1

CS 268: Lecture 24 Sensor Network

Architecture (SNA)

Ion StoicaComputer Science Division

Department of Electrical Engineering and Computer SciencesUniversity of California, Berkeley

Berkeley, CA 94720-1776

2

Sensor Network Protocols Today

Phy

Link

Topology

Routing

Transport

Appln

RadioMetrixRFM

CC1000Bluetooth 802.15.4

eyesnordic

WooMacSMAC

TMACWiseMAC

FPS

MintRoute

ReORg

PAMAS

CGSR

DBF

MMRP

TBRPF

BMAC

DSDV

ARADSR

TORA

GSR GPSR GRAD

Ascent

SPIN

SPAN

Arrive

AODV

GAF

SchedulingResynch

Yao

Diffusion

Deluge Trickle Drip

RegionsHood

EnviroTrackTinyDB

PC

TTDD

Pico

FTSP

Obligatory David Culler Slide…

What if I want to use any two protocols together??

3

Network Model

Transit Network

Basestation

Sensor Patch

Patch Network

Base-Remote Link

Data Service

Internet

Client Data Browsingand Processing

Sensor Node

Gateway

Dense patches of sensing nodes- Many resource constrained- Non-homogeneous- Modalities, roles, HW, SW- Power, BW

Transit tier- Often specialized wireless- Provides gateways

Internet Tier- Multiple connections to infra- Deep storage, proc. Viz

SNA should not require unconstrained nodes

Should utilize unconstrained nodes to reduce burden on constrained ones

Mobility within physically embedded context

4

What is an Architecture?

Architecture is how to “organize” implementations- What interfaces are supported

- Where functionality is implemented

Architecture is the modular design of the network

Architecture is not the implementation itself

5

Internet vs Sensor Nets

Internet goals Interconnect separate networks Resilience to loss and failure Support many comm. services Accommodate variety Distributed management Cost effective Low effort attachment Resource accountability Network Architecture

Sensor Nets Resource efficiency Data centric design Deal with intermittent

connectivity Self-managed Observation, monitoring of

various environments Cost effective Scalability

6

Internet vs Sensor Nets

Internet goals Interconnect separate networks Resilience to loss and failure Support many comm. services Accommodate variety Distributed management Cost effective Low effort attachment Resource accountability Network Architecture

Sensor Nets Dense real world monitoring Resilience to loss, failure and

noise Support many applications Scale to large, small, long Cost effective Evolvable in resources Composable Security

7

Why not IP?

One or very few applications running on a sensornet vs huge number running in the Internet

Large variety of traffic patterns (most not point-to-point):- Any-to-any, many-to-one, many-to-few, one-to-many- Inneficient to impl. these patterns over point-to-point

IP does not address (well):- Resource and energy constraints- Unattended operation - Intermittent connectivity- Space embeded nodes- ...

8

A Sensor Network Architecture (SNA)

Narrow waist: Sensornets Protocol (SP)- Goals: generality and efficiency

- Position: between data-link and network layers

- Service: best-effort, single hop

- Common to both single- vs multiple-hop deployments

9

Properties of SP

SP provides mechanisms for network protocols to operate- Network protocols may introduce policy

Three key elements of SP:- Data Reception

- Data Transmission

- Neighbor Management

10

Collaborative Interface

Control- Reliability Best effort to transmit the msg

- Urgency Priority mechanism

Feedback- Congestion Was the channel busy?

• Should I slow down?

- Phase Was there a better time to send?

• Decouple appl sampling from communication

11

Message Reception

Message arrives from link SP dispatches Network protocols establish

- naming/addressing

- filtering

SP

Receive

12

Message Transmission

Messages placed in shared message pool- All entries are a promise to send a

packet in the future Messages include

- Pointer to first packet and # of packets

- Control information: reliability and urgency

- Feedback information: congestion and phase

SP

Msg Pool

Send Receive

13

Neighbor Management

SP provides a shared neighbor table- Cooperatively managed

- SP mediates interaction using table

• No policy on admission/eviction by SP

• Scheduling information

SP

Neighbor Table Msg Pool

Neighbors Send Receive

14

SP

Data Link A Data Link B

SP Adaptor A SP Adaptor B

NetworkProtocol 1

PHY A PHY B

NetworkProtocol 2

NetworkProtocol 3

NetworkService

Manager

LinkE

stimator

LinkE

stimator

Neighbor Table Msg Pool

Neighbors Send Receive

SP Architecture

15

SP

Message Pool

sp_message_t

destinationmessagequantityurgentreliability

phasecongestion

address_t1st TOSMsg to send# of pkts to sendon or offon or off

adjustmenttrue or false

con

tro

lfe

edb

ack

Neighbor Table Msg Pool

Neighbor Table

2

1

NetworkLinkRequiredNeighbor

addresstime ontime offlistenquality

address_tlocal time node wakeslocal time node sleepstrue or falseestimated link quality

16

SP

Msg Pool

Network Protocol

msg*

Next PacketHandler

packetsSP Message

MessageDispatch

transmit

completed

(1)

(2)

(3) (4)

(5)

(6)

1st packet

Send

insp

ect

Link Protocol

SP Message Futures

1) Submit an SP Message for Transmission

2) Message added to message pool3) SP requests the link transmit the

1st packet4) Link tells SP the transmission

completed5) SP asks protocol for next packet6) Protocol updates packet entry in

message pool

17

What SP Isn’t

SP does not dictate any header fields- Messages are opaque to SP

Instead, rely on abstract data types- Can query for address, length, etc

No explicit security mechanism- Message content opaque to SP

- Link, Network, and App security can be built transparently to SP

18

Benchmarks

Minimal performance reduction in single hop- Compare to B-MAC paper

- Compare to IEEE 802.15.4

Simpler multihop/network protocol code

Power consumption

Network protocol co-existence

19

Results: mica2 Throughput

0 5 10 15 200

2000

4000

6000

8000

10000

12000

14000

16000

Th

rou

gh

pu

t (kb

ps)

Nodes (n)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Pe

rce

nta

ge

of C

ha

nn

el C

ap

aci

ty

B-MACSPSP + CCSP + LPL + CCSP + LPL + CC + PhaseChannel Capacity

20

Results: mica2 Throughput

0 5 10 15 200

2000

4000

6000

8000

10000

12000

14000

16000

Th

rou

gh

pu

t (kb

ps)

Nodes (n)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Pe

rce

nta

ge

of C

ha

nn

el C

ap

aci

ty

B-MACSPSP + CCSP + LPL + CCSP + LPL + CC + PhaseChannel Capacity

21

Results: mica2 Throughput

0 5 10 15 200

2000

4000

6000

8000

10000

12000

14000

16000

Th

rou

gh

pu

t (kb

ps)

Nodes (n)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Pe

rce

nta

ge

of C

ha

nn

el C

ap

aci

ty

B-MACSPSP + CCSP + LPL + CCSP + LPL + CC + PhaseChannel Capacity

22

Results: mica2 Throughput

0 5 10 15 200

2000

4000

6000

8000

10000

12000

14000

16000

Th

rou

gh

pu

t (kb

ps)

Nodes (n)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Pe

rce

nta

ge

of C

ha

nn

el C

ap

aci

ty

B-MACSPSP + CCSP + LPL + CCSP + LPL + CC + PhaseChannel Capacity

23

Results: mica2 Throughput

0 5 10 15 200

2000

4000

6000

8000

10000

12000

14000

16000

Th

rou

gh

pu

t (kb

ps)

Nodes (n)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Pe

rce

nta

ge

of C

ha

nn

el C

ap

aci

ty

B-MACSPSP + CCSP + LPL + CCSP + LPL + CC + PhaseChannel Capacity

24

Results:Single Hop Benchmarks (802.15.4)

25

Conclusion

SNA: provide context for sharing our community work and accelerate the development and deployment of sensornet applications

Effective link abstraction, SP, allows network protocols to run efficiently on varying power management schemes