Eric L' Morgan

1
Implementing Remote Real-time Biosensing in
Watershed Management Historical Perspectives
(1970 to 2005)

• by
• Eric L. Morgan
• Dennis. B. George, Ester. T.Ososanya, Anil. U.
  Kukreja, Ninetha Thirunavukkarasu, Subramanian S.
  Meiyappan, and Erik Suffridge
• Center for the Management, Utilization, and
  Protection of Water Resources,
• Department of Electrical Computer Engineering
  and Department of Biology
• Tennessee Technological University, Cookeville,
  TN 38505
• and
• W. Thomas Waller , Kenneth L. Dickson, and Joel
  H. Allen,
• Institute of Applied Sciences and Department of
  Biological Sciences
• University of North Texas, Denton, Texas 86203

2
(No Transcript)
3
Introduction

• Biological Monitoring
• Orderly use of biological responses to evaluate
  changes in environment with intent to use the
  information in quality control (Morgan et al
  1999)
• Automated Biomonitoring
• Designed to support quality control by
  continuously recording biological responses of
  organisms while subjected to in-situ, ambient
  environmental conditions
• Providing real-time data on physiological and
  behavioral status of organism

4
Advantages of Automated Biosensing

• Provides multi-species assessments
• Provides continuous, real-time monitor of
  biological responses
• Operates at remote locations on solar cells
• Transmits data from remote sites via satellite,
  cell phone or radio
• Gives an Early Warning of Toxic stress
• Intelligent System provides emergency response

5
Project Objective

• Objective To design and implement an Early
  Warning Biosensing System that will detect toxic
  stress in aquatic animals at remote river sites
  and transmit sensed data to a distant data
  processing center for stress detection/toxic
  prevention.

6
Automated Biosensing System

• Automated acquisition and storage of data from a
  small population of aquatic organisms and data
  transmission to a distant data processing system
  (for stress detection)

7
Remote In-stream Fish Chamber(with Rainbow Trout)
8
Example of Bioelectric Signals Monitored

9
Multiple Species- Water Quality Monitoring Logic

10
Automated Biological Monitoring Network for
Watershed Assessment 1970 (Concept Proposed by
Cairns, et. al.)

11
Automated Biological Monitoring Network for
Watershed Assessment 1976 (Remote
Platforms Proposed by Morgan, et. al)
12
COMPLEX WATERSHEDS
13
REGIONAL STUDY AREA

14
RESEARCH WATERSHED- Tellico, TN

15
Instrumented Watershed- Tellico (1986-91)
16
Remote Platform (view upstream) - Tellico

17
Remote Platform (Side view)- Tellico

18
Remote Platform (Fish Chambers)- Tellico

19
Remote Platform (Fish Chambers)- Stilling Well

20
Remote Platform (Instrumentation)
21
Remote Platform (Water Quality Instruments)
22
Date Collection Platform (DCP)- Tellico
23
Data Collection Platform Logistics

24
Effects of Hydrograph on Trout Breathing Rates

25
Effects of Acidic Flows on Trout Breathing Rates

26
Little Miami River System

• Prototype Multi-species Automated Biosensing
  System installed at the Little Miami River,
  Cincinnati, Ohio. (collaborative effort between
  Univ. North Texas, Tenn. Tech. Univ. and the
  US-EPA)
• Two biosensing systems on one remote river
  platform
• One common data communication device
• Handshaking between two system controllers
• Two important system design requirements
• Low power consumption
• Small Size/configuration

27
System Requirements

• Sensor Designed to Detect Bioelectric Responses
  from Aquatic Animals
• Signal Conditioning System
• Multi-channel data Acquisition System
• Compact Low-power Consuming System Controller
• Remote Data Communication System
• Remote Data Transmission via Cell Phone

28
Biosensing System

• Sensors Pairs of Stainless Steel Probes
• Signal Conditioning System Rack with plug-in
  cards
• Keithley PC-add on Data Acquisition Board with
  Channel Expansion Modules
• System Controller - 133MHz Pentium PC
• Data Transmission from Remote Platform to Biology
  Electrical Engineering Department at Tennessee
  Tech. Univ. and Made Available over the Internet

29
Proposed New-Generation Biosensing System for
Little Miami River, OH
30
In-stream Sensor Housing (torpedo)
31
On-site Experimental System Logic
32
Experimental Platform On-site
33
Field Work Review

• Coaxial cables used to carry the signal from the
  torpedo to the cabinet found to be expensive and
  difficult to work with
• Signal conditioning system found to be bulky
• Poor signal resolution of bioelectric signals
  after digitization
• Loss of data during transmission Data flow
  control incorporated between the computer and the
  modem

34
System Modifications

• Build a signal conditioning system made up of
  small modules
• Attach potted signal conditioning modules to the
  fish chambers
• Use simple instrumentation wire to carry the
  signals
• Build a multi-channel data acquisition system
  with automatic gain control for each channel

35
System Components - 3rd Generation Remote
Biosensing Network

• Probe, Chamber, and Animal Module
• Signal Conditioning Module
• Data Acquisition Module
• Signal Processing Module
• Pattern Recognition Module
• Early Warning Control Module
• Proposed Coordinated Watershed Network Module

36
3rd - Generation Biosensing System Logic
37
Potted Signal Conditioning Module

• Original Design by US-Tennessee Valley Authority
• Two components
• Fixed-gain Instrumentation Amplifier,
• Bandpass Filter

38
Experimentally Generated Magnitude Response
39
System Components - 3rd Generation Remote
Biosensing Network

• Probe, Chamber, and Animal Module
• Signal Conditioning Module
• Signal Processing Module
• Pattern Recognition Module
• Early Warning Control Module
• Proposed Coordinated Watershed Network Module

40
AGC Multi-Channel Data Acquisition System
41
AGC 16-channel Data Acquisition Board
42
System Operation

• System operation split up into three phases
• Gain Setting Phase
• Data Acquisition Phase
• Data Transmission Phase
  

43
Gain Setting Phase

• Calculate Vmax, Vmin, delta_v, and V_average
• Select gain_value from Table and calculate new
  range,
• range (delta_v gain_value) V_average
• Store gain_values to be used in the data
  acquisition phase
  

44
Data Acquisition phase

• Data is acquired simultaneously from all the
  channels
• Based on the channel selected, the appropriate
  gain control bits , bit1..0 are fed to the
  programmable gain amplifier
• The data is stored in a 2-dimensional array in
  the system memory

45
System Testing

• System testing split into two parts
• Data acquisition system tested in the laboratory
• Data communication system tested in the field
  

46
Data Acquisition System
47
System Components - 3rd Generation Remote
Biosensing Network

• Probe, Chamber, and Animal Module
• Signal Conditioning Module
• Data Acquisition Module
• Pattern Recognition Module
• Early Warning Control Module
• Proposed Coordinated Watershed Network Module

48
Example of Bioelectric Signals Monitored

49
Waveforms (1)
50
Waveforms (2)
51
Data gt Fast Fourier gt Power Spectral Density
52
PSD gt Artificial Neural Network
53
Data Management for Early Warning Alert

54
System Components - 3rd Generation Remote
Biosensing Network

• Probe, Chamber, and Animal Module
• Signal Conditioning Module
• Data Acquisition Module
• Signal Processing Module
• Early Warning Control Module
• Proposed Coordinated Watershed Network Module

55
Pattern Recognition / Classification Module

• Bayes Statistical Classifier
• Artificial Neural Network Classifier

56
Bayes Classification Module
57
ANN Classification Module

•  

58
Comparison of Artificial Neural Networks and
Bayes Classification Techniques.
59
System Components - 3rd Generation Remote
Biosensing Network

• Probe, Chamber, and Animal Module
• Signal Conditioning Module
• Data Acquisition Module
• Signal Processing Module
• Pattern Recognition Module
• Proposed Coordinated Watershed Network Module

60
Data Transmission Phase

• Standard modem AT commands are outputted to the
  CDPD modem with one difference, instead of the
  tel , give IP address of host system
• Command Response Status
  of modem
• AT OK modem
  is ON
• ATL192008N1 OK
  modem settings 19200 baud,
• 
  8-data bits, no parity,
  1 stop
• ATS57? 161
  cell-tower available
• (channel acquired)
• ATDT149.149.42.85/4001 CONNECT
  connection established with host system at
  TTU with the IP address

61
Communication System (1)
62
Communication system (2)

• Types of Communication
• Between the two single board computers (Tenn.
  Tech Univ. North Texas), for sharing the modem
• Between the Tenn. Tech, computer and the modem,
  to establish flow control
• Between the single board computer and the Tenn.
  Tech. host system for synchronous data
  transmission

63
System Limitations

• Maximum overall frequency of operation of the AGC
  data acquisition system for all channels 20 KHz
• (the time delay between the SOC and EOC pulses
  is 50 usec)
• Limited amount of data can be acquired and stored
  
• (single board computer memory 0.5MB)
• Once the modules are potted, they cannot be
  modified to be used with any other species

64
Summary of Early Warning Control Module
65
Proposed Coordinated Monitoring Network for Water
Resource Protection

66
Summary

• Main objective to design and implement an
  automated biosensing system for continuous
  biological monitoring on a remote river platform
• Existing biosensing systems and remote data
  communication options were studied
• Second generation biosensing system built and
  installed on two experimental watersheds
  Laurel Branch, Tellico, Tennessee and Little
  Miami River, Cincinnati, Ohio
• Field work reviewed and some modifications in
  the system were proposed.

67
Summary (Contd)

• Proposed modifications were designed into a
  prototype 3rd generation biosensing system built
  with potted signal conditioning modules and
  automatic gain control, multi-channel data
  acquisition system
• Two major tasks performed by the system, data
  acquisition and data communication were tested

68
Observations

• Strength of the fish bioelectric signals varies
  with animal size, species, age, movements, and
  health, so module gain for each channel needs to
  set adaptively, depending on the signal output of
  the animal
• Rainbow trout breathing rates were monitored by
  stream-side CDCPs and transmitted via satellite,
  experiencing no power problems at Laurel Branch
• The CDPD modem used at the Little Miami River
  site for data communication consumed a more than
  50 of platform power (7.2W, estimate) compared
  with other modules of the system (12.552W)
• Continuous river stage, temperature, and pH were
  simultaneously monitored and transmitted via
  satellite at Laurel Branch

69
3rd Generation Biosensor Conclusions

• Automatic gain control features improve signal
  resolution and avoids clipping and enhances
  system performance
• 3rd generation biosensing system is compact
  consisting of
• potted signal conditioning modules that are
  individually distributed and attached to the fish
  chambers in-situ
• AGC Data acquisition system (4.5 x 3.5 x 0.5)
• Single board computer (6.3 x 6.4 x 0.5)
• CDPD modem (6.3 x 3.4 x 1.0)
• 3rd generation system is readily compatible with
  remote CDCP

70
3rd Generation - Conclusions (Contd)

• 3rd generation biosensing systems with embedded
  PC based controllers are less expensive than
  earlier biosensing designs
• In-situ, Torpedo devises protect the animals and
  at the same time facilitates installation
  simplicity and provide easy maintenance
• Use of instrumentation wire (instead of bulky
  co-axle cables) reduces system cost and provides
  additional installation simplicity

71
Future Recommendations

• Fixed-gain signal conditioning modules can be
  modified to include a programmable control (PGA)
  in every module
• Batteries with better stored charge to weight
  ratio should be selected for floating platform
  installations
• Wireless radio can be looked at as an alternate
  data communication option, in addition to
  satellite and cell phone applications
• Implementing remote biosensing platforms in
  coordinated, watershed monitoring networks

72
Acknowledgements

• The Water Resources Center, Tennessee
  Technological University, Cookeville, Tennessee
• Institute of Applied Sciences Group, University
  of North Texas, Texas, Denton, Texas
• US-Environmental Protection Agency, Cincinnati,
  Ohio
• US-Tennessee Valley Authority, Knoxville,
  Tennessee
• The University of Tennessee Water Resources
  Center, Knoxville, Tennessee

73
Literature Cited

• See Similar Publications

74
THE END