Research Thrusts
This project mathematically characterizes
structural dynamic properties of sensor networks
in order to engineer their design and operation for
dependable information fusion in urban operations.
Four research thrusts address our comprehensive
approach towards formulating an engineering science
for operational dependability of sensor networks
in this context.
III. Dynamic Control of Network Operations
Architectural design variables may not completely
determine structure. Both design and adaptive
operation may be necessary to maintain structure
and performance. The nondeterministic components
of network structure will be modeled as
high level behavior specifications in a network
model of hybrid interacting automata. A nonlinear
distributed adaptive network control mechanism
will be formulated to, in-situ, monitor and mitigate
observable deviations from normalcy during operations.
This multi-layered discrete event control
mechanism interacts with the information space
via the interacting automata models at each layer.
The control mechanism at each sensor node is
located at the lowest level 0. Level 1 autonomously
forms and controls dynamic sensor clusters in
response to stimuli in the operational environment.
Level 2 exerts inter-cluster control for regional
tracking of spatial-temporal phenomena and level
3 provides network wide control to correlate events
across regions and monitor structural parameters.
Tactical Control at the highest level maintains
structural boundaries, predicts and mitigates
emerging structural breakdown, and coordinates
recovery.
| Dr. R. Brockett, Harvard | brockett@hrl.harvard.edu, http://hrl.harvard.edu/~brockett/brockett.html |
| Dr. A. Ray, PSU | axr2@psu.edu |
| Dr. V. Tarkoh, Harvard Dr. I. Chattopadhyay, PSU |
ixc128@psu.edu |
| Mr. S. Kim, Harvard Ms. N. Devroye, Harvard Mr. B. Babadi, Harvard Mr. A. O’Connor, Harvard |
|
Summary
Research Thrusts
-
Modeling
-
Network Design
-
Network Operations
-
Experimentation