Geoscience Reference
In-Depth Information
server side (without notifying the client about the data sent). Regular browsers do
not make use of PUT and DELETE methods of HTTP, which can be invoked by
specific REST clients/APIs that are developed for communicating with all HTTP
methods. Several recent research projects implement RESTful Web services for
smart things within (what has become to be known as) the Web of Things.
As mentioned earlier, interconnected devices (i.e. Things) today are capable of
providing real time information about them on the web. The information can be
acquired from various sensors that are connected to the device and broadcasted
through the embedded web servers. The information broadcasted by these serv-
ers is usually in form of XML or JSON documents. Several loosely coupled web
architectures can be designed to reach and utilize this information, where,
• Each sensor is represented by an URI (Sensor Endpoint)
• Each device is represented by an URI (Device Endpoint)
• A facade that interacts with devices is represented by and URI (Facade
Endpoint).
In such situation, web services can be used to interact with the (Sensor/Device/
Facade) endpoints. Due to its lightweight and loosely coupled nature, REST archi-
tectural style is preferred for device to web and web to device interactions. In
recent research REST is viewed as ideal candidate to build a “universal” archi-
tecture and Application Programming Interface (API) for smart things. The use of
Facade (i.e. wrapper layer) is recommended for communicating and interacting
with devices and sensors. Usländer et al. (
2010
) suggest that Virtual Sensors (i.e.
soft-sensors that are used to gather and abstract data from diverse sets of sensor
network nodes) can act as a middleware between the Sensors and Services.
The client side of the architecture as indicated in Usländer et al. (
2010
) can
be composed of visualisation, reporting and other sensor applications. In terms of
visualisation of geospatial information, recent tools offer a huge potential. Since
embedded web servers in an Internet of Things architecture generally have fewer
resources than Web clients, browsers or mobile phones, Asynchronous JavaScript
and XML (AJAX) clients has proven to be a good way of transferring some of
the server workload to the client (Guinard et al.
2011a
). As stated by Abernathy
(
2011
), the advent of Web 2.0 tools such as AJAX have given online mapping
tools increased versatility and visual appeal. Virtual globes such as Google Earth
provide users access to geospatial information in three dimensions, allowing them
to turn on and off various layers, download datasets, and create their own spatial
data complete with text, photos, Web links, and even video clips.
5 The Web Service
Isikdag and Zlatanova (
2011
) proposed a Sensor Acquisition Framework, as a web
service architecture for information acquisition from indoor sensor networks, and for
uniting information acquired from Things with 3D building model representations in
