04-06-2013, 04:47 PM
Cooperative download in vehicular environments
Cooperative download.pdf (Size: 423.69 KB / Downloads: 29)
Abstract
We consider a complex (i.e., non-linear) road scenario
where users aboard vehicles equipped with communication
interfaces are interested in downloading large files from road-side
Access Points (APs). We investigate the possibility of exploiting
opportunistic encounters among mobile nodes so to augment the
transfer rate experienced by vehicular downloaders. To that end,
we devise solutions for the selection of carriers and data chunks at
the APs, and evaluate them in real-world road topologies, under
different AP deployment strategies. Through extensive simulations,
we show that carry&forward transfers can significantly
increase the download rate of vehicular users in urban/suburban
environments, and that such a result holds throughout diverse
mobility scenarios, AP placements and network loads.
INTRODUCTION
Vehicles traveling within cities and along highways are regarded
as most probable candidates for a complete integration
into mobile networks of the next generation. Vehicle-toinfrastructure
and vehicle-to-vehicle communication could indeed
foster a number of new applications of notable interest and critical
importance, ranging from danger warning to traffic congestion
avoidance. It is however easy to foresee that the availability of onboard
communication capabilities will also determine a significant
increase in the number of mobile users regularly employing
business and infotainment applications during their displacements.
As a matter of fact, equipping vehicles with WiMAX/LTE and/or
WiFi capabilities would represent a clear invitation for passengers
on cars or buses to behave exactly as home-based network users.
The phenomenon would thus affect not only lightweight services
such as web browsing or e-mailing, but also resource-intensive
ones such as streaming or file sharing.
RELATED WORK
The cooperative download of contents from users aboard vehicles
has been first studied in [3], that introduced SPAWN,
a protocol for the retrieval and sharing of contents vehicular
environments. SPAWN is designed for unidirectional traffic over
a highway, and is built on the assumption that all on-road vehicles
are active downloaders of a same content. Instead, we target urban
environments where users may be interested in different contents.
Similar considerations hold for the works in [4] and [5].
In [6], the highway scenario is replaced by a circular bus route
within a campus, which however implies again easily predictable
vehicular contacts: indeed, the focus of the work is on the
prefetching and multi-hop transfer of data at each individual AP,
while carry&forward communications are not taken into consideration.
Conversely, [7] and [8] examine urban environments.
Carriers selection
The first problem we address is that of the selection of data
chunk carriers at APs that are idle, i.e., that are currently not
transferring data directly to vehicular downloaders. As previously
discussed, these APs can opt to employ their spare airtime to
delegate, to mobile users within range, portions of files being
downloaded. Taking such a decision means to answer to two
questions: (i) which, among the vehicles in range of an idle
AP, should be picked as carriers, if any? and (ii) which of the
downloaders should these carriers transport data for?
The key to the answers is to know in advance whether (and
possibly when) one or more cars currently within coverage of an
AP will meet a specific downloader vehicle, so to perform the
selection that maximizes the download rate. Also, by choosing
carriers depending on their future contacts, the destination of the
data becomes constrained to the elected carriers, and the second
question above is inherently solved along with the first one.
However, assuming that the roadside infrastructure has perfect
knowledge of the future route of each user is unrealistic, other
than raising privacy issues. At the same time, the movement of
individual vehicles over urban road topologies cannot be easily
predicted as in unidimensional highways. We then adopt a probabilistic
approach, by leveraging the fact that large-scale urban
vehicular flows tend to follow common movement patterns [23],
[24], [25]. More precisely, the solution we propose leverages
contacts maps, that are built by exploiting historical data on
contacts between car flows, and then used to estimate the meeting
probability between downloaders and candidate data carriers.
Carriers selection algorithms
Contacts maps can be exploited by APs to select local cars as
data carriers in the cooperative download process, by retrieving
their contact probability estimates with respect to downloader
vehicles. Firstly, it is necessary that APs know which cars in their
surroundings are interested in some content. Thus, every time a
downloader vehicle starts a production phase, the fact that it is
requesting data, as well as the nature of the desired content, is
attached to the usual information on the production phase that
the local AP shares with other APs. This way, each AP can track
downloaders through their production phases history.
Thanks to such knowledge, an AP that has active local production
phases can compute the delivery potential pa resulting
from electing one (or some, or all) of the local vehicles as
carrier(s) for data destined to a specific downloader vehicle a.
CONCLUSIONS AND OPEN PROBLEMS
We presented a complete study of cooperative download in urban
vehicular environments. We identified and proposed solutions
to the problems of carriers selection and chunk scheduling, and
extensively evaluated them. The main contribution of this work
lies in the demonstration that vehicular cooperative download in
urban environments can bring significant download rate improvements
to users traveling on trafficked roads in particular.