CN104378797A - Collaborative awareness node scheduling method for Internet of Things for manufacturing - Google Patents

Abstract

The invention discloses a collaborative awareness node scheduling method for Internet of Things for manufacturing. The collaborative awareness node scheduling method is designed according to the dynamic change features of uncertain events of the Internet of Things for manufacturing and the availability dynamic change characteristics of awareness nodes, and mainly includes a multi-node collaborative awareness model under a probability awareness model, a collaborative awareness scheduling method based on authority nodes, and a node selection strategy based on cost performance. By means of the method, the awareness precision in the complex application environment is effectively improved, and the network lifecycle is effectively prolonged.

Description

A kind of manufacture Internet of Things collaborative perception node scheduling method

Technical field

The present invention relates to Internet of Things field, more specifically, relate to a kind of manufacture Internet of Things collaborative perception node scheduling method.

Background technology

For sensing node isomery distribution in manufacture internet of things deployment scenario, the dynamic change characterization of uncertain event and sensing node availability dynamic change characteristic, the efficient optimization method studied towards the node scheduling Optimization Mechanism of the reliable perception of uncertain event and Fast Convergent is necessary.

Summary of the invention

The object of the invention is a kind of manufacture Internet of Things collaborative perception node scheduling method proposing perceived accuracy original text and network lifecycle length, is the collaborative perception that foundation event location sensing node adaptive scheduling completes to object event.

Technical scheme of the present invention is:

A kind of manufacture Internet of Things collaborative perception node scheduling method, be the collaborative perception that foundation event location sensing node adaptive scheduling completes to object event, method comprises the following steps:

S1. after there is nodal test to event, broadcast, whether each node decision event P is self sensing region;

S2. based on the mode of flooding, continuous exchange message learns the node set S of capable perception around event P _p;

S3. authoritative node s is selected ₀, be responsible for the collaborative perception of scheduling events P;

S4. authoritative node s ₀exchange message in event node set, separately report and incident distance, dump energy, self current task, authoritative node s ₀the capable nodes of event is sorted according to cost performance;

S5. according to cost performance from high in the end, select the collaborative perception value of wherein n node to be greater than threshold value η, n-1 node is then less than threshold value η, using these nodes as cooperative nodes queue.

In the preferred scheme of one, step S3 selects nearest node as authoritative node s according to the distance of event area ₀be responsible for the cooperative scheduling of perception task.

In the preferred scheme of one, in step s 4 which, have the node of perception according to distance to event, dump energy, self task current defines cost performance, and its cost performance defined formula is φ representation node is selected as the cost performance of event perception node; Γ (d, e) represents that node is for event e, and distance is the perceived accuracy of d; E _rfor residue energy of node; T is self other perception task; C is sensing node communications cost; The weights that α, β are arranged according to demand.

In the preferred scheme of one, in step s 5 for the selection of cooperative nodes, choose the node that wherein cost performance is the highest and work in coordination with.

The invention has the beneficial effects as follows: the one proposed manufactures Internet of things node collaborative perception dispatching method, according to task current location, dump energy, current task and perception select node to work in coordination with.According to probability sensor model, research multi-node collaborative sensor model.In conjunction with node perceived distance, the cost performance that dump energy and current task carry out cost performance sequencing selection the highest makes network task energy consumption balance, prolong network lifetime.Additionally by the setting of perception probability threshold value, guarantee the perceived accuracy of multi-node collaborative perception, meet the monitoring requirements to uncertainty event.Compared with existing reliable perception dispatching method, its perceived accuracy and network lifecycle all have a clear superiority in.

Accompanying drawing explanation

Fig. 1 is random node distribution schematic diagram of the present invention.

Fig. 2 is random node collaborative perception precision comparison figure of the present invention.

Life cycle comparison diagram during the perception of Fig. 3 network cooperating.

Embodiment

Below in conjunction with accompanying drawing, the present invention will be further described, but embodiments of the present invention are not limited to this.Embodiment 1

For manufacturing industry Internet of Things application characteristic, there is unreliability in sensing node, and primary study multinode is to the collaborative perception dispatching method of uncertainty event.The present invention mainly includes: the multi-node collaborative sensor model under probability sensor model, based on the collaborative perception dispatching method of authoritative node, based on the sensor selection problem strategy of cost performance.

Collaborative covering method is the angle from application demand, considers the fusion of node perceived probability, by the sensing results of comprehensive multiple node, provides the monitoring result of monitored impact point, makes it meet the monitoring perception probability requirement of application.Here the perception probability of our counterpart node introduces the false dismissal probability of node, is defined as

m(i,j)＝1-c(i,j) (1)

Here c (i, j) represents the perception probability of node i to monitored target j.Like this when multiple node is monitored impact points some in area to be monitored simultaneously, according to the thought of the common factor in probability, now network can be defined as the collaborative loss of this impact point

$m_c\left(i\right)=\underset{j\∈S}{\mathrm{\Π}}m(i,j)=\underset{j\∈S}{\mathrm{\Π}}(1-c(i,j))---\left(2\right)$

Wherein S is the set of all nodes in area to be monitored, supposes that the perception of node to target area is separate simultaneously.Formula is known thus, can by the collaborative reduction of multiple node to the loss of impact point.

In probabilistic model, the probability that target is arrived by nodal test is relevant with the distance between target and node and direction, and the Mathematical Modeling that the present embodiment adopts is general probability model, and monitoring point j is placed on an i place type, and to be the probability that k sensor node detects be:

$p_{\mathrm{ijk}}=e^{-{\mathrm{\α}}_k{d}_{\mathrm{ij}}}---\left(3\right)$

Wherein α _kfor the parameter that type is k transducer, its value depends on the type of transducer.D _ijfor the Euclidean distance between an i and some j.Point j fails to be placed on i point place type that to be the probability that k sensor node detects be:

Point can not be by the probability of all the sensors nodal test:

$p^{\′}=\underset{i=1}{\overset{N}{\mathrm{\Π}}}\underset{k=1}{\overset{K}{\mathrm{\Π}}}(1-x_{\mathrm{ik}}{p}_{\mathrm{ijk}})\≤L---\left(5\right)$

L is take the logarithm and carry out conversion and can obtain in above formula both sides by probability upper bound 0<L<1.

$\underset{i=1}{\overset{N}{\mathrm{\&Pi;}}}\underset{k=1}{\overset{K}{\mathrm{\&Pi;}}}\ln (1-x_{\mathrm{ik}}{p}_{\mathrm{ijk}})\&le;\ln L---\left(6\right)$

Due to so can be changed into

$\underset{i=1}{\overset{N}{\mathrm{\&Pi;}}}\underset{k=1}{\overset{K}{\mathrm{\&Pi;}}}\ln (1-p_{\mathrm{ijk}})x_{\mathrm{ik}}\&le;\ln L---\left(7\right)$

Then under probabilistic model:

a _ijk＝-ln(1-p _ijk)，b _j＝-ln L (8)

As i=j, p _ijk=p _iik=1, (1-p _ijk)=0, logarithm is meaningless, makes p _ijk=0.999.

Upon detecting an event, the node definition nearest apart from event is authoritative node, its other node collaborative perception of scheduling surroundings nodes.

1., if when the enough resource and competence of authoritative node can carry out effective perception to event, now collaborative perception deteriorates to single perception, namely only authoritative node carries out perception to object event.

If authoritative node cannot effectively perception time, scheduling surroundings nodes is needed to carry out collaborative perception to the object time, and the task of surroundings nodes own, energy, the combined factors such as time gap are considered, collaborative perception maximizing efficiency and network lifecycle are maximized.

Think that node is all selfish, namely when working in coordination with without other nodes or machine-processed scheduling events surroundings nodes, node only perception from self week recent events information, even if or gather other and also can not transmit or process away from event information.Therefore collaborative dispatch mechanism is needed to complete reliable perception to event.

Collaborative perception thought is around event, have several have perception sensing node to event, selects one of them as authoritative node, and its leader as collaborative perception is responsible for scheduling surroundings nodes and works in coordination with object event perception.Need to consider to make to ensure that the perceived accuracy of overall coordination perception events is greater than threshold value η when collaborative, also need to consider nodal distance incident distance d when secondly selecting node to work in coordination with, distance is relevant with perception energy consumption with perceived accuracy; Need to consider other perception tasks of sensing node self T; Need to consider sensing node dump energy E _r; And sensing node communications cost C.

The sequence of comprehensive above consideration node cost performance, then by cumulative perception probability summation from high in the end, stops when perception probability meets threshold value.

Define about cost performance φ formula:

$\mathrm{\&phi;}=\frac{\mathrm{\&Gamma;}(d,e)+E_r}{(E_r-\mathrm{\&alpha;T})+\mathrm{\&beta;C}}---\left(9\right)$

φ representation node is selected as the cost performance of event perception node; Γ (d, e) represents that node is for event e, and distance is the perceived accuracy of d; E _rfor residue energy of node; T is self other perception task; C is sensing node communications cost; The weights that α, β are arranged according to demand.

Collaborative perception node scheduling method:

Input: event coordinates P (x, y)

Export: cooperative nodes queue S _c

1. whether each node decision event P is self sensing region;

2. if so, then node broadcasts towards periphery;

3. constantly exchange message learns the node set S of capable perception around event P _p;

4. select node nearest in wherein event area as authoritative node s ₀, be responsible for the collaborative perception of scheduling events P;

5.s ₀exchange message in event node set, separately report and incident distance, dump energy, self current task;

6. authoritative node sorts according to cost performance to the capable nodes of event;

7. according to cost performance from high in the end, select the collaborative perception value of wherein n node to be greater than threshold value η, n-1 node is then less than threshold value η, if by these node cooperative nodes queues S _c;

8. return cooperative nodes queue S _c;

The one that the present embodiment proposes manufactures Internet of things node collaborative perception dispatching method, and according to task current location, dump energy, current task and perception select node to work in coordination with.According to probability sensor model, research multi-node collaborative sensor model.In conjunction with node perceived distance, the cost performance that dump energy and current task carry out cost performance sequencing selection the highest makes network task energy consumption balance, prolong network lifetime.Additionally by the setting of perception probability threshold value, guarantee the perceived accuracy of multi-node collaborative perception, meet the monitoring requirements to uncertainty event.Compared with existing reliable perception dispatching method, its perceived accuracy and network lifecycle all have a clear superiority in.

Above-described embodiments of the present invention, do not form limiting the scope of the present invention.Any amendment done within spiritual principles of the present invention, equivalent replacement and improvement etc., all should be included within claims of the present invention.

Claims (4)

1. manufacture an Internet of Things collaborative perception node scheduling method, be the collaborative perception that foundation event location sensing node adaptive scheduling completes to object event, it is characterized in that, said method comprising the steps of:

S1. after there is nodal test to event, broadcast, whether each node decision event P is self sensing region;

S2. based on the mode of flooding, continuous exchange message learns the node set S of capable perception around event P _p;

S3. authoritative node s is selected ₀, be responsible for the collaborative perception of scheduling events P;

S4. authoritative node s ₀exchange message in event node set, separately report and incident distance, dump energy, self current task, authoritative node s ₀the capable nodes of event is sorted according to cost performance;

S5. according to cost performance from high in the end, select the collaborative perception value of wherein n node to be greater than threshold value η, n-1 node is then less than threshold value η, using these nodes as cooperative nodes queue.

2. Internet of Things collaborative perception node scheduling method according to claim 1, is characterized in that, step S3 selects nearest node as authoritative node s according to the distance of event area ₀be responsible for the cooperative scheduling of perception task.

3. Internet of Things collaborative perception node scheduling method according to claim 1, is characterized in that, in step s 4 which, have the node of perception according to distance to event, dump energy, self task current defines cost performance, and its cost performance defined formula is φ representation node is selected as the cost performance of event perception node; Γ (d, e) represents that node is for event e, and distance is the perceived accuracy of d; E _rfor residue energy of node; T is self other perception task; C is sensing node communications cost; The weights that α, β are arranged according to demand.

4. Internet of Things collaborative perception node scheduling method according to claim 1, is characterized in that, in step s 5 for the selection of cooperative nodes, chooses the node that wherein cost performance is the highest and works in coordination with.