5. application examples

20
1 5. Application Examples 5.1. Programmable compensation for analog circuits (Optimal tuning) 5.2. Programmable delays in high-speed digital circuits (Clock skew compensation) 5.3. Automated discovery Invention by Genetic Programming (Creative Design) 5.4. EDA Tools, analog circuit design 5.5. Adaptation to extreme temperature electronics (Survivability by EHW) 5.6. Fault-tolerance and fault-recovery 5.7. Evolvable antennas (In-field adaptation to changing environment) 5.8. Adaptive filters (Function change as result of mission change) 5.9 Evolution of controllers

Upload: hiroko

Post on 15-Jan-2016

51 views

Category:

Documents


1 download

DESCRIPTION

5. Application Examples. 5.1. Programmable compensation for analog circuits (Optimal tuning) 5.2. Programmable delays in high-speed digital circuits (Clock skew compensation) 5.3. Automated discovery – Invention by Genetic Programming (Creative Design) 5.4. EDA Tools, analog circuit design - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: 5. Application Examples

1

5. Application Examples

5.1. Programmable compensation for analog circuits (Optimal tuning)5.2. Programmable delays in high-speed digital circuits (Clock skew

compensation)5.3. Automated discovery – Invention by Genetic Programming

(Creative Design)5.4. EDA Tools, analog circuit design 5.5. Adaptation to extreme temperature electronics (Survivability by

EHW)5.6. Fault-tolerance and fault-recovery 5.7. Evolvable antennas (In-field adaptation to changing environment) 5.8. Adaptive filters (Function change as result of mission change)5.9 Evolution of controllers

Page 2: 5. Application Examples

2

Other ideas about evolvable antennas • Use real reconfigurable antenna, morphing in real time under evolutionary control • Components of reconfigurable antennas could be macroscopic (e.g., actuated wires) or MEMS• Evolution of antenna arrays• Co-evolution of antenna & electronic interface (e.g., for matching impedance, etc.)

• A Genetic Algorithm was used in conjunction with the Numerical Electromagnetic Code, Version (NEC) (as simulator) to create and optimize atypical wire antenna designs with impressive characteristics [LIN97]. Evolutionary techniques may revolutionize the design of wire antennas.• GA-optimized Yagi antennas surpass by ~1dB the gain of conventional Yagis• Crooked-wire antennas, consisting of wires joined at various locations and with various lengths (both determined by GA), evolved to unusual shapes, unrealizable using conventional design, and demonstrated excellent performance both in simulations and physical implementation

Fig. 1 Possible novel designs, e.g. tree like antennas [LIN97]

Genetic design of antennas – Linden’s crooked wire antennas

Page 3: 5. Application Examples

3

ST5 Antenna Requirements

• Transmit Frequency: 8470 MHz• Receive Frequency: 7209.125 MHz• Antenna RF Input: 1.5W = 1.76 dBW = 31.76 dBm• VSWR: < 1.2 : 1 at the antenna input port at Transmit Freq, < 1.5 : 1 at the antenna input

port at Receive Freq• Antenna Gain Pattern: Shall be 0 dBic or greater for angles 40 <= theta <=80; 0 <= phi <= 360• Antenna pattern gain (this shall be obtained with the antenna mounted on the ST5 mock-up)

shall be 0.0 dBic (relative to anisotropic circularly polarized reference) for angles 40 <= theta <=80; 0 <= phi <= 360, where theta and phi are the standard spherical coordinate angles as defined in the IEEE Standard Test Procedures for Antennas, with theta=0 to direction perpendicular to the spacecraft top deck. The antenna gain shall be measured in reference to a right hand circular polarized sense (TBR).

• Desired: 0 dBic for theta = 40, 2 dBic for theta = 80, 4 dBic for theta = 90, for 0 <= phi <= 360• Antenna Input Impedance: 50 ohms at the antenna input port• Magnetic dipole moment: < 60 mA-cm^2• Grounding: Cable shields of all coaxial inputs and outputs shall be returned to RF ground at the

transponder system chasis. The cases of all comm units will be electrically isolated from the mounting surface to prohibit current flow to the spacecraft baseplate.

• Antenna Size: diameter: < 15.24 cm (6 inches), height: < 15.24 cm (6 inches)• Antenna Mass: < 165 g

Page 4: 5. Application Examples

4

Antenna Genotype

• Genotype specifies design of 1 arm in 3-space

• Genotype is tree-structured computer program that builds a wire form

• Commands:

• forward(length radius)

• rotate_x(angle)

• rotate_y(angle)

• rotate_z(angle)

• Branching in genotype

branching in wire form

2.5 cm

2.5 cm

5 cm

Feed Wire

f

rx f

rz rxf

Genetic representation is a small programming language with 3 instructions• place wire• place support• branch

Work by Lohn and Linden

Page 5: 5. Application Examples

5

Fitness Function

• Antenna designs are evaluated by NEC4 running on Linux Beowulf supercomputer

• 3 copies w/added noise are evaluated for each design

• Fitness function (to be minimized):F = VSWR_Score *

Gain_Score * Penalty_Score• VSWRs from both freqs are

scaled and multiplied

Gain is sampled at 5 degree increments between theta=40 and theta=90

f = 0 if gain > 0.5 dBf = 0.5 – gain if gain < 0.5 dB

• Penalty: proportional to # gain samples less than 0.01 dB

VSWR

VSWR stands for Voltage Standing Wave RatioThe antenna is usually located some difference from the transmitter and requires a feedline to transfer power between the two. If the feedline has no loss, and matches BOTH the transmitter output impedance AND the antenna input impedance, then - and only - then will maximum power be delivered to the antenna. In this case the VSWR will be 1:1 and the voltage and current will be constant over the whole length of the feedline. Any deviation from this situation will cause a "standing wave“ of voltage and current to exist on the line.

Work by Lohn and Linden 03

Page 6: 5. Application Examples

6

Evolvable antenna system

• Goal: to explore the in-situ optimization of a reconfigurable antenna (vs. optimization on a controlled antenna range)

• Software evolves relay configurations, user subjectively ranks designs based on signal quality (future work will automate this process)

• 30-relay antenna created, along with all necessary software to control and evolve it

• System is able to optimize effectively for frequencies in the upper portion of the VHF broadcast TV band (177 - 213 MHz)

• 1.5m diagonal length (about 1 wavelength at above frequencies)

• Continuing work: improve optimization effectiveness, expand number of relays and antenna size to enhance low frequency performance

Incoming Signal

Received Signal

Control Signals (up to 96 bits)

Evolutionary Optimization

Software

Reconfigurable Antenna

RF Receiver

Antenna Control Interface

Antenna Evaluation Interface

30-relay antenna

Relay module

• Evolvable antenna concept developed by Linden & Stoica• Antenna work performed by D. Linden. JPL funding

1999-2001

Page 7: 5. Application Examples

7

AudioOutput

IncomingSignal

Received Signal

ControlSignals

EvolutionaryOptimization

Software

ReconfigurableAntenna RF Receiver

AntennaControlInterface Genetic

Information

Evolvable antenna systemMotivation: adapt to changes in spacecraft configuration and orientation as well as damage

Evolvable antenna concept by Linden & StoicaWork by Dr. Linden of LIR with JPL funding

RF Generator

RF Receiver PCAudio cable

DAQPad 6507USB cable

RF cable

Control cables

AntennaBarrier

0.430cm

10 750 cm

Demonstrated•Automatic evolution of reconfigurable antennas•Antenna adaptation to different barriers, orientations, frequencies, and loss of control•Superior performance over conventional antenna

1999-2000

Page 8: 5. Application Examples

8

Rec’d Signal Strength Indicator (RSSI) Voltage

Incoming & Outgoing Signals

RF Signals

Control Signals

EvolutionaryOptimization

Software

Reconfigurable Antenna

RF Transceiver

Antenna Control Interface

Genetic Information

Computer receiving /

transmitting data

A/D converter

Digitized RSSI Voltage

Digital Signals

Work by Derek Linden with JPL [email protected]

Evolvable AntennaObjective: show that an evolved reconfigurable antenna, in-situ adapted to environment, can outperform conventional antennas.Frequency: 2.4 GHz. Transceiver sends and receives IP data, can send data while EvAn is being optimized, provides a Received Signal Strength Indicator (RSSI) voltage output

2000-2001

Page 9: 5. Application Examples

9

Physical Setup

RF Transceiver

RF TransceiverData cable

USB cable

RF cable

Control cables

Antenna Barrier (Al sheet, 40 cm x 46.4 cm x 0.5 mm)

1.012.5 cm

30 375 cm

A/D Conv

RSSI Data cable

Wire(RSSI

Voltage)

DAQPad 6507

PC

Dipole Antenna

Page 10: 5. Application Examples

10

Reconfigurable Grid Antenna

•Tuned 5/8 wave antenna used as baseline•50 ohm input impedance• 3.0 dBi nominal gain at the horizon• ~23 cm in length• Omnidirectional in azimuth

•Reconfigurable grid antenna•48 Reed relay switches •13.2 cm x 10.5 cm overall size

•2.1 cm cell size •SMA coaxial cable connection for RF signal•Coaxial control lines feeds control coils•Control lines are roughly perpendicular to plane of

antenna to limit interference

Page 11: 5. Application Examples

11

DEvAn System Details

Antenna control interface

• National Instruments DAQPad 6507• Connects to USB port on PC• Directly drives antenna switches (48)

with TTL output• Receives digital RSSI input (8 bits)• Photo shows DAQPad open, displaying

its screw terminals• Software drivers convert genetic code

into the command strings• Same unit as for previous EvAn project• DEvAn uses ribbon cable for controls

and input connected to ADC unit

• Transceiver • Aerocomm PKLR2400S 2.4 GHz transceiver

•10 mW• OEM Developer version• Serial comm port on board

• Wires are for ground and RSSI voltage to ADC

A/D ConverterChanges analog RSSI voltage from transceiver into 8-bit digital signal into NI DAQPad• LEDs indicate status of digital outputs and read/hold input• Ribbon cables connect to DAQPad, wires connect to switches • Range 0 – 5 v used, transceiver uses 1.4 – 4.5 v for RSSI

Page 12: 5. Application Examples

12

Chromosome Mapping

• Binary chromosome used• The 1s and 0s in the

chromosome correspond to closed and open switches (1 closed, 0 open)

• The connections between the DAQPad and the switches define the chromosome mapping

• Wire connections are essentially fixed in place

• Mapping is done in software• Used spiraling to help

correspond physical distance with distance on chromosome

Page 13: 5. Application Examples

13

Objective function and GA Parameters

• Objective Function Maximize RSSI level

• Uses digitized voltage

• Favors designs with increased values

• Very noisy, bi-modal signal

• Overcome with repeated testing, averaging, rejection regions

• GA Parameters

• Fitness-proportional selection with scaling

• One-point crossover

• All individuals evaluated in each generation

• Top 50% carried to next generation

• Mutation rate = 20%

• Population size = 10

• 10 generation limit (+ initial generation)

Page 14: 5. Application Examples

14

Evolution and Test Matrix

3 Orientations• Broadside (plane of antenna |_ to

signal)• 45 deg • Endfire (plane of antenna // to

signal) 2 Barrier configurations (placed before optimization begins)• None• Solid metal (Al) sheetPolarization is vertical for all tests

Objective: Evolve to maximize RSSI levelHypothesis: Higher signal input => better data transmission

Page 15: 5. Application Examples

15

Results

Score: higher is better Stdev: lower is better

Without Barrier

With Barrier

Page 16: 5. Application Examples

16

Adaptation to Environment

Perfect conducting wire mesh 40 cm x 46.4 cm, 12.5 cm above antenna, with 2 cm x 2 cm cells

(Aluminum sheet was used in the hardware test)Showing optimal configuration found in hardware

5/8-wave AntennaOptimized Antenna

Antenna adaptation to different barriers and orientations,

outperforms conventional commercial antenna.

Page 17: 5. Application Examples

17

Scores vs dB gain• A rough calibration exists for voltage (score) vs dBm

• For baseline vs DEvAn broadside:

– Baseline score: 119.0 score => -59.1 dBm

– DEvAn score: 129.5 score => -56.4 dBm

– Difference: 2.7 dBm

• This indicates that almost twice as much power was delivered by DEvAn than by the baseline antenna

Page 18: 5. Application Examples

18

General Observations

• Time to optimize

– ~8 minutes per 10-generation run

– Antenna reconfigured in <0.1 seconds

– Time mostly due to re-sampling RSSI voltage

• Repeat runs of the same antenna and test configuration gave very similar performance

• Noise was a major challenge, as it was in the original EvAn system

Page 19: 5. Application Examples

19

Relationship Btw RSS and Data Transmission

Antenna Type Avg score Errors

Dipole 115 (± 5) 574 (± 318)

DEvAn 128 (± 2) 509 (± 121)

DEvAn (sub-opt) 119 (± 4) 296 (± 162)

There appears to be a “sweet spot” for RSSI. Highest is notnecessarily the best for data transmission! Needs further exploration.

• Hypothesis: Higher signal input => better data transmission

• Testing done with packet-level data communications program with test data sent over network

• Having fewer errors translates into having greater bandwidth

• Tests performed with packet-level send and receive program

– Script used to continuously transmit packets across the network

– Number of errors per 2400 packets recorded (repeated 5 times)

– Separation: 16.5

Page 20: 5. Application Examples

20

Conclusion

Accomplishments

• Testing of 2.4 GHz reconfigurable antenna that can both receive and transmit

• Automatic evolution of reconfigurable antenna using RSSI while leaving data channel usable

• Antenna adaptation to different barriers and orientations

• Performance above that of a conventional commercial antenna