In my thesis work, I also performed in depth
analysis of the temporal properties of wireless links. These
properties have high impact on the performance of routing
algorithms. I studied short term temporal issues, like lagged
autocorrelation of individual links, lagged correlation of reverse
links, and consecutive same path links. I also studied long term
temporal aspects, gaining insight on the length of time the channel
needs to be measured and how often we should update our models.
As part of my thesis, I identified a set of properties that are the
most relevant for the design of efficient routing protocols. For
example, high temporal correlation implies the need to use a newly
developed link quality metric called Required
Number of Packets (RNP) instead of the commonly used Reception Rate and implies the
importance of using only high quality links. High variance in
time lagged correlation of forward and reverse links implies the need
for immediate sending of acknowledgments, while low short time variance
of links favors communication using long packets. Correlations
between links on the same multi-hop paths imply a need for the
development of new types of shortest path algorithms, while high
consistency of high quality links over time implies the rare need to
update the models of communication links.
Guided by the obtained insights, I have
developed and analyzed two new routing algorithms: (i) a generalized
Dijkstra algorithm with centralized execution that considers
correlation of successive links in multi-hop communication, and (ii) a
localized probabilistic algorithm that uses statistics about the
reverse forwarding paths to establish probabilistic gradients on the
forwarding path. I also performed simulations to analyze the overhead
of the distributed solution with respect to the optimal solution in the
scenario were a subset of nodes is powered down in order to save energy.
This work has been implemented in the UCLA/CENS EmStar software
environment running on Linux-class hardware platforms---Stargates,
iPAQs, PC104s, as well as in smaller Mote-class hardware platforms
using EmStar in
emulation/hybrid mode.
In my thesis work, I also performed in depth
analysis of the temporal properties of wireless links. These
properties have high impact on the performance of routing
algorithms. I studied short term temporal issues, like lagged
autocorrelation of individual links, lagged correlation of reverse
links, and consecutive same path links. I also studied long term
temporal aspects, gaining insight on the length of time the channel
needs to be measured and how often we should update our models. 
Guided by the obtained insights, I have
developed and analyzed two new routing algorithms: (i) a generalized
Dijkstra algorithm with centralized execution that considers
correlation of successive links in multi-hop communication, and (ii) a
localized probabilistic algorithm that uses statistics about the
reverse forwarding paths to establish probabilistic gradients on the
forwarding path. I also performed simulations to analyze the overhead
of the distributed solution with respect to the optimal solution in the
scenario were a subset of nodes is powered down in order to save energy.