Download Document

Chapter 4
Routing Protocols
1
Overview

Routing in WSNs is challenging due to the inherent
characteristics that distinguish these networks from other
wireless networks like mobile ad hoc networks or
cellular networks.


2
First, due to the relatively large number of sensor nodes, it is
not possible to build a global addressing scheme for the
deployment of a large number of sensor nodes. Thus, traditional
IP-based protocols may not be applied to WSNs. In WSNs,
sometimes getting the data is more important than knowing the
IDs of which nodes sent the data.
Second, in contrast to typical communication networks, almost
all applications of sensor networks require the flow of sensed
data from multiple sources to a particular BS.
Overview (cont.)





3
Third, sensor nodes are tightly constrained in terms of
energy, processing, and storage capacities. Thus, they require
carefully resource management.
Fourth, in most application scenarios, nodes in WSNs are
generally stationary after deployment except for, may be, a
few mobile nodes.
Fifth, sensor networks are application specific, i.e., design
requirements of a sensor network change with application.
Sixth, position awareness of sensor nodes is important since
data collection is normally based on the location.
Finally, data collected by many sensors in WSNs is typically
based on common phenomena, hence there is a high
probability that this data has some redundancy.
Overview (cont.)

The task of finding and maintaining routes in WSNs is
nontrivial since energy restrictions and sudden
changes in node status (e.g., failure) cause frequent
and unpredictable topological changes.
To minimize energy consumption, routing techniques
proposed for WSNs employ some well-known routing
strategies, e.g., data aggregation and in-network
processing, clustering, different node role assignment,
and data-centric methods were employed.

4
Outline

4.1 Routing Challenges and Design Issues in WSNs
4.2 Flat Routing
4.3 Hierarchical Routing
4.4 Location Based Routing
4.5 QoS Based Routing
4.6 Data Aggregation and Convergecast
4.7 Data Centric Networking
4.8 ZigBee
4.9 Conclusions








5
Chapter 4.1
Routing Challenges and Design Issues in
WSNs
6
Overview

The design of routing protocols in WSNs is influenced by
many challenging factors. These factors must be overcome
before efficient communication can be achieved in WSNs.












7
Node deployment
Energy considerations
Data delivery model
Node/link heterogeneity
Fault tolerance
Scalability
Network dynamics
Transmission media
Connectivity
Coverage
Data aggregation/convergecast
Quality of service
Node Deployment

Node deployment in WSNs is application dependent and
affects the performance of the routing protocol.
The deployment can be either deterministic or randomized.
In deterministic deployment, the sensors are manually placed
and data is routed through pre-determined paths.
In random node deployment, the sensor nodes are scattered
randomly creating an infrastructure in an ad hoc manner. If the
resultant distribution of nodes is not uniform, optimal
clustering becomes necessary to allow connectivity and enable
energy efficient network operation.



8
Energy Considerations

Sensor nodes can use up their limited supply of energy
performing computations and transmitting information in a
wireless environment. Energy conserving forms of
communication and computation are essential.
Sensor node lifetime shows a strong dependence on the battery
lifetime. In a multi-hop WSN, each node plays a dual role as
data sender and data router. The malfunctioning of some sensor
nodes due to power failure can cause significant topological
changes and might require rerouting of packets and
reorganization of the network.

9
Data Delivery Model

Time-driven (continuous)


Event-driven



React immediately to sudden and drastic changes
Query-driven


Suitable for applications that require periodic data
monitoring
Respond to a query generated by the BS or another node in
the network
Hybrid
The routing protocol is highly influenced by the data
reporting method in terms of energy consumption and
route stability.
10
Node/Link Heterogeneity



Depending on the application, a sensor node can have
a different role or capability.
The existence of a heterogeneous set of sensors raises
many technical issues related to data routing.
Even data reading and reporting can be generated
from these sensors at different rates, subject to diverse
QoS constraints, and can follow multiple data
reporting models.
11
Fault Tolerance


Some sensor nodes may fail or be blocked due to lack
of power, physical damage, or environmental
interference.
It may require actively adjusting transmit powers and
signaling rates on the existing links to reduce energy
consumption, or rerouting packets through regions of
the network where more energy is available.
12
Scalability



The number of sensor nodes deployed in the sensing
area may be on the order of hundreds or thousands, or
more.
Any routing scheme must be able to work with this
huge number of sensor nodes.
In addition, sensor network routing protocols should
be scalable enough to respond to events in the
environment.
13
Network Dynamics



Routing messages from or to moving nodes is more
challenging since route and topology stability become
important issues.
Moreover, the phenomenon can be mobile (e.g., a
target detection/ tracking application).
On the other hand, sensing fixed events allows the
network to work in a reactive mode while dynamic
events in most applications require periodic reporting
to the BS.
14
Transmission Media


The traditional problems associated with a wireless
channel may also affect the operation of the sensor
network.
In general, the required bandwidth of sensor data will
be low, on the order of 1-100 kb/s. Related to the
transmission media is the design of MAC.


15
TDMA (time-division multiple access)
CSMA (carrier sense multiple access)
Connectivity



High node density in sensor networks precludes them
from being completely isolated from each other.
However, may not prevent the network topology from
being variable and the network size from shrinking
due to sensor node failures.
In addition, connectivity depends on the possibly
random distribution of nodes.
16
Coverage



In WSNs, each sensor node obtains a certain view of
the environment.
A given sensor’s view of the environment is limited in
both range and accuracy.
It can only cover a limited physical area of the
environment.
17
Data Aggregation/Convergecast



Since sensor nodes may generate significant
redundant data, similar packets from multiple nodes
can be aggregated to reduce the number of
transmissions.
Data aggregation is the combination of data from
different sources according to a certain aggregation
function.
Convergecasting is collecting information “upwards”
from the spanning tree after a broadcast.
18
Quality of Service


In many applications, conservation of energy, which is
directly related to network lifetime.
As energy is depleted, the network may be required to
reduce the quality of results in order to reduce energy
dissipation in the nodes and hence lengthen the total
network lifetime.
19
Routing Protocols in WSNs: A taxonomy
Routing protocols in WSNs
Network Structure
Flat routing
•
•
SPIN
Directed Diffusion (DD)
Hierarchical routing
•
•
•
LEACH
PEGASIS
TTDD
Location based routing
•
•
GEAR
GPSR
Protocol Operation
Negotiation based routing
•
SPIN
Multi-path network routing
•
DD
Query based routing
•
DD, Data centric routing
QoS based routing
•
TBP, SPEED
Coherent based routing
•
DD
Aggregation
•
20
Data Mules, CTCCAP
Reference

J. N. Al-Karaki and A. E. Kamal, “Routing techniques in
wireless sensor networks: a survey,” IEEE Wireless
Communications, vol. 11, no. 6, pp. 6-28, Dec. 2004.
21
Chapter 4.2
Flat Routing
22
Overview



In flat network, each node typically plays the same role and
sensor nodes collaborate together to perform the sensing task.
Due to the large number of such nodes, it is not feasible to
assign a global identifier to each node. This consideration has
led to data centric routing, where the BS sends queries to
certain regions and waits for data from the sensors located in
the selected regions. Since data is being requested through
queries, attribute-based naming is necessary to specify the
properties of data.
Prior works on data centric routing, e.g., SPIN and directed
diffusion, were shown to save energy through data negotiation
and elimination of redundant.
23
4.2.1
SPIN
Sensor Protocols for Information via Negotiation
24
SPIN -Motivation

Sensor Protocols for Information via Negotiation,
SPIN


A Negotiation-Based Protocols for Disseminating
Information in Wireless Sensor Networks.
Dissemination is the process of distributing individual
sensor observations to the whole network, treating all
sensors as sink nodes



25
Replicate complete view of the environment
Enhance fault tolerance
Broadcast critical piece of information
SPIN (cont.)- Motivation





Flooding is the classic approach for dissemination
Source node sends data to all neighbors
Receiving node stores and sends data to all its
neighbors
Disseminate data quickly
Deficiencies



26
Implosion
Overlap
Resource blindness
SPIN (cont.)-Implosion
A
Node
The direction
of data sending
The connect
between nodes
x
x
B
C
x
x
D
27
SPIN (cont.)- Overlap
r
q
Node
The direction
of data sending
The connect
between nodes
The searching
range of the
node
s
A
B
(q, r)
(s, r)
C
28
SPIN (cont.)- Resource blindness

In flooding, nodes do not modify their activities based
on the amount of energy available to them.

A network of embedded sensors can be resourceaware and adapt its communication and computation
to the state of its energy resource.
29
SPIN (cont.)
Sensor Protocols for Information via Negotiation

Negotiation




Before transmitting data, nodes negotiate with each other to
overcome implosion and overlap
Only useful information will be transferred
Observed data must be described by meta-data
Resource adaptation



30
Each sensor node has resource manager
Applications probe manager before transmitting or
processing data
Sensors may reduce certain activities when energy is low
SPIN (cont.)- Meta-Data

Completely describe the data



Must be smaller than the actual data for SPIN to be
beneficial
If you need to distinguish pieces of data, their meta-data
should differ
Meta-Data is application specific


31
Sensors may use their geographic location or unique node ID
Camera sensor may use coordinate and orientation
SPIN (cont.)- SPIN family

Protocols of the SPIN family
 SPIN-PP
 It is designed for a point to point communication, i.e., hopby-hop routing
 SPIN-EC
 It works similar to SPIN-PP, but, with an energy heuristic
added to it
 SPIN-BC
 It is designed for broadcast channels
 SPIN-RL
 When a channel is lossy, a protocol called SPIN-RL is
used where adjustments are added to the SPIN-PP protocol
to account for the lossy channel.
32
SPIN (cont.)- Three-stage handshake protocol

SPIN-PP: A three-stage handshake protocol for pointto-point media



33
ADV – data advertisement
 Node that has data to share can advertise this by
transmitting an ADV with meta-data attached
REQ – request for data
 Node sends a request when it wishes to receive some
actual data
DATA – data message
 Contain actual sensor data with a meta-data header
 Usually much bigger than ADV or REQ messages
SPIN (3-Step Protocol)
A
B
34
SPIN (3-Step Protocol)
A
B
Notice the color of the data packets sent by node B
35
SPIN (3-Step Protocol)
A
B
SPIN effective when DATA sizes are large :
REQ, ADV overhead gets amortized
36
SPIN (cont.)- SPIN-EC (Energy-Conserve)





Add simple energy-conservation heuristic to SPIN-PP
 SPIN-EC: SPIN-PP with a low-energy threshold
Incorporate low-energy-threshold
Works as SPIN-PP when energy level is high
Reduce participation of nodes when approaching low-energythreshold
 When node receives data, it only initiates protocol if it can
participate in all three stages with all neighbor nodes
 When node receives advertisement, it does not request the
data
Node still exhausts energy below threshold by receiving ADV
or REQ messages
37
SPIN (cont.)- Conclusion



SPIN protocols hold the promise of achieving high
performance at a low cost in terms of complexity, energy,
computation, and communication
Pros
 Each node only needs to know its one-hop neighbors
 Significantly reduce energy consumption compared to
flooding
Cons
 Data advertisement cannot guarantee the delivery of data
 If the node interested in the data are far from the source,
data will not be delivered
 Not good for applications requiring reliable data delivery,
e.g., intrusion detection
38
SPIN (cont.)- Reference

J. Kulik, W.R. Heinzelman, and H. Balakrishnan, “Negotiationbased protocols for disseminating information in wireless
sensor networks,” Wireless Networks, Vol. 8, pp. 169-185, 2002.
39
4.2.2
Directed Diffusion
A Scalable and Robust Communication
Paradigm for Sensor Networks
40
Overview

Data-centric communication



Data is named by attribute-value pairs
Different form IP-style communication
 End-to-end delivery service
E.g.
 How many pedestrians do you
observe in the geographical region X?
A sensor field
Sources
Event
Directed
Diffusion
Sink Node
41
Overview (cont.)

Data-centric communication (cont.)





Human operator’s query (task) is diffused
Sensors begin collecting information about query
Information returns along the reverse path
Intermediate nodes aggregate the data
 Combing reports from sensors
Directed Diffusion is an important milestone in the
data centric routing research of sensor networks
42
Directed Diffusion

Typical IP based networks



Requires unique host ID addressing
Application is end-to-end
Directed diffusion – use publish/subscribe


43
Inquirer expresses an interest, I, using attribute values
Sensor sources that can service I, reply with data
Directed Diffusion (cont.)

Directed diffusion consists of




44
Interest - Query which specifies what a user wants
Data - Collected information
Gradient
 Direction and data-rate
 Events start flowing towards the originators of interests
Reinforcement
 After the sink starts receiving events, it reinforces at least
one neighbor to draw down higher quality events
Data Naming

Expressing an Interest



Using attribute-value pairs
E.g.,
Other interest-expressing schemes possible

45
E.g., hierarchical (different problem)
Interests and Gradients

Interest propagation




The sink broadcasts an interest
Exploratory interest with low data-rate
Neighbors update interest-cache and forwards it
 Flooding
 Geographic routing
 Use cached data to direct interests
Gradient establishment


46
Gradient set up to upstream neighbor
Low data-rate gradient
 Few packets per unit time needed
Gradient Set Up

Inquirer (sink) broadcasts exploratory interest, i1

Intended to discover routes between source and sink

Neighbors update interest-cache and forwards i1

Gradient for i1 set up to upstream neighbor



47
No source routes
Gradient – a weighted reverse link
Low gradient  Few packets per unit time needed
Exploratory Gradient
Exploratory Request
Gradient
Event
Low Data-rate Interest
Low Data-rate Interest
48
Low Data-rate Interest
Data Propagation


A sensor node that detects a target
 Search its interest cache
 Compute the highest requested data-rate among all
its outgoing gradients
 Data message is unicast individually
A node that receives a data message
 Find a matching interest entry in its cache
 Check the data cache for loop prevention
 Re-send the data to neighbors
49
Event-Data Propagation



Event e1 occurs, matches i1 in sensor cache
 e1 identified based on waveform pattern matching
Interest reply diffused down gradient (unicast)
 Diffusion initially exploratory (low packet-rate)
Cache filters suppress previously seen data
50
Reinforcement (1/4)

Positive reinforcement
 Sink selects the neighboring node
 Original interest message but with high data-rate
 Neighboring node must also reinforce at least one neighbor
 Low-delay path is selected
 Exploratory gradients still exist: useful for faults
Event
Reinforced gradient
Reinforced gradient
Source
A sensor field
Sink
51
Reinforcement (2/4)

Path establishment for multiple sources and sinks


Node reinforce all neighbors from which new events were
recently received
Ex: Multiple sources A and B
Event B
D
A
C
Sink
Multiple sources
52
Reinforcement (3/4)

Path failure and recovery
 Link failure detected by reduced rate, data loss
 Choose next best link (i.e., compare links based on
infrequent exploratory downloads)
 Negatively reinforce lossy link
 Either send interest with base (exploratory) data rate or
allow neighbor’s cache to expire over time
Event
D
M
Source
A
C
53
B
Sink
Link A-M lossy
A reinforces B
B reinforces C
C reinforces D
or
A negative reinforces M
M negative reinforces D
Reinforcement (4/4)

Multipath routing


Consider each gradient’s link quality
B
Source
Event
Using negative reinforcement




54
Path Truncation
Loop removal
 For resource saving
Ex:
 B gets same data from both A and D, but
D always delivers late due to looping
 B negative reinforces D, D negative
reinforces E, E negative reinforces B
 Loop B→E →D →B eliminated
Conservative negative reinforces useful for
fault resilience
A
Sink
Multiple paths
D
A
E
B
C
A removable loop
Design Considerations

Design Space for Diffusion
Diffusion element
Design Choices
Interest Propagation
•Flooding
•Constrained or directional flooding based on location
•Directional propagation based on previously cached data
Data Propagation
•Reinforcement to single path delivery
•Multipath delivery with selective quality along different
paths
• Multipath delivery with probabilistic forwarding
Data caching and
aggregation
•For robust data delivery in the face of node failure
•For coordinated sensing and data reduction
• For directing interests
Reinforcement
•Rules for deciding when to reinforce
•Rules for how many neighbors to reinforce
•Negative reinforcement mechanisms and rules
55
Directed Diffusion: Pros & Cons



Different from SPIN in terms of on-demand data querying
mechanism
 Sink floods interests only if necessary (lots of energy savings)
 In SPIN, sensors advertise the availability of data
Pros
 Data centric: All communications are neighbor to neighbor
with no need for a node addressing mechanism
 Each node can do aggregation & caching
Cons
 On-demand, query-driven: Inappropriate for applications
requiring continuous data delivery, e.g., environmental
monitoring
 Attribute-based naming scheme is application dependent
 For each application it should be defined a priori
 Extra processing overhead at sensor nodes
56
Conclusions





57
Directed diffusion, a paradigm proposed for event monitoring
sensor networks
Directed Diffusion has some novel features - data-centric
dissemination, reinforcement-based adaptation to the
empirically best path, and in-network data aggregation and
caching.
Notion of gradient (exploratory and reinforced)
Energy efficiency achievable
Diffusion mechanism resilient to fault tolerance
 Conservative negative reinforcements proves useful
References


C. Intanagonwiwat, R. Govindan, and D. Estrin, “Directed
Diffusion: A Scalable and Robust Communication Paradigm
for Sensor Networks,” in the Proceedings of the Sixth Annual
International Conference on Mobile Computing and Networks
(MobiCom’00), August 2000.
C. Intanagonwiwat, R. Govindan, D. Estrin, J. Heidemann, and
F. Silva, “Directed Diffusion for Wireless Sensor Networking,”
IEEE/ACM Transactions on Networking, Vol. 11, No. 1, Feb.
2003.
58
Chapter 4.3
Hierarchical Routing
59
Overview



In a hierarchical architecture, higher energy nodes can be used
to process and send the information while low energy nodes
can be used to perform the sensing of the target.
Hierarchical routing is mainly two-layer routing where one
layer is used to select cluster heads and the other layer is used
for routing.
Hierarchical routing (or cluster-based routing), e.g., LEACH,
PEGASIS, TTDD, is an efficient way to lower energy
consumption within a cluster and by performing data
aggregation and fusion in order to decrease the number of
transmitted messages to the base stations.
60
4.3.1
LEACH
Low-Energy Adaptive Clustering Hierarchy
61
LEACH




LEACH (Low-Energy Adaptive Clustering Hierarchy), a
clustering-based protocol that minimizes energy dissipation in
sensor networks.
LEACH outperforms classical clustering algorithms by using
adaptive clusters and rotating cluster-heads, allowing the
energy requirements of the system to be distributed among all
the sensors.
LEACH is able to perform local computation in each cluster to
reduce the amount of data that must be transmitted to the base
station.
LEACH uses a TDMA/CDMA MAC to reduce inter-cluster
and intra-cluster collisions.
62
LEACH (cont.)




Sensors elect themselves to be local cluster-heads at any given
time with a certain probability. These cluster-head nodes
broadcast their status to the other sensors in the network.
Each sensor node determines to which cluster it wants to
belong by choosing the cluster-head that requires the minimum
communication energy.
Once all the nodes are organized into clusters, each clusterhead creates a schedule for the nodes in its cluster.
A cluster-head drains the battery of that node. In order to
spread this energy usage over multiple nodes, the cluster-head
nodes are not fixed; rather, this position is self-elected at
different time intervals.
63
LEACH: Adaptive Clustering




Periodic independent self-election
 Probabilistic
CSMA MAC used to advertise
Nodes select advertisement with strongest signal strength
Dynamic TDMA cycles
All nodes marked with a given symbol belong to the same cluster, and
the cluster head nodes are marked with a ●.
64
Algorithm


Periodic process
Three phases per round:



65
Advertisement
 Execute election algorithm
Setup
 Schedule creation
 the clusters are organized and cluster heads are selected
Steady-State
 Data transmission
 the data transfers to the BS (Base Station)
Homework Assignment - LEACH
Fixed-length cycle
Setup phase
Steady-state phase
Time slot Time slot Time slot
1
2
3
Advertisement phase
66
Self-election of cluster
heads
Cluster heads compete
with CSMA
Cluster setup phase
Members
compete with
CSMA
Broadcast schedule
Cluster head Broadcast
CDMA code to members
Algorithm (cont.)

Set-up phase


Node n choosing a random number m between 0 and 1
If m < T(n) for node n, the node becomes a cluster-head where
P

if n G

T (n )  1  P [r * mod(1/ P )]
0
otherwise ,


67
where P = the desired percentage of cluster heads (e.g., P= 0.05), r=the
current round, and G is the set of nodes that have not been cluster-heads
in the last 1/P rounds. Using this threshold, each node will be a clusterhead at some point within 1/P rounds. During round 0 (r=0), each node
has a probability P of becoming a cluster-head.
Algorithm Details (cont.)



Set-up phase
 Cluster heads assign a TDMA schedule for their members
where each node is assigned a time slot when it can transmit.
 Each cluster communications using different CDMA codes to
reduce interference from nodes belonging to other clusters.
TDMA intra-cluster
CDMA inter-cluster
 Spreading codes determined randomly
 Broadcast during advertisement phase
68
Algorithm (cont.)

Steady-state phase



69
All source nodes send their data to their cluster heads
Cluster heads perform data aggregation/fusion through local
transmission
Cluster heads send them back to the BS using a single direct
transmission
An Example of a LEACH Network

While neither of these diagrams is the optimum scenario, the
second is better because the cluster-heads are spaced out and
the network is more properly sectioned
Good case scenario
70
Bad case scenario
Node
Cluster-Head Node
X Node that has been cluster-head in the last 1/P rounds
Cluster Border
Conclusions

Advantages
 Increases the lifetime of the network
 Even drain of energy
 Distributed, no global knowledge required
 Energy saving due to aggregation by CHs

Disadvantages
 LEACH assumes all nodes can transmit with enough power
to reach BS if necessary (e.g., elected as CHs)
 Each node should support both TDMA & CDMA
 Need to do time synchronization
 Nodes use single-hop communication
71
Reference

W. Heinzelman, A. Chandrakasan, and H. Balakrishnan,
“Energy-efficient communication protocol for wireless sensor
networks,” Proceedings of the 33rd Hawaii International
Conference on System Sciences, January 2000.
72
Chapter 4.4
Location Based Routing
73
Overview

Sensor nodes are addressed by means of their locations.





The distance between neighboring nodes can be estimated on the basis of
incoming signal strengths.
Relative coordinates of neighboring nodes can be obtained by
exchanging such information between neighbors.
To save energy, some location based schemes demand that
nodes should go to sleep if there is no activity.
More energy savings can be obtained by having as many
sleeping nodes in the network as possible.
Hereby, two important location based routing protocols, GEAR
and GPSR, are introduced.


74
Geographical and Energy Aware Routing (GEAR)
Greedy Perimeter Stateless Routing (GPSR)
4.4.1
GEAR
Geographical and Energy Aware Routing
75
Geographical and Energy Aware Routing (GEAR)



The protocol, called Geographic and Energy Aware Routing
(GEAR), uses energy aware and geographically-informed
neighbor selection heuristics to route a packet towards the
destination region.
The key idea is to restrict the number of interests in directed
diffusion by only considering a certain region rather than
sending the interests to the whole network. By doing this,
GEAR can conserve more energy than directed diffusion.
The basic concept comprises of two main parts


76
Route packets towards a target region through geographical and energy
aware neighbor selection
Disseminate the packet within the region
Energy Aware Neighbor Computation




Each node N maintains state h(N, R) which is called learned
cost to region R, where R is the target region
Each node infrequently updates neighbor of its cost
When a node wants to send a packet, it checks the learned cost
to that region of all its neighbors
If the learned cost of a neighbor to a region is not available, the
estimated cost is computed as follows:
c(Ni, R) = αd(Ni, R) + (1-α)e(Ni)
where
α = tunable weight, from 0 to 1.
d(Ni, R) = normalized distance of neighbor to region
e(Ni) = normalized consumed energy at node i
77
Energy Aware Neighbor Computation (cont.)



When a node wants to forward a packet to a destination, it
checks to see if it has any neighbor closer to destination than
itself
In case of multiple choices, it aims to minimize the learned cost
h(Nmin, R)
It then sets its own cost to:
h(N, R) = h(Ni, R) + c(N, Ni)
c(N, Ni) = combination of remaining energy of N and Ni and the
distance between them
78
Forwarding Around Holes
K
L
T
C–T=2
B–T= 5
F
A
G
B
xH
I
J
h(C,T) = h(B,T)+c(C,B)
5
C
D
E
S
α is set to 1. Initially, at time 0, at node S, among all neighbors of S, B, C, D
are closer to T than S. h(B,T)=c(B,T)= 5 , h(C,T)=c(C,T)=2, h(D,T)=c(D,T)= 5 .
79
Recursive Geographic Forwarding


Once the target region is reached, the packets are disseminated
within the region by recursive geographic forwarding
Forwarding stops when a node is the only one in a sub-region
Ni
80
Recursive Geographic Forwarding (cont.)
Pathologies

Inefficient Transmission

Recursive geographic forwarding vs. Restricted flooding
A
Restricted flooding 1 times for
Recursive Geographic
sending and 4 times for receiving
Forwarding 3 times for sending
= consuming
and 3 times for receiving =
5 units of energy
consuming 6 units of energy
C
E
B
D
F
81
Recursive Geographic Forwarding (cont.)
Pathologies

Non-Termination

When network density is low compared to (sub) target region size
K
E
H
B
A
C
L
F
82
Recursive Geographic Forwarding (cont.)
Proposed solution for pathologies


Node degree is used as a criteria to differentiate low density
networks from high density ones
Choice of restricted flooding over recursive geographic
forwarding is made accordingly
83
Conclusion


GEAR strategy attempts to balance energy consumption and
thereby increase network lifetime
GEAR performs better in terms of connectivity after initial
partition
84
References





Y. Yu, D. Estrin, and R. Govindan, “Geographical and Energy-Aware
Routing: A Recursive Data Dissemination Protocol for Wireless
Sensor Networks”, UCLA Computer Science Department Technical
Report, UCLA-CSD TR-01-0023, May 2001.
Nirupama Bulusu, John Heidemann, and Deborah Estrin. “Gps-less
low cost outdoor localization for very small devices”. IEEE Personal
Communications Magazine, 7(5):28-34, October 2000.
L. Girod and D. Estrin. “Robust range estimation using acoustic and
multimodal sensing”. In IEEE/RSJ International Conference on
Intelligent Robots and Systems (IROS 2001), Maui, Hawaii, October
2001.
Nissanka B. Priyantha, Anit Chakraborty, and Hari Balakrishnan.
“The cricket location-support system”. In Proc. ACM Mobicom,
Boston, MA, 2000.
Andreas Savvides, Chih-Chieh Han, and Mani B. Strivastava.
“Dynamic fine-grained localization in adhoc networks of sensors”. In
Proc. ACM Mobicom, 2001.
85
4.4.2
GPSR
Greedy Perimeter Stateless Routing
86
Greedy Perimeter Stateless Routing (GPSR)




Greedy Perimeter Stateless Routing (GPSR) proposes the
aggressive use of geography to achieve scalability
GEAR was compared to a similar non-energy-aware routing
protocol GPSR, which is one of the earlier works in geographic
routing that uses planar graphs to solve the problem of holes
In case of GPSR, the packets follow the perimeter of the planar
graph to find their route.
Although the GPSR approach reduces the number of states a
node should keep, it has been designed for general mobile ad
hoc networks and requires a location service to map locations
and node identifiers.
87
Algorithm & Example

The algorithm consists of two methods:
greedy forwarding + perimeter forwarding

Greedy forwarding, which is used wherever possible,
and perimeter forwarding, which is used in the regions
greedy forwarding cannot be done.
88
Greedy Forwarding (cont.)




Under GPSR, packets are marked by their originator
with their destinations’ locations
As a result, a forwarding node can make a locally
optimal, greedy choice in choosing a packet’s next
hop
Specifically, if a node knows its radio neighbors’
positions, the locally optimal choice of next hop is the
neighbor geographically closest to the packet’s
destination
Forwarding in this regime follows successively closer
geographic hops, until the destination is reached
89
Greedy Forwarding (cont.)
D
x
y
90
Greedy Forwarding (cont.)



A simple beaconing algorithm provides all nodes with
their neighbors’ positions: periodically, each node
transmits a beacon to broadcast MAC address,
containing its own identifier (e.g., IP address) and
position
Position is encoded as two four-byte floating point
quantities, for x and y coordinate values
Upon not receiving a beacon from a neighbor for
longer than timeout interval T, a GPSR router assumes
that the neighbor has failed or gone out-of-range, and
deletes the neighbor from its neighbor table
91
Greedy Forwarding (cont.)
The Problem of Greedy Forwarding
D
v
z
|xD|<|wD|and|yD|
x will not choose to
forward to w or y
using greedy
forwarding
void
w
y
x
x
x
92
The Right-Hand Rule: Perimeters


Use the right-hand rule to map perimeters by sending packets
on tours of them. The state accumulated in these packets is
cached by nodes, which recover from local maxima in greedy
forwarding by routing to a node on a cached perimeter closer to
the destination.
This approach requires a heuristic, the no-crossing heuristic, to
force the right-hand rule to find perimeters that enclose voids
in regions where edges of the graph cross
x
z
y
93
Right-Hand Rule Does Not Work with
Cross Edges
z
u
D

w
x originates a packet to u
Right-hand rule results in the
tour x-u-z-w-u-x

x
94
Remove Crossing Edge
z
u
v
Make
w
the graph planar
Remove
x
(w,z) from the graph
Right-hand rule results in the
tour x-u-z-v-x

95
Make a Graph Planar




A graph in which no two edges cross is known as
planar. A set of nodes with radios, where all radios
have identical, circular radio range r, can be seen as a
graph: each node is a vertex, and edge (n, m) exists
between nodes n and m if the distance between n and
m, d(n, m)≦r.
Convert a connectivity graph to planar non-crossing
graph by removing “bad” edges
Ensure the original graph will not be disconnected
Two types of planar graphs:


Relative Neighborhood Graph (RNG)
Gabriel Graph (GG)
96
Planarized Graphs (cont.)
Gabriel Graph (GG)
w
u
97
v
Planarized Graphs (cont.)
Relative Neighborhood Graph (RNG)
u
98
w
v
Planarized Graphs (cont.)


An algorithm for removing edges from the graph that
are not part of the RNG or GG would yield a network
with no crossing links
The RNG is a subset of the GG



Because RNG removes more edges
Hereby, the RNG is used
If the original graph is connected, RNG is also
connected
99
Connectedness of RNG Graph

Key observation
 Any edge on the minimum spanning
tree of the original graph is not
removed
 Proof by contradiction: Assume
(u,v) is such an edge but removed in
RNG
w
u
v
100
Planarized Graphs (cont.)
Original
Gabriel Graph (GG)
The full graph of a radio
The GG subset of
network, 200 nodes, uniformly
the full graph
randomly placed on a 2000 x
2000 meter region, with a radio
range of 250 m.
101
Relative
Neighborhood Graph
(RNG)
The RNG subset of the
full and GG graphs.
Combining Greedy and Planar Perimeters






All data packets are marked initially at their originators as
greedy mode
GPSR packet headers include a flag field indicating whether
the packet is in greedy mode or perimeter mode
Packet sources also include the geographic location of the
destination in packets
Only a packet’s source sets the location destination field, it is
left unchanged as the packet is forwarded through the network
Upon receiving a greedy-mode packet for forwarding, a node
searches its neighbor table for the neighbor geographically
closest to the packet’s destination
When no neighbor is closer, the node marks the packet into
perimeter mode
102
GPSR
greedy fails
Greedy Forwarding
Perimeter Forwarding
have left local maxima
greedy works
greedy fails
103
Combining Greedy and Planar Perimeters (cont.)

GPSR packet header fields used in perimeter mode forwarding
Field
D
Lp
Lf
e0
M
104
Function
Destination Location
Location Packet Entered Perimeter Mode
Point on xV Packet Entered Current Face
First Edge Traversed on Current Face
Packet Mode: Greedy or Perimeter
Combining Greedy and Planar Perimeters (cont.)
D
Lf
e0
x
105
Lp
If forwarding node to D < Lp to D,
returns a packet to greedy mode
Conclusion


GPSR’s benefits all stem from geographic routing’s
use of only immediate-neighbor information in
forwarding decisions.
GPSR keeps state proportional to the number of its
neighbors, while both traffic sources and intermediate
DSR routers cache state proportional to the product of
the number of routes learned and route length in hops.
106
References





B. Karp and H. T. Kung, “Greedy Perimeter Stateless Routing
for Wireless Networks”, Proc. 6th Annual ACM/IEEE Int'l.
Conf. Mobile Comp. Net., Boston, MA, pp. 243-54, August
2000.
G. G. Finn, “Routing and addressing problems in large
metropolitan-scale internetworks”, Tech. Rep. ISI/RR-87-180,
Information Sciences Institute, March 1987.
S. Floyd and V. Jacoboson, “The synchronization of periodic
routing messages”, IEEE/ACM Transactions on Networking,
Vol. 2, pp. 122-136, April 1994.
B. Karp “Greedy perimeter state routing”, Invited Seminar at
the USC/Information Sciences Institute, July 1998.
J. Saltzer, D. P. Reed, and D. Clark, “End-to-end arguments in
system design”, ACM Transactions on Computer Systems, Vol.
2, No. 4, Pages: 277-288, November 1984.
107
Chapter 4.5
QoS Based Routing
108
Overview




QoS is the performance level of service offered by a
network to the user.
The Goal of QoS is to achieve a more deterministic
network behavior so that the information carried by
the network can be better delivered and the resources
can be better utilized.
In QoS-based routing protocols, the network has to
balance between energy consumption and data quality.
In particular, the network has to satisfy certain QoS
metrics, e.g., delay, energy, bandwidth, etc. when
delivering data to the BS.
109
Parameters of QoS Networks

Different services require different QoS parameters

Multimedia


Emergency services


Network availability
Group communications


Bandwidth, delay jitter & delay
Battery life
Generally the parameters that are important are:





110
bandwidth
delay jitter
battery charge
processing power
buffer space
Challenges in QoS Routing






Dynamically varying network topology
Imprecise state information
Lack of central coordination
Hidden node problem
Limited resource
Insecure medium
111
4.5.1
TBP (Ticket-Based Probing)
QoS of Bandwidth
112
Ticket-Based Probing

Distributed multi path QoS routing scheme





Bandwidth-constrained routing and delay-constrained routing
There are numerous paths from source to destination,
we shall not randomly pick several paths to search
We shall not use any flooding path-discovery
approaches, which may send routing messages to the
entire network
Multipath search is tolerant to imprecise information
We want to make an intelligent hop-by-hop path
selection to guide the search along the best candidate
paths
113
Ticket-Based Probing (cont.)
S
D
114
Ticket-Based Probing (cont.)




A ticket is the permission to search one path. The source node
issues a number of tickets based on the available state
information
Utilizes tickets to limit the number of paths searched during
route discovery
 A ticket is the permission to search a single path
 More tickets, more QoS constraints are required
Probes (routing messages) are sent from the source toward the
destination to search for a low-cost path that satisfies the QoS
requirement
Each probe is required to carry at least one ticket
115
Ticket-Based Probing (cont.)
i
S
D
j
k
116
Ticket-Based Probing (cont.)
A
Demand = 3
S
3
B
2
2
6
D
x2
C
3
117
3
3
5
E
Ticket-Based Probing (cont.)
A
Demand = 4
(1.1,3)
3
S
C
3
(1.2,1)
2
B
118
2
(1.1,3)
3
D
2
2
6
(1.2,1)
5
E
(1.2,1)
Ticket-Based Probing (cont.)
A
Demand = 4
S
(1.1,3)
C
3
(1.2,1)
2
B
119
3
3
D
(1.1.1,2)
2
(1.1.2,1)
(1.1.2,1)
2
2
6
(1.2,1)
5
(1.2,1)
E
Ticket-Based Probing (cont.)
T1
T1
D
S
D
S
T2
T2
T1
S
120
x
T2
D
Ticket-Based Probing (cont.)
A
Demand = 4
x
3
(1,4)
4
S
(2.1,3)
(2.2,1)
x2
3
C
121
2
(2.1,3)
4
3
6
B
D
5
(2.1,3)
(2.1,3)
(2.2,1)
E
(2.2,1)
Conclusion



The routing overhead is controlled by the number of tickets,
which allows the dynamic tradeoff between the overhead and
the routing performance. Issuing more tickets means searching
more paths, which results in a better chance of finding a
feasible path at the cost of higher overhead.
A distributed routing process is used to avoid any centralized
path computation that could be very expensive for QoS routing
in large networks.
This approach not only increases the chance of success but also
improves the ability to tolerate the information imprecision
because the intermediate nodes may gradually correct a wrong
decision made by the source.
122
Conclusion (cont.)


Ticket-based probing scheme achieves a balance between the
single-path routing algorithms and the flooding algorithms. It
does multipath routing without flooding.
The basic idea is to achieve a near-optimal performance with
modest overhead by using a limited number of tickets and
making intelligent hop-by- hop path selection.
123
References





S. Chen and K. Nahrstedt, “On finding multi-constrained paths,” in Proc.
IEEE ICC’98, pp. 874-879.
R. Guerin and A. Orda, “QoS-based routing in networks with inaccurate
information: Theory and algorithms,” in Proc. IEEE INFOCOM’97,
Japan, pp. 75-83.
Q. Ma and P. Steenkiste, “Quality-of-service routing with performance
guarantees,” in Proc. 4th Int. IFIP Workshop Quality of Service, May
1997, pp. 115-126.
Z. Wang and J. Crowcroft, “QoS routing for supporting resource
reservation,” IEEE J. Select. Areas Commun., Sept. 1996.
S. Chen and K Nahrstedt, “Distributed Quality-of-Service Routing in Ad
Hoc Networks,” IEEE J. Select. Areas Commun, vol.17, no. 8, pp.
1488-1505, Aug. 1999.
124
References


T. Hea, J. A Stankovic, C. Lu, and T. Abdelzaher, “SPEED: a
stateless protocol for real-time communication in sensor
networks,” in Proc. IEEE International Conference on
Distributed Computing Systems, pp. 46-55, May 2003.
G. S. Ahn, A. T. Campbell, A. Veres, and L.H. Sun. “SWAN:
Service Differentiation in Stateless Wireless Ad Hoc Networks,”
In Proc. IEEE INFOCOM'2002, June 2002.
125