is time interval set, and S is the collection of base stations comprising mesh routers and mesh gateway devices,For mesh gateways Set,Reduce the collection of base stations of i-th TP for covering receiving power request for covering to have, h reduces covering for covering to have The Base Station Identification of i-th TP of receiving power request, c_jFor the installation cost of base station j, p_jFor the connection between gateway and core net Cost, d_itIt is the business demand in i-th TP of time t, u_jlFor the link capacity between base station j and base station l, v_jIn for base station j Distribute to the access capacity of TP, L_iFor cover i-th TP number of base stations, M be towards mesh gateways backbone network incoming end hold Amount, BS is base station, and TP is test node, l₁Higher limit for h.

3. method according to claim 1, for step B it is characterized in that：For reducing room and time complexity, carry out Engineering equivalent process, specially：

$\begin{array}{cc}min& \underset{j\∈S}{\Σ}\underset{t\∈T}{\Σ}{z}_{jt}{\mathrm{\ξ}}_j\mathrm{\Δ}T\end{array}$

$\begin{array}{ccc}s.t.& \underset{l\∈S}{\Σ}(f_{ljt}-f_{jlt})+\underset{i\∈I}{\Σ}{d}_{it}{x}_{ijt}=f_{j0t}& \∀j\∈S,\∀t\∈T\end{array}$

$\begin{array}{cc}\underset{i\∈I}{\Σ}{d}_{it}{x}_{ijt}\≤v_j,& \∀j\∈S,t\∈T\end{array},$

$\begin{array}{cc}{f}_{ljt}+f_{jlt}\≤u_{jl}{b}_{jl}{z}_{jt},& \∀j,l\∈S,t\∈T\end{array},$

$\begin{array}{cc}{f}_{j0t}\≤M,& \∀j\∈G\⊆S,t\∈T\end{array},$

$\underset{h=l+1}{\overset{l_1}{\Σ}}{x}_{{\mathrm{iJ}}_h^{\left(i\right)}t}+z_{J_l^{\left(i\right)}t}\≤1,\∀l=1,...,L_i-1,i\∈I,t\∈T,$

$\begin{array}{cccc}{x}_{ijt},z_{jt}\∈\{0,1\},& \∀i\∈I,& j,l\∈S,& t\∈T\end{array},$

$\begin{array}{cc}\underset{j\∈S}{\Σ}{x}_{ijt}=h_{it},& \∀i\∈I,t\∈T\end{array}$

Wherein h_itFor decision variable, if sending service request d in i-th TP of moment t_itThen h_it=1, on the contrary then h_it=0, x_ijtFor Decision variable, x if i-th TP to be distributed to j-th BS in moment t_ijt=1, on the contrary then x_ijt=0, z_jtFor decision variable, if In the in running order then z of moment t base station j_jt=1, on the contrary then z_jt=0, f_jltIt is the industry in moment t wireless link (j, l) Business stream, f_j0tIt is the Business Stream between j-th base station of moment t and the 0th node, ξ_jThe power consumed by base station j ∈ S.

4. method according to claim 1, for step C it is characterized in that：By through engineering approaches process obtain optimization solution with The optimization solution obtained by wireless mesh network link optimizing model is contrasted, and adaptive correction wireless mesh network link Optimized model, specially：Application service o is based on, by the optimization solution of through engineering approaches process acquisition and by wireless mesh network link Optimized model obtain optimization solution contrasted, if the optimization solution set of the two is completely the same, using the optimization solution as towards The optimization solution of application service o；If the optimization solution set of the two is inconsistent, the optimization solution that through engineering approaches process is obtained is used as application The final optimization pass solution set of service g, and revise the adjustable ginseng of wireless mesh network in wireless mesh network link optimizing model Number makes which obtain the optimization solution set that through engineering approaches are processed, and using revised wireless mesh network link optimizing model as next The wireless mesh network link optimizing model of individual application service o+1, the answering with related application attribute in wireless mesh network It is O={ 1,2 ..., o, o+1 } with set of service, optimizing solution set by o+1 of acquisition carries out statistical average process, and should Statistical average processes to optimize to solve to gather to optimize to solve to gather with o+1 for obtaining revises wireless mesh network link optimizing model, and Using the model as the application service set O+1 of other related application attributes in wireless mesh network system next time priori chain Road Optimized model.

5. method according to claim 1, for step C it is characterized in that：Set up through-put power and time delay combined optimization Model, specially：

$\begin{array}{cc}min& \{\underset{0\≤q\≤\mathrm{\ω}}{max}\[s_q+\mathrm{\θ}\left({\mathrm{\η}}_q\right)\]\}\end{array}$

$\begin{array}{cc}s.t.& \mathrm{\θ}\left({\mathrm{\η}}_{{\mathrm{\π}}_q}\right)-s_q\≤0,q=1,2,...,\mathrm{\ω}\end{array}$

(s_r+θ(η_r)-s_q)(s_q+θ(η_q)-s_r)H(η_q·η_r) >=0, q ≠ r, q, r=0,1 ..., ω

η_q∈ W, q=0,1 ..., ω

θ(η_q)>0, q=0,1 ..., ω

Wherein θ (η_q) it is the node v when network adopts multicast mechanism_qTransmission business flow to the propagation delay time of its child node, η_qFor v_q Through-put power, I (η_q) it is by node v_qThe node set of interference, H (η_q,η_r) it is decision variable, if { π_e∈C_q}∪{π_e∈I (η_q) and { π_e∈C_r}∪{π_e∈I(η_r) while setting up, then H (η_q,η_r)=1, otherwise then H (η_q,η_r)=0, C_rFor v_rSon section Point set, π_rFor v_rChild node, η_qFor v_qThrough-put power, s_qFor v_qBusiness Stream send initial time, W is nodes Transmit power set,For v_qChild node through-put power, ω be network in node number.

6. method according to claim 1, for step C it is characterized in that：By through engineering approaches process obtain optimization solution with By set up wireless mesh network link optimizing model acquisition optimization solution contrasted, and adaptive correction through-put power and when Prolong combined optimization model.