Reporting Road Problems in Smart Cities Using OpenIoT Framework - Interoperability and Open-Source Solutions for the Internet of Things - page 173

The design of user-friendly interface will be included in future work. CityWatcher
Middleware is based on OpenIoT Platform [ 24 ]. OpenIoT project is a new open source
middleware platform for intelligent IoT applications. It is an extension to cloud
computing implementations and provides functions for managing IoT resources. In this
way users can get IoT services including Sensing-as-a-Service. OpenIoT platform is
discussed in [ 24 26 ].

The CityWatcher project relates to urban crowdsensing. Therefore, the city
administrations are to become service providers and a host to an OpenIoT platform.
Using an OpenIoT platform gives us an ef

-

cient and advanced way to fusing data
sources. Client software located in Utility App Plane is a web-based application, which
allows an authorized user both to make requests to the system and to process reports.
One of the technical problems needed to be solved is identifying the device, on
which video is recorded and stored. Firstly, middleware has to be aware of all par-
ticipating ICO s in the system, as it has to broadcast/multicast a request to all of them,

register that the task has been sent and match the answer with the device ID. Secondly,
middleware needs to have information about user personal preferences, as the user
obtains some incentives for participating and sharing his/her videos. Besides, having
information about the ICO owner makes the video fragment more legal in the court. We
must also bear in mind that one user can have several devices and one device can be
used by different users.
We propose to use the following scheme: every user gets an account in the City-
Watcher system. It is a classical login/password pair. With this account a user can start
the CityWatcher app. On the middleware side login is linked to full user description/
pro

'

le, so the user can get his/her incentives from city services. As the video is stored
on the users devices and one user can have multiple devices, we need to uniquely

identify every device. To solve this problem a unique device ID can be generated
following the CityWatcher installation. It is like a license number for software. Then,
all users of one device use this unique ID, but log in with their own accounts. This
enables the middleware to distinguish devices for video searching possibilities and also,
distinguish users for incentive purposes.
When the video fragment is loaded into the cloud, it should be stored and registered
in the database. Additional information about the fragment includes task ID, on which
it was found, user ID, device ID and obtaining time. Besides, some information about
device like camera resolution and record format is also attached. OpenIoT middleware
with CityWatcher Application and their connection with city authorities are shown
in Fig. 6 .
We use SQLite as an engine for storing data about timestamps and coordinates. The
CityWatcher app instantly adds record to the database. When a task arrives and there is
a need to understand if there is a video on local storage

'

- a request to the database is

made.
The table structure of the local database that is used for both scenarios is shown in
Fig. 7 . For the content search scenario, when the request is made, we search not exactly
the same time and coordinates, but add delta values and perform range search. For the
“

scenario needed data is derived from the same local database,
before sending it to the cloud service. The middleware database structure is omitted in
this paper due to space limitations.

driver-initiated alerts ”