Addressing Information Processing Needs of Digital Agriculture with OpenIoT Platform - Interoperability and Open-Source Solutions for the Internet of Things - page 141

are described using the Semantic Sensor Network Incubator Group (SSN-XG)
ontology [ 10 ], which is used to describe the physical and technical nature of
the installed sensor network (e.g. what phenomenon a sensor observes or what
platform the sensor is on and, therefore, where it is located).
Once deployed, the sensors observe a sampling feature that samples either
the plot as a whole, the experimental site or some sub-feature of the plot (e.g.
a layer of soil within plot). The sampling feature allows sensors to be linked to
plots and from the plot to the combination of genotype, treatments and events
such as sowing dates that define the crop. Features can be related, so that a soil
layer can be built from the individual plot/layer segments or plots linked into
an experimental site. The ability to relate features means that measurements of
larger systems, such as the site weather, can be seen to apply to sub-parts such
as individual plots. The ontology allows mapping of the Phenonet data model
to the OpenIoT system enabling on-the-fly annotation of sensor data streams.
The ontology allows queries expressed in the domain of the experiment to
navigate to sensors that measure appropriate information. For example, select-
ing the sensors that observe plots sown with a specific genotype; selecting the
sensors that observe plots in a specific block; or selecting the sensors that sample
a specific depth of soil. Sensors can also be queried using sensor-specific infor-
mation, such as location. Conversely, the ontology can be used to navigate from
a sensor property to experiment-specific information. For example: the types of
treatments that have canopy temperature data available.
4.4 Architecture
Figure 6 presents a detailed architecture of Phenonet application using OpenIoT
middleware. The sensed data captured from the field via the OpenIoT X-GSN
component is pushed to the cloud storage via LSM. The X-GSN component
performs basic error checking such as identifying outliers. Once the annotated
data is available in the cloud store, the OpenIoT Request Definition, Request
Presentation, Scheduler and SDUM services are employed to discover relevant
data streams based on the ontology, compose a service with operators like min,
max, avg and deploy the composed service for reuse.
Phenonet Ontology Integration with X-GSN and LSM-Light. The Ope-
nIoT X-GSN component is responsible to interface with the sensors in the field,
collect data, semantically annotate the data based on the extended OpenIoT-
Phenonet ontology and push the data into the cloud RDF store via LSM-Light.
The X-GSN wrapper connects to the sensors in the field. We implemented Phe-
nonet specific X-GSN wrapper to interfaces with the sensor data streams. The
wrapper is also responsible to generate RDF's descriptions of sensors by encod-
ing domain specific information based on the Phenonet ontology. A sample RDF
description of a sensor and a plot is listed in Listings 1.1 and 1.2 .
In Listing 1.1 , we define a sensor of type /ontology/phenonet#ArduCrop
with identifier sensor/arducrop/20140611-1962-0012 , which is deployed in a