Centre for Wireless Communications

Media Access and Routing Protocols for Power Constrained Ad Hoc Networks

Carlos Pomalaza-Ráez

Centre for Wireless Communications – University of Oulu and

Indiana University - Purdue University, USA

carlos@ee.oulu.fihttp://www.cwc.oulu.fi

Outline

• Introduction• Main features of power constrained

networks• Design considerations• MAC layer• Routing algorithms• Physical layer issues• Cross-Channel design• Final observations

Main Features of Ad Hoc Networks

• Dynamic topology• Bandwidth-constrained and

variable capacity links• Energy-constrained operations• Limited physical security

Mobility in Ad Hoc Networks

Dynamic Routing

Route Maintenance

Wireless Sensor Networks (WSN)

Sensing

Computation

Networking

Circulatory Net

New technologies have reduced the cost, size and power of micro-sensors and wireless interfaces

Environmental Monitoring

Benefits from 3 technologies• digital circuitry• wireless communication• silicon micro-machining

Applications

•BattlefieldDetection, classification

and tracking

• Habitat MonitoringMicro-climate and

wildlife monitoring Examples:

– ZebraNet (Princeton)– Seabird monitoring in Maine’s

Great Duck Island(Berkeley & Intel)

Applications

• Structural, seismic Bridges, highways, buildings

• Examples: Coronado Bridge San Diego (UCSD), Factor Building (UCLA)

• Smart roadsTraffic monitoring, accident

detection, recovery assistance• Examples: ATON project (UCSD)

highway

camera microphone

• Contaminants detection

Sensor Nodes

Sensor Node Evolution

Mote Type WeC Rene Rene2 Dot Mica

Date Sep-99 Oct-00 Jun-01 Aug-01 Feb-02

Microcontroller (4MHz)

Type AT90LS8535 ATMega163 ATMega103/128

Prog. mem. (KB) 8 16 128

RAM (KB) 0.5 1 4

Communication

Radio RFM TR1000

Rate (Kbps) 10 10/40

Modulation Type OOK OOK/ASK

Typical Features of WSN• Relatively large number of nodes• Low cost, size, and weight per node• Energy constrained• Prone to failures• Almost static topology• More use of broadcast

communications instead of point-to-point

• Nodes do not have a global ID• Limited security

• Fault tolerance• Scalability• Cost• Power consumption• Hardware and software constraints• Topology maintenance• Deployment• Environment

Design Considerations

Node Energy Consumption Projections

20002000 20022002 20042004

10,0010,0000

1,0001,000

100100

1010

11

.1.1

Ave

rag

e P

ow

er

(mW

)

• Deployed (5W)

• (50 mW)

(1mW)

Node Hardware

sensors CPU radio

battery

Acoustic, seismic, magnetic, etc.

interface

Electro-magnetic interface

Limited battery supply

Eventdetection

Wireless communication with neighboring nodes

In-node processing

Energy Limitations

Power consumption of node subsystems

0

5

10

15

20

Pow

er

(mW

)

Sensing

CPU TX RX

IDLE SLEEP

• Each sensor node has limited energy supply

• Nodes may not be rechargeable• Energy consumption in

Sensing Data processing Communication (most energy intensive)

A Layered Approach

PHY

MAC

Network

Media Access Control

Let multiple radios share the same communication media

MAC:• Local Topology Discovery and

Management

• Media Partition By Allocation or Contention

• Provide Logical Channels to Upper Layers

Physical

MAC

Network

ApplicationTime

Code

Frequen

cy

Channel Access in Multi-hop Networks

A

BC

E D

Large number of short range radios in a wide area

Pros: Channel Reuse

Hidden Terminal - CSMA is not appropriate -No Global Synch

Cons:

Which MAC is Good for WSNs?• Most existing MACs are targeted for

One-hop, centralized control network: cellular network, 802.11, Bluetooth…

Bandwidth hungry application, strict QoS requirement

• Existing MACs are based on existing radiosMore than 90% of power is burned when radio

is idle

“Energy efficient” WSNs built using existing MACs might not be that

efficient

Two Possible Approaches

• Modify or enhance current protocols to make them more energy awareFor example use the power control

mechanism present in the IEEE 802.11 standard

• Develop and implement new protocols that take into account full consideration of the network constraints

IEEE 802.11 Standard

• PCF (Point Coordination Function)Centralized medium access control

• DCF (Distributed Coordination Function)Distributed medium access control

802.11 DCF: RTS-CTS-DATA-ACK

A (Sender)

B (Receiver)

Four Way Handshake RTS-CTS-DATA-ACK

• A sender node waits for DIFS( Distributed Inter-Frame Space) before making an RTS attempt

• A node enters a SIFS ( Short Inter Frame Space) before sending an ACK, DATA or CTS frame

• NAV (Network Allocation Vector) indicates the duration of the current transmission

Four Way Handshake RTS-CTS-DATA-ACK

Sender node

Receiver node

Others

RTS

CTS

DIFS

DATASIFS

SIFS

ACKSIFS

NAV(RTS)NAV(CTS)

NAV- Network Allocation Vector

Simple Power Control

BA C DRTS

CTS

Simple Power Control

DATA

ACK

BA C D

Power Levels (simple power control)

RTS CTS DATA ACK

Pmax

Pi

0

Ranges• Transmission range

Receive and correctly decode packets

• Carrier sensing rangeSensing the signal

• Carrier sensing zoneSensing the signal, but cannot decode

it correctlyCan interfere with on-going

transmission

Ranges

D E

TransmissionRange

Carrier SensingRange

Carrier SensingZone

BA C

Variation of 802.11 DCFSRC RTS

CTS

NAV (RTS)

DATA

ACK

NAV (CTS)

NAV (EIFS)

NAV (EIFS)

NAV (EIFS)

DIFS

DIFSSIFS

SIFS

SIFS

Defer Channel Access

DST

TX range

CS zone

Improved Power Control

RTS CTS DATA ACK

Pmax

Pi

0

Less than EIFS

Revisiting the CDMA Multi-Channel Problem

1

3

2

4

5

6 7

8

Nodes use different channels (codes) to transmit dataThe codes are locally unique with global reuseParallel transmission without synchImplicit local address is the channel

Channel Assignment in Cellular Networks

• Same frequency can be used in all cells of the same color• Minimize number of frequencies (colors)• The topology is static

Code Assignment = Graph Coloring

For any node, all its neighbors have different colors

ORAll two-hop neighbors have different colors

Graph G = (V,E)

Δ is the maximum

degree

Number of colors needed <= min {Δ(Δ -1)+1, |V|}

Brook and Vizing theorem

Code Assignment in Ad Hoc Networks

• There is no base station• Nodes are free to connect or disconnect• Nodes move about• Increase or decrease their transmission

range

These features call fordistributed, dynamic, power

aware code assignment algorithms

Routing

Multihop Routing due to limited transmission range Routing

Issues• Low mobility• Power aware• Irregular topology• MAC aware• Limited buffer spacePhysical

MAC

Network

Application

Routing Proactive vs. Reactive

Proactive routing maintains routes to every other node in the network

Regular routing updates impose large overhead

Suitable for high traffic networks

Reactive routing maintains routes to only those nodes which are needed

Cost of finding routes is expensive since flooding is involved

Good for low/medium traffic networks

Ad-hoc On-demand Distance Vector Protocol (AODV)

nodes discard thepackets havingbeen seen

sourcedestination

Sourcebroadcastsa route packet

neighbors re-broadcastthe packet till it reachesthe destination

reply packet follows thereverse path of the routerequest packet recordedin broadcast packet

RREQ

RREP

Traditional Reactive Protocols

Source Destination

But that is NOT a good solution!Energy depletion in certain nodesCreation of hotspots in the network

Finds the best route

and then uses it as much as possible

New Approaches• Application aware communication

primitives (expressed in terms of named data not in terms of node who requests data)

• Achieve locality for decision making (and reduce the communication)

• Application centric, data-driven networks

• Achieve desired global behavior through localized interactions, without global state

Gradient represents both direction towards data matching and status of demand with desired update rate

Probability 1/energy costThe choice of path is made locally at every node for every packet

Directed Diffusion

Sink

Source

Application-aware communication primitivesexpressed in terms of named data

Consumer of data initiates interest in data with certain attributes

Nodes diffuse the interest towards producers via a sequence of local interactions

This process sets up gradients in the network to draw events matching the interest

Collect energy metrics along the wayEvery route has a probability of being chosen

Four-leggedanimal

Directed Diffusion

Sink

Reinforcement and negative reinforcement used to converge to efficient distributionHas built in tolerance to nodes moving

out of range or dying

Source

Directed Diffusion• Pros

Energy – much less traffic than flooding –

Latency – transmits data along the best path –

Scalability – local interactions only –Robust – retransmissions of interests –

• ConsThe set up phase of the gradients is

expensive

SPINSensor Protocol for Information via Negotiation

• Basic ideaExchange data when neededSave energy by being resource aware

• Data negotiation Meta-data (data naming) Application-level control

SPIN

A B

A B

A B

ADV

REQ

DATA

•SPIN messages ADV- advertise data REQ- request specific data DATA- requested data

•Resource management Nodes decide their capability of participation

in data transmissions

The process repeats itself across the network

SPIN-BC (broadcast)

DATAREQADV

A node senses something “interesting”It sends meta-data to neighborsNeighbor sends a REQ listing all of the data it would like to acquireSensor broadcasts dataNeighbors aggregate data and broadcast(advertise) meta-data

SPIN-BC (broadcast)

I am tired I need to sleep …

Advertise meta-data

Request data

Send dataAdvertise

Advertise

Nodes do need not to participate in the process

Request data

Send data

Send data

Advertise meta-data

Request data

Send data

SPIN• Pros

Energy – more efficient than flooding –

Latency – converges quickly –Scalability – local interactions only –Robust – immune to node failures –

• ConsNodes always participating

Some Physical Layer Issues

• Frequency selection• Carrier frequency generation• Signal detection• Modulation

Binary and M-ary modulation schemesBinary modulation scheme is deemed to

be more energy-efficient• Low transmission power and simple

transceiver circuitry make Ultra Wideband (UWB) an attractive candidate

• Hardware design, e.g. wake-up radio

Physical

MAC

Network

Application

Wake-up Radio

Sleeping nodes

Communicating nodes

• Sleeping mode based on Ultra Low Power wake-up radio• Sleeping nodes have to wake-up to broadcast signals, and

not to any signal from surrounding communicating nodes• Broadcast signals should not disrupt data transmission

Cross-Layer Design

• Optimizing single layer might not be enough

• Scheduling, adaptability, and diversity are most powerful in the context of a cross-layer design

• Energy consumption must be addressed across all protocol layers

Final Observations• The constraints imposed by

factors such as power consumption, costs, fault tolerance, multihop topology, etc., are more stringent in sensor type networks than in conventional ad-hoc networks

• This calls for new techniques and protocols at different layers of the protocol stack

References• I.F. Akyildiz, W. Su, Y. Sankarasubramaniam, E. Cayirci, “Wireless Sensor

Networks: A Survey,” Computer Networks, vol. 38, 2002, pp. 393-422.• C.-Y. Chong and S.P. Kumar, “Sensor Networks: Evolution, Opportunities,

and Challenges,” Proceedings of the IEEE, vol. 91, no. 8, August 2003, pp. 1247-1256.

• K.A. Delin and S.P. Jackson, “Sensor Web for In Situ Exploration of Gaseous Biosignatures,” IEEE Aerospace Conference, Big Sky, Montana, March 2000.

• A.J. Goldsmith, S.B. Wicker, “Design Challenges for Energy-Constrained Ad Hoc Wireless Networks, IEEE Wireless Communications, August 2002, pp. 8-27.

• C. Guo, L.Z. Zhong, J.M. Rabaey, “Low Power Distributed MAC for Ad Hoc Sensor Radio Networks,” IEEE Global Telecommunications Conference, San Antonio, November 2001, vol. 5, pp. 2944-2948 .

• C. Itanagonwiwat, R. Govindan, D. Estrin, J. Heidemann, and F. Silva, “Directed Diffusion for Wireless Sensor Networking,” IEEE/ACM Transactions on Networking, vol. 11, no. 1, February 2003, pp.2-16.

• C. Itanagonwiwat, R. Govindan, D. Estrin, J. Heidemann, and F. Silva, “Impact of Network Density on Data Aggregation in Wireless Sensor Networks,” Proceedings of the 22nd International Conference on Distributed Computing Systems, Vienna, Austria, July 2002, pp. 457-458.

References• C. Itanagonwiwat, R. Govindan, D. Estrin, “Directed Diffusion: A Scalable and

Robust Communication Paradigm for Sensor Networks,” Proceedings of the ACM/IEEE Conference on Mobile Computing and Networking, Boston, August 2000, pp. 56-67.

• E.-S. Jung and N.H. Vaidya, “A Power Control MAC Protocol for Ad Hoc Networks,” ACM/IEEE Int. Conf. on Mobile Computing and Networking, Atlanta, Georgia, September 2002, pp. 36-47.

• J. Kulik, W. Rabiner Heinzelman, and H. Balakrishnan, “Negotiation-Based Protocols for Disseminating Information in Wireless Sensor Networks,” ACM/IEEE Int. Conf. on Mobile Computing and Networking, Seattle, WA, Aug. 1999.

• J. Rabaey, J. Ammer, J. da Silva, D. Patel, S. Roundy, “Picoradio Supports Ad-hoc Ultra-low Power Wireless Networking”, IEEE Computer Magazine, July 2000.

• W. Rabiner Heinzelman, A. Chandrakasan, and H. Balakrishnan, “Energy-Efficient Communications Protocols for Wireless Microsensor Networks,” Proceedings of the 33rd International Conference on System Sciences, January 2000.

• K. Sohrabi, J. Gao, V. Ailawadhi, and G.J. Pottie, “Protocols for Self-Organization of a Wireless Sensor Network,” IEEE Personal Communications, October 2000, pp. 16-27.

• C.-K. Toh, “Maximum Battery Life Routing to Support Ubiquitous Mobile Computing in Wireless Ad Hoc Networks,” IEEE Communications Magazine, June 2001, pp. 2-11.

• S. Toumpis and A.J. Goldsmith, “Performance, Optimization, and Cross-Layer Design of Media Access Protocols for Wireless Ad Hoc Networks,” International Conference on Communications (ICC), Anchorage, Alaska, May 2003, pp. 2234-2240.