CN103546948A - Method and system for scheduling node sleeping of energy capture sensor network based on graph theory - Google Patents

Abstract

The invention discloses a method and system for scheduling node sleeping of an energy capture sensor network based on a graph theory. The method comprises the steps that the weighted directed graph G equal to (V, S) of the energy capture sensor network is constructed in a monitor area; the weight of each link is calculated according to the weighted directed graph G equal to (V, S); the nodes forming the links are colored according to the weight values of the links, coloring is carried out in the sequence of the weight values of the links, and when the monitor area is completely covered with the colored nodes, the residual nodes are not colored, and the nodes which are not colored are in a sleeping state. The system achieved through the method can adapt to energy capture randomness and the dynamic change of a sensor network topological structure, captured energy can be used more effectively, and the system is suitable for multiple kinds of sensor network types and is more suitable for field monitor and other application environments with no power supplied and batteries hard to replace.

Description

Energy harvesting sensor network nodes dormancy dispatching method and system based on graph theory

Technical field

The present invention relates to sensor network technique field, be specifically related to a kind of energy harvesting sensor network nodes dormancy dispatching method and system based on graph theory.

Background technology

Sensor network is application oriented special wireless self-organization network, in fields such as Smart Home, intelligent transportation, industry monitoring, environmental monitoring, tele-medicine, battlefield surveillances, has potential using value.In sensor network, a large amount of microsensor node deployments are in observation area, image data and data are sent to fusion center in the mode of multi-hop.Sensor node powered battery, it is limited carrying energy, and battery is difficult to change.The characteristic feature of sensor network is energy constraint for this reason, and it is primary goal that the design of various algorithms, agreement all be take energy-conservation, and do the best balance quality and energy consumption.

Energy harvesting sensor network becomes the focus of domestic and international educational circles in recent years.In energy harvesting sensor network, node obtains energy (as solar energy) from the external world, has alleviated to a certain extent the energy constraint of self, has extended network lifecycle.But extraneous retrievable energy is unsettled, not to have in any moment, and can amount to obtain temporal evolution.In addition, the optimization utilization that the efficiency that solar energy transforms and energy spatial cache also have influence on solar energy.This just requires us when design energy is obtained the algorithm of sensor network and agreement, takes into full account the randomness of energy harvesting and the task of sensor network, improves efficiency as far as possible.Energy harvesting sensor network is particularly suitable for wild environment monitoring.When the energy content of battery that carries when sensor node is not enough, node obtains the new forms of energy such as solar energy, wind energy and they is converted into electric energy from the external world.

When sensor network is monitored for wild environment, node dormancy scheduling allows redundant node enter resting state to save energy while making sensor network occur covering redundancy, and all right like this balance node energy consumption, extends network lifecycle.The exemplary process of sensor network nodes dormancy dispatching has the dormancy dispatching method of dormancy dispatching method, the dormancy dispatching method based on nodal distance and jumping figure, the dormancy dispatching method covering based on angle, the dormancy dispatching method based on Voronoi figure, ensuring coverage degree and connectedness based on sub-clustering etc.These methods are the fixing sensor network of nodes oriented energy all, although extended network lifecycle, but the sensor network topological structure dynamic change of not considering the randomness of energy harvesting and causing thus, may cause the improper use of obtained energy.

Summary of the invention

For existing sensor network dormancy dispatching method, be not suitable for the situation of energy harvesting sensor network, the present invention proposes a kind of energy harvesting sensor network nodes dormancy dispatching method based on graph theory, and provide the dormancy dispatching system that realizes the method, adapted to the dynamic change of randomness and the sensor network topological structure of energy harvesting, can more effectively utilize obtained energy, be applicable to multiple sensors network type and be more suitable for that field monitoring etc. cannot be powered and be difficult to change the applied environment of battery.

Technical solution of the present invention is as follows:

An energy harvesting sensor network nodes dormancy dispatching method for graph theory, comprising:

In monitored area, construct the weighted digraph G=(V, S) of energy harvesting sensor network;

According to weighted digraph G=(V, S), calculate the weight of each link;

According to the weighted value of each link, to forming the node of link, carry out painted, painted order is carried out successively according to the weighted value size of each link, when monitored area is covered completely by painted node, remaining node does not carry out painted, and the node not being colored enters resting state;

According to the dormancy dispatching system of setting, start next round dormancy dispatching.

Wherein: V is the set of sensor node, S is the set of link between sensor node.

Described dormancy dispatching system, for when energy harvesting sensor network can not meet monitored area coverage requirement, starts next round dormancy dispatching.

Described link weight is obtained by following formula:

${\mathrm{\λ}}_{\mathrm{ij}}=\frac{{\mathrm{\ϵ}}_i{\left|h_{\mathrm{ij}}\right|}^2}{d_{\mathrm{ij}}^v};$

Wherein: λ _ijnode x _i(i=1,2,3 ...) and node x _jform the weight of link; Node x _jit is link crossed node; ε _inode x _idump energy; h _ijnode x _iwith node x _jbetween channel coefficients; d _ijnode x _iwith node x _jbetween distance; V is constant, v=2-4.

An energy harvesting sensor network nodes dormancy dispatching system for graph theory, comprising:

In monitored area, construct the device of the weighted digraph G=(V, S) of energy harvesting sensor network;

According to weighted digraph G=(V, S), calculate the device of the weight of each link;

According to the weighted value of each link, to forming the node of link, carry out painted, painted order is carried out successively according to the weighted value size of each link, when monitored area is covered completely by painted node, remaining node does not carry out painted, and the node not being colored enters the device of resting state;

According to the dormancy dispatching system of setting, start the device of next round dormancy dispatching.

Accompanying drawing explanation

Fig. 1 is the graph theory model of energy harvesting sensor network.

Embodiment

Below in conjunction with drawings and Examples, the present invention is described in further detail.

1. technical scheme is described:

(1) in energy harvesting sensor network, it is different that node obtains the moment of energy, and the energy value obtaining is also different.If have N node in sensor network, be designated as { x ₁..., x _n.N node x _nprimary power be designated as E _{n, 0}(Jiao).At moment t _n,l(l=1 ..., L), x _nfrom the external world, obtain energy E _n,l(Jiao) (convenient for statement, establishing all nodes here, to obtain moment of energy identical).Every, take turns the moment that dormancy dispatching starts, according to node x _nprimary power and the energy that obtains from the external world dump energy ε that can calculate it _n(Jiao).

(2) data of sensor node collection send to fusion center in multi-hop mode, and the multi-hop Routing Protocol of the formation of link and employing is relevant.Structure weighted digraph G=(V, S), wherein V is the set of sensor node, S is the set of link between sensor node.Only consider the energy consumption of node communication, because the energy consumption of perception, calculating is conventionally much smaller.If two node x _iand x _jbetween there is a link, the weight λ of this link _ijand x _idump energy ε _i(Jiao), x _iand x _jbetween distance d _ij(rice) and x _iand x _jbetween channel coefficients h _ijrelevant, be expressed as λ _ij=f (ε _i, d _ij,h _ij), wherein

be a definite function, v is a constant.Function f () has been included the key parameters such as dump energy, communication distance and channel coefficients, and meets the following conditions: dump energy is more, and weight is larger; Distance is far away, and channel conditions is poorer, and weight is less.Because node obtains energy from the external world, the value of link weight is dynamic change.

(3) according to the painted theory of graph theory, carry out the node dormancy scheduling of energy harvesting sensor network.As a plurality of node x ₁..., x _ithere is link to arrive x simultaneously _jtime, according to the weighted value of these links to node x ₁..., x _icarry out painted.The node of weighted value maximum is colored at first, and then second largest node of weighted value is colored, and the rest may be inferred.When observation area is covered completely by painted node, remaining node is not colored, and enters resting state.Sensor network while starting to next round dormancy dispatching, re-constructs weighted digraph and link generation weight, then according to the painted theory of graph theory, carries out node dormancy scheduling, until can not meet region coverage requirement.

2. use concrete parametric description:

If there are 5 node x in sensor network ₁..., x ₅there is link to arrive node x simultaneously ₆.Their primary power is all 1J, i.e. E _1,0=E _2,0=E _3,0=E _4,0=E _5,0=E _6,0=1.At moment t _{n, 1}, x ₁..., x ₆from the external world, obtain respectively energy 0.1J, 0.2J, 0.3J, 0.4J, 0.5J, 0.6J.Before dormancy dispatching starts, x ₁..., x ₆dump energy be respectively 1.1J, 1.2J, 1.3J, 1.4J, 1.5J, 1.6J.If x ₁..., x ₅to x ₆distance be respectively 1m, 1.5m, 2m, 2.5m, 3m.Channel coefficients is obeyed the multiple Gaussian Profile that average is 0, variance is 1.If | h ₁₆| ²=| h ₂₆| ²=| h ₃₆| ²=| h ₄₆| ²=| h ₅₆| ²=0.5.Link weight is

v=2 wherein.Obtaining thus the node that respective links weight sorts from big to small, is x ₁..., x ₅.When dormancy dispatching starts, first to x ₁carry out painted, more successively to x ₂, x ₃, x ₄carry out painted.If these four nodes cover observation area, so x completely ₅to enter resting state.When next round dormancy dispatching starts, the dump energy according to the energy newly obtaining and when the first round, the work period finished recalculates the dump energy of node, then regenerates x ₁..., x ₅link weight and find out dormancy node according to painted theory.

Claims (6)

1. the energy harvesting sensor network nodes dormancy dispatching method based on graph theory, comprising:

In monitored area, construct the weighted digraph G=(V, S) of energy harvesting sensor network;

According to weighted digraph G=(V, S), calculate the weight of each link;

According to the weighted value of each link, to forming the node of link, carry out painted, painted order is carried out successively according to the weighted value size of each link, when monitored area is covered completely by painted node, remaining node does not carry out painted, and the node not being colored enters resting state;

Wherein: V is the set of sensor node, S is the set of link between sensor node.

2. dormancy dispatching method according to claim 1, wherein said link weight is obtained by following formula:

${\mathrm{\λ}}_{\mathrm{ij}}=\frac{{\mathrm{\ϵ}}_i{\left|h_{\mathrm{ij}}\right|}^2}{d_{\mathrm{ij}}^v};$

Wherein: λ _ijnode x _i(i=1,2,3 ...) and node x _jform the weight of link; Node x _jit is link crossed node; ε _inode x _idump energy; h _ijnode x _iwith node x _jbetween channel coefficients; d _ijnode x _iwith node x _jbetween distance; V is constant, v=2-4.

3. dormancy dispatching method according to claim 1 and 2, also comprises: according to the dormancy dispatching system of setting, start next round dormancy dispatching.

4. dormancy dispatching method according to claim 3, wherein said dormancy dispatching system be for when energy harvesting sensor network can not meet monitored area coverage requirement, beginning next round dormancy dispatching.

5. the energy harvesting sensor network nodes dormancy dispatching system based on graph theory, comprising:

In monitored area, construct the device of the weighted digraph G=(V, S) of energy harvesting sensor network;

According to weighted digraph G=(V, S), calculate the device of the weight of each link;

According to the weighted value of each link, to forming the node of link, carry out painted, painted order is carried out successively according to the weighted value size of each link, when monitored area is covered completely by painted node, remaining node does not carry out painted, and the node not being colored enters the device of resting state;

According to the dormancy dispatching system of setting, start the device of next round dormancy dispatching.

6. dormancy dispatching system according to claim 5, wherein said dormancy dispatching system be for when energy harvesting sensor network can not meet monitored area coverage requirement, beginning next round dormancy dispatching.

Images