Design and Development of Underwater Acoustic Modem for Shallow Waters and Short Range Communication

M.S.Ramaiah School of Advanced Studies 1

Final Project Presentation

Vinay DivakarCWB0912001 , FT-12

M. Sc. [Engg.] in Electronic Systems Design

Engineering

Proposed Academic Guides :

Mr. Sreekrishna R.

Ms. Preetham Shankpal

Underwater Acoustic Modem.

Design and Development of Underwater Acoustic

Modem for Shallow Waters and Short Range

Communication

M.S.Ramaiah School of Advanced Studies

Introduction A modem is responsible for transmitting and receiving data.

In air electromagnetic waves are used for the transmission and

reception of the data, however EMW cannot be used for data

transmission underwater, since the bandwidth is severely limited to

KHz range.

RF signals are easily absorbed underwater.

And the antenna required to transmit RF signals at KHz will be

very large with high power consumption, therefore acoustics was

seen as the most reliable choice of underwater communications.

2Underwater Acoustic Modem.


Introduction continued

Due to the unique characteristics of water, its not been an easy task to

set up UWSN’s.

Available BW’s for different ranges in UWA channel


Range [Km] Bandwidth [BW]

KHz

Very Long 1000 <1

Long 10 - 100 2 - 5

Medium 1 - 10 ~10

Short 0.1 - 1 20 -50

Very Short <0.1 <0.1

Table 1. BW available at different ranges[4]


Summary of Literature ReviewAuthors Data Rate Bit Error Rate

(BER)

Signal-to-Ratio

(SNR)(dB)

Ratings Range

B. Benson et al 200 bps 10-2 10dB Transmit:2W to

40W

Receive:

375mW(Standby)

& 750mW(active)

<350m

Jack Wills, Wei Ye

and John

Heidemann

- 10-5 - 5V 50m

Saunvit Pandya et

al

- - 20.2dB(air)

250dB(underwater)

5V 155cm(air)

20cm(underwater)

Raja Jurdak et al 24 bps - 6.95dB - 13m

Heungwoo Num

and Sunshin

100 bps - - 3.3V 3m

Jun-Ho-Jeon and

Sung-Joon-Park

48 Kbps(watertank)

5 Kbps(Pond)

1 Kbps(Pond)

1 Kbps(offshore)

- - 14.8V 1m

1m

30m

20m

Underwater Acoustic Modem. 4


Existing systemsThe significant deficiencies identified in the existing underwater commercial acoustic

modems such as Linquest and Benthos[15] are as follows

• The physical structure of these commercial modems is bulk and also heavy in

weight

• Since these modems are basically designed for long range communication,

therefore they consume immense amounts of power, hence the overall power

consumption is high

• The cost of development of one such commercial modem is very high, therefore

not a cost effective option for deployment of underwater sensor networks

• These modems are generally designed for deployment in the deep oceans, thus

making it difficult to access these modems, and frequent charging or replacement

of the batteries is necessary which includes high maintenance cost.



Target SystemTo overcome the stated deficiencies,

• The target system should be designed for short range communication (<100 m).

• The target system should be designed for shallow waters of depth (<100m).

• The cost of development per node should be low for making it reliable for

deployment of underwater sensor network using several such low cost nodes.

• The target system should be small in size i.e. the size per node should be small.

• The overall power consumption per node or system should be low in order to

increase the lifetime of the system, and reduce the maintenance cost

• The system should be able to efficiently transmit and receive data adapting the

underwater environmental characteristics.



Problem Definition• The customer problems with existing modems are, they are

designed for long range communication and for deep waters.

Therefore the power consumption and the maintenance cost is

high.

Hence the chosen problem is to, Design and Development of

Underwater Acoustic Modem for shallow waters and short range

communication



Aim of the Project

The aim of the project is to Design and Develop a Underwater

Acoustic Modem for shallow waters (<100m) and short range

(<100m) communication.



Project Objectives• To conduct literature review on acoustic signals, existing underwater acoustic

modems, underwater channel characteristics, modulation schemes, error

detection and correction schemes.

• To arrive at a system level specifications and functional block diagram of the

low rate underwater acoustic transceiver system.

• To model the underwater acoustic channel and integrate noise models to the

designed acoustic modem for analysing the system performance.

• To design, model and simulate the underwater acoustic modem using

appropriate simulation tool for the desired specifications.

• To realize the designed system in hardware for low data rate communication

• To test and analyse the performance of the developed system



Concept Development


Fig 1. Divergent and Convergent Thinking


Design SpecificationsParameters Specification

Modulation Scheme BFSK

Receiver Non-coherent BFSK Detection

Error Protection and Detection Cyclic Redundancy Check (CRC-6420)

Bandwidth (BW) 3200 Hz

Frequency Separation (Δf) 1 KHz

Data rate (N) 1600 bps (1300bps + Redundancy)

Redundancy Bits ~300 bps

Transmit signal Power (P) 0.1 W

Sampling Time ,Ts 1/1600

Symbol period,Tsym 0.625 ms

Frames per second 100 Fps

Operating Range (R), meters Below 100 m

Operating Depth (D), meters Below 100 m


Table 1- Design specifications


Advanced Development


Fig 2. Detailed Block Diagram


Underwater/Aquatic Channel Block


Fig 3. Underwater Channel Block Description


Underwater Channel Model• Propagation Delay block

• Path Loss (Absorption Loss + Spreading Loss)

• AWGN

• Ambient Noise (Turbulence, Wave, Shipping and Thermal)

• Multipath delays for surface and sea bottom reflections



Complete Underwater Channel


Fig 4. Underwater Acoustic Channel


Propagation Delay Block


Fig 5. Propagation Delay underwater


Spreading Loss


Plspreading (r) = k*10*log (r) (dB)

r: Range in meters (m)

k: Spreading Factor (bounded=1, unbounded = 2)

Fig 6. Spreading Loss


Absorption Loss


Fig 7. Boric Acid Component


Absorption Loss contd…



Path Loss


Fig 8. Path Loss (Absorption + Spreading Loss


Ambient Nosie



Ambient Noise contd…


Fig 9. Ambient Noise Models


Multipath Delays

Underwater Acoustic Modem.23

• 1st delay due to reflection from surface (4ms)

• 2nd delay due to reflection from sea bottom (15ms)

Fig 10. Multipath Delays


Multipath Delays contd…

The parameters of the two delay signals used in the simulations are as

follows;

• Delayed Signal1 : Time shift, T1 : 9600 samples (4Ts)

Phase shift, ⱷ : 20 degrees

Frequency shift, FD1: 20 Hz (2 m/s)

Gain, α1: 0.25

• Delayed Signal2 : Time shift, T1 : 24000 samples (15Ts)

Phase shift, ⱷ : 30 degrees

Frequency shift, FD1: 40 Hz (4 m/s)

Gain, α1: 0.5



System Model in Matlab/Simulink


Fig 11. System Model


Model ExperimentsThe Systems End to End performance is simulated for different

mediums;

• Model 1 – System with AWGN

• Model 2 – System with AWGN + Path Loss

• Model 3 – System with AWGN + Path Loss + 2 Multipath delays

• Model 4 – System with AWGN + Path Loss + 2 Multipath delays

+ Shipping Noise (complete underwater channel)



Model 1 -AWGN


Fig 12. System with AWGN


Simulation Results – Model 1

• Eb/No vs BER • System is showing excellent

performance for AWGN medium

BER Eb/No (dB)

10-5 11

10-4 10

10-3 9


Fig 13. ber vs eb/no for AWGN


Model 2-AWGN + Path Loss


Fig 14. System with AWGN + Path Loss


Model 2 – Simulation Results• Depth – 10 m • System operates well upto a range

of 30 m, Eb/No required is 19 dB

for BER 10-4

Depth (m) Range (m) BER EbNo

(dB)

10 50 10-3 22

30 10-3 18

10-4 19

10 10-5 20

10-4 19

10-3 18


Fig 15. Depth 10m


Model 2 – Simulation Results contd…

• Depth – 30 m • System operates well upto a range

of 100 m, Eb/No required is 12

dB for BER 10-5

Depth (m) Range (m) BER Eb/No

(dB)

30 100 10-5 12

10-4 11

10-3 10

50, 30 and

10

10-5 11

10-4 10

10-3 9


Fig 16. Depth 30m


Model 2 – Simulation Results contd…

• Depth – 50m • An Eb/No of 6 dB can attain a

transmission range upto 100m for

a BER of 10-4.


Depth (m) Eb/No

(dB)

Range (m) BER

50

6

10 10-4

30 10-4

50 10-4

100 10-4

Fig 17. Depth 50m


Model 3-AWGN + Path Loss + Multipath


Fig 18. System with AWGN + Path Loss +2 Multipath delays


Model 3 - Simulation Results• Depth – 10m

• As multipath induced, system

cannot operate for a depth of 10m

anymore, performance totally

degraded

Depth (D)

in meters

Range in

meters

Eb/No BER

10

10 22 10-2

30 22 10-2

50 22 10-2


Fig 19. Depth 10m


Model 3 - Simulation Results contd…

• Depth – 50m • System can achieve a BER of

10-3 for a Eb/No of 19 dB, with a

transmission range of 10m.

Depth (D)

in meters

Range in

meters

Eb/No BER

50

10 24 10-4

19 10-3

30 25 10-3

50 25 10-3


Fig 20. Depth 50m



• Depth - 100m • BER of 10-3, range upto 30m for

Eb/No of 16 dB

Depth (D)

in meters

Range in

meters

Eb/No BER

100

10 18 10-5

16 10-4

12 10-3

30 24 10-4

16 10-3

50 25 10-3


Fig 21. Depth 100m


Model 4 – underwater channel with shipping noise


Fig 21. System with AWGN + Path Loss + Multipath + Shipping noise


Model 4 - Simulation Results

• Depth – 10m• BER is high & system

performance degraded

Depth

(m)

Range

(m)

Shipping

Factor

(S)

Eb/No

(dB)

BER

10 10 0 23 10-2

0.5 25 10-2

1 17 10-1


Fig 22. Depth 10m




• Depth – 30m

• Ranges – 10m, 30m, 50m

• S – 0, 0.5, 1

Fig 23. Depth 30



Depth (m) Range (m) Ship Factor (S) Eb/No (dB) BER

30 10 0 22 10-5

20 10-4

17 10-3

0.5 23 10-5

21 10-4

18 10-3

1 4 10-1

30 0 22 10-4

0.5 22 10-3

1 5 10-1

50 0 23 10-2

0.5 23 10-2

1 6 10-1


• system shows good performance for a range of 10 m and 30m, however the

system performance degrades as the shipping activity is at peak




• Depth – 50m

• Ranges – 10m, 30m, 50m

• S – 0, 0.5, 1

Fig 24. Depth 50m




50 10 0 18 10-5

15 10-4

13 10-3

0.5 18 10-5

21 10-4

16 10-3

1 13 10-2

30 0 18 10-5

16 10-4

13 10-3

0.5 20 10-5

18 10-4

14 10-3

1 13 10-1

50 0 21 10-5

18 10-4

14 10-3

0.5 23 10-5

21 10-4

17 10-3

1 6 10-1





• Depth – 70m

• Ranges – 10m, 30m, 50m

• S – 0, 0.5, 1

Fig 25. Depth 70m




70 10 0 14 10-5

12 10-4

10 10-3

0.5 15 10-5

13 10-4

11 10-3

1 17 10-2

30 0 15 10-5

13 10-4

11 10-3

0.5 15 10-5

14 10-4

14 10-3

1 20 10-2

50 0 16 10-5

14 10-4

11 10-3

0.5 17 10-5

15 10-4

11 10-3

1 22 10-2



Performance behavior of system

• Model 1 - The system provides excellent performance of BER 10-5 for a signal

strength of 11 dB, for AWGN channel.

• Model 2 – For a depth of 10m, the system can operate upto a range of 30m for a

signal strength of 19 dB. As the depth increases to 50m, systems range extends

to 100m for a low signal strength of 6 dB

• Model 3 – Multipath are very significant shallow waters, therefore for a depth

of 10m, the systems performance is degraded with high BER. For a depth of

50m, system operates efficiently for range of 10m, BER of 10-4 for signal

strength of 19 dB. For a depth of 100m, the system can achieve a BER of 10-4

for Eb/No of 18 dB with range 10m.

• Model 4 – For a depth of 10m, the system performance is degraded. For a depth

of 30m, with s-0, 0.5, the system can operate efficiently upto a range of 30m

with BER 10-4 for EbNo of 20 dB. For a depth of 50 m ,the system shows

excellent performance for a range upto 30m with BER of10-4, for EbNo of 18

dB. For a depth of 70m, the system is operating at its best for a range up to

50m, BER 10-5 for a Eb/No of 17 dB.



Challenges faced• Modeling and integrating the complex characteristics resembling the

underwater acoustic channel.

• Identifying the appropriate frequency separation (Ϫf) required in a Bandwidth

of 3.2 KHz for BFSK. Therefore the system was simulated for different Ϫf of

2.5KHz, 2KHz and 1 KHz, thus the system provided an optimum performance

for 1KHz.

• The systems performance degraded when multipath was introduced, for a

CRC-4. Therefore the system was simulated for different CRC’s, such as

CRC-6, CRC-8 and CRC-16. Hence CRC-6 was used to increase the

redundancy between packets in order to minimize ISI effects and achieve an

optimum performance.



ConclusionFrom the simulation results, it is concluded that;

• The performance of the system improves as the depth increases, since, as the

depth increases the affects of multipath, absorption loss and shipping noise is

minimized.

• From the Table, seen that, system achieves excellent performance for a depth

of 30 m and above, up to a range of 50m with a BER of 10-4 for a maximum

required Eb/No of 20 dB, for a low shipping factor of 0 and 0.5.


Depth (m) Range (R) Eb/No (dB) BER

10 10 23 10-2

30 30 20 10-4

50 30 18 10-4

70 50 17 10-5

• The systems performance is degraded when operated at a depth of 10m

and the system cannot operate efficiently while the shipping activities

are at peak i.e. S-1.

Table – Final System Performance Specifications


Conclusion• The system performed efficiently for a depth of 10m up to model-2, but the

introduction of multipath (4ms and 14.33ms), degraded the signal strength

drastically while the systems performance degraded. Therefore multipath

significantly affects the performance of the system for shallow waters.

• The requirement of Eb/No to achieve a BER of 10-4 can be further reduced by

increasing the depth of the system placed underwater, above 100m, but this

will conflict with the requirement of the system to operate at shallow waters.

• Therefore using the CRC-6 scheme and Non-coherent BFSK scheme to design

an underwater acoustic modem, can achieve a good performance operating at

shallow depths of 30m, 50m and 70m, providing a transmission range up to

50m.



Future Work

• Errors can be minimized further by using, error correction by

retransmission of the non-critical data frames using the Selective ARQ

(Automatic Repeat Request) protocol.

• In order to correct some critical data, an error correcting scheme can be

used in conjunction with CRC, to correct critical data on spot and transmit.

This will provide excellent error detection as well correction capabilities.



Hardware Details

Transmitter :

• Input Data – Terminal 1 from PC

• Level Converter – Max 232

• Microcontroller-CRC

• Ultrasound Transducer (40 KHz)

Receiver:

• Ultrasound Transducer (40 KHz)

• Microcontroller-CRC

• Level Converter

• Output Data – Terminal 2 to PC



Block Diagram of System


Fig 26 – Hardware System Block Diagram


Hardware Setup



Transmission Process



Data Reception Process



ReferencesBenson.B, Y.Li, Kastner, Faunce, Domond, Kimball and Schurgers (2010).

‘Design of Low-Cost Underwater Acoustic Modem for Short Range Sensor

Networks’. Oceans 2010 IEEE, San Diego.

Akyilidz, Pompili, Melodia (2005)‘Underwater Acoustic Sensor Networks :

research Challenges’ Elsevier, Ad Hoc Networks3,257-279,February.

Wills, Wei Ye and Heidemann (2006) ‘Low-Power Acoustic Modem for Dense

Underwater Sensor Networks’. USC information Sciences Institute.

Jullo Diago Miranda Xavier (2012) ‘Modulation Analysis for an Underwater

Communication Channel’. FEUP, October.

Gunilla Burrowes and Jamil.Y.Khan (2011) ‘Short-Range Undewater Acoustic

Communication Networks’. InTech, ISBN: 978-953-307-432-0, Australia.



ReferencesJurdak, Agmas, Beldi and lopes (2007) ‘Software Modems for Underwater

Sensor Networks’. Oceans 2007 IEEE, ISBN-978-1-4244-0635-7, 1 – 6.

Saunvit Pandya, Jonathan Engel, Jack Chen, Zhifang Fan, Chang Liu (2005)

‘CORAL: Miniature Acoustic Communication Subsystem Architecture for

Underwater Wireless Sensor Networks’. SensorsIEEE.

Heungwoo Nam and Sunshin An (2007) ‘An Ultrasonic Sensor Based Low-

Power Acoustic Modem for Underwater Wireless Sensor Network’. IEEE

Secon, 2010, 1-3.

Babu, Dr. Rao (2011) ‘Evaluation of BER for AWGN, Rayleigh and Rician

Fading Channels under various Modulation Schemes’. International Journal

of Computer Applications, Volume 26-9.



ReferencesBabu, Dr. Rao (2011) ‘Evaluation of BER for AWGN, Rayleigh and Rician Fading

Channels under various Modulation Schemes’. International Journal of Computer Applications, Volume 26-9.

Jeon and Park (2010) ‘Implementation of a Low-Power Acoustic Modem for Underwater Wireless Sensor Networks’. IEEE Seacon.

Nejah NASRI , Laurent ANDRIEUX , Abdennaceur KACHOURI , and MounirSAMET (2009), ‘Behavioral Modelling and Simulation of Underwater Channel” WSEAS TRANSACTIONS on COMMUNICATIONS , February.

Forouzan (2007) ‘Data Communications and Networking’. 4th ed, New York McGraw-Hill.

Haykins (2001) ‘ Communication Systems’. 4thed, U.S.A, McMaster University.

LinkQuest, Inc., Underwater Acoustic Modems,

http://www.link-quest.com/html/uwm hr.pdf


http://www.link-quest.com/html/uwm hr.pdf


Activity Weeks 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Pre-Project

Presentation

Literature

Survey

Concept

Exploration

Concept

Selection

Underwater

Channel

Modelling

Design and

Modelling of Tx

& Rx

Model

&Simulations

Model Analysis

Interim Review

Model

Optimization

Preparation of

Final Report

and

Presentation

Final

Presentation

Project Schedule


Thank You


Design and Development of Underwater Acoustic Modem for Shallow Waters and Short Range Communication

Engineering

Transcript of Design and Development of Underwater Acoustic Modem for Shallow Waters and Short Range Communication