CN103442382B - Reduce the topology control method disturbing between link - Google Patents

Abstract

The invention discloses a kind of topology control method disturbing between link that reduces, mainly solve and in prior art, can not ensure to reduce the problem of disturbing between link. Its implementation procedure is: first, build peak power topology by center control nodes; Secondly, by center control nodes, taking the degree of disturbance on limit as standard, all limits in the limit set of traversal peak power topology, delete optimization to this limit set, build full mesh topology; Again, by center control nodes all node broadcasts full mesh topology information except center control nodes in network; Finally, determine the logic neighbors of oneself according to the full mesh topology information of receiving by all nodes except center control nodes in network, and adjust the transmitting power of oneself. The present invention has reduction node energy consumption, and the advantage that reduces to disturb between link, can be used for radio sensing network.

Description

Reduce the topology control method disturbing between link

Technical field

The invention belongs to wireless communication field, particularly a kind of method that builds network topology structure, can be used forRadio sensing network.

Background technology

Wireless sensor network is the wireless network consisting of Ad hoc mode the small node of one group of random distribution,Each small node is made up of sensor, data processing unit and communication unit. Wireless sensor network combines sensingDevice, low-power consumption, communication and the technology such as micro electronmechanical, have real-time data acquisition, Supervised Control and information sharing and depositThe functions such as storage management are a kind of network communicating systems that is full of development potentiality.

Energy reserving is the research aspect of two outbalances in wireless sensor network with reducing interference. In topology controlAspect, a large amount of research work in recent years mainly focuses on how algorithm for design is to reduce the energy consumption of network node, severalIndividual comparatively typical algorithm, if the authors such as N.Li are at IEEETransactionsonWirelessCommunicationsOn the article " DesignandAnalysisofanMST-BasedTopologyControlAlgorithm " delivered mentionLMST algorithm, the authors such as Y.Shen are on ACMSIGCOMMComputerCommunicationReviewArticle " the Ashortest-path-basedtopologycontrolalgorithminwirelessmu lti-hop deliveringNetworks " in the LSP algorithm mentioned etc., be subjected to and paid close attention to widely, due to the common tool of low energy consumption topological structureHave the characteristics such as sparse property, node degree be low, therefore people generally believe that the topology of low energy consumption also means low interference, soIn this class algorithm, seldom have in reducing node energy consumption and take explicitly into account the interference how reducing between node.

But, the article " Dosetopology that the authors such as M.Burkhart deliver on ACMMobiHoc2004Controlreduceinterference? " in be disclosed in energy reserving and reduce disturb between do not have definite relation, also, a low energy consumption Topology Control Algorithm can not guarantee that its final topological structure is low interference.

At present, in order to reduce interference, researcher has proposed some Topology Control Algorithms. According to interference model notWith, these algorithms roughly can be divided into two large classes: a class be consideration a pair of node in communication process to its transmission rangeThe interference that interior node causes, i.e. the low interference Topology Control Algorithm of the interference model based on limit noise spot, as M.Article " the Dosetopologycontrolreduce that the authors such as Burkhart deliver on ACMMobiHoc2004Interference? " the LISE algorithm of middle proposition; Another kind of is that consideration individual node is made the node in its transmission rangeThe interference becoming, i.e. the low interference Topology Control Algorithm of the interference model based on a noise spot, as the authors such as G.Feng existArticle " the Interferenceminimumnetwork delivering on Wirel.Commun.Mob.Comput2012TopologiesforNetworksadhocnetworks " the middle BIMA algorithm proposing. Owing to opening up based on bottom in researchWhen the Channel Assignment Problems of flutterring or scheduling problem, need to consider a pair of node in communication process to can in its transmission rangeThe interference causing with the link of communication, and above-mentioned two class algorithms all cannot guarantee to reduce the interference between link, Jin ErwuMethod provides a good bottom topology.

Summary of the invention

The object of the invention is to the problem for above-mentioned prior art, propose a kind of topology control of disturbing between link that reducesMethod processed, to reduce node energy consumption, reduces the interference between link, realizes bottom topology.

To achieve these goals, network topology control method of the present invention comprises the steps:

(1) in network, except center control nodes o, all node u send a Hello number to center control nodes oAccording to bag, center control nodes o obtains peak power topology G_max＝(V(G_max),E(G_max)), wherein V (G_max) forHigh-power topological G_maxNode set, E (G_max) be peak power topology G_maxLimit set:

(2) the center control nodes o in network calculates the degree of disturbance on every limit in peak power topology:

(2a) center control nodes o judges limit set E (G_max) in any two limit (x_i,y_i) and (x_j,y_j) whether mutualDisturb, if limit (x_j,y_j) on have at least one node x_jOr y_jAt node x_iOr node y_iTransmission radius r within the scope of,Limit (x_i,y_i) interfere with limit (x_j,y_j)；

(2b) center control nodes o calculates limit set E (G_max) in the degree of disturbance on every limit (x, y), gather on limitE(G_max) total number on the limit that statistics limit (x, y) interferes with in scope;

(3) center control nodes o builds full mesh topology G=(V (G), E (G)), and wherein V (G) is full mesh topology G'sNode set, comprises in center control nodes o and network all node u except center control nodes o, and E (G) is instituteThe limit set of the full mesh topology G building;

(4) center control nodes o is recorded in the node set V of described full mesh topology G (G) and limit set E (G) to open upFlutter more in new information MSG, and wide with the mode all node u except center control nodes o in network that floodBroadcast this MSG message;

(5) in network, except center control nodes o, all node u, according to above-mentioned MSG message, determine patrolling of oneselfCollect neighbors, and adjust the transmitting power of oneself.

Tool of the present invention has the following advantages:

1) the present invention, owing to having considered the phase mutual interference of limit with limit in the time calculating interference, therefore adopts the inventive method structureThe bottom topology of building, is more conducive to the link interference problem in processing channel distribution or link scheduling process;

2) the present invention is because the interference taking between link is weight, and in topology constructing, considered arbitrary node combinationBetween shortest path length, make the present invention ensureing, under the prerequisite of Euclidean distance t-Spanner characteristic, reducing nodeEnergy consumption, reduces the interference between link.

Brief description of the drawings

Fig. 1 is the applicable wireless network scenario schematic diagram of the present invention;

Fig. 2 is the peak power topological diagram forming when 50 meshed network scenes in Fig. 1;

Fig. 3 is realization flow figure of the present invention;

Fig. 4 is the sub-process figure that center control nodes of the present invention builds full mesh topology;

Fig. 5 is the present invention and the simulation comparison figure of existing Topology Control Algorithm to node maximum interference degree;

Fig. 6 is the present invention and the simulation comparison figure of existing Topology Control Algorithm to node maximum transmitted radius.

Detailed description of the invention

Below in conjunction with accompanying drawing, embodiment of the present invention is described in further detail.

With reference to Fig. 1, the wireless network that the present invention uses is made up of n the homogeneity node being distributed in two dimensional surface region,Suppose to control node o centered by one of them node, remain as controlling all node u outside joint o in network Zhong Chu center.Each node has No. ID, unique authentication number, and can obtain by GPS or other location technologiesIt self positional information. In network, each internodal wireless channel is additive white Gaussian noise channel. All jointsPoint is by omnidirectional antenna and node communication around, and maximum transmission power is P_max. The transmitting power P of all nodes canTo regulate continuously between minimum and maximum, i.e. 0≤P≤P_max. The transmission range corresponding to transmitting power of each nodeFor the transmission radius r of this node, maximum transmission power P_maxCorresponding transmission radius is maximum transmitted radius R, twoThe necessary and sufficient condition that has Radio Link between node is that the Euclidean distance between them is less than or equal to transmission radius r.

When all nodes in network all use maximum transmission power P_maxThe topological structure forming when transmission is that peak power is opened upFlutter, as shown in Figure 2, peak power topological representation is: G_max＝(V(G_max),E(G_max)), wherein V (G_max) forHigh-power topological G_maxNode set, E (G_max) be peak power topology G_maxLimit set. The peak power hereTopology is undirected topology, at limit set E (G_max) in (x, y) and (y, x) be same limit, in explanation afterwards,All represent with (x, y).

With reference to Fig. 3, performing step of the present invention is as follows:

Step 1, center control nodes o builds peak power topology G_max。

(1a) all node u except center control nodes o in center control nodes o and network, form maximum workRate topology G_maxNode set V (G_max)；

(1b) all node u that save outside o are controlled with maximum transmission power P in network Zhong Chu center_maxBy the side floodingFormula sends a Hello packet, the authentication that comprises node u in this Hello packet to center control nodes oNo. ID, number and positional information;

(1c) center control nodes o receives the Hello packet that in network, all node u except own send, andBy the information recording in Hello packet in the whole network informational table of nodes of oneself;

(1d) center control nodes o, according to the node location information in the whole network informational table of nodes of oneself, calculates the whole networkIn any two node x, the Euclidean distance d (x, y) between y:

$d(x,y)=\sqrt{{(X_x-X_y)}^2+{(Y_x-Y_y)}^2},$

Wherein, (X_x,Y_x) be the position coordinates of node x, (X_y,Y_y) be the position coordinates of node y;

(1e) center control nodes o judges between any two points according to node maximum transmitted radius R and described Euclidean distanceAnnexation, obtain limit set E (G_max) information, concrete determination methods is as follows:

If the Euclidean distance d (x, y) of limit (x, y) is less than or equal to node maximum transmitted radius R, when d (x, y)≤R,(x, y) ∈ E (G_max)；

If the Euclidean distance d (x, y) of limit (x, y) is greater than node maximum transmitted radius R, i.e. d (x, y) > when R,Wherein, ∈ represents subordinate relation,Represent non-subordinate relation;

(1f) with node set V (G_max) and limit set E (G_max), jointly form peak power topology:G_max＝(V(G_max),E(G_max))。

Step 2, the center control nodes o in network calculates peak power topology limit set E (G_max) in every limit dryDegree of disturbing.

(2a) center control nodes o is according to the limit set E (G of peak power topology_max) in every limit (x, y) EuropeanApart from d (x, y), judge limit set E (G_max) in any two limit (x_i,y_i) and (x_j,y_j) whether phase mutual interference:

If limit (x_j,y_j) on have at least one node x_jOr y_jAt node x_iTransmission radius r within the scope of, limit (x_i,y_i)Interfere with limit (x_j,y_j)；

If limit (x_j,y_j) on have at least one node x_jOr y_jAt node y_iTransmission radius r within the scope of, limit (x_i,y_i)Interfere with limit (x_j,y_j), otherwise (x_i,y_i) interference edge (x not_j,y_j)；

(2b) center control nodes o calculates limit set E (G_max) in the degree of disturbance on every limit (x, y), gather on limitE(G_max) total number on the limit that statistics limit (x, y) interferes with in scope.

Step 3, center control nodes o builds full mesh topology G=(V (G), E (G)).

With reference to Fig. 4, being implemented as follows of this step:

(3a) center control nodes o initializes full mesh topology and shortest path collection:

The node set V of full mesh topology G (G) is initialized to all joints in peak power topology by center control nodes oPoint, is initialized to all limits in peak power topology by the limit set E (G) of full mesh topology G, and is limit set E (G)In every limit (x, y) set up traversal mark, be initialized as 1, represent not ergodic state, i.e. label_(x,y)＝1；

Center control nodes o sets up set of minimal paths P={p_G(x, y) }, wherein p_G(x, y) is in node set V (G)Any a pair of node combination<x, y>shortest path in full mesh topology G, if limit (x, y) ∈ E (G), For empty set, if limitCenter control nodes o computing node combination<x, y>Shortest path p in full mesh topology G_G(x, y), and be recorded to set of minimal paths P;

(3b) center control nodes o checks in limit set E (G), whether to exist in the limit of ergodic state not, if do not deposit, full mesh topology G has built, otherwise center control nodes o gathers all traversals in E (G) by limit and is labeled asSorting from big to small by degree of disturbance in 1 limit, selects the limit of degree of disturbance maximumTraveled through mark and be made as 0, tableShow ergodic state,And set up test topology For node setWhole nodes in V (G),For limit set E (G) removes limitAfter set;

(3c) center control nodes o, taking Euclidean distance as weight, adopts shortest path tree algorithm, as Dijkstra calculatesMethod, bellman-ford algorithm, computing node combinationIn test topologyIn shortest pathAnd decision node combinationIn test topologyIn shortest pathCan meet Euclidean distanceT-Spanner characteristic:

$\left|p_{\hat{G}}(\hat{x},\hat{y})\right|\&le;t\&times;\left|p_{G_{max}}(\hat{x},\hat{y})\right|,$

WhereinFor node combinationIn test topologyIn shortest path length,For node combinationAt peak power topology G_maxIn shortest path length, t is the parameter of t-Spanner characteristicThe factor, value is 2 or 4;

If (3d) node combinationIn test topologyIn shortest pathCan meet Euclidean distanceT-Spanner characteristic, center control nodes o execution step (3e), if can not meet, center control nodes oReturn to step (3b);

(3e) center control nodes o finds out in set of minimal paths P and uses limitThe combination of all nodes<m, n>, form and quote set T,If quoting set T is empty set,Center control nodes execution step (3g), if set T is not empty set, center control nodes execution step (3f);

(3f) center control nodes o, taking Euclidean distance as weight, adopts shortest path tree algorithm, as Dijkstra calculatesMethod, bellman-ford algorithm, calculate all node combination<m that quote in set T, n>in test topologyInShortest path, and judge that can it meet Euclidean distance t-Spanner characteristic, if quote all node groups in set TClose and can meet, center control nodes o execution step (3g), has at least one pair of node to combine if quote in set TIn test topologyIn shortest path cannot meet Euclidean distance t-Spanner characteristic, center control nodes o returnsStep (3b);

(3g) center control nodes o upgrades topological state:

(3g1) center control nodes o is by limitFrom limit set E (G), delete, and node is combined?Test topologyIn shortest pathWith quote set T in all nodes be combined in test topologyInShort path is updated in set of minimal paths P;

(3g2) center control nodes o upgrades the degree of disturbance on every limit in limit set E (G), according to step (2) instituteThe degree of disturbance that the method for stating recalculates every limit (x, y) in limit set E (G), afterwards, returns to step (3b).

Step 4, center control nodes o is to more new information MSG message of the whole network broadcast topology.

(4a) center control nodes o is recorded in the node set V of described full mesh topology G (G) and limit set E (G)Topology is more in new information MSG;

(4b) center control nodes o maximum transmission power, controls joint o in the mode flooding to network Zhong Chu centerOuter all node u broadcast described MSG message.

Step 5, all node u that network Zhong Chu center is controlled outside joint o determine patrolling of oneself according to described MSG messageCollect neighbors, and adjust the transmitting power of oneself.

(5a) all node u that save outside o are controlled according to the full mesh topology in described MSG message in network Zhong Chu centerInformation, one on full mesh topology G jumped to neighbors v as logic neighbors, and form logic neighbors collection:LN_u=v ∈ V (G) | (u, v) ∈ E (G) }, wherein, V (G) is the node set of full mesh topology G, E (G) is the whole networkThe limit set of topology G;

(5b) all node us of joint outside o are controlled according to described logic neighbors collection in network Zhong Chu center, by oneselfTransmitting power is adjusted into the needed minimum power of logic neighbors that can cover farthest?

$P_{{\mathrm{LN}}_u}=max\left\{p_{u,v}\right|v\&Element;{\mathrm{LN}}_u\}.$

Effect of the present invention can further illustrate by emulation:

(1) simulated conditions

Emulation adopts MATLAB simulation software to realize.

In simulating scenes, network node is evenly distributed on a 1000 × 1000m at random²On two dimensional surface region. NetIn network, all nodes adopt identical maximum transmitted radius R=250m. Number of network node excursion is 50～100.

(2) emulation content and result

Emulation 1, is respectively at t under 2,4 condition, by the inventive method and existing LISE algorithm, peak powerTopology Control Algorithm NONE carries out simulation comparison to the maximum interference degree of network, and result as shown in Figure 5.

Fig. 5 shows: with increasing of nodes, and the network that adopts LISE algorithm and peak power Topology Control Algorithm to generateMaximum interference degree be obvious ascendant trend, and adopt the maximum interference degree of network that the inventive method generates with nodesThe ascendant trend increasing is slow, even tends to be steady; And in situation for same node point number, adopt the inventive methodThe maximum interference degree value of the network generating and average degree of disturbance value are much smaller than other two kinds of algorithms.

Emulation 2, is respectively at t under 2,4 condition, by the inventive method and existing LISE algorithm, peak powerTopology Control Algorithm NONE carries out simulation comparison to the maximum transmitted radius of network, and result as shown in Figure 6.

Fig. 6 shows: adopt the maximum transmitted radius of the network of the inventive method generation to be better than other Topology Control Algorithm,Because maximum transmitted radius is directly proportional to maximum transmission power, therefore the inventive method can effectively reduce the transmitting merit of nodeRate, this can prove the validity of the inventive method.

Claims (7)

1. reduce the topology control method disturbing between link, comprise the steps:

(1) in network, except center control nodes o, all node u send a Hello number to center control nodes oAccording to bag, center control nodes o obtains peak power topology G_max＝(V(G_max),E(G_max)), wherein V (G_max) forHigh-power topological G_maxNode set, E (G_max) be peak power topology G_maxLimit set:

(2) the center control nodes o in network calculates the degree of disturbance on every limit in peak power topology:

(2a) center control nodes o judges limit set E (G_max) in any two limit (x_i,y_i) and (x_j,y_j) whether mutualDisturb, if limit (x_j,y_j) on have at least one node x_jOr y_jAt node x_iOr node y_iTransmission radius r within the scope of,Limit (x_i,y_i) interfere with limit (x_j,y_j)；

(2b) center control nodes o calculates limit set E (G_max) in the degree of disturbance on every limit (x, y), gather on limitE(G_max) total number on the limit that statistics limit (x, y) interferes with in scope;

(3) center control nodes o builds full mesh topology G=(V (G), E (G)), and wherein V (G) is full mesh topology G'sNode set, comprises in center control nodes o and network all node u except center control nodes o, and E (G) is instituteThe limit set of the full mesh topology G building;

(4) center control nodes o is recorded in the node set V of described full mesh topology G (G) and limit set E (G) to open upFlutter more in new information MSG, and wide with the mode all node u except center control nodes o in network that floodBroadcast this MSG message;

(5) in network, except center control nodes o, all node u, according to above-mentioned MSG message, determine patrolling of oneselfCollect neighbors, and adjust the transmitting power of oneself.

2. topology control method according to claim 1, the wherein described network Zhong Chu center control of step (1)The outer all node u of node o processed send a Hello packet to center control nodes o, refer to that node u is with maximumTransmitting power P_maxSend the Hello packet of oneself to center control nodes o in the mode flooding.

3. topology control method according to claim 1, the wherein described center control nodes o of step (1)Obtain peak power topology G_max＝(V(G_max),E(G_max)), obtain as follows:

(1a) in center control nodes o and network, except center control nodes o, all node u formation peak powers are opened upFlutter G_maxNode set V (G_max)；

(1b) in network, except center control nodes o, all node u send oneself Hello to center control nodes oPacket, No. ID, the authentication number and the positional information that in this Hello packet, comprise node u;

(1c) center control nodes o receives the Hello data that in network, all node u of other except own sendBag, and by the information recording in Hello packet in the whole network informational table of nodes of oneself;

(1d) center control nodes o, according to the positional information in the whole network informational table of nodes, calculates any two points in the whole networkX, the Euclidean distance d (x, y) between y;

(1e) center control nodes o is any according to node maximum transmitted radius R and described Euclidean distance d (x, y) judgementThe annexation of point-to-point transmission, obtains peak power topology G_maxLimit set E (G_max)；

(1f) with node set V (G_max) and described limit set E (G_max), jointly form peak power topology:G_max＝(V(G_max),E(G_max))。

4. topology control method according to claim 1, the wherein described center control nodes o of step (3)Build full mesh topology G=(V (G), E (G)), build as follows:

(3a) initialize full mesh topology:

The node set V of full mesh topology G (G) is initialized to all joints in peak power topology by center control nodes oPoint, is initialized to all limits in peak power topology by the limit set E (G) of full mesh topology G;

Center control nodes o sets up set of minimal paths P={p_G(x, y) }, wherein p_G(x, y) is node set V (G)In any a pair of node combination<x, y>shortest path in full mesh topology G, if limit (x, y) ∈ E (G), For empty set, if limitCenter control nodes o calculates based on full mesh topology Gp_G(x,y)；

(3b) center control nodes o gathers in E (G) all limits that were not traversed by degree of disturbance from big to small by limitSort, select the limit of degree of disturbance maximumAnd set up test topology For set of nodeClose the whole nodes in V (G),For limit set E (G) removes limitAfter set, if in limit set E (G)Limit all traveled through, full mesh topology G has built;

(3c) center control nodes o is taking Euclidean distance as weight, and computing node combinesIn test topologyInShortest pathAnd judgementCan meet Euclidean distance t-Spanner characteristic, if can not meet,Center control nodes o returns to step (3b), if can meet, and center control nodes o execution step (3d);

(3d) center control nodes o finds out in set of minimal paths P and uses limitThe combination of all nodes<m, n>, form and quote set: $T_{(x,y)}=\{<m,n>|(\hat{x},\hat{y})\&Element;p(m,n),p(m,n)\&Element;P\},$ If quote set T_(x,y)For empty set, center control nodes o performs step (3f), is not empty set if quote set T, and center control nodes is heldRow step (3e);

(3e) center control nodes o, taking Euclidean distance as weight, calculates and quotes every pair of node combination in set T<m, n>in test topologyIn shortest pathAnd judgementCan meet Euclidean distanceT-Spanner characteristic, if there is at least one pair of node combination not meet, center control nodes o returns to step (3b),All can meet center control nodes o execution step (3f) if quote all node combinations in set T;

(3f) center control nodes o is by limitFrom limit set E (G), leave out, if exist not in limit set E (G)The limit being traversed, center control nodes o upgrades the degree of disturbance on limit in limit set E (G), and node is combinedIn test topologyIn shortest pathAnd quote and gather every couple of node combination < m in T, n in testTopologyIn shortest pathBe updated in set of minimal paths P, and return to step (3b), if limit collectionThe limit of closing in E (G) all traveled through, and full mesh topology G has built.

5. topology control method according to claim 1, the wherein described network Zhong Chu center control of step (5)The outer all node u of node o processed determine the logic neighbors of oneself according to described MSG message, and adjust sending out of oneselfPenetrate power, carry out as follows:

(5a) node u, according to described MSG message, jumps neighbors v as logic neighbour using one on full mesh topology GNode, and form logic neighbors collection: LN_u＝{v∈V(G)|(u,v)∈E(G)}；

(5b) node u, according to logic neighbors, is adjusted into the transmitting power of oneself logic that can cover farthestThe needed minimum power of neighbors? $P_{{\mathrm{LN}}_u}=max\left\{p_{u,v}\right|v\&Element;{\mathrm{LN}}_u\}.$

6. topology control method according to claim 4, wherein said step (3a), step (3c) and stepSuddenly the shortest path in (3e) is by using dijkstra's algorithm or bellman-ford algorithm to calculate.

7. topology control method according to claim 4, the Euclidean distance in wherein said step (3c)T-spanner characteristic, refers to node combinationIn test topologyIn shortest pathMeet

$\left|p_{\hat{G}}(\hat{x},\hat{y})\right|\&le;t\&times;\left|p_{G_{max}}(\hat{x},\hat{y})\right|,$

WhereinFor node combinationIn test topologyIn shortest path length,For node combinationAt peak power topology G_maxIn shortest path length, t is the parameter of t-Spanner characteristicThe factor, value is 2 or 4.