56 / 64
p ro j e ct 1 5
Rapid response teams have to deal with
unpredictableRF conditions, sketchy spectrum
availability, and system incompatibility.
Collapsed buildings and damaged subway
tunnels degrade signal quality inside.
Power outages affecting cellular base stations
are common after earthquakes and weather
disasters.
Ubiquitous smartphones with IEEE 802.11
adapters and GPS localization provide a
powerful
platform
for
opportunistic
communication in difficult environments.
Smartphones with IEEE 802.11 adapters
configured in ad-hoc mode can form a wireless
network that does not need any infrastructure
to communicate end-to-end using multi-hop
routes. Still, TCP/IP protocols only work if
there is a continuously available path between
endpoints. With intermittent connectivity,
traditional IP MANET routing and transport
protocols inevitably fail. The solution is to use
Delay Tolerant Networking (DTN) techniques
that can tolerate intermittent lack of end-to-end
connectivity by storing and carrying messages.
This project develops DTN protocols for rapid
response applications that has 802.11 ad-hoc
networks self-organize and deliver packets
end-to-end when the network topology is
dynamic from node mobility and sporadic link
availability. The protocols rely on cognitive
networking techniques. Channel allocation is
aware of availability from primary users (access
points), and usage policies in order to prevent
interference and to satisfy message quality
of service. Message forwarding is scheduled
based on contextual information derived
from the user’s mission plan and resource
availability. Conversely, the mission plan and
node trajectory of some nodes is adjusted to
serve communication demands using message
ferrying.
These technologies will improve the effectiveness
of communications for emergency response and
for other ad-hoc scenarios using readily available
smartphones when circumstances prohibit use
of cellular or specialized government radio
systems.
Smartphone Cognitive Networking for Rapid Response
Ionut Cardei, PI
5 6