An Analysis of a Large An Analysis of a Large John Anderson, - - PDF document

an analysis of a large an analysis of a large
SMART_READER_LITE
LIVE PREVIEW

An Analysis of a Large An Analysis of a Large John Anderson, - - PDF document

The Problem The Problem An Analysis of a Large An Analysis of a Large John Anderson, College of the John Anderson, College of the Scale Habitat Monitoring Scale Habitat Monitoring Atlantic Atlantic Application Application


slide-1
SLIDE 1

1

An Analysis of a Large An Analysis of a Large Scale Habitat Monitoring Scale Habitat Monitoring Application Application

Presented by: Mike Presented by: Mike Leib Leib

The Problem The Problem

  • John Anderson, College of the

John Anderson, College of the Atlantic Atlantic

  • Studying distribution and abundance of

Studying distribution and abundance of sea birds on Great Duck Island, Maine sea birds on Great Duck Island, Maine

  • Wanted to:

Wanted to:

  • Measure occupancy of burrows

Measure occupancy of burrows

  • Understand role of micro

Understand role of micro-

  • climactic

climactic factors in habitat selection factors in habitat selection

The Problem The Problem Goals Goals

1.

  • 1. System capable of producing animal

System capable of producing animal density GIS plots density GIS plots 2.

  • 2. Continuous real

Continuous real-

  • time data collection

time data collection 3.

  • 3. Minimal impact on birds and surrounding

Minimal impact on birds and surrounding environment environment

The Solution The Solution – – A Wireless Sensor A Wireless Sensor Network Network

1.

  • 1. System capable of producing animal

System capable of producing animal density GIS plots density GIS plots

  • Record temperature of burrows

Record temperature of burrows

  • Hot

Hot-

  • spots are likely caused by a resident bird

spots are likely caused by a resident bird

The Solution The Solution -

  • Continued

Continued

2. 2. Continuous real Continuous real-

  • time data collection

time data collection

  • Real

Real-

  • time sensor readings

time sensor readings

  • Internet Connectivity

Internet Connectivity

  • Continuously updated logs available via web

Continuously updated logs available via web

3. 3. Minimal impact on birds and surrounding Minimal impact on birds and surrounding environment environment

  • Remote m onitoring

Remote m onitoring

  • Both habitat and instrumentation can be monitored

Both habitat and instrumentation can be monitored

  • Minimizes observer effects and alteration of

Minimizes observer effects and alteration of environment environment

  • Long life of sensor network means fewer

Long life of sensor network means fewer visits by humans visits by humans

slide-2
SLIDE 2

2

System Architecture System Architecture

Base-Remote Link Data Service Internet Client Data Browsing and Processing Transit Network Basestation Gateway Sensor Patch Patch Network Sensor Node

Mote Networks Mote Networks

  • Mica2Dot motes

Mica2Dot motes

  • Repackaged Mica2 w/ 1

Repackaged Mica2 w/ 1” ” diam eter diam eter

  • 2 kinds of sensors

2 kinds of sensors

  • Burrow motes

Burrow motes

  • Detect occupancy via infrared temperature sensors

Detect occupancy via infrared temperature sensors

  • Ambient temperature/ humidity sensors

Ambient temperature/ humidity sensors

  • Weather motes

Weather motes

  • Monitor surface microclimates

Monitor surface microclimates

  • Temperature, humidity and barometric pressure

Temperature, humidity and barometric pressure sensors sensors

Motes Motes Mote Networks Mote Networks

  • 2 Networks

2 Networks

  • Single hop (433 MHz)

Single hop (433 MHz)

  • Multi

Multi-

  • hop (435 MHz)

hop (435 MHz)

  • In both networks, data was sent in

In both networks, data was sent in streaming fashion over an unreliable link streaming fashion over an unreliable link

  • Required maintaining very little information

Required maintaining very little information about network state about network state

  • Oversampled

Oversampled environment to m itigate effect of environment to m itigate effect of dropped packets dropped packets

Mote Networks Mote Networks

  • Single hop

Single hop

  • Sent packets directly to gateway (no

Sent packets directly to gateway (no routing) routing)

  • Sampled sensors every 5 minutes

Sampled sensors every 5 minutes

  • Multi

Multi-

  • hop

hop

  • Routed packets toward gateway

Routed packets toward gateway

  • Sampled sensors every 20 minutes

Sampled sensors every 20 minutes

  • Gateway sent out routing beacons to

Gateway sent out routing beacons to seed network discovery seed network discovery

Media Access and Routing Media Access and Routing

  • Low Power Listening (LPL)

Low Power Listening (LPL)

  • Discussed last week in B

Discussed last week in B-

  • MAC presentation

MAC presentation

  • Node periodically wakes up to sample radio for

Node periodically wakes up to sample radio for activity activity

  • If idle, returns to sleep

If idle, returns to sleep

  • Else wakes up and receives packet

Else wakes up and receives packet

  • Preamble must be longer than sleep interval

Preamble must be longer than sleep interval

  • Gateway always awake and listening

Gateway always awake and listening

  • Single hop network can send normally sized

Single hop network can send normally sized packets packets

slide-3
SLIDE 3

3

Media Access and Routing Media Access and Routing

  • Multi

Multi-

  • hop

hop

  • Nodes selected parent based on most

Nodes selected parent based on most reliable link reliable link – – Window Mean with Window Mean with Exponentially Weighted Moving Average Exponentially Weighted Moving Average (WMEWMA) (WMEWMA)

  • Link quality broadcast to neighbors

Link quality broadcast to neighbors every 20 minutes every 20 minutes

  • Long packets required to support LPL

Long packets required to support LPL

Deployed System Deployed System

  • Stages

Stages

  • First deployment: June 8

First deployment: June 8th

th, 2003

, 2003

  • Single hop, 1 week deployment

Single hop, 1 week deployment

  • Second deployment: July 8

Second deployment: July 8th

th

  • Multi

Multi-

  • hop, 4 week deployment

hop, 4 week deployment

  • Peak deploym ent

Peak deploym ent

  • Single hop: 49 motes

Single hop: 49 motes

  • Multi

Multi-

  • hop: 98 motes

hop: 98 motes

  • 115 days of operation yielded 650,000

115 days of operation yielded 650,000

  • bservations
  • bservations
  • Hurricane Isabel forced base station shutdown

Hurricane Isabel forced base station shutdown

  • Sept 15

Sept 15-

  • Oct 9

Oct 9

  • Operation continued through Oct 20

Operation continued through Oct 20 th

th

Power Consumption Power Consumption Results vs. Estimates Results vs. Estimates

Median Lifetime of Motes (in days) Median Lifetime of Motes (in days)

Multi Multi-

  • hop

hop 63 63 90 90 Weather Weather 34 34 80 80 Burrow Burrow 52 52 127 127 Burrow Burrow 120+ 120+ 140 140 Weather Weather Actual Actual Estimated Estimated Single Hop Single Hop

Analysis Analysis

  • Why were estimates so far off???

Why were estimates so far off???

  • SH burrow

SH burrow

  • Power drawn exceeded battery rating!

Power drawn exceeded battery rating!

  • Multi

Multi-

  • hop motes

hop motes

  • Ignored forwarding & overhearing cost

Ignored forwarding & overhearing cost

  • Power draw from overhearing 8x more than cost

Power draw from overhearing 8x more than cost

  • f single transmission!
  • f single transmission!
  • Bad seals

Bad seals

  • Significant base station outages

Significant base station outages

Multi Multi-

  • hop Analysis

hop Analysis

  • LPL lowers cost of listening

LPL lowers cost of listening

  • But at higher cost of sending/ receiving

But at higher cost of sending/ receiving

  • Long preamble means longer transmission times

Long preamble means longer transmission times

  • Overhearing is very costly

Overhearing is very costly

  • No way to reject packets early

No way to reject packets early

  • All neighboring nodes receive the packet

All neighboring nodes receive the packet

  • Possible improvements:

Possible improvements:

  • Leaf nodes don

Leaf nodes don’ ’t need to generate routing beacons t need to generate routing beacons

  • Could explicitly reject packet forwarding

Could explicitly reject packet forwarding

  • 32% of motes were leaf nodes

32% of motes were leaf nodes

  • Burrow motes heavily taxed by routing & overhearing

Burrow motes heavily taxed by routing & overhearing

  • Prevent them from routing packets

Prevent them from routing packets

  • Reduce overhearing

Reduce overhearing

slide-4
SLIDE 4

4

Routing Stability Routing Stability

  • 80% of packets

80% of packets routed over just routed over just 15% of all links 15% of all links

  • Most links short

Most links short -

  • lived

lived

Packet Delivery Effectiveness Packet Delivery Effectiveness

  • Results:

Results:

  • Single hop > 70%

Single hop > 70%

  • Multi

Multi-

  • hop = 52%

hop = 52% and 28% and 28%

Node Reclamation Node Reclamation

  • Why node reclamation?

Why node reclamation?

  • Potential for pollution

Potential for pollution

  • Post mortem

Post mortem

  • 150 motes deployed

150 motes deployed

  • 78 recovered

78 recovered

  • Observations

Observations

  • Only 13 retained fully intact antennas

Only 13 retained fully intact antennas

  • 6 had bite marks

6 had bite marks

  • Remaining shortened or removed by animals

Remaining shortened or removed by animals

  • Condensation problems

Condensation problems

Improvements Improvements

  • Physical

Physical

  • Dimensions lim it battery choice

Dimensions lim it battery choice

  • Lack of external

Lack of external LEDs LEDs

  • Deploying large number of sensors takes a lot

Deploying large number of sensors takes a lot

  • f resources
  • f resources
  • Need procedures that work on groups of motes in

Need procedures that work on groups of motes in aggregate aggregate

  • Software

Software

  • Power monitoring

Power monitoring

  • Data Logging

Data Logging

  • Buffer readings locally and send later when network

Buffer readings locally and send later when network unavailable (could store up to 113 days of readings) unavailable (could store up to 113 days of readings)

  • Useful for post mortem

Useful for post mortem

More Improvements More Improvements

  • Identified several tools useful for real

Identified several tools useful for real-

  • world deployment

world deployment

  • Field tools

Field tools

  • Self check

Self check

  • Show neighbors & network stats

Show neighbors & network stats

  • Client tools

Client tools

  • Ability to re

Ability to re-

  • task nodes

task nodes

  • Network monitoring capability

Network monitoring capability

  • Backend analysis tools

Backend analysis tools

  • GUI network monitoring tools available at time of

GUI network monitoring tools available at time of deployment deployment

Critique Critique

  • Estimates of mote lifetime are very

Estimates of mote lifetime are very

  • ptimistic
  • ptimistic
  • Knowingly ignore overhearing and routing

Knowingly ignore overhearing and routing costs costs

  • Assume best case scenarios

Assume best case scenarios

  • LPL alone not enough

LPL alone not enough

  • Scheduled Channel Polling (SCP)

Scheduled Channel Polling (SCP)

  • Reduce overhearing = save energy

Reduce overhearing = save energy

  • Low packet delivery effectiveness

Low packet delivery effectiveness

  • Link Symmetry Detection

Link Symmetry Detection

  • Support retransmission?

Support retransmission?

slide-5
SLIDE 5

5

Comparison Comparison

  • VigilNet

VigilNet surveillance application requires surveillance application requires higher bandwidth higher bandwidth

  • Tracking vehicles in real

Tracking vehicles in real-

  • time

time

  • GDI is a real

GDI is a real-

  • world deployment vs.

world deployment vs. VigilNet VigilNet prototype system prototype system

  • Packaging issues not addressed

Packaging issues not addressed

  • GDI uncovered many practical issues with

GDI uncovered many practical issues with deploying a mote network (tools) deploying a mote network (tools)

  • Similar requirements for mote longevity

Similar requirements for mote longevity

Comparison Comparison

  • Sentries not practical for GDI

Sentries not practical for GDI

  • VigilNet

VigilNet incorporates some improvements incorporates some improvements suggested in GDI paper suggested in GDI paper

  • Mote reprogramming

Mote reprogramming – – BIG time saver BIG time saver

  • Visualization tools for deployment

Visualization tools for deployment

  • VigilNet

VigilNet has more sophisticated network has more sophisticated network initialization & routing initialization & routing

  • Beaconing is not stealthy or energy

Beaconing is not stealthy or energy-

  • efficient

efficient

  • Synchronization allows for more efficient use of radio =

Synchronization allows for more efficient use of radio = energy efficiency energy efficiency

  • Link Symmetry Detection = higher reliability links

Link Symmetry Detection = higher reliability links

  • Supports retransmission

Supports retransmission