Protocols in Wireless Sensor Networks

From Vision to Reality

MAC in Wireless Sensor Networks

IEEE 802.15.4 Basics• 802.15.4 is a simple packet data protocol:

– CSMA/CA - Carrier Sense Multiple Access with collision avoidance

– Optional time slotting and beacon structure– Three bands, 27 channels specified

• 2.4 GHz: 16 channels, 250 kbps• 868.3 MHz : 1 channel, 20 kbps• 902-928 MHz: 10 channels, 40 kbps

• Works well for:– Long battery life, selectable latency for

controllers, sensors, remote monitoring and portable electronics

MAC Options• Two channel access mechanisms

– Non-beacon network• Standard CSMA-CA communications + ACK

– Beacon-enabled network• Superframe structure

– For dedicated bandwidth and low latency– Set up by network coordinator to transmit

beacons at predetermined intervals» 15ms to 252sec » 16 equal-width time slots between

beacons» Channel access in each time slot is

contention free

IEEE 802.15.4 standard• Includes layers up to and including Link Layer Control

– LLC is standardized in 802.1• Supports multiple network topologies including Star,

Cluster Tree and Mesh• Channel scan for beacon is included, but it is left to the

network layer to implement dynamic channel selection

IEEE 802.15.4 MAC

IEEE 802.15.4 LLC IEEE 802.2LLC, Type I

IEEE 802.15.42400 MHz PHY

IEEE 802.15.4868/915 MHz PHY

Data Link Controller (DLC)

Networking App Layer (NWK)

ZigBee Application Framework

• Low complexity: 26 service primitives

versus 131 service primitives for 802.15.1 (Bluetooth)

IEEE 802.15.4 Device Types• Three device types

– Network Coordinator• Maintains overall network knowledge; most

memory and computing power– Full Function Device

• Carries full 802.15.4 functionality and all features specified by the standard; ideal for a network router function

– Reduced Function Device• Carriers limited functionality; used for network

edge devices• All of these devices can be no more complicated

than the transceiver, a simple 8-bit MCU and a pair of AAA batteries!

ZigBee Topology Models

ZigBee coordinatorZigBee RoutersZigBee End Devices

Star

Star topology The communication is established between devices and a single central controller, called the PAN coordinator.

ZigBee Topology Models

ZigBee coordinator

ZigBee Routers

ZigBee End Devices

Mesh

Cluster Tree

Mesh topology There is also one PAN coordinator. In contrast to star topology, any device can communicate with any other device as long as they are in range of one another.

Cluster-tree network is a special case of a Mesh network

IEEE 802.15.4 PHY

• Features– Activation/Deactivation of radio Transceiver– Energy Detection (ED)– Link Quality Indication (LQI)– Channel Selection– Clear Channel Assessment (CCA)– Transmission/Reception of packets over physical

medium

Operating frequency bands

Co-exist with 802.11

Surviving Wi-Fi Interference in Low Power ZigBee

NetworksChieh-Jan Mike Liang, Nissanka Bodhi Priyantha,

Jie Liu, Andreas TerzisJohns Hopkins University, Microsoft Research

Sensys 2010

Experiment

• In Parking garage• 802.11

– 802.11 b/g access point and a laptop– A stream of 1,500-byte TCP segments

• 802.15.4– One sender, five receivers– Sends one max-size packet every 75 ms– Broadcast 2000 packets– Predefined byte pattern– Record every packets

Packet Reception Rate

Overlay of 802.11 and 802.15.4

Asymmetric Region

symmetric Region

Bit-error Distributionsymmetric Region

symmetric Region

Asymmetric Region

S-MAC Sensor Medium Access Control

ProtocolAn Energy Efficient MAC protocol

for Wireless Sensor Networks

Wireless Sensor Networks

• Application specific wireless networks for monitoring, smart spaces, medical systems and robotic exploration

• Battery operated and power limited sensor devices

• Large number of distributed nodes deployed in an ad-hoc fashion

Existing MAC Design• Contention-based protocols

• IEEE 802.11 – Idle listening• PAMAS – heavy duty cycle of the

radio, avoids overhearing, idle listening

• TDMA based protocols Advantages - Reduced energy

consumption Problems – requires real clusters, and

does not support scalability

Design ConsiderationsPrimary attributes:

Energy Efficiency often difficult to recharge or replace

batteries prolonging the network lifetime is important

Scalability Some nodes may die or new nodes may

join

Secondary attributes:Fairness, latency, throughput and

bandwidth

Sources of Energy Inefficiency

• Collision

• Overhearing

• Control packet overhead

• Idle listening

S-MAC

• Tries to reduce wastage of energy from all four sources of energy inefficiency Collision – by using RTS and CTS Overhearing – by switching the radio off

when transmission is not meant for that node

Control Overhead – by message passing Idle listening – by periodic listen and sleep

Components of S-MAC• Periodic listen and sleep

– Each node goes into periodic sleep mode during which it switches the radio off and sets a timer to awake later

– When the timer expires, it wakes up

• Collision and Overhearing avoidance– using RTS/CTS mechanism– Interfering nodes go to sleep after they hear the

RTS or CTS packet

• Message passing– Only one RTS packet and one CTS packet are

used– ACK would be sent after each data fragment

S-MAC (Sensor-Networks)

• Testbed– Used Rene Motes– TinyOS– 3 working modes: receiving,

transmitting and sleep

• Topology used in the experiment– 3 MAC modules on the mote and

TinyOS platform

1. Simplified IEEE802.11 DCF

2. Message passing with overhearing avoidance

3. The complete S-MAC

S-MAC (Sensor-Networks)

• The energy consumption result on the source nodes A and B– When the traffic is heavy (the

inter-arrival time<4s), S-MAC achieves energy saving mainly by avoiding overhearing and efficiently transmitting a long message

– When the traffic is light, the periodic sleep plays a key role for energy savings

相关研究• 考虑的因素

节点能量有限且难以补充具备良好的可扩展性能量效率以外的公平性一般不作为设计目标

• 协议通常采用“侦听 /休眠”交替的信道访问策略 , 以减少 collision 、 overhearing 和 idle listening;

• 通过限制控制分组长度和数量减少控制开销 ;尽量延长节点休眠时间 ,减少状态切换次数 .

• 为了避免MAC协议本身开销过大 ,消耗过多的能量 , MAC协议尽量做到简单、高效 .

Routing in Wireless Sensor Networks

Directed Diffusion:A Scalable and Robust

Communication Paradigm for Sensor Networks

Motivation

• Properties of Sensor Networks– Data centric– No central authority– Resource constrained– Nodes are tied to physical locations– Nodes may not know the topology– Nodes are generally stationary

• How can we get data from the sensors?

Directed Diffusion

• Data centric – Individual nodes are unimportant

• Request driven– Sinks place requests as interests– Sources satisfying the interest can be found– Intermediate nodes route data toward sinks

• Localized repair and reinforcement• Multi-path delivery for multiple sources,

sinks, and queries

Motivating Example• Sensor nodes are monitoring animals

• Users are interested in receiving data for all 4-legged creatures seen in a rectangle

• Users specify the data rate

Interest and Event Naming• Query/interest:

1. Type=four-legged animal2. Interval=20ms (event data rate)3. Duration=10 seconds (time to cache)4. Rect=[-100, 100, 200, 400]

• Reply:1. Type=four-legged animal2. Instance = elephant3. Location = [125, 220]4. Intensity = 0.65. Confidence = 0.856. Timestamp = 01:20:40

• Attribute-Value pairs, no advanced naming scheme

Directed Diffusion

• Sinks broadcast interest to neighbors– Initially specify a low data rate just to find sources

for minimal energy consumptions

• Interests are cached by neighbors• Gradients are set up pointing back to where

interests came from • Once a source receives an interest, it routes

measurements along gradients

Interest Propagation• Flood interest

• Constrained or Directional flooding based on location is possible

• Directional propagation based on previously cached data

Source

Sink

Interest

Gradient

Data Propagation

• Multipath routing – Consider each gradient’s link quality

Source

Sink

Gradient

Data

Reinforcement

• Reinforce one of the neighbor after receiving initial data.– Neighbor who consistently performs better than others– Neighbor from whom most events received

Source

Sink

Gradient

Data

Reinforcement

Summary of the protocol

Evaluation

• ns2 simulation• Modified 802.11 MAC for energy use calculation

– Idle time: 35mW– Receive: 395mw– Transmit: 660mw

• Baselines– Flooding – Omniscient multicast: A source multicast its event to all

sources using the shortest path multicast tree – Do not consider the tree construction cost

• Simulate node failures• No overload• Random node placement

– 50 to 250 nodes (increment by 50)– 50 nodes are deployed in 160m * 160m

• Increase the sensor field size to keep the density constant for a larger number of nodes

– 40m radio range

Metrics

• Average dissipated energy– Ratio of total energy expended per node to number of

distinct events received at sink– Measures average work budget

• Average delay– Average one-way latency between event transmission and

reception at sink– Measures temporal accuracy of location estimates

• Both measured as functions of network size

Average Dissipated Energy

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

0.018

0 50 100 150 200 250 300

Ave

rag

e D

issi

pat

ed E

ner

gy

(Jo

ule

s/N

od

e/R

ecei

ved

Eve

nt)

Network Size

DiffusionDiffusion

Omniscient MulticastOmniscient Multicast

FloodingFlooding

They claim dThey claim diffusion iffusion can can outperform omniscient multicastoutperform omniscient multicast due to due toin-network processing & suppression. For example, multiple in-network processing & suppression. For example, multiple

sources can detect a four-legged animal in one area.sources can detect a four-legged animal in one area.

Impact of In-network Processing

0

0.005

0.01

0.015

0.02

0.025

0 50 100 150 200 250 300

Ave

rag

e D

issi

pat

ed E

ner

gy

(Jo

ule

s/N

od

e/R

ecei

ved

Eve

nt)

Network Size

Diffusion With Diffusion With SuppressionSuppression

Diffusion Without Diffusion Without SuppressionSuppression

LEACH [HICSS00]

• Proposed for continuous data gathering protocol

• Divide the network into clusters• Cluster head periodically collect &

aggregate/compress the data in the cluster using TDMA

• Periodically rotate cluster heads for load balancing

 LEACH’s hierarchical routing architecture

Geographic Routing for Sensor Networks

Motivation• A sensor net consists of hundreds or thousands of nodes

– Scalability is the issue– Existing ad hoc net protocols, e.g., DSR, AODV, ZRP, require

nodes to cache e2e route information– Dynamic topology changes– Mobility

• Reduce caching overhead– Hierarchical routing is usually based on well defined, rarely

changing administrative boundaries– Geographic routing

• Use location for routing

• Assumptions – Every node knows its location

• Positioning devices like GPS • Localization

– A source can get the location of the destination

Geographic Routing: Greedy Routing

S D

Closest to D

A

- Find neighbors who are the closer to the destination- Forward the packet to the neighbor closest to the destination

Greedy Forwarding does NOT always work

If the network is dense enough that each interior node has a neighbor in every 2/3 angular sector, GF will always succeed

GF fails

Dealing with Void

Apply the right-hand rule to traverse the edges of a voidPick the next anticlockwise edgeTraditionally used to get out of a maze

TTDD: A Two-tier Data Dissemination Model for Large-scale Wireless Sensor Networks

Haiyun Luo

Fan Ye, Jerry Cheng

Songwu Lu, Lixia Zhang

UCLA CS Dept.

Sensor Network Model

Source

Stimulus

Sink

Sink

Mobile Sink

Excessive PowerConsumption

Increased WirelessTransmissionCollisions

State MaintenanceOverhead

TTDD Basics

Source

Dissemination Node

Sink

Data Announcement

Query

Data

Immediate DisseminationNode

TTDD Mobile Sinks

Source

Dissemination Node

Sink

Data Announcement

Data

Immediate DisseminationNode

Immediate DisseminationNode

TrajectoryForwarding

TrajectoryForwarding

TTDD Multiple Mobile Sinks

Source

Dissemination Node

Data Announcement

Data

Immediate DisseminationNode

TrajectoryForwarding

Source

Conclusion

• TTDD: two-tier data dissemination Model– Exploit sensor nodes being stationary and

location-aware– Construct & maintain a grid structure with low

overhead

• Proactive sources– Localize sink mobility impact

• Infrastructure-approach in stationary sensor networks– Efficiency & effectiveness in supporting mobile

sinks

Double Cross for Data Dissemination in Sensor Networks

Basic Principle

Up to 69%

A

B

98%

Reachability Numerical Analysis

98.3% 99.75%

57%

67.7%

Probability of Unreach highest at perimeters and corners

NS2 Simulations with MAM show

around 99% reachability

Main Problem

• How to use this principle if the nodes have no location ?

• How to forward the message along a Line?

Simulation Results

Simulation Results

Range-Based and Range-Free Localization Schemes for Sensor

Networks

Localization• Critical service

– A sensor reading consists of <time, location, measurement>

– E.g., target tracking, disaster recovery, fire detection, patient location in a smart hospital, …

– Needed for geographic routing

• Too expensive for an individual sensor to have a GPS (Global Positioning System)– Reference nodes (called anchor or beacon

nodes) + sensor nodes

Range-based localization schemes• TOA (Time of Arrival)

– Get range info via signal propagation delay– E.g., GPS– Expensive, power consuming, inaccurate

• TDOA (Time Difference of Arrival)– Transmit both radio and ultrasonic signals at the

same time to observe the arrival time difference– Extra hardware, i.e., ultrasonic channel, is

required– Not only radio but also sound signals have

multipath effects affected by humidity, temperature, …

• Received signal strength (RSS) – Distance estimation based on RSS– Hard due to radio signal vagaries

• AoA (Angle of Arrival)– A node estimates the relative angles

between neighbors– Requires directional antennae

Range-based localization schemes

Range-free localization

• Centroid algorithm– Anchors beacon their positions to

neighbors (single hop broadcast)– A sensor node computes the centroid using

all received beacon messages

• DV-HOP– Anchor locations are flooded through the

network– Keep the running hop count– Estimate average one hop distance

• Amorphous Positioning– Similar to DV-HOP– Use offline one hop distance estimation

Range-Free Localization Schmes for Large Scale Sensor NEtworks

- APIT (Approximate Point In Triangulation)

Mobicom 2003

PIT (Point In Triangulation)

• A node chooses three anchors from all audible anchors

• Test whether it’s inside the triangle• Repeat for all possible combinations of

audible three anchors• Compute the COG of the intersection of

all the triangles

Perfect PIT test• For three given anchors, A, B, C, determine

whether a point M with an unknown position is inside the triangle ABC or not

• Proposition I: If M is inside the triangle, when M is shifted, the new position is nearer to (or farther from) at least one anchor A, B, or C

A

CB

M

Continued…

• Proposition II: If M is outside the triangle, when M is shifted, there must exist a direction in which the position of M is farther from or closer to all three anchors A, B and C

A

CB

M

Problems with Perfect PIT test

• How can a sensor node perform the PIT test w/o actually moving?

• How to do exhaustive tests considering all possible directions of departure?

APIT (Approximate PIT test)• In a certain propagation direction, the

received signal strength is assumed to monotonically decrease in an environment w/o obstacles

• Departure test: further away a node is from the anchor, weaker the received signal strength.

Appropriate PIT Test.• Use neighbor information to emulate the movements

of the nodes in the perfect PIT test. • If no neighbor of M is further from/ closer to all

three anchors A, B and C simultaneously, M assumes that it is inside triangle ABC. Otherwise, M assumes it resides outside this triangle.

Inside CaseOutside Case

Signal strength at different distances

• to justify the departure test

Localization error for varying AH

• APIT works better as AH increases. • Large errors when AH < 8

• It’s relatively less sensitive to random deployment.

Localization error impact on geographic forwarding

Summary

• APIT is resilient to irregular radio patterns and random deployment

• Relatively low overhead compared to DV-Hop & Amorphous localization (but more overhead than Centroid)

• Localization has been well studied but still needs more work