Database Reference
In-Depth Information
ture location within a specific time period may be limited both by
the time elapsed and the layout of that location. This can be used
to estimate the location more accurately with the use of a smaller
number of samples [40].
One major issue, which has been observed in [183] is that the
requests for GPS may come from Location Based Applications
(LBA), and therefore, it is critical to design the energy saving
strategies in the context of the LBA requests, which may not nec-
essarily be synchronized with one another. The framework uses
four design principles corresponding to substitution . Substitu-
tion makes use of alternative location-sensing mechanisms, such
as network-based location sensing, which consume less power than
GPS. Suppression uses less power-intensive sensors such as an ac-
celerometer to suppress unnecessary GPS sensing for a statitionary
user. Piggybacking synchronizes the location sensing requests from
multiple running LBAs. Adaptation aggressively adjusts system-
wide sensing parameters such as time and distance, when battery
level is low.
In addition to software solutions, it is also possible to implement
hardware solutions. For example, simple operations can be directly
performed in main memory with dedicated hardware, without ac-
tually using the (more energy-intensive) main processor [135].
In addition to the power eciency of the sensing process, issues also
often arise about the power-eciency of the data transmission process.
Typically, data transmission is significantly more expensive, as com-
pared to the sensing process itself. For example, many applications are
enabled by the ability to capture videos on a smartphone and to have
these videos uploaded to an internet connected server. This capability
requires the transfer of large volumes of data from the phone to the
infrastructure. Typically smartphones have multiple transfer interfaces
such as 3G, Edge, and Wifi, all of which vary considerably in terms
of availability, data transfer rates and power consumption. In many
cases, the underlying applications are naturally delay-tolerant, so that
it is possible to delay data transfers until a lower-energy WiFi connection
becomes available. This tradeoff is explored in some detail in the SALSA
system proposed in [137]. An online algorithm is proposed, which can
automatically adapt to channel conditions and it requires only local in-
formation to decide whether and when to defer a transmission. Such an
approach has been shown to result in considerable power savings without
significantly affecting the operation of the underlying system.
Search WWH ::




Custom Search