As part of my thesis work, I focused on the
design, implementation, and
evaluation of adaptive self-configurable algorithms for sensor
networks. I have concentrated my efforts on adaptive
self-configuring
sensor networks topologies. The large number of nodes working
unattended and untethered in a wide range of environments make
self-configuration a desirable property for sensor network
systems. The increase number of nodes leads to maintenance and
configuration complexity, which precludes manual configuration.
Moreover, the environmental and system dynamics produce regions with
non-uniform connectivity, precluding design time
pre-configuration.
While the nodes in these sensor networks need to coordinate and
communicate to perform the distributed sensing task, the radio
component dominates the use of energy resources. Furthermore,
several studies have shown that idle energy dissipation of the radio
subsystem has the same order of magnitude than energy dissipation when
transmitting or receiving messages, and cannot be ignored. These
studies suggest that energy optimizations must turn the radio off, and
not just simply reduce the number of packets sent and/or received.
Topology Control Algorithm
In my thesis work, I suggest that one of the
ways system designers can
address such challenging operating conditions is by taking advantage of
the redundant communication capacity available and designing the system
algorithms to make use of that redundancy over time to extend the
systems life. ASCENT is
a distributed localized
algorithm that allows applications to configure the underlying topology
based on their needs while trying to save energy to extend network
lifetime. In addition, ASCENT
uses self-configuring adaptive techniques that react to the operating
conditions measured
locally. A subset of nodes is chosen to form an active topology
that establishes the communication backbone, while the rest of the
nodes turn their radios' off to save energy. I have shown through
analysis, simulation, and real
experimentation that the system achieves linear increase in energy
savings as a function of density and the convergence time required in
case of nodes failures while still providing adequate
connectivity. As part of my thesis work, I am also exploring the
interactions between self-configuring topology control schemes and
different routing and data dissemination mechanisms.
This work has been implemented in the UCLA/CENS EmStar software
environment running on Linux-class hardware platforms---Stargates,
iPAQs, PC104s. In addition, an implementation of ASCENT in TinyOS
running natively in Mica 1 and Mica 2 motes has been done by Thanos Stathopoulos.
As part of my thesis work, I focused on the
design, implementation, and
evaluation of adaptive self-configurable algorithms for sensor
networks. I have concentrated my efforts on adaptive
self-configuring
sensor networks topologies. The large number of nodes working
unattended and untethered in a wide range of environments make
self-configuration a desirable property for sensor network
systems. The increase number of nodes leads to maintenance and
configuration complexity, which precludes manual configuration.
Moreover, the environmental and system dynamics produce regions with
non-uniform connectivity, precluding design time
pre-configuration. 
In my thesis work, I suggest that one of the
ways system designers can
address such challenging operating conditions is by taking advantage of
the redundant communication capacity available and designing the system
algorithms to make use of that redundancy over time to extend the
systems life. ASCENT is
a distributed localized
algorithm that allows applications to configure the underlying topology
based on their needs while trying to save energy to extend network
lifetime. In addition, ASCENT
uses self-configuring adaptive techniques that react to the operating
conditions measured
locally. A subset of nodes is chosen to form an active topology
that establishes the communication backbone, while the rest of the
nodes turn their radios' off to save energy. I have shown through
analysis, simulation, and real
experimentation that the system achieves linear increase in energy
savings as a function of density and the convergence time required in
case of nodes failures while still providing adequate
connectivity. As part of my thesis work, I am also exploring the
interactions between self-configuring topology control schemes and
different routing and data dissemination mechanisms.