Resource Management ofHeterogeneous Wireless Systems

Supported by: NSF, CNS, Sun Micro., Qualcomm, HP

HPWREN connected topology agenda

http://hpwren.ucsd.edu

UCSD

toSCI

SDSU

PL

MLO

MPO

SMER

CE

KSW

BVDALVA2

FRD

BZNSND

CRY

RDM WMCPFO

BDC

SantaRosa

DHL

to CI andPEMEX

CWC

PSAP

WIDCKYVW

COTDKNW

GVDA

45Mbps FDX 11GHz FCC licensed45Mbps FDX 6GHz FCC licensed45Mbps FDX 5.8GHz license-exempt45Mbps-class, HDX, license-exempt~8Mbps HDX 2.4GHz license-exempt~3Mbps HDX 2.4GHz license-exempt115kbps HDX 900MHz license-exempt

WLA

Backbone/relay nodeAstronomy science siteBiology science siteEarth science siteUniversity siteResearcher locationNative American siteIncident management site

GLRS

SLMS

USGC

CRRS

p480

AZRY

SO

HPWREN

HPWREN - three tier networkWireless MESH

QoS scheduling and routingFast wireless connectivity

Sensor Cluster HeadsKey issue:

Delivering good QoS With long battery lifetime

Use faster radio to support QoS requirements

Sensor NetworkQoS

not considered in traditional sensor net research

Battery lifetime

Earthquake sensors in the desert – 10kbps

Palomar Observatory – 150 Mbps

U.S. Navy Deep Submergence Unit

http://www.csp.navy.mil/csda5/dsu/dsu.htm

Volcan Fire HPWREN connection, September 2005Incident Command Postsite

Volcan relay site

Trigger email/pager/….if:

condition A + condition B +condition C

occurs

several San Diego fire officers are currently being paged during alarm conditions, based on HPWREN data parameterization by a CDF Division Chief

Real-time data based alerts

Mountain fire video camera

Motion detect camera

Acoustic sensors:Wolf howls at the California Wolf Center

Sponsored by LANLIn collaboration with Uof Bologna

Damage

No Damage

A pair of PZTs: actuator & sensor

PZTs

PZTs

• Sensing – 10ms• Sampling– 10MSPS• 10K samples• Data occupies 20KB

Time to analyze of a pair of PZTs = 3.5 seconds

To charge a 100F super capacitor using 100 cm2 solar panel takes less than 90 min

DSP can process up to 260 2xPZT at 150MIPS with one charge

ReconfigurableLow PowerEasily ProgrammableSimple Hardware Integration

Joint with EPFL and IMEC Supported by Sun Microsystems

EKG Monitoring with SunSPOT

Reconfigurable Programming Paradigm:•Library of functions•General code template•Real-time compilation•Reconfigurable intelligence•Transmission Cleanup

Reconfigurability inEKG Monitoring with SunSPOT

PDA

3d ultrasonic anemometer

Temperature, humidity

HPWREN

Animal Monitoring

Notebook Cellular Phone PC

Ship Monitoring

Wireless Sensor Network

Data Distribution

Network

Precipitation

Solar radiation

In-flight camera

Weather station

Mobile and Stationary Operations

Stationary camera

Seismic

Storage

Data Acquisition

Network

• Objective:• Design an adaptive, distributed and low power QoS

scheduling & routing methodology

• Main Challenges:• Devise a good scheduler:

• Understand and characterize the incoming traffic• Improve delay and throughput • Reduce the power consumption

• Devise a good routing algorithm:• Characterize and devise simple & accurate metrics• Low power -> route changes occur frequently -> fast adaptation

Research Topics: Energy-efficient & QoS-aware Scheduling & Routing

Initial Project Testbed - SMER: Santa Margarita Ecological Reserve

80 Cluster heads connected with WLAN

Sensor Node and Cluster Head Power Consumption

TransmitReceive

Encode Decode Transmit

Receive

EncodeDecode

Energy breakdown for voice Energy breakdown for MPEG video

Lucent WLAN & SA-1100 CPU at 150 MPISSource : Mobicom’01 SensorsTutorial

Power consumption of sensor node subsystems

0

5

10

15

20Po

wer

(mW

)

SENSORS CPU TX RX IDLE SLEEP

RADIO

QoS issues: 802.11 contention

t

busy

station1

station2

station3

station4

station5

packet arrival at MAC

DIFS

busy

elapsed backoff time

residual backoff timebusy medium busy

DIFSDIFS

busy

busy

DIFSbusy

collision

shortest backoff time

exponentialbackoff

WLAN Bandwidth vs. Contention

3

3 .5

4

4 .5

5

5 .5

0 5 1 0 1 5 2 0 25 30

Number o f nodes

Thro

ughp

ut (M

bps)

k=3

This suggests that in finite traffic:Throughput improvements are possible with bursts of packetsScheduling k clients at a time can be beneficial

TDMA with CBR on WLAN

Proposed TDM fixes contenders at 2-4 Lower contention means higher throughputAnother benefit is more sleep

3.0

3.5

4.0

4.5

5.0

5.5

0 5 10 15 20 25 30Number of nodes

Thro

ughp

ut (M

bps)

No schedulingScheduling

0%

2%

4%

6%

8%

10%

12%

14%

0 5 10 15 20 25 30Number of nodes

Col

lisio

n ra

te

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

0 5 10 15 20 25 30Number of nodes

Pow

er c

onsu

mpt

ion

(W)

...

)(1 nTa

)(2 nTa

)(nTaN

Transportation/Network layer

scheduler

scheduler

scheduler

......

Application Layer Proxy Layer

IEE

E 802.11b M

AC

Station 1

Station 2

Station 2

)(1 nTλ

)(2 nTλ

)(nTNλ

WAKE-UP DELAY

PREMPTEDLOW-POWER MODE

WAKE-UP

BROADCAST

BROADCAST

WAKE-UP

CH 1 CH 2

DELAY

BURST BURSTBURST

LOW-POWER MODE

Hybrid distributed scheduling

Combines cell and node level scheduling

Multi-cell wireless network

Cell-level scheduling

Node scheduling in a cell

Distributed node scheduling

Distributed scheduling with minimal overheadLess vulnerable to a node failureFlexible to the change of network topologyRequires two-hop connectivity information

107 13

19 22

5 16 8

122426

17

11

3 2

23

15

42021

18

1

14

625

10

7 13

19 22

5 16 8

122426

17

11

3

Distributed cell schedulingActivate cells that will not interfere with each other → Improve the overall throughput

Unscheduled cells

Scheduled cells

Recent results

XScale PXA27x DVK representing sensor node cluster heads (CH)NS2 simulator for multiple nodesThe applications used are

Various sensor traffic from SMER/HPWRENMPEG4 videoMP3 audioEmail, Telnet, WWW Data consumption rate of applications kbps

WLAN

BT

Traffic characterization

WWW Trace

0.0001

0.001

0.01

0.1

1

0.01 0.1 1 10

Interarrival Time (s)

Experimental

Exponential

Pareto

Distributions

atbPareto −⋅−= 1

teeExp λ−−= 1

105 106 107 10810-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1

100

101

Interarrival time(usec)

Tail distr. of burst inter-arrival time(Tth:250ms)

experimental2p-exponentialbounded paretopareto

106 107 10810-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1

100

101

Interarrival time(usec)

Tail distr. of burst inter-arrival time(Tth:500ms)

experimental2p-exponentialbounded paretopareto

Video cluster heads Sensor cluster heads

Node scheduling

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

0.0 1.0 2.0 3.0 4.0 5.0 6.0Traffic from nodes (Mbps)

Thro

ughp

ut (M

bps)

No scheduling (real)Scheduling (real)No scheduling (2-param exp)Scheduling (2-param exp)No scheduling (Pareto)Scheduling (Pareto)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0 2.0 4.0 6.0 8.0Traffic from nodes (Mbps)

Pow

er c

onsu

mpt

ion

(W)

No scheduling (real)Scheduling (real)No scheduling (2-param exp)Scheduling (2-param exp)No scheduling (Pareto)Scheduling (Pareto)

0.00

0.01

0.02

0.03

0.0 2.0 4.0 6.0 8.0Traffic from nodes (Mbps)

Del

ay (s

ec)

0.00

2.00

4.00

6.00

8.00

10.00

0.0 2.0 4.0 6.0 8.0Traffic from nodes (Mbps)

Del

ay (s

ec)

Significant improvements in throughput, MAC delay and power consumption regardless of the traffic model used

Distributed cell and node scheduling

Nodes scheduled with a distributed algorithmResults show large power savings with throughput improvement

Average throughput

1.0

1.2

1.4

1.6

1.8

2.0

5 10 15 20 25 30Node density

Thro

ughp

ut p

er c

ell

No schedulingScheduling: 0.3 sec

Power per node

0.0

0.2

0.4

0.6

0.8

1.0

1.2

5 10 15 20 25 30Node density

Pow

er (W

)No schedulingScheduling: 0.3 sec

Conclusion

Scheduling communication at sensor node cluster heads has significant benefits

Lower energy consumptionBetter bandwidth utilization

Benefits of scheduling measured forSensor node traffic Multimedia trafficStandard web traffic