Hello,
I am a Ph.D. student at the University of California, Davis working on
Embedded Software and specifically addressing power consumption on
mobile platforms. We recently had a paper accepted into the
"International Conference on Embedded Software" (EMSOFT 2009 -
http://www-verimag.imag.fr/EMSOFT09/dates.shtml) called "Markov
Decision Process (MDP) Framework for Optimizing Software on Mobile
Phones". This framework was tested using data from an Android G1 and
we are currently working on integrating it into the email client.
We proposed a framework for tasks that are not time critical (such as
email synchronization), but treat phone calls as the highest priority.
The reason we did this is because we believe that smart phones are
phones first, computers second. Our optimization works like this:
1) First, we look at the call records to figure out how likely a
phone call is to be received.
2) Next we create an optimal decision table which will tell
wether you should synchronize or not depending on how much time since
you last synchronized, the amount of battery life left and the current
time.
3) We create this table by predicting your charge time and then
working backwards to figure out the best decision table. The goal is
to synchronize as much as possible, but still ensure that you can
receive phone calls.
4) We do this by using a hard to explain set mathematical
equations to figure out if it is "worth it" to synchronize or not. (I
can post more details about the math if people are interested, but I'm
not sure how much interest there would be in that level of detail ;-)
I am very interested in getting questions, comments and general
feedback from you, the Android developer community, on this technique.
You can find the full paper here:
http://www.ece.ucdavis.edu/mcsg/pubs/emsoft09.pdf
Thanks for your feedback!
-Frank
ABSTRACT
We present a framework based on Markov decision process
to optimize software on mobile phones. Unlike previous
approaches in literature that focus on energy optimization
while meeting a speci c task-related time constraint, we
model the desired talk-time as an explicit user given pa-
rameter and formulate the optimization of resources such
as battery-life on a mobile phone as a decision processes
that maximizes a user speci ed application speci c reward
or utility metric while meeting the talk-time constraint. We
propose e cient techniques to solve the optimization prob-
lem based on dynamic programming and illustrate how it
can be used in the context of realistic applications such as
WiFi radio power optimization and email synchronization.
We present a design methodology to use the proposed tech-
nique and experimental results using the Android platform
from Google running on the HTC mobile phone.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to android-developers@googlegroups.com
To unsubscribe from this group, send email to
android-developers-unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/android-developers?hl=en
-~----------~----~----~----~------~----~------~--~---
