SOCIAL SENSING - Managing and Mining Sensor Data - page 275

which can be picked up by satellite. RFID technology has even found
application in a number of medical applications, in which RFID chips
are embedded in patients in order to track their case history. RFID
Technology has lead to the general vision of the internet of things [16],
in which uniquely identifiable objects can be continuously tracked over
time. In the case of commercial applications, the products may have im-
plicit links among them which correspond to shared batches or processes
during the production and transportation process. Such tracking data
can be used in conjunction with linkage analysis in order to determine
the causality and origin of tainted products. It can also be used to track
the current location of other products which may be tainted. Such data
is typically quite noisy, error-prone, incomplete, and massive in volume.
Thus, this leads to numerous challenges in data compression, storage
and querying. A detailed tutorial on RFID methods may be found in
[81]. The technology is also discussed in some detail in a later chapter
of this topic [4, 5].
8.3 Vehicular Participatory Sensing
In vehicular participatory sensing, a variety of sensor data from vehi-
cles such as mobile location, or other vehicular performance parameters
may be continuously transmitted to users over time. Such data may be
shared with other users in the aggregate in order to preserve privacy.
This is the social aspect of such applications, since they enable useful
individual decisions based on global patterns of behavior. In addition,
vehicular participatory sensing may be used in order to enable quick
responses in case of emergencies involving the vehicle operation. We
note that much of the work discussed above for animal and moving ob-
ject trajectory mining [104, 105, 112-115, 109, 110] are also applicable
to the case of vehicular data. In addition, vehicular data poses unique
challenges in terms of data collection, sensing, transmission and privacy
issues. Classic examples of vehicular participatory sensing include the
CarTel [88] and GreenGPS systems [64]. While we will focus on a de-
tailed discussion of these systems as the most well known representatives
of vehicular participatory sensing, a number of other sensing systems
have been designed for different applications such as trac monitoring
and road conditions [124], cyclist experience mapping [55, 142], and the
determination of transportation modes [144].
The problem of sharing bike track paths by different users has been
explored in [142]. The problem of finding bike routes is naturally a trial-
and-error process in terms of finding paths which are safe and enjoyable.
The work in [142] designs Biketastic , which uses GPS-based sensing on