CN106304243A - A kind of wireless sensor network topology control method based on gesture game - Google Patents

Abstract

The invention discloses a kind of wireless sensor network topology control method based on gesture game, determine network model, sensor network is expressed as flat topology figure, establish the network of a connection, then set up topology and control betting model (TCBPG).On topology betting model, Nash Equilibrium Solution is there is in the revenue function established in game in order to ensure game, the betting model that definition topology controls is an ordinal number gesture game, after establishing betting model according to preceding step, carry out distributed energy consumption balance Topology Control Algorithm again to realize, it is divided into neighbor discovery phase, game performs the stage, topology maintenance phase, obtain a stable network topology structure of wireless sensor, from power control and the angle of energy consumption balance, the dump energy and the node transmitting power that consider node devise a kind of new utility function, propose a kind of Distributed Topology Control Algorithm based on gesture game.Theory analysis and simulation result show, can reach Nash Equilibrium and ensure that the superperformances such as the connectedness of network, energy consumption balance, extend the life cycle of network in the topology of this algorithm construction.

Description

A kind of wireless sensor network topology control method based on gesture game

Technical field

The present invention relates to Topology Control Algorithm, a kind of wireless sensor network topology control method based on gesture game.

Background technology

Wireless sensor network has the characteristic of multi-hop, and node may show and comes to save energy in data forwarding process Private behavior.Therefore, recent years, theory of games starts by many scholars for studying topology control problem.EIDENBENZ Et al. using topology control problem as a non-cooperative game, and analyze the complexity finding Nash Equilibrium algorithm, and propose 3 connective games, but the existence of Nash Equilibrium cannot be ensured, the most do not prove the energy efficiency of constructed topology. Therefore Komali et al. utilizes the topology control of Non-Cooperative Game Analysis on Trade ad hoc network, devises one and can ensure that convergence Distributed optimal response algorithm MIA, its basic thought is: each node performs game in turn, when node performs game Time, other node keeps current power constant, and the node selection performing game maximizes the power of oneself income, and in network All node broadcasts games perform after power, other node recalculates neighbor list, the most repeatedly, until any one Till individual node cannot be by changing oneself power additional income, this algorithm ensure that and minimizes network maximum transmission power, But the different execution sequences of algorithm interior joint result in different topologys.Author proposes based on MIA Topology Control Algorithm subsequently Innovatory algorithm, DIA and Local-DIA.DIA algorithm is based on more excellent response policy, and its basic thought is: all nodes are by merit Rate selects set discretization, it is ensured that at most disconnecting a link from a power drop to adjacent power network, each node is passed successively Subtract oneself launches power until any one node cannot be by changing the power of oneself and till additional income, this algorithm is raw The network topology become makes each node minimize transmitting power in the case of ensureing network bi-directional connection, and DIA algorithm is possible not only to Reach Nash Equilibrium, but also can guarantee that the uniqueness building topology, eliminate the impact on topological structure of the node execution sequence. Local-DIA is local DIA algorithm, effectively reduces the exchange times of information during algorithm performs.But DIA and Local-DIA algorithm does not all consider the residue energy of node impact on network lifetime.One is proposed based on this Sajjad et al. Planting topology based on link power consumption and control game playing algorithm MLPT, the main thought of this algorithm is: all nodes run with peak power Minimum hop routing algorithm, node calculates the number of routed path through its each link, then calculate each link power consumption because of Son (minimum emissive power of this link of dissipation factor=routed path quantity × maintain), each node performs game in turn, when one When node performs game, maximize the link of its benefit function with optimal response policy selection, and broadcast the power after game, directly No longer changing its power to all nodes.MLPT algorithm can reduce the energy consumption of whole network and equalize the load of node link, MLPT algorithm can recalculate the link power consumption factor according to the statistic record mutually sending data between node.HAO proposes one Planting Distributed Topology Control Algorithm VGEB based on virtual game, in this algorithm, each node is only according to the information of neighbor nodes collected Found the power after carrying out game and broadcasting game, although this algorithm reduces the number of times of information exchange during game performs, but Node is needed to have stronger computing capability and bigger memory space to support the execution of game.

Summary of the invention

Feature and great majority topology based on game for wireless sensor network node energy constraint control not consider node The situation of energy, the present invention provides a kind of wireless sensor network topology control method based on gesture game, and this method is from power Controlling and the angle of energy consumption balance is set out, the dump energy and the node transmitting power that consider node devise a kind of new effectiveness Function, proposes a kind of Distributed Topology Control Algorithm based on gesture game.Theory analysis and simulation result show, this algorithm construction Topology in can reach Nash Equilibrium and ensure that the superperformances such as the connectedness of network, energy consumption balance, extend network Life cycle.

The technical scheme realizing the object of the invention is:

A kind of wireless sensor network topology control method based on gesture game, comprises the steps:

1) determine network model, sensor network is expressed as flat topology figure, establish the network of a connection, then build Vertical topology controls betting model (TCBPG)；

2) in step 1) on topology betting model, establish the revenue function in game:

$u_i(p_i,p_{-i})=f_i(p_i,p_{-i})(\mathrm{\α}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\β}\stackrel{\‾}{E_i\left(p_i\right)})-\mathrm{\α}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}$

In formula, α and β is weight factor and is positive number, p_iRepresent the transmitting power of joint i, p_-iRepresent other N-1 The transmitting power of node, f_i(p_i, p_-i) represent network connectedness, work as f_i(p_i, p_-iDuring)=1, represent that network is Connection, i.e. node i can be by two-way link and other all node communications, otherwise f_i(p_i, p_-i)=0 network is not Connection, it is clear that f_i(p_i, p_-i) it is dull non-decreasing function, i.e. for arbitrary nodeAnd its transmitting power p_i>q_iTime, f_i(p_i, p_-i)≥f_i(q_i, p_-i),Represent the maximum transmission power of node i, J represents that node i is at power p_iTime a hop neighbor node, E_rJ () represents the dump energy of node j, E_o(j) table Show the primary power of j,Represent the income obtained when network-in-dialing；

3) there is Nash Equilibrium Solution in order to ensure game, the betting model that definition topology controls is an ordinal number gesture game, its Potential function is defined as:

$V(p_i,p_{-i})=\underset{i\∈N}{\mathrm{\Σ}}[f_i(p_i,p_{-i})(\mathrm{\α}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\β}\stackrel{\‾}{E_i\left(p_i\right)})-\mathrm{\α}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}]$

In formula, function V is referred to as the ordinal number potential function of Г；

4) after establishing betting model according to preceding step, then the realization of distributed energy consumption balance Topology Control Algorithm is carried out, point Three phases is become to realize；

5) step 4 is realized) neighbor discovery phase of three phases in algorithm；

6) realize step 4) in algorithm the game of three phases perform the stage；

7) step 4 is realized) the topological maintenance phase of three phases in algorithm；

A stable network topology structure of wireless sensor can be obtained after above-mentioned steps.

Step 1) in sensor network is expressed as flat topology figure

H=(N, E, Ω) (1)

Wherein H represents topological diagram plane, set of node N={1, and 2 ..., n} represents the set of all nodes in network, limit collectionRepresent the set of all communication links between node, Ω=[ω_ij] then represent it is a matrix.

N sensor node random placement, in two dimension monitored area, has identical maximum transmission powerWherein ω: E → R⁺, (i j) is communication link (i, j) minimum power required for connection to ω.If Ω is a symmetrical matrix, i.e. from node i to joint Minimum power required for some j transmission message is identical with from node j to node i.Definition vector G (p) is expressed as all of two-way link set figure, and G (p) is a subgraph of H, if each node uses Peak power communicates, and note G (p) is G_max, G_maxReferred to as peak power network, G_maxIt is connection,

Network-in-dialing is the basic demand that topology controls, and well ensures the energy consumption balance of network, existence week to set up a characteristic The network topology that phase is relatively long, it is necessary to set up the topological Controlling model of a gesture game,

Definition gesture game topology framework is Г=<N, P, u>:

Wherein, Г represents that standard game accords with, N={1,2 ..., n} represents that participant gathers, and P represents strategy set, set of strategies Closing P is all node power level set P_iCartesian product,P_iRepresent the power stage collection that node i can select Closep_i∈P_i, p_iRepresent the power stage selected by node, u_i(p_i, p_-i) represent revenue function.

Step 2) in betting model design revenue function

$u_i(p_i,p_{-i})=f_i(p_i,p_{-i})(\mathrm{\&alpha;}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\&beta;}\stackrel{\&OverBar;}{E_i\left(p_i\right)})-\mathrm{\&alpha;}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}---\left(2\right)$

In formula, α and β is weight factor and is positive number, p_iRepresent the transmitting power of joint i, p_-iRepresent other N-1 node Transmitting power, f_i(p_i, p_-i) represent network connectedness, work as f_i(p_i, p_-iDuring)=1, represent that network is connection, i.e. Node i can be by two-way link and other all node communications, otherwise f_i(p_i, p_-i)=0 network does not connects, it is clear that f_i(p_i, p_-i) it is dull non-decreasing function, i.e. for arbitrary nodeAnd its transmitting power p_i>q_iTime, f_i(p_i, p_-i)≥f_i(q_i, p_-i),Represent the maximum transmission power of node i,J represents that node i is at power p_iTime one jumping neighbour Occupy node, E_rJ () represents the dump energy of node j, E_oJ () represents the primary power of j,Represent The income obtained during network-in-dialing, it is clear thatShow to maintain network-in-dialing ratio to save node energy consumption More important.

In order to choose node that dump energy is many as in its neighbor node revenue function as far as possibleControl, from And make the consumption of whole network keep balance.

Step 3) in betting model design ordinal number potential function

$V(p_i,p_{-i})=\underset{i\&Element;N}{\mathrm{\&Sigma;}}[f_i(p_i,p_{-i})(\mathrm{\&alpha;}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\&beta;}\stackrel{\&OverBar;}{E_i\left(p_i\right)})-\mathrm{\&alpha;}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}]---\left(3\right)$

In formula, function V is referred to as the ordinal number potential function of Г.

Step 4) in distributed energy consumption balance Topology Control Algorithm characteristic:

G_maxBeing a connected network, DEBA algorithmic statement is in keeping network G_maxThe Nash Equilibrium state of connection characteristic.By step Rapid 3) define TCBPG model is ordinal number gesture game, it is known that a finite ordinal number gesture game, more excellent response strategy is bound to In limited step, convergence reaches Nash Equilibrium.Additionally, in DEBA algorithm, each node is straight in the benefit constantly increasing oneself To can not increasing.According to definition 1, all cannot pass through to change certainly it is known that Nash Equilibrium state is exactly a kind of any node Oneself strategy increases the steady statue of benefit, it is clear that DEBA algorithm can be restrained and Nash Equilibrium state.For all nodes For, it is the most meaningful to the Nash Equilibrium state keeping network-in-dialing that node reduces the transmitting power of oneself by more excellent response strategy,

Defining according to Nash Equilibrium, neither one node can obtain higher benefit by continuing to reduce the power of oneself, if Continue to reduce power and only can reduce the benefit of oneself, even make network not connect.The most just run counter to lemma 1.Secondly, the most not There is m (m > 2) individual node and reduce power to increase the benefit of oneself on the premise of ensureing network-in-dialing simultaneously.If this is because If certain node can increase benefit by reducing the power of oneself, network is inevitable not to be connected, then be accomplished by other node Increasing power to maintain network-in-dialing, like this benefit of other nodes will reduce.Therefore determining according to Pareto optimality Justice, DEBA algorithmic statement is in Pareto optimality Nash Equilibrium.

Step 5) neighbor discovery phase in algorithm:

Definition node has unique ID, and each node i initializes the power of oneself and isThen withBroadcast " neighbor uni-cast Information ", wherein comprise node ID,And dump energy, and collect from receiving its " discovery information " sent joint The return information that point sends.Node i is according to the ACK information from node j received, by ID, the residual energy of node j Amount, p_ijIt is added in the neighbor list of oneself.Free space model is used to calculate the minimum power required for i to node j $p_{\mathrm{ij}}=\frac{p_{\mathrm{th}}{\left(4\mathrm{\&pi;}\right)}^{2}d(i,j)L}{G_t{G}_r{\mathrm{\&lambda;}}^2},$ Exchange each node by information and generate the set of strategies of oneself $P_i=\{p_i^{\max }=p_i^0,p_i^1,...,p_i^k=p_i^{\min }\},$ Its Middle k is a hop neighbor node number of i

Step 6) game in algorithm performs the stage:

Node obtains the connectivity of whole network, and all nodes are held in turn by random or rising descending according to node ID Row game determines its power and often takes turns its power of only one of which knot adjustment, and other node keeps its power constant, when a node Change its power, notify that other node recalculates neighbor node set by sending control information.Assorted in order to converge to receive Equilibrium, uses more excellent reaction (Better-Response) policy update scheme, a finite ordinal number gesture game, more excellent response strategy It is bound to converge on Nash Equilibrium in limited step.During game performs, if node selects the low power of grade can be than current merit The benefit that rate obtains is big, then node can select the power lower than current power and recalculate neighbor node set, otherwise, and joint Point does not change power, say, that for arbitrary node i, its current power isM=1,2 ..., k-1, then joint The power that some i selects is:It is joint i currently optional power, the process that game performs False code is as follows:

Step 7) topological maintenance phase in algorithm:

Along with the increase of network operation time, it is more and more unbalanced that the dump energy of the node in network can become.In order to balance joint Energy expenditure between point, network topology should be adjusted extending Network morals dynamically.Each node both knows about certainly The dump energy that oneself is current, if the energy that can set a time cycle or node is less than certain limit, as less than just The 30% of beginning energy, re-executes the topology generation algorithm load with balance node, it is achieved thereby that the wireless senser of gesture game Network topology control.

Beneficial effect:

This wireless sensor network topology control method based on gesture game, from power control and the angle of energy consumption balance, The dump energy and the node transmitting power that consider node devise a kind of new utility function, it is proposed that a kind of based on gesture game Distributed Topology Control Algorithm, theory analysis and simulation result show, can reach Nash Equilibrium also in the topology of this algorithm construction And ensure that the superperformances such as the connectedness of network, energy consumption balance, extend the life cycle of network.

Detailed description of the invention

Hereinafter present invention is further elaborated, but is not limitation of the invention.

Embodiment:

A kind of wireless sensor network topology control method based on gesture game, comprises the steps:

1) sensor network is expressed as flat topology figure H=(N, E, Ω), wherein set of node N={1,2 ..., n} represents in network The set of all nodes, it is assumed that n sensor node random placement, in two dimension monitored area, has identical emission maximum PowerIts limit collectionRepresent the set of all communication links between node, Ω=[ω_ij] it is a matrix, Wherein ω: E → R⁺, ω (i, j) be communication link (i, j) connection required for minimum power. assume that Ω is a symmetrical square Battle array, i.e. from node i to node j transmission message required for minimum power identical with from node j to node i, definition VectorG (p) is expressed as all of two-way link set figure, and G (p) is H's One subgraph, if each node uses peak power to communicate, then remembers that G (p) is G_max, referred to as peak power network, Determining network model, establish the network of a connection, network-in-dialing is the basic demand that topology controls, in order to set up one Individual characteristic well ensures the energy consumption balance of network, the network topology that life cycle is relatively long, it is necessary to sets up a gesture and wins The topological Controlling model (TCBPG) played chess, definition gesture game topology framework is Г=<N, P, u>；

2) in step 1) on topology betting model, establish the revenue function in game:

$u_i(p_i,p_{-i})=f_i(p_i,p_{-i})(\mathrm{\&alpha;}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\&beta;}\stackrel{\&OverBar;}{E_i\left(p_i\right)})-\mathrm{\&alpha;}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}$

In formula, α and β is weight factor and is positive number, p_iRepresent the transmitting power of joint i, p_-iRepresent other N-1 The transmitting power of individual node, f_i(p_i, p_-i) represent network connectedness, work as f_i(p_i, p_-iDuring)=1, represent Network is that connection, i.e. node i can be by two-way link and other all node communications, otherwise f_i(p_i, p_-i)=0 Network does not connects, it is clear that f_i(p_i, p_-i) it is dull non-decreasing function, i.e. for arbitrary nodeAnd its Penetrate power p_i>q_iTime, f_i(p_i, p_-i)≥f_i(q_i, p_-i),Represent the maximum transmission power of node i.J represents that node i is at power p_iTime a hop neighbor node, E_rJ () represents node j's Dump energy, E_oJ () represents the primary power of j,Represent the acquisition when network-in-dialing Income；

3) there is Nash Equilibrium Solution in order to ensure game, the betting model that definition topology controls is an ordinal number gesture game, its gesture letter Number is defined as:

4) after establishing betting model according to preceding step, then the realization of distributed energy consumption balance Topology Control Algorithm is carried out, this punishment Three phases is become to realize；

5) step 4 is realized) neighbor discovery phase in algorithm, it is assumed that node has unique ID, and each node i initializes oneself Power beThen withBroadcast " neighbor uni-cast information " (wherein comprise node ID,And dump energy), and And collect from receiving the return information that its " discovery information " node sent sends；Node i according to receive from The ACK information of node j, by the ID of node j, dump energy, p_ijIt is added in the neighbor list of oneself.Use the most empty Between model calculate the minimum power required for i to node jExchange each node by information and generate oneself Strategy $P_i=\{p_i^{\max }=p_i^0,p_i^1,...,p_i^k=p_i^{\min }\},$ Wherein k is a hop neighbor node number of i $p_i^j\&GreaterEqual;p_i^{j+1}\&ForAll;j\&Element;[0,k-1];$

6) step 4 is realized) game in algorithm performs the stage, it is assumed that and node obtains the connectivity of whole network, all joints Point determines its power by random or performing game in turn according to node ID liter descending and often takes turns only one of which node tune Its power whole, other node keeps its power constant；When a node changes its power, lead to by sending control information Know that other node recalculates neighbor node set；In order to converge to Nash Equilibrium, use more excellent reaction (Better-Response) Policy update scheme, a finite ordinal number gesture game, more excellent response strategy is bound to converge on Nash Equilibrium in limited step； During game performs, if node selects the low power of grade can be bigger than the benefit that current power obtains, then node can select The power lower than current power also recalculates neighbor node set, and otherwise, node does not change power, say, that for Arbitrary node i, its current power isM=1,2 ..., k-1, then the power that node i selects is:Being joint i currently optional power, the process false code that game performs is as follows:

7) step 4 is realized) topological maintenance phase in algorithm, along with the increase of network operation time, the residual energy of the node in network It is more and more unbalanced that amount can become, and in order to balance internodal energy expenditure, network topology should be adjusted prolonging dynamically Long Network morals, each node both knows about oneself current dump energy, can set time cycle or such as When really the energy of node is less than certain limit, as less than the 30% of primary power, re-execute topology generation algorithm to balance joint The load of point, it is achieved thereby that the wireless sensor network topology of gesture game controls.

A stable network topology structure of wireless sensor can be obtained after above-mentioned steps.

Theorem used in this method and definition statement:

Theorem 1: if strategy game Г=<N, S, u>is ordinal number gesture game and V is its ordinal number potential function, then can make its V Maximized strategy combination s^*It it is exactly a Nash Equilibrium of game Г.Therefore, if we can determine that the ordinal number gesture of a game Function, it is possible to by asking the strategy combination making its ordinal number potential function maximum to be i.e. its a Nash Equilibrium.

Theorem 2: Pareto optimality (Pareto Optimal, PO): if for arbitrary s ∈ S, strategic vector s^*∈ S is full FootSo strategic vector s^*It it is exactly Pareto optimality.

Definition 1: Nash Equilibrium a: strategy combinationIt is that of a game Г=<N, S, u>receives assorted Equilibrium, if forWithOne game may have more than one to equalize, or root This does not exists.At least there is a Nash Equilibrium in some type of game.In order to ensure the existence of Nash Equilibrium, D. Monderer and L.Shapley analyzes and have studied the tactful formula game gesture game (Potential that a class is special Games), the main feature of this kind of game is that it at least exists a Nash Equilibrium.

Definition 2: ordinal number gesture game (Ordinal Potential Game, OPG) and ordinal number potential function (Ordinal Potential Function, OPF): tactful game Г=<N, S, a u>is an ordinal number gesture game, if there is a function V:S → R, Right $\&ForAll;i\&Element;N,\&ForAll;s_{-i}\&Element;S_{-i}$ And $\&ForAll;a_i,b_i\&Element;S_{i}V(a_i,a_{-i})-V(b_i,a_{-i})>0\&DoubleLeftRightArrow;u_i(a_i,a_{-i})-u_i(b_i,a_{-i})>0$ Function V is referred to as The ordinal number potential function of Г.

Lemma 1: if G_maxBeing a connected network, DEBA algorithmic statement is in keeping network G_maxReceiving of connection characteristic is assorted Equilibrium state.

Claims (7)

1. a wireless sensor network topology control method based on gesture game, it is characterised in that comprise the steps:

1) determine network model, sensor network is expressed as flat topology figure, establish the network of a connection, then set up Topology controls betting model (TCBPG)；

2) in step 1) on topology betting model, establish the revenue function in game:

$u_i(p_i,p_{-i})=f_i(p_i,p_{-i})(\mathrm{\&alpha;}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\&beta;}\stackrel{\&OverBar;}{E_i\left(p_i\right)})-\mathrm{\&alpha;}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}$

In formula, α and β is weight factor and is positive number, p_iRepresent the transmitting power of joint i, p_-iRepresent other N-1 The transmitting power of individual node, f_i(p_i, p_-i) represent network connectedness, work as f_i(p_i, p_-iDuring)=1, represent net Network is that connection, i.e. node i can be by two-way link and other all node communications, otherwise f_i(p_i, p_-i)=0 Network does not connects, it is clear that f_i(p_i, p_-i) it is dull non-decreasing function, i.e. for arbitrary nodeAnd its Penetrate power p_i>q_iTime, f_i(p_i, p_-i)≥f_i(q_i, p_-i),Represent the maximum transmission power of node i,J represents that node i is at power p_iTime a hop neighbor node, E_rJ () represents node j's Dump energy, E_oJ () represents the primary power of j,Represent the acquisition when network-in-dialing Income；

3) there is Nash Equilibrium Solution in order to ensure game, the betting model that definition topology controls is an ordinal number gesture game, its gesture Function is defined as:

$V(p_i,p_{-i})=\underset{i\&Element;N}{\mathrm{\&Sigma;}}[f_i(p_i,p_{-i})(\mathrm{\&alpha;}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\&beta;}\stackrel{\&OverBar;}{E_i\left(p_i\right)})-\mathrm{\&alpha;}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}];$

4) after establishing betting model according to preceding step, then carry out the realization of distributed energy consumption balance Topology Control Algorithm, be divided into Three phases realizes；

5) step 4 is realized) neighbor discovery phase in algorithm；

6) step 4 is realized) game in algorithm performs the stage；

7) step 4 is realized) topological maintenance phase in algorithm.

Wireless sensor network topology control method based on gesture game the most according to claim 1, it is characterised in that step 1) in, sensor network topological is become a plane graph

H=(N, E, Ω) (1)

Wherein H represents topological diagram plane, set of node N={1, and 2 ..., n} represents the set of all nodes in network, limit collectionRepresent the set of all communication links between node, Ω=[ω_ij] then represent it is a matrix,

N sensor node random placement, in two dimension monitored area, has identical maximum transmission powerWherein ω: E → R⁺, (i is j) that (i, j) minimum power required for connection, if Ω is a symmetrical matrix to communication link, i.e. from node i to joint to ω Minimum power required for some j transmission message is identical with from node j to node i, definition vector G (p) is expressed as all of two-way link set figure, and G (p) is a subgraph of H, if each node uses Peak power communicates, and note G (p) is G_max, G_maxReferred to as peak power network, G_maxIt it is connection；

Network-in-dialing is the basic demand that topology controls, and well ensures the energy consumption balance of network, existence week to set up a characteristic The network topology that phase is relatively long, it is necessary to set up the topological Controlling model of a gesture game,

Definition gesture game topology framework is Г=<N, P, u>:

Wherein, Г represents that standard game accords with, N={1,2 ..., n} represents that participant gathers, and P represents strategy set, strategy Set P is all node power level set P_iCartesian product,P_iRepresent the power stage that node i can select Setp_i∈P_i, p_iRepresent the power stage selected by node, u_i(p_i, p_-i) represent revenue function.

Wireless sensor network topology control method based on gesture game the most according to claim 1, it is characterised in that step 2) In $u_i(p_i,p_{-i})=f_i(p_i,p_{-i})(\mathrm{\&alpha;}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\&beta;}\stackrel{\&OverBar;}{E_i\left(p_i\right)})-\mathrm{\&alpha;}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}$ It it is betting model design Revenue function, determines the income obtained when each node is connected to network and is connected to the balance of the cost that network is paid.

Wireless sensor network topology control method based on gesture game the most according to claim 1, it is characterised in that step 3) in $V(p_i,p_{-i})=\underset{i\&Element;N}{\mathrm{\&Sigma;}}[f_i(p_i,p_{-i})(\mathrm{\&alpha;}{p}_i^{\max }\frac{E_o\left(i\right)}{E_r\left(i\right)}+\mathrm{\&beta;}\stackrel{\&OverBar;}{E_i\left(p_i\right)})-\mathrm{\&alpha;}{p}_i\frac{E_o\left(i\right)}{E_r\left(i\right)}]$ It is that ordinal number potential function determines The certainly existing property of Nash Equilibrium in game.

Wireless sensor network topology control method based on gesture game the most according to claim 1, it is characterised in that step 5) Definition node has unique ID, and it is pmax i that each node i initializes the power of oneself, then " adjacent with pmax i broadcast Occupy discovery information ", wherein comprise node ID, pmax i and dump energy, and collect from receiving " sending out of its transmission Existing information " return information that node sends, node i is according to the ACK information from node j received, by node j ID, dump energy, pij be added in oneself neighbor list, needed for using free space model to calculate i to node j The minimum power wantedExchange each node by information and generate oneself tactful Pi={pmax i=p0i, p1 I ..., pk i=pmin i}, wherein k is a hop neighbor node number of i

Wireless sensor network topology control method based on gesture game the most according to claim 1, it is characterised in that step 6) Interior joint obtains the connectivity of whole network, and all nodes come in turn by random or rising descending according to node ID Performing game determine its power and often take turns its power of only one of which knot adjustment, other node keeps its power constant, when one Individual node changes its power, notifies that other node recalculates neighbor node set by sending control information, in order to Converge to Nash Equilibrium, use more excellent reaction (Better-Response) policy update scheme, a finite ordinal number gesture game, More excellent response strategy is bound to converge on Nash Equilibrium in limited step, during game performs, if node selects grade low Power can than current power obtain benefit big, then node can select the power lower than current power and recalculate neighbour Occupying node set, otherwise, node does not change power, say, that for arbitrary node i, and its current power is pm i, M=1,2 ..., k-1, then the power that node i selects is:Ph i is joint i Current optional power, the process false code that game performs is as follows:

$\mathrm{Algorthm}1\left(G\max \right)\&RightArrow;(G\max ,\hat{p})$

1: for all of node i, its its parameter h=0 is set；

2:// initialization node power is peak power

3:WhileIt it not Nash Equilibrium state do

4:For all i∈N do

5:h=h+1

6: ${\hat{p}}_i=\arg {\max }_{q_i\&Element;\{{\hat{p}}_i,p_i^h\}}{u}_i(q_i,{\hat{p}}_{-i})$

7:End for

8:End while。

Wireless sensor network topology control method based on gesture game the most according to claim 1, it is characterised in that step 7) In along with the increase of network operation time, it is more and more unbalanced that the dump energy of the node in network can become, in order to balance Internodal energy expenditure, network topology should be adjusted extending Network morals, each node dynamically Know oneself current dump energy, if the energy that can set a time cycle or node is less than certain limit, As less than the 30% of primary power, re-execute the topology generation algorithm load with balance node, it is achieved thereby that gesture game Wireless sensor network topology controls.