CN103533573A - Wireless sensor network for monitoring greenhouse environment and energy-saving transmission method thereof - Google Patents

Abstract

The invention discloses a wireless sensor network for monitoring the greenhouse environment and an energy-saving transmission method thereof. The method is characterized in that data preprocessing is carried out on a sensing node by adopting an event driving model, and data fusion of multiple sensors is carried out on an aggregation node by adopting a support-degree model; i.e., the sensing node carries out accumulated transmission or hopping real-time transmission on acquired greenhouse environment data according to a set threshold, and on the premise of meeting the real-time processing of an event, the data transmission quantity of the sensing node is reduced. After the sensing node transmits the data to the aggregation node, the aggregation node calculates out optimal weight of all data according to the support degree among the sensor data and finally calculates out a fusion value of the data. The wireless sensor network disclosed by the invention has the advantages that a reliable basis is provided for decision of the system, the accuracy of control is ensured and energy-saving transmission of the system is realized.

Description

A kind of greenhouse monitoring wireless sensor network and energy-conservation transmission method thereof

Technical field

The invention belongs to industrialized agriculture greenhouse observation and control technology field, be specifically related to a kind of energy-conservation transmission method of greenhouse wireless sensor device network.

Background technology

Wireless sensor network integrates sensor technology, micro electro mechanical system (MEMS) technology, wireless communication technology, embedded computing technique and distributed information processing, for people provide the approach of a kind of brand-new obtaining information, process information.The application of wireless sensor network in greenhouse can be obtained efficiently and in real time and be made substance environment and Crop Information, effectively reduce manpower consumption, accelerate to advance modernization.Due to wireless sensor network by the limited sensor node of one group of function with the Ad hoc mode extensive perception task that cooperated; It is limited and difficult supplementary that each sensor node carries the energy content of battery.Therefore the main purpose of data transmission policies research is rationally to use the finite energy resource of wireless sensor network node, obtains the information of the required accuracy of user, the life cycle of the whole wireless network of maximized prolongation with minimum energy consumption cost.

Greenhouse environment parameter has the advantages that to change slowly, on time and space, have larger redundancy, under normal circumstances, can there is not saltus step phenomenon in the perception data of sensor node in close several collection period, therefore can be by sensing node data be carried out to preliminary treatment, to reduce the volume of transmitted data of node, reach and reduce node energy consumption, the object that extends node life cycle.Simultaneously because data preliminary treatment meeting causes the loss of data, and in data transmission procedure, may occur the situations such as packet loss of data, be the accuracy that improves Monitoring Data, conventionally adopts Fusion to obtain optimal value.This research, for the feature of environmental information in greenhouse, proposes the energy-conservation data transfer model of data of the greenhouse monitoring wireless sensor network based on event-driven and support fusion.

Summary of the invention

For realizing the energy-conservation transmission object of greenhouse wireless sensor device network system, set up greenhouse WSN Energy-saving Data transmission structure as shown in Figure 1, mainly comprise the preliminary treatment of sensing node data and aggregation node data fusion two parts.

The energy-conservation transmission method of greenhouse wireless sensor device network of the present invention, is to adopt event-based model to carry out data preliminary treatment at sensing node, at aggregation node, adopts support model to carry out the data fusion of multisensor.

For sensing node event-based model, carry out data preliminary treatment, be that sensing node accumulates transmission or saltus step real-time Transmission according to the threshold value of setting to the greenhouse data that collect, meeting under the prerequisite of event processing in real time, reducing the volume of transmitted data of sensing node.Sensing node transmits data to after aggregation node, and aggregation node, according to the degree of support between each sensing data, calculates the optimal weights of each data, finally calculates the fusion value of data.

The invention also discloses the greenhouse monitoring wireless sensor network that utilizes said method, comprise some transducers, sensing node, aggregation node, wherein:

Sensing node, adopts event-based model to carry out data preliminary treatment for the environmental data that transducer is collected, and transmits data to aggregation node; Aggregation node adopts support model to carry out the data fusion of multisensor after receiving the data of sensing node transmission.

Described employing event-based model carries out data preliminary treatment, refer to that sensing node accumulates transmission or saltus step real-time Transmission according to the threshold value of setting to the greenhouse data that collect, meeting under the prerequisite of event processing in real time, reducing the volume of transmitted data of sensing node.

Described employing support model carries out the data fusion of multisensor, is that sensing node transmits data to after aggregation node, and aggregation node, according to the degree of support between each sensing data, calculates the optimal weights of each data, finally calculates the fusion value of data.

The present invention utilizes the energy-conservation transmission of the greenhouse monitoring wireless sensor network of event-driven and support fusion.For greenhouse environment parameter, have and change slowly, on time and space, have the feature of larger redundancy and the perception of greenhouse wireless sensor device network and transmission double-layer structure, carry out respectively the preliminary treatment of sensing node data and aggregation node data fusion.The present invention can provide reliable basis for the decision-making of system, guarantees the accuracy of control, realizes the energy-conservation transmission of system.

Accompanying drawing explanation

Fig. 1 is greenhouse wireless sensor device network energy-saving data transmission method schematic diagram.

Embodiment

1, based on the preliminary treatment of event driven sensing node data

The preliminary treatment of sensing node data belongs to the rudimentary integration technology that wireless sensor network data merges, and fusion results, directly to aggregation node transmission, is the basis that follow-up data merges.Data perception node for wireless sensor network, set two and processed threshold value: ε and τ, wherein ε defines the difference between perception data, when the difference in the i time data gathering and i-1 data that gather before surpasses threshold epsilon, just think that jumping phenomenon appears in data, for guaranteeing the real-time monitoring requirement of greenhouse data, need to carry out saltus step transmission; τ has set the times of collection of sensing node, and in τ collection period, data are not undergone mutation, and this secondary data and front carrying out for i-1 time are transferred to aggregation node after mean value computation.Based on event driven sensing node preprocessing algorithms, specific as follows:

(1) establish a _1jthe 1st data for node j collection.

(2) every collection one secondary data a _ij, respectively with the data a of first i-1 time _1j, a _2j..., a _i-1jcompare, when | a _ij-a _nj| > ε, (n=1,2 ..., in the time of i-1), sensor node will send current accidental data a in time _ij.

(3), when times of collection surpasses τ time, calculate the average of a current i data and send

value to aggregation node.

(4) repeat (1)～(3).

2, aggregation node data fusion method

If i the transducer in the t moment and the greenhouse data of j transducer are respectively X through preliminary treatment _iand X (t) _j(t) (i, j=1,2,3 ..., N), the support that defines two measurement value sensors is

$\sup (a,b)=D(a,b,K,\mathrm{\β})=\frac{K}{1+\mathrm{\β}|X_i\left(t\right)-X_j\left(t\right)||X_i\left(t\right)-X_j\left(t\right)||X_i\left(t\right)-X_j\left(t\right)||X_i\left(t\right)-X_j\left(t\right)|}$ According to above formula, can build Support matrix γ

$\mathrm{\&gamma;}=\left[\begin{array}{cccc}{\mathrm{\&gamma;}}_{11}& {\mathrm{\&gamma;}}_{12}& ...& {\mathrm{\&gamma;}}_{1N}\\ {\mathrm{\&gamma;}}_{21}& {\mathrm{\&gamma;}}_{22}& ...& {\mathrm{\&gamma;}}_{2N}\\ ...& ...& ...& ...\\ {\mathrm{\&gamma;}}_{\mathrm{<figure-callout\; id="1"\; label="N"\; filenames="HDA0000391184760000011.png"\; state="\{\{state\}\}">N</figure-callout>}1}& {\mathrm{\&gamma;}}_{N2}& ...& {\mathrm{\&gamma;}}_{\mathrm{NN}}\end{array}\right]$

Describe the mutual degree of support between each transducer, the support summation that i transducer obtains other all the sensors is

$T\left(X_i\left(t\right)\right)=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{\&gamma;}}_{\mathrm{ij}}$

According to calculating the transducer support summation T (X obtaining _i(t) the optimum fusion weight that) can obtain i transducer i is suc as formula being

ω _i＝1+T(X _i(t))

Greenhouse environment information optimal estimation value after merging is

$X\left(t\right)=\frac{\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}({\mathrm{\&omega;}}_i\&times;X_i\left(t\right))}{\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{\&omega;}}_i}$

Implementation result

The data transmission method of the greenhouse monitoring wireless sensor network merging based on event-driven and support that utilizes that the present invention sets up has carried out verification experimental verification in certain greenhouse, obtains following effect:

(1) with interval 30S timing transmission method comparison, within a period of time between 17:30 to 18:37, the data transmission times of each data threshold preliminary treatment node will be significantly less than the data transmission times of non-processor node, regularly transmission is for transmission 136 times, the transmission method driving based on the time transmits 18 times, reduce transmission 114 times, account for 83.8% of total the number of transmissions.Illustrate that threshold value preconditioning technique can effectively reduce the volume of transmitted data of sensing node, reach the effect that extends network lifecycle.

(2) result that the support of aggregation node merges shows: support function can, according to the fusion weighted value of the correlation degree reasonable distribution data between data, make fusion results lay particular emphasis on the data that the degree of association is larger.When the relative difference between each node data is larger, the fusion value based on support function is greater than mean value, and data result is compared with initial data closer to both larger data of the degree of association.When each node data is worth relative difference hour, the fusion value based on support function is identical with mean value size, plays the effect of averaging.

Claims (6)

1. an energy-conservation transmission method for greenhouse monitoring wireless sensor network, is characterized in that adopting event-based model to carry out data preliminary treatment at sensing node, at aggregation node, adopts support model to carry out the data fusion of multisensor.

2. method according to claim 1, it is characterized in that, described sensing node adopts event-based model to carry out data preliminary treatment, refer to that sensing node accumulates transmission or saltus step real-time Transmission according to the threshold value of setting to the greenhouse data that collect, meeting under the prerequisite of event processing in real time, reducing the volume of transmitted data of sensing node.

3. according to claim, require the method described in 1, it is characterized in that, described data fusion of carrying out multisensor at aggregation node employing support model, that sensing node transmits data to after aggregation node, aggregation node is according to the degree of support between each sensing data, calculate the optimal weights of each data, finally calculate the fusion value of data.

4. a greenhouse monitoring wireless sensor network that utilizes method described in claim 1, comprise some transducers, sensing node, aggregation node, it is characterized in that, described sensing node, for the environmental data that transducer is collected, adopt event-based model to carry out data preliminary treatment, and transmit data to aggregation node; Aggregation node adopts support model to carry out the data fusion of multisensor after receiving the data of sensing node transmission.

5. wireless sensor network according to claim 4, it is characterized in that: described employing event-based model carries out data preliminary treatment, refer to that sensing node accumulates transmission or saltus step real-time Transmission according to the threshold value of setting to the greenhouse data that collect, meeting under the prerequisite of event processing in real time, reducing the volume of transmitted data of sensing node.

6. wireless sensor network according to claim 4, it is characterized in that: described employing support model carries out the data fusion of multisensor, that sensing node transmits data to after aggregation node, aggregation node is according to the degree of support between each sensing data, calculate the optimal weights of each data, finally calculate the fusion value of data.

Images