IEEE Consumer Electronics Magazine - May 2018 - 71

CA computing is still enabled to filter the list of SOs shown,
displaying those relevant to the user and the environment.
2) View/edit: The user can still find the SOs in the surrounding area but when clicking on one of them, the system
provides more detailed information about them. Furthermore, the user is able to manage the selected SO according to his or her privileges by using the control buttons
shown on the AR screen.
3) Search/edit: The user searches for a specific SO inside the
area of interest. The AR system maintains a possible location of that object (from the last stored coordinates) and
directs the user to physically locate it. When found, the
system functions like the previous case, allowing the user
to manipulate the object through the AR interface.
Standardization efforts have been directed toward universal
protocols that will encompass all IoT procedures. Still, it is
possible that multiple solutions will emerge and be implemented into the SOs by manufacturers. In such a scenario, the
devices scanning the area will also need to request protocol
information after initiating communication with a smart thing.
After determining the interaction protocol, the user's
device will need to retrieve the application programming
interface (API) used by the selected object, possibly from
the middleware. The wide range of SOs does raise some
challenges in designing and implementing APIs that are
usable by many similar objects. Standardization will again
be needed to make the interaction between objects of the IoT
quick, without the need to route it through middleware [18].
Our proposal to introduce natural interactions with interfaces that will accompany the IoT applications of the (near) future
is the AR visualization of the smart environment (Figure 4).
The abstraction offered by the AR layer allows a simpler view

of the smart things, hiding technical details irrelevant to the
context. By allowing interaction with them through the touch
screens of their devices, we succeed in moving from traditional
ways to more natural ones. The wide range of today's CE
products functioning as user devices can also allow voice and
gesture interfaces and other abstract ways of communication to
fulfill our initial requirement for natural interaction.

ARCHITECTURE EVALUATION
Various tests and experiments will need to be conducted to
evaluate user experience and the impact that AR visualization
will have on a user's interaction with the IoT services. Apart
from the technical issues that need to be examined (e.g., the
efficiency of the BLE beacons, the performance of the AR
software, and security/privacy issues), it is important to measure the difference of the user's perception of the IoT infrastructure, with and without the AR visualization.
Considering that our proposed architecture refers to large
IoT systems in smart cities, we must test our prototype in
large, public city regions. A characteristic subsystem of our
implementation could be consisted of a middleware, a few
sensors, and BLE beacons referring to the SOs of its command area. Our goal is to compare users' reactions and
behaviors during their interaction with the IoT infrastructure,
with and without the AR visualization. It is important to evaluate the user's feedback and identify the impact of AR techniques in the IoT perception, along with our architecture's
effectiveness and feasibility.

DISCUSSION
This article discusses an IoT case from the near future and
analyzes a proposed architecture that addresses the issue of

Approximate
Distance: 300 m

Approximate
Distance: 350 m

Approximate
Distance: 100 m

Approximate
Distance: 150 m

Approximate
Distance: 250 m

FIGURE 4. An AR visualization in the IoT through our proposed application.

MAy 2018

IEEE Consumer Electronics Magazine

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - May 2018