ibul 2004
hin the next
, are either
specifically
stry, cloud
se satellites
vill provide
synthesize
ding of the
st of these
o the top of
04) and its
netrate the
T chemical
n from this
d provide
1 satellites.
1€ as Aqua:
cimately 15
R-sensitive
ozone. Às
| cloud-free
portunity to
mple, data
)ectrometer
ice, in real
ns can be
d of course
s operating
originally
| still take
cale event-
processing
nts can be
perating in
I', LandSat
ultispectral
nstruments
tral and 36
some event
system and
1 look and
via ground
via remote
wever, this
anding its
et, air-sea
and the
ve to study
; regarding
n balloons
altitude (>
ds of time
ıry at one
n intricate
be able to
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXX V, Part Bl. Istanbul 2004
The basic concept in this configuration as shown in Figure 7 is to
deploy a series of ULDBs with atmospheric sensing payloads
around the globe. For example, these ULDBs can be distributed:
— In polar regions to constantly monitor the ice caps
- In mid latitudes for chemical species and transport
phenomena
— To study invasive species in the tropical regions
— For coastal zone management
- To monitor vegetation and crop growth patterns
— To monitor Earth faults and surface deformation
— To monitor volcanoes
Balloons can provide constant surveillance with high resolution in
the troposphere and communicate the relevant data to processing
sites either in space or on the ground. The LEO platforms can
gather upper atmospheric data and similarly send it to the
processing sites. — Geosynchronous satellites can provide the
communication pathway and balloon constellation management
and control. All of this can be integrated in high-resolution
temporal models for analytical and predictive studies.
The ultimate goal is to have an integrated web, as depicted in
Figure 2, of sensors sweeping the globe with active and passive
Figure 2. Sensor web concept in LEO, GEO and suborbital plains.
sensors in a variety of spectral bands. This requires constellations
to be intra-connected and interconnected with extensive
communication networks. These sensors can be in various orbits
to provide a global coverage with high temporal and hyperspectral
resolution for both ground and atmospheric observations.
Additionally, suborbital sensors can be mounted on ULDBs and
aircrafts for discrete process studies and observations. The major
task will be to coordinate all of these sensors and maintain a close
measurement synergy. This is not only a challenge to come up
with the operational scenarios, but is even a greater challenge for
the users to coordinate and decipher the multitude of
measurements from a mix of sensors.
6. GENERIC TOPOLOGIES
There are many potential defining concepts for designing the
sensor web configurations. However, one of the most difficult
challenges is to develop the efficient protocols and
communication methodologies for making effective observations.
Sensors can come from many classes of technologies. Sensor can
be an electrical detector, an optical detector, a magnetic detector,
a radiation sensor, a biological or a chemical sensor; or a
combination of any one of these devices connected or deployed in
the environment under study. However, there should be some
attributes of any of such sensing devices and these should be:
— Detector system and a signal conversion mechanism
- Communication system and processing
307
— Intelligent sensing for decision making
— Event driven observations
- Selforganizing and adaptable to new configurations
— Easily deployable schemes
= "Cost effective
- Minimize communication overhead
- Reliable for remote or space based applications
— Capable to operate both in autonomous and/or supervised
modes
— Standardized interfaces
- Maintainable with modular replacement units
— Programmable spectral bands
— Easily mountable on their respective platforms
— Self organizing
— Self configuring
Majority of the above criteria can be easily met for in situ type
sensors and their mounting platforms. However, this problem is
much more complex and complicated to deal with for space-based
systems. This is because of multiple constraints such as: severe
thermal and radiative environment, poor signal to noise ratio,
power limitations, platform stability, space debris impact, and
maintenance inaccessibility. The following paragraphs present
basic building blocks of any possible observing architecture.
Ring Controller
s3 Gs £552
s(n-1) C C.
sn
Figure 3. Sensors connectivity via a communication Ring.
Ring Topology: This configuration in Figure 3, has all the
observers linked in a Ring. The Ring provides basic means for
. communication, control and data flow. However, in case of smart
sensors, the activities on the Ring can be kept to a minimum level
because of the autonomy of decision making, presumably it can
be delegated to each independent sensor. This is relatively easier
for the in situ or ground based sensors. However, the problem
becomes rather acute when majority of the sensors are space
based. This is due to the fact that the data products are based on
the global observations. It therefore requires comprehensive
schemes for data collection, and archiving so integrated products
can be generated. For example, let us consider an example of
multiple sensors deployed in a sun synchronous orbit at an
altitude of around 800km and 99° inclination. The idea is to
increase a spatial, spectral and temporal resolutions so high
quality products can be produced showing land cover/use and also
report on some disastrous situations such as volcanic eruption or
floods. The idea would be to have a much greater field of view so
a wide area is observed in real time and critical conditions are
reported in a very short time to take corrective actions. This can
be accomplished by deploying a large network of sensors which is
capable of both providing data for integrated maps and specific
signals for emergencies. This further requires a sophisticated
coordination scheme among such sensors that would use time,
ephemeris, satellites conditions (viewing angle, exact altitude,
