Transcript File

Nanosatellite Communication
And MEMS Technology
Overview
• Changing satellite architecture
– Smaller, distributed systems
– Require RF communication
• MEMS communication devices
–
–
–
–
Switches
Antennas
Signal Filters
Phase Shifters
• Completed picosatellite experiment
• Suggestions for future
Faster, Better, Cheaper
• NASA Administrator
Daniel Goldin sought
new methods for space
exploration
• Reduced mass results in significant gain in shrinking
launching cost
• Less expensive to launch small components individually
rather than monolithic device
– Low Earth Orbit (LEO) ~ $10k per kilogram
– Geosynchronous Orbit (GEO) ~ $50k per kilogram
• Situation perfectly suited for MEMS devices
– Low mass, resistant to inertial and vibration damage
– Endurance in high radiation environments
Distributed Satellite Architecture
• Spread component capabilities to separate
vehicles
– Individualized vehicles faster to produce
because of less system integration
– Easily replaceable for component failure
– Eliminate physical hardware connections and
reduce overall mass
Capabilities of DSA
Chandra X-Ray Observatory
• Increase aperture size
for interferometer and
distributed radar
systems
– Hubble, Chandra
limited size due to
launch constraints
– Failures aboard Hubble
are repairable by
humans; Chandra out
of reach
PKS 0637
3C273
CXC
Hubble Space Telescope
Planned DSA Missions
Terrestrial Planet
Finder (JPL)
TechSat 21 Distributed
Radar (AFRL)
Space Technology 5 & 6 (NASA - NMP) –
First to use primarily MEMS components
Consideration for DSA
• Actively control relative positions and velocities
• Robust, reliable feedback from sensors possibly
onboard separate vehicle
• Remote RF communication necessary
– RF comm. requires sender/receiver pair with signal
processing hardware
• MEMS RF devices explored:
–
–
–
–
Switches
Antennas
Signal Filters
Phase Shifters
DC-Contact Coplanar Waveguide Shunt Switch
• Switches used for beam
shaping and steering
• RF MEMS switches have
better efficiency and lower
insertion losses than
conventional switches
• Ideal for space:
–
–
–
–
–
Rapid response
Good power handling
Wide bandwidth
Good EM isolation
High open isolation
• May experience stiction and
slow response time
DC-Contact Coplanar Waveguide Shunt Switch
• Force balance can be used to calculate
the electrostatic force
• The restoring force is found from spring
equation
• For deflections greater than 1/3 d, pull in
occurs
• Pull-in voltage not affected by dielectric
layer
• A, 0, V, d, x, and k are the projected
area of the electrodes, permittivity of the
free space, applied voltage, gap between
the line and bridge, deflection of the
bridge, and spring constant respectively
  0 AV 2
FE 
2(d  x)2
8d 3 k
Vp 
27 0 A
DC-Contact Coplanar Waveguide Shunt Switch
S 21  2 Rs Z 0  , L  Rs 
2
2
 2L Z 0  , L  Rs 
2
• Model of RsL
circuit in isolation
• Rc, Rl are contact
resistance and
contact line
resistance
S 21 
2
4Rb2  1 / 2Cb2 
2R
2
b
 Z 0   4 / 2Cb2 
2
• Model of R-C
circuit, insertion
• Capacitors model
coupling between
switch and pull-in
electrodes
DC-Contact Coplanar Waveguide Shunt Switch
• Process similar to other MEMS devices
manufactured by batch lithographic processing
• 1.7 mm PECVD SiO2 grown as sacrificial layer; dimples created by partially
etching 5500 Å; 0.8 mm sputtered Au creates bridge; buffered HF solution
used to remove SiO2
• RSC microrelay: micromachine @ 250 C; SiO2 removed by dry release
etching in oxygen plasma
Signal Filters
•
CPW filter
structure
All RF communication circuits require at
least one filter to pull out a desired signal
or insert one to be transmitted
– Currently done with solid state electronics
– Surface Acoustic Wave (SAW) filters or
back-end digital signal processing
•
MEMS offers passive front-end signal
processing capability
– Compact one-chip design
– High fidelity signal handling
– Tunable configuration
•
•
CPW Filter
Filter sensitivity and quality factor
(Q) would be greatly increased
Different MEMS filter designs are
possible
– Coplanar waveguide (CPW) layout
on thin GaAs membrane
– Flexural beam resonator
Beam Resonator
Antennas
• Improved performance of components achieved
by integrating antenna design with other
components on same chip
1600 mm
– “Smart” antennas
• Double-folded shot antenna
– 2.5 mm gold deposited on silicon oxide
dielectric membrane
– Cross members placed half wavelength apart for
optimal performance
1600 mm
77 GHz Double-Folded Slot Antenna
• Reconfigurable V-Antenna
– Arms of antenna can be moved independently
with comb-drive actuators
– Structure fabricated using silicon multi-layer
surface micromachining
– When both arms moved at fixed angle, antenna
can steer beam to focus reception or
transmission
– Adjusting relative angle of arms can modify
shape of beam
17.5 GHz V-Antenna
Phase Shifters
• Phased-array antenna
– Able to transmit or receive signals
from different directions without
being physically re-oriented
– Currently this is done with FET or
diode technology
• Low power consumption but high
signal loss
– MEMS design would cut down on
signal loss, especially at frequency
range 8-120 GHz
– Not as many amplifiers are needed to
boost signal, resulting in power
savings
Phase shifter composed of array of RF-MEMS switches
• Straightforward design: MEMS switches used in place of solid-state components
– Large body of research already exists in phase shifter design and application
– Proper placement of switches is known
• Significant cost benefits
– MEMS-based array could cut cost of complex phase-shifter by an order of magnitude
Picosatellite Mission
• Satellite mission has been completed proving feasibility of MEMS devices in
space
– RF-MEMS switches in picosatellite (< 1 kg)
• Stanford-designed Orbiting Picosatellite Automated Launcher (OPAL)
launched in 2000 released testing platform
– Two tethered picosatellites in LEO containing four RF-MEMS switches in series
• Switches developed by Rockwell Science Center (RSC)
– Each satellite measures 3 x 4 x 1 in3 and weighs less than half a pound
– Actual communication system made with standard radio components
• MEMS switches used only for experiment
Switch Experiment
• RF switches cycled through on and off states at 500 Hz
• During contact time with ground station, test data and statistics downloaded
– Relatively inexpensive test was successful
• Unfortunately, due to difficulties establishing initial contact with the
picosatellites, mission was prematurely ended when power ran out
– If communication system had been composed of MEMS devices, mission could
have been lengthened!
• Future missions are planned: AFRL MightySat 2.1 and beyond
Figure 15. Picosatellite system architecture [21]
Conclusions
• Network of MEMS satellites for continual base
station communication
– Tap into network much like Internet
– Eliminates remote control stations
• MEMS are ideal for reducing cost of space
exploration
–
–
–
–
Reduced overall mass (cheaper launch)
Increased efficiency
Adaptability
Robust to space environment
• Faster, Better, Cheaper… and Smaller