Database Reference
In-Depth Information
a mobile phone application in order to create a platform which enables
rich sharing of biker experiences with one another. The microphone and
the accelerometer embedded on the phone are sampled to infer route
noise level and roughness. The speed can also be inferred directly from
the GPS sensing abilities of the mobile phone. The platform combines
this rich sensor data with mapping and visualization in order to provide
an intuitive and visual interface for sharing information about the bike
routes. A different application uses the time-stamped location infor-
mation in order to determine the mobility profiles of individuals [144].
Next, we will discuss the
Cartel
and
GreenGPS
systems.
8.3.1 CarTel System.
The
CarTel
project at MIT [88] is de-
signed for mining and managing large amounts of sensor data, which are
derived from vehicular participatory sensing. The most common data is
vehicular position data, from which large amounts of information about
road congestion, conditions, and other violations may be determined.
The project focusses on the collection and use of such data in an ef-
ficient and privacy-preserving way. The actual data may be collected
either from mobile phones in the car or from embedded devices within
the car itself. For example, the Onboard Diagnostics Interface (OBD-II)
equipped on modern cars can be used to collect tremendous amounts
of useful data in this context. The OBD-II is a diagnostic system that
monitors the health of the automobile using sensors that measure ap-
proximately 100 different engine parameters. Examples of monitored
measurements include fuel consumption, engine RPM, coolant tempera-
ture and vehicle speed. Vehicles that have been sold in the United States
after 1996 are mandatorily equipped with a sensing subsystem called the
On-Board Diagnostic (OBD-II) system. A number of key components
of the
CarTel
system are as follows:
Tra
c Mitigation:
In this case, two systems
VTrack
and
CTrack
[154, 155] have been proposed for processing error-prone position streams
for estimating trajectory delays accurately. Since the location data is
typically error-prone as a result of transmission errors, or outages, the
technique is designed to be resistant to errors. In particular, the
CTrack
system [154] can work with the position data from cellular base stations,
in which the location error is much higher than GPS data. The system
continuously collects the data, and combines real-time and historic de-
lay estimates to produce predictions of future delays at various points
in time in the future. The results of the predictive model are sent to a
commute portal where users can view the data along with appropriate
trac routing strategies.
Road Conditions:
The idea in this approach [58] is to use the oppor-
