UCSB Aquamodem Design and Development of a Design and Development of a Software-Defined Underwater Software-Defined Underwater Acoustic Modem for Sensor Acoustic Modem for Sensor Networks for Environmental Networks for Environmental and Ecological Research and Ecological Research Tricia Fu, Daniel Doonan, Chris Utley, Ronald Iltis, Ryan Kastner, and Hua Lee
24
Embed
UCSB Aquamodem Design and Development of a Software-Defined Underwater Acoustic Modem for Sensor Networks for Environmental and Ecological Research Design.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
UC
SB
A
quam
odem
Design and Development of a Design and Development of a Software-Defined Underwater Software-Defined Underwater Acoustic Modem for Sensor Acoustic Modem for Sensor Networks for Environmental Networks for Environmental and Ecological Research and Ecological Research
Tricia Fu, Daniel Doonan, Chris Utley, Ronald Iltis, Ryan Kastner, and Hua Lee
UC
SB
A
quam
odem
OutlineOutline
Summary of MP Algorithm and Summary of MP Algorithm and Aquamodem ImplementationAquamodem Implementation
Comparison w/ other Acoustic Comparison w/ other Acoustic ModemsModems
Hardware ImplementationHardware Implementation Test Results: wire, air, tankTest Results: wire, air, tank Future ideas (in the works)Future ideas (in the works)
UC
SB
A
quam
odem
Design ObjectivesDesign Objectives
Operating range: >= 100 mOperating range: >= 100 m Operating depth: up to 20 mOperating depth: up to 20 m 90-day deployed lifetime90-day deployed lifetime Handle multipath spread of Handle multipath spread of
several ms (up to 11.2 ms)several ms (up to 11.2 ms) What else to add here?What else to add here?
UC
SB
A
quam
odem
MP AlgorithmMP Algorithm
UC
SB
A
quam
odem
MP Algorithm: SimulationMP Algorithm: Simulation
Sym
bol
Delay (in ms)
Simulated MIP, 5 Training Symbols, SNR = -4 dB, Estimated NF = 5 to 16
1 2 3 4 5 6 7 8 9 10 11
200
400
600
800
1000
1200
1400
1600
Sym
bol
Delay (in ms)
Simulated MIP, 2 Training Symbols, SNR = -4 dB, Estimated NF = 3 to 17
1 2 3 4 5 6 7 8 9 10 11
200
400
600
800
1000
1200
1400
1600
Estimated MIP from simulations, with actual number of multipaths = 10 and SNR = -4 dB.
• Left: 2 training symbols, SER = 0.6645.
• Right: 5 training symbols, SER = 0.1023.
UC
SB
A
quam
odem
Matching Pursuits vs Matching Pursuits vs FSKFSK Direct Sequence Spread Spectrum Direct Sequence Spread Spectrum
(DSSS) preferred over frequency (DSSS) preferred over frequency hopping when channel exhibits hopping when channel exhibits deep, narrow fades so that signal deep, narrow fades so that signal energy spread over wide bandwidth.energy spread over wide bandwidth.
MP produces channel estimates MP produces channel estimates during symbol demodulation during symbol demodulation process.process.
cosine filter, 30 % excess bandwidthcosine filter, 30 % excess bandwidth 24 kHz center frequency24 kHz center frequency Guard interval, normally same length as Guard interval, normally same length as
symbol duration.symbol duration. At Fs = 96 kHz rate, symbol rate is ~ 54 At Fs = 96 kHz rate, symbol rate is ~ 54
Hz, bit rate ~ 161 bps.Hz, bit rate ~ 161 bps.
UC
SB
A
quam
odem
Hardware Hardware ImplementationImplementation Texas Instruments’ C6713 DSP BoardTexas Instruments’ C6713 DSP Board Sonatech custom built transducers: Sonatech custom built transducers:
horizontal omnidirectional coverage in horizontal omnidirectional coverage in 20 kHz – 30 kHz range20 kHz – 30 kHz range
System power: alkaline primary cellsSystem power: alkaline primary cells Commercial single-chip class D power Commercial single-chip class D power
amp driving matching network, amp driving matching network, automatic transmit/receive switch.automatic transmit/receive switch.– Tx transducer also connected to low-noise Tx transducer also connected to low-noise
variable gain amplifier when power amp variable gain amplifier when power amp not operating.not operating.
UC
SB
A
quam
odem
Hardware Hardware ImplementationImplementation Blah need djd input!Blah need djd input!
UC
SB
A
quam
odem
Wire Test: resultsWire Test: results
Multipath Intensity Profile, NF MAX = 5
Sy
mb
ol
Delay (in ms) 2 4 6 8 10
500
1000
1500
2000
2500
3000
3500
4000
-25 -20 -15 -10 -5 0 5 10 15 20 250
2000
4000
6000
8000
10000
12000
Frequency (Hz)
Ma
gn
itu
de
Doppler Spread, Wire Test
UC
SB
A
quam
odem
Air Test in Lab: resultsAir Test in Lab: results
Multipath Intensity Profile, NF MAX = 20
Sy
mb
ol
Delay (in ms) 25 50 75 100
50
100
150
200
250
-2 -1.5 -1 -0.5 0 0.5 1 1.5 20
50
100
150
200
250
300
350
400
450
500
Frequency (Hz)M
ag
nit
ud
e
Doppler Spread, In-Lab Air Test
UC
SB
A
quam
odem
Air Test in Hallway: Air Test in Hallway: setupsetup
Earlier Tests Later Tests
Range 67.5 ft 117 ft
NF Fixed Estimated
# training symbols 2 5
UC
SB
A
quam
odem
Air Test in Hallway : Air Test in Hallway : resultsresults
Lower SNR due to Lower SNR due to Tx power amp Tx power amp problems (-1.0 dB problems (-1.0 dB vs. 6.1 dB)vs. 6.1 dB)
Only 0.60 m Only 0.60 m results collectedresults collected
UC
SB
A
quam
odem
Tank Test Series B: Tank Test Series B: resultsresults
Series A: 6.1 dB vs. Series B: -1.0 dB SNR Series A: 6.1 dB vs. Series B: -1.0 dB SNR Note weak path at 6 ms not as apparent in Series B Note weak path at 6 ms not as apparent in Series B
MIP compared with Series A MIP.MIP compared with Series A MIP.
Test Series A: Multipath Intensity Profile, 0.6 m distance, NF = 5
Sym
bol
Delay (in ms) 2 4 6 8 10
200
400
600
800
1000
1200
1400
1600
Test Series B: Multipath Intensity Profile, 0.6 m distance, NF = 1-2