CN106412935A - Network topological structure establishing method based on complex network theory - Google Patents

Abstract

The invention discloses a network topological structure establishing method based on a complex network theory. The method comprises the following steps: 1) finding all common nodes in the communication radius of a sink node and calculating selection probability of the common nodes; 2) selecting a plurality of common nodes, the probability values of which are maximum, from the selection probability of the common nodes, and connecting the plurality of common nodes and the sink node to form an initial topological structure; 3) selecting a to-be-added node, finding all common nodes within the communication radius of the to-be-added node and in the initial topological structure, and calculating selection probability thereof; 4) selecting the common node, the probability value of which is maximum, from the selection probability of the common nodes, and connecting the selected common node and the to-be-added node to form an updated topological structure; and 5) taking the updated topological structure as the initial topological structure, and carrying out the step 3) until all common nodes are added to the topological structure of a wireless sensor network. The method can establish an energy-efficient network topological structure, and thus energy utilization rate of the nodes is improved and the service life of a network is prolonged.

Description

A kind of network topology structure method for building up based on Complex Networks Theory

Technical field

The invention belongs to communication technical field is and in particular to a kind of opened up based on the wireless sensor network of Complex Networks Theory Flutter structure method for building up.

Background technology

Wireless sensor network is a kind of self-organizing, distributed being made up of sink node and substantial amounts of sensor node Network system, can be used for the fields such as military surveillance, environment measuring, medical monitoring, urban traffic control, warehousing management.Sensor The volume of node is small, typically battery-powered, leads to that the energy that it carries is very limited, and sensor node is often by cloth simultaneously Put in rugged environment or no man's land, be difficult to node is changed after sensor node energy expenditure, thus section Point is easy to cause inefficacy because of depleted of energy and environmental disruption.Wireless sensor network is related to multidisciplinary height intersection, current Receive much concern in the world, how to build network topology structure, improve fault-tolerance and the life cycle of network, the always heat of research Point.

Due to the topological structure of wireless sensor network and the similarity of complex network, much research dividing complex network Analysis method is applied in the middle of network topology structure of wireless sensor.When Barabasi et al. studies to WWW topological structure Find that power law distribution is obeyed in node degree (connection number) distribution, in network, only have the node of minority to have extremely many connections, mostly Several sections of points only have minimal amount of connection, and this characteristic being referred to as uncalibrated visual servo has very strong fault-tolerance to random fault, sees literary composition Offer Barabasi A L, Albert R.Emergence of Scaling in Random Networks [J] .Science, 1999,286(5439):509-512；On this basis, Jian et al. constructs the wireless sensor network topology of Energy Efficient Structure, the probability that the new node adding is connected with the node having existed in network, in addition to the degree depending on node, is gone back Including the dump energy of node, see document JianY, Liu E, WangY, et al.Scale-free model for wireless sensor networks[J].2013:2329-2332；Peng et al. gives the wireless biography based on complex network Sensor network cluster dividing scheme, while schemes synthesis consider degree and the dump energy of node, combines the characteristic of sensor network, Further increase the capacity usage ratio of node, improve Network morals, see document Peng H, Si S, Shen X S, et al.Energy-efficient and scalable clustering scheme for wireless sensor networks[C].International Conference on Wireless Communications&Signal Processing.IEEE,2015.

For the topological structure building wireless sensor network, the existing research method based on complex network can be effective Ground improve node capacity usage ratio, but in the distant scene of the distance between sensor node, energy expenditure due to The change of data transmission distance sharply increases greatly, and the life cycle of node can greatly shorten, and existing research approach is in this scene Lower effect is no longer obvious.

Content of the invention

The present invention is the weak point overcoming in the presence of above-mentioned prior art, propose a kind of based on Complex Networks Theory Network topology structure method for building up, the wireless sensor network to setting up Energy Efficient in the case of a kind of telecommunication is opened up Flutter structure, thus improving capacity usage ratio and the Network morals of node.

For reaching above-mentioned purpose, the technical solution used in the present invention is：

A kind of feature of the network topology structure method for building up based on Complex Networks Theory of the present invention is to be applied to by one In the wireless sensor network that the ordinary node of sink node and N number of random distribution is constituted, with described wireless sensor network Any one summit of boundary rectangle, is set up respectively as X-axis and Y-axis as initial point O, two sides adjacent with described initial point O Coordinate system XOY；Described network topology structure method for building up is to carry out as follows：

Step 1, according to described sink node the coordinate position in described coordinate system XOY, find described sink node exist All n ordinary nodes in self communication radius, and in sink node communication radius any i-th is calculated using formula (1) The select probability ∏ of individual ordinary node_i-sink, thus obtaining the select probability { ∏ of n ordinary node_1-sink,П_2-sink..., ∏_i-sink..., ∏_n-sink}：

In formula (1), E_jRepresent the residual energy of any j-th ordinary node in self communication radius for the described sink node Amount；E_iRepresent the dump energy of any i-th ordinary node in self communication radius for the described sink node；D_jRepresent arbitrarily The distance of the coordinate position of j-th ordinary node and described sink node；D_iRepresent any i-th ordinary node and described sink The distance of the coordinate position of node；A and b represents the weight of dump energy and distance respectively, and has a+b=1；1≤i≤n ＜ N, 1 ≤ j≤n ＜ N；

Step 2, from the select probability { ∏ of described n ordinary node_1-sink,∏_2-sink..., ∏_i-sink..., П_n-sink} The maximum ordinary node corresponding to front m select probability of middle select probability value is attached with described sink node, thus shape Become initial primary topology；

Step 3, randomly choose l-th ordinary node in described initial primary topology, in described l-th ordinary node Select any k-th ordinary node as pre-add ingress in communication radius, find in the communication radius of described pre-add ingress, And be concurrently present in all s ordinary nodes in described initial primary topology, and obtain the logical of pre-add ingress using formula (2) The select probability ∏ of c-th ordinary node in letter radius_k-c, thus obtaining the select probability { ∏ of s ordinary node_k-1, ∏_k-2..., ∏_k-c..., ∏_k-s}：

In formula (2), E_fRepresent in the communication radius of described pre-add ingress, and be concurrently present in described initial primary topology In any f-th ordinary node dump energy；E_cRepresent described s node in self communication radius, and exist simultaneously The dump energy of any c-th ordinary node in described initial primary topology；D_fRepresent any f-th ordinary node and institute State the distance of the coordinate position of pre-add ingress；D_cRepresent the coordinate bit of any c-th ordinary node and described pre-add ingress The distance put；d_fRepresent the degree of any f-th ordinary node；d_cRepresent the degree of any c-th ordinary node；A ', b ' and c ' are respectively Represent dump energy, the weight away from divergence factor, and have a '+b '+c '=1；1≤k≤N-m, 1≤s≤m, 1≤l≤m, 1≤c≤ S, 1≤f≤s；

Step 4, from the select probability { ∏ of described s ordinary node_k-1,∏_k-2..., ∏_k-c..., ∏_k-sMiddle selection is Ordinary node corresponding to greatest is connected with described pre-add ingress, thus forming the topological structure after renewal；

Step 5, using the topological structure after described renewal as initial primary topology, and after the value of m increases by 1, return to step 3 Execution, till described N number of ordinary node is all added in the topological structure of wireless sensor network.

Compared with the prior art, the present invention has the beneficial effect that：

1st, the present invention is based on BA model, has considered wireless sensor network during natural increase and optimum selecting Node communication radius are limited, and the characteristic such as energy constraint is it is adaptable to the foundation of wireless senser topological structure, and can improve section The capacity usage ratio of point and Network morals.

2nd, the select probability formula that the present invention is given adds distance factor, in other conditions identical feelings when node connects Under condition prioritizing selection distance relatively closer to node, it is to avoid node energy consumes excessively because of the increase of communication distance, therefore may be used Larger for node communication radius, distant scene.

3, the invention belongs to self-organizing network, realize simple and effect are obvious, can be used for military surveillance, environment measuring, doctor Treat the structure of the dual-use sensor networks such as monitoring, urban traffic control, warehousing management.

Brief description

Fig. 1 is the network topology structure Establishing process figure of the present invention；

Fig. 2 is that the pre-add ingress of the present invention adds network development process schematic diagram.

Specific embodiment

In the present embodiment, a kind of network topology structure method for building up based on Complex Networks Theory, it is applied to by one Ordinary node in the wireless sensor network that the ordinary node of sink node and N number of random distribution is constituted, to N number of random distribution Numbering, label is from 1,2 ..., N, and the primary power of ordinary node is E, any one with the boundary rectangle of wireless sensor network Individual summit sets up coordinate system XOY as initial point O, two sides adjacent with initial point O respectively as X-axis and Y-axis；Referring to Fig. 1, net Network topological structure method for building up is to carry out as follows：

Step 1, according to sink node the coordinate position in coordinate system XOY, find sink node in self communication radius Interior all n ordinary nodes, sink node sends a small request signal to the node in communication radius, and receiving please The numbering by itself for the ordinary node of signal, dump energy, co-ordinate position information is asked to be sent to sink node by feeding back signal, Sink node is ranked up according to the size of numbering from small to large to ordinary node, and is calculated logical in sink node using formula (1) The select probability ∏ of any i-th ordinary node in letter radius_i-sink, thus obtaining the select probability of n ordinary node {∏_1-sink,∏_2-sink..., ∏_i-sink..., П_n-sink}：

In formula (1), E_jRepresent the dump energy of any j-th ordinary node in self communication radius for the sink node；E_i Represent the dump energy of any i-th ordinary node in self communication radius for the sink node；D_jRepresent any j-th common The distance with the coordinate position of sink node of node；D_iRepresent the coordinate position of any i-th ordinary node and sink node Distance；A and b represents the weight of dump energy and distance respectively, and has a+b=1；1≤i≤n ＜ N, 1≤j≤n ＜ N；Using The transmission energy consumption of the difference calculate node of formula (1-1) and formula (1-2) and reception energy consumption：

E_Rx(K)=KE_elec(1-2)

In formula (1-1) and formula (1-2), K represents bit number, E_elecIndication circuit energy consumption, ε_fsAnd ε_mpRepresent two kinds respectively Amplifier energy consumption under pattern.

Step 2, from the select probability { ∏ of n ordinary node_1-sink,∏_2-sink..., ∏_i-sink..., ∏_n-sinkMiddle choosing Select the maximum ordinary node corresponding to front m select probability of probit to be attached with sink node, thus formed initially opening up Flutter structure；

Step 3, referring to Fig. 2, initial primary topology is that 2,16,22 ordinary node forms by numbering, according to numbering by little To big order sequence, initial primary topology randomly chooses l-th ordinary node, l is equal to 1 here, selection common Node is the ordinary node that numbering is equal to 2, selects any k-th ordinary node to make in the communication radius of l-th ordinary node For pre-add ingress, equal to 2, pre-add ingress is the node that numbering is 23 to k, finds the communication radius of pre-add ingress here Interior, and be concurrently present in all s ordinary nodes in initial primary topology, refer herein to number and be two of 2,16 and common save Point, and the select probability ∏ of c-th ordinary node in the communication radius of pre-add ingress is obtained using formula (2)_k-c, thus obtaining s Select probability { the ∏ of individual ordinary node_k-1,∏_k-2..., Π_k-c..., ∏_k-s}：

In formula (2), E_fRepresent in the communication radius of pre-add ingress, and be concurrently present in any in initial primary topology The dump energy of f-th ordinary node；E_cRepresent s node in self communication radius, and be concurrently present in initial topology knot The dump energy of any c-th ordinary node in structure；D_fRepresent the coordinate bit of any f-th ordinary node and pre-add ingress The distance put；D_cRepresent the distance of any c-th ordinary node and the coordinate position of pre-add ingress；d_fRepresent any f-th general The degree of logical node；d_cRepresent the degree of any c-th ordinary node；A ', b ' and c ' represent dump energy, the power away from divergence factor respectively Weight, and have a '+b '+c '=1；1≤k≤N-m, 1≤s≤m, 1≤l≤m, 1≤c≤s, 1≤f≤s；

Step 4, from the select probability { ∏ of s ordinary node_k-1,∏_k-2..., ∏_k-c..., П_k-sIn select the most general Ordinary node corresponding to rate value is connected with pre-add ingress, thus forming the topological structure after renewal；

, as initial primary topology, the value of m increases by 1, and return to step 3 executes for step 5, the topological structure after to update, Till N number of ordinary node is all added in the topological structure of wireless sensor network；

Primary network topological structure is set up after completing, and enters data collection and merges the transmission stage, it is big that each connects number Ordinary node in 1 is that all ordinary nodes being attached with it as pre-add ingress distribute time slot, and sensor node is adopted The data collecting is eventually sent to sink node through data fusion in the form of multi-hop, treat the data of transmission in need all arrive After reaching sink node, disconnect all connections, recurrent network topological structure sets up process, carry out network reconfiguration, maximize and extend Network lifecycle.

Claims (1)

1. a kind of network topology structure method for building up based on Complex Networks Theory, is characterized in that being applied to by a sink node In the wireless sensor network constituting with the ordinary node of N number of random distribution, with the boundary rectangle of described wireless sensor network Any one summit as initial point O, coordinate system, respectively as X-axis and Y-axis, is set up in two sides adjacent with described initial point O XOY；Described network topology structure method for building up is to carry out as follows：

Step 1, according to described sink node the coordinate position in described coordinate system XOY, find described sink node at itself All n ordinary nodes in communication radius, and it is general to calculate in sink node communication radius any i-th using formula (1) The select probability ∏ of logical node_i-sink, thus obtaining the select probability { ∏ of n ordinary node_1-sink,∏_2-sink..., ∏_i-sink..., ∏_n-sink}：

${\mathrm{\Π}}_{i-\sin k}=a\frac{E_i}{{\mathrm{\Σ}}_{j=1}^n{E}_j}+b\frac{\frac{1}{D_i}}{{\mathrm{\Σ}}_{j=1}^n\frac{1}{D_j}}---\left(1\right)$

In formula (1), E_jRepresent the dump energy of any j-th ordinary node in self communication radius for the described sink node；E_i Represent the dump energy of any i-th ordinary node in self communication radius for the described sink node；D_jRepresent any j-th The distance of the coordinate position of ordinary node and described sink node；D_iRepresent any i-th ordinary node and described sink node Coordinate position distance；A and b represents the weight of dump energy and distance respectively, and has a+b=1；1≤i≤n ＜ N, 1≤j ≤ n ＜ N；

Step 2, from the select probability { ∏ of described n ordinary node_1-sink,∏_2-sink..., ∏_i-sink..., ∏_n-sinkMiddle choosing Select the maximum ordinary node corresponding to front m select probability of probit to be attached with described sink node, thus being formed just Beginning topological structure；

Step 3, randomly choose l-th ordinary node in described initial primary topology, in the communication of described l-th ordinary node Select any k-th ordinary node as pre-add ingress in radius, find in the communication radius of described pre-add ingress, and with When be present in all s ordinary nodes in described initial primary topology, and obtain the communication half of pre-add ingress using formula (2) The select probability ∏ of c-th ordinary node in footpath_k-c, thus obtaining the select probability { Π of s ordinary node_k-1,∏_k-2..., ∏_k-c..., ∏_k-s}：

${\mathrm{\Π}}_{k-c}=a^{\′}\frac{E_c}{{\mathrm{\Σ}}_{f=1}^s{E}_f}+b^{\′}\frac{\frac{1}{D_c}}{{\mathrm{\Σ}}_{f=1}^s\frac{1}{D_f}}+c^{\′}\frac{d_c}{{\mathrm{\Σ}}_{f=1}^s{d}_f}---\left(2\right)$

In formula (2), E_fRepresent in the communication radius of described pre-add ingress, and be concurrently present in described initial primary topology The dump energy of any f-th ordinary node；E_cRepresent described s node in self communication radius, and be concurrently present in institute State the dump energy of any c-th ordinary node in initial primary topology；D_fRepresent that any f-th ordinary node is pre- with described Add the distance of the coordinate position of node；D_cRepresent any c-th ordinary node and the coordinate position of described pre-add ingress Distance；d_fRepresent the degree of any f-th ordinary node；d_cRepresent the degree of any c-th ordinary node；A ', b ' and c ' represent respectively Dump energy, the weight away from divergence factor, and have a '+b '+c '=1；1≤k≤N-m, 1≤s≤m, 1≤l≤m, 1≤c≤s, 1≤ f≤s；

Step 4, from the select probability { П of described s ordinary node_k-1,П_k-2..., П_k-c..., П_k-sIn select the most general Ordinary node corresponding to rate value is connected with described pre-add ingress, thus forming the topological structure after renewal；

Step 5, using the topological structure after described renewal as initial primary topology, and after the value of m increases by 1, return to step 3 is held OK, till described N number of ordinary node is all added in the topological structure of wireless sensor network.