Data Collection Storage and Retrieval with an Underwater

  • Slides: 17
Download presentation
Data Collection, Storage, and Retrieval with an Underwater Sensor Network I. Vasilescu, K. Kotay,

Data Collection, Storage, and Retrieval with an Underwater Sensor Network I. Vasilescu, K. Kotay, and D. Rus MIT CSAIL 2007. 09. 27 1: 00 PM bwkim@dependable. kaist. ac. kr

Contents Introduction The Hardware Infrastructure Networking Data Collection Data Retrieval using Mobility Conclusion &

Contents Introduction The Hardware Infrastructure Networking Data Collection Data Retrieval using Mobility Conclusion & Discussion 2020 -09 -10 2/16

Introduction Why underwater sensor network(UWSN)? § The oceans cover 70% of our planet and

Introduction Why underwater sensor network(UWSN)? § The oceans cover 70% of our planet and are critical to our well-being § Monitoring these environments is difficult and costly for humans hours, depth, weather condition, boat § UWSN could monitor physical variables water temperature, turbidity , pollutant Goal § The novel platform for UWSN hardware and software architecture Optical and acoustic networking protocol 2020 -09 -10 3/16

Key Issues of UWSN Power efficiency, deployment and repair § Common to wireless sensor

Key Issues of UWSN Power efficiency, deployment and repair § Common to wireless sensor network deployments on land but more difficult in the underwater environments Communications § Ultrasonic § Optical Mobility § A mean for deploying, reconfiguring, and retrieving the nodes § Large area coverage with sparse network § Mobile nodes can act as data mules No hot spot problem More time-efficient and energy-efficient 2020 -09 -10 4/16

Motivation Scenario Data Collection (6. 86 Mbytes) § AUV as a data mule using

Motivation Scenario Data Collection (6. 86 Mbytes) § AUV as a data mule using short optical communication Transmission speed : 320 kbytes/sec Time consumption: 21 sec. Total energy : 120 J § An acoustic communication network with node-to-node routing Transmission speed : 480 bits/sec Time consumption: 1. 3 days Total energy : 247 k. J if we consider multi-hop, more energy is needed. Static sensor node(Aquafleck) § Recording of data Mobile sensor node(Amour AUV) § Can locate the static nodes § Visits static nodes periodically to upload the data § Have Communications links for data transmit § Can be rechargeable 2020 -09 -10 5/16

The Hardware Infrastructure Static sensor node Aquafleck § § § § Fleck CPU Unit

The Hardware Infrastructure Static sensor node Aquafleck § § § § Fleck CPU Unit 4 kbytes RAM 512 kbytes flash memory for storage Optical communications boards Interface board for analog electronics Sensor board 170 x 100 x 90 mm yellow watertight box High speed optical communication module 532 nm light, a cone of 30 degrees, 320 kbits/s § Acoustic communication module A range of 20 m omnidirectional, 50 bits/s For ranging(the distance between them) § CMUCam Camera § Pressure/temperature sensor 2020 -09 -10 6/16

The Hardware Infrastructure Mobile sensor nodes Amour AUV § § § dock and transport

The Hardware Infrastructure Mobile sensor nodes Amour AUV § § § dock and transport the Aquafleck Hover above Aquafleck for data muling 4 external thrusters for moving Magnetic compass for grid and spiral search 140 Wh lithium polymer battery Optical communication link Starbug AUV § § dock , transport, hover Aquafleck Docking with Amour Designed to optimize endurance 6. 4 Ah lithium polymer battery 2020 -09 -10 7/16

The Hardware Infrastructure comparison of unit size 2020 -09 -10 8/16

The Hardware Infrastructure comparison of unit size 2020 -09 -10 8/16

Networking Acoustic communication § Can be used for much longer range communication than optical

Networking Acoustic communication § Can be used for much longer range communication than optical § Omnidirection § Low propagation velocity Permit accurate timing of signals to determine the distance between nodes § Suffers from attenuation and reflection Optical communication § Higher data rate § Short communication range § Line-of-sight operation Tradeoff between communication range and data rate § Acoustic system signal events, small amounts of data Localization § Optical System Large amounts of data close communication(AUV upload command to static nodes) 2020 -09 -10 9/16

Optical communication Range § Factors light absorption of water, scattering, beam divergence, ambient light

Optical communication Range § Factors light absorption of water, scattering, beam divergence, ambient light § absorption Output current § S : photodiode sensitivity, P : power Choice of green light green § The severe attenuation of infrared § Attenuation of Red light beyond 1 m infrared 2020 -09 -10 10/16

Optical communication Protocol § Modified VFIR(Very Fast Infra. Red) Pulse position modulation § four

Optical communication Protocol § Modified VFIR(Very Fast Infra. Red) Pulse position modulation § four pulse position determined by the value of two bits of data § communication Start pulse Interval by the value of the first two data bits The next two data bits determine the offset from the previous pulse Experimental results 2020 -09 -10 11/16

Acoustic Communication the message length as short as possible § Reflection Hardware § Under

Acoustic Communication the message length as short as possible § Reflection Hardware § Under consideration Small for Aquafleck Inexpensiveness Distance (25 m) § Receiver : Panasonic EFR-RQB 40 K 5 § Transmitter : EFR-TQB 40 K 5 Experimental results § in tow tank 2020 -09 -10 12/16

Data Collection Experiments § (a) : Pressure data collected by three different sensor nodes

Data Collection Experiments § (a) : Pressure data collected by three different sensor nodes for three days § (b) : Image of the Starbug AUV taken from a sensor node (a) (b) 2020 -09 -10 13/16

Data Retrieval using Mobility The key challenges for data muling § § § Locating

Data Retrieval using Mobility The key challenges for data muling § § § Locating the first node Locating the next node in the sequence Controlling the hover mode for the mobile node Data transfer Synchronizing clocks Time stamp of collected data 2020 -09 -10 14/16

Conclusion A first prototype for an underwater sensor network § developed § built §

Conclusion A first prototype for an underwater sensor network § developed § built § Used An effective way to collect, store, retrieve large data over long time § Data muling, that is, mobility Software(and hardware) reliability are extremely import § Controlling the mobile nodes in the presence of currents 2020 -09 -10 15/16

Any Question? Thank you for your listening 2020 -09 -10 16/16

Any Question? Thank you for your listening 2020 -09 -10 16/16

References http: //groups. csail. mit. edu/drl/publications/paper-pub-0826. pdf http: //www. csiro. au/files/pf 8 z. pdf

References http: //groups. csail. mit. edu/drl/publications/paper-pub-0826. pdf http: //www. csiro. au/files/pf 8 z. pdf http: //www. ict. csiro. au/page. php? cid=87 http: //www. csail. mit. edu/index. php 2020 -09 -10 17/16