Database Reference
In-Depth Information
tunistic mobility of sensor-equipped vehicles to detect and report the
surface conditions of roads. Each car in the system carries 3-axis ac-
celeration and GPS sensors, gathering location-tagged vibration data.
The system uses uses
CarTel
's opportunistic wireless protocols to de-
liver the data over whatever wireless network is available to a back-end
server (discussed in detail below). The server processes this vibration
data using machine learning techniques in order to predict the surface
conditions.
Data Muling and Networking:
The data collected in a vehicle (such
as information about the road surface conditions) may sometimes need
to be routed to a back-end server, even in cases where a continuous
mobile connection is not available. In such cases, intermittent wifi ac-
cess points may be available along the route of the vehicle. should use
wireless networks opportunistically [57, 29]. The idea is to use a com-
bination of WiFi, Bluetooth, and cellular connectivity, using whatever
mode is available, while being completely transparent to underlying ap-
plications. In some cases, cars may be used as mules in order to carry
the data, when direct connectivity is not available [29].
Query Processing of Intermittently Connected Data:
Participa-
tory sensing sensor network applications must cope with a combination
of node mobility and high data rates when media-rich data such as au-
dio, video or images are being captured by a sensors. As a result of
the mobility, the sensor networks may display intermittent and variable
network connectivity, and often have to deliver large quantities of data
relative to the bandwidth available during periods of connectivity. In or-
der to handle this challenge, a system known as
ICEDB (Intermittently
Connected Embedded Database)
[178] was proposed, which incorporates
a delay-tolerant continuous query processor, coordinated by a central
server and distributed across the mobile nodes. The system contains
algorithms for prioritizing certain query results to improve application-
defined utility metrics.
Privacy Protection:
The process of tracking the position of individ-
ual vehicles is fraught with numerous challenges from a privacy perspec-
tive. Therefore, techniques are needed to be able to compute appropri-
ate functions on the location data, without violating individual privacy.
The
CarTel
system provides excellent privacy protection of user location
data, while being able to compute aggregate functions on the location
statistics. This is called the
VPriv
system [134]. More details on this
system are discussed in the section on privacy in this chapter.
8.3.2 Green GPS.
Green GPS [64] is a participatory sensing
navigation service that allows drivers to find the most fuel-ecient routes
