Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor...
Transcript of Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor...
![Page 1: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/1.jpg)
Top-Down Clustering Based
Self-Organization of Collaborative
Wireless Sensor Networks
M.Sc Final ExamM.Sc Final Exam
Dilum BandaraDepartment of Electrical and Computer Engineering
July 21, 2008
Graduate Committee
� Prof. Anura P. Jayasumana (Advisor)� Prof. V. Chandrasekar� Dr. Daniel F. Massey
� This work is supported in part by the
grant from Environmental Sciences
Division, Army Research Office
![Page 2: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/2.jpg)
Overview
� Motivation
� Virtual Sensor Networks (VSNs)
� Cluster & cluster tree formation algorithm
� Simple Hierarchical Clustering (SHC)
� Hop-ahead Hierarchical Clustering (HHC)� Hop-ahead Hierarchical Clustering (HHC)
� Routing
� Hierarchical addressing scheme
� Cross-links based routing
� Circular path based routing
� Forming Virtual Sensor Networks
� Secure backbone design
2MSc Final Exam - 7/21/2008
![Page 3: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/3.jpg)
Wireless Sensor Networks (WSNs)
� Sensing of physical world at a far greater temporal &
spatial granularity
� Novel applications
� Habitat monitoring, earthquake monitoring, disaster response, eldercare, battlefield intelligence, etc. eldercare, battlefield intelligence, etc.
� Sensor nodes are resource constrained
� Battery powered, low processing, memory, & transmission power
� In most cases
� Randomly deployed, location unaware, no time synchronization,
unreliable, & dense networks
3MSc Final Exam - 7/21/2008
![Page 4: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/4.jpg)
Motivation – Collaborative WSNse-SENSE project [1]
Underwater Acoustic Sensor Networks project [2]
Plume tracking
[1] www.ist-esense.org[2] http://www.ece.gatech.edu/research/labs/bwn/UWASN/work.html 4
![Page 5: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/5.jpg)
Virtual Sensor Networks (VSNs)
� An emerging concept that
� supports collaborative, resource efficient, & multipurpose WSNs
� involve dynamically varying subset of sensor & users
� Provide protocol support for
� formation, usage, adaptation, & maintenance� formation, usage, adaptation, & maintenance
� Realization of VSNs require
� some structure within the network
� many-to-many communication
� VSN management functions
� The solutions should be independent of
� neighbourhood information, location awareness, network topology, time synchronization, etc.
5MSc Final Exam - 7/21/2008
![Page 6: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/6.jpg)
Contributions
� Clustering algorithm
� 2 clustering schemes� Cluster tree formation� Performance analysis
� Routing
� Hierarchical addressing scheme� 3 routing mechanisms� Performance analysis
� Virtual Sensor Networks
� Formation� Communication
� Secure backbone design
6MSc Final Exam - 7/21/2008
![Page 7: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/7.jpg)
Generic Top-Down Cluster & cluster
tree formation (GTC) algorithm
C
C2
C4
C1
7MSc Final Exam - 7/21/2008
C3
![Page 8: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/8.jpg)
GTC - Cluster tree formation
� Cluster tree is formed by keeping track of parent & child relationships
C2
C4 C1
C1
C3
C4C3C2
C10C8C7C6C5C9
8MSc Final Exam - 7/21/2008
![Page 9: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/9.jpg)
GTC algorithm
Form_Cluster(NIDCH, CIDCH, delay, nCCHs, hopsmax, TTLmax, depth)
Wait(delay)
TTL ← TTLmax
Broadcast_Cluster(NIDCH, CIDCH, hopsmax, TTLmax, TTL, depth)
ack_list ← Receive_ACK(NIDchild, hops, p1, p2, timeoutACK)
IF(ack_list = NULL)
Join_Cluster()Join_Cluster()
FOR i = 1 TO nCCHs
CCHi ← Select_Candidate_CHs(ack_list)
CIDi ← Select_next_CID(i)
delayi ← Select_delay(i)
depthi ← depth + 1
Request_Form_Cluster(CCHi, CIDi, delayi, nCCHs, hopsmax, TTLmax, depthi)
9MSc Final Exam - 7/21/2008
![Page 10: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/10.jpg)
GTC algorithm (cont.)
Join_Cluster()Listen_Broadcast_Cluster(NIDCH, CIDCH, hopsmax, TTLmax, TTL, depth)
TTL ← TTL – 1, hops ← TTLmax – TTL
IF (hops ≤ hopsmax AND my_CID = 0)
my_CID ← CIDCH, my_CH ← NIDCH, my_depth ← depth + 1
Send_ACK(my_NID, hops, p1, p2)
IF(TTL > 0)IF(TTL > 0)
Wait(Random(backoff_time))
Forward_Broadcast_Cluster(NIDCH, CIDCH, hopsmax,, TTLmax, TTL, depth)
IF(hops ≤ hopsmax)
Exit()
ELSE
IF(Wait_Listen_Neighbors(Random(backoff_time) = FALSE)
Send_ACK(my_NID, hops, p1, p2)
IF(Listen_Form_Cluster(CCH, CID, delay, nCCHs, hopsmax, TTLmax, depth, timeout) = TRUE)
Form_Cluster(my_NID, CID, delay, nCCHs, hopsmax, TTLmax, depth)
Exit()
Join_Cluster()
10MSc Final Exam - 7/21/2008
![Page 11: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/11.jpg)
GTC algorithm (cont.)
� SHC – Simple Hierarchical Clustering
� HHC – Hop-ahead Hierarchical Clustering
CID 1
73
10
CID 3CID 7
CID 10
SHC clustershopsmax = TTLmax = 1
Similar to IEEE 802.15.4 cluster tree
HHC clustersTTLmax = 2×hopsmax + 1
1
CID 1
5
4
3
2
9
8
6
CID 2
CID 4
CID 5
CID 6CID 9
CID 8
11MSc Final Exam - 7/21/2008
![Page 12: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/12.jpg)
Clusters
� HHC
� 5000 nodes
� Grid - 201×201
� Grid spacing – 5m
� R = -20dBm
� Root node in the middle12
![Page 13: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/13.jpg)
Cluster tree
13
![Page 14: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/14.jpg)
Performance analysisC
ircu
lari
ty
20
30
40
50
60
70
80
90
SHC
HHC
RHHC
Hexagonal
Nu
mb
er
of
clu
ste
r h
ead
s
60
80
100
120
140
160
SHC
HHC
RHHC
Hexagonal
Transmission power (dBm)
-20 -18 -16 -14 -12 -10
Nu
mb
er
of clu
ste
rs
0
300
600
900
1200
1500
1800
2100
SHC
HHC
RHHC
Hexagonal
14MSc Final Exam - 7/21/2008
Transmission power (dBm)
-20 -18 -16 -14 -12 -10
0
10
Depth
0 2 4 6 8 10 12 14 16N
um
be
r o
f clu
ste
r h
ead
s0
20
40
60
� HHC� Uniform clusters
� Better circularity
� Lower number of clusters
� Lower depth
� Message complexity O(n)
![Page 15: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/15.jpg)
Optimization phase –Handling disconnected nodes & optimizing cluster tree
PT = -12dBm, 5000 nodes
15MSc Final Exam - 7/21/2008
![Page 16: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/16.jpg)
Overview
� Motivation
� Cluster & cluster tree formation algorithm
� Simple Hierarchical Clustering (SHC)
� Hop-ahead Hierarchical Clustering (HHC)
� Routing� Routing
� Hierarchical addressing scheme
� Cross-links based routing
� Circular path based routing
� Forming Virtual Sensor Networks
� Secure backbone design
16MSc Final Exam - 7/21/2008
![Page 17: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/17.jpg)
Cluster tree based routing17
� Need some sort of an addressing scheme to route
17MSc Final Exam - 7/21/2008
![Page 18: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/18.jpg)
Hierarchical addressingA(0)
D(20)C(10)B(00) F(40) G(50)E(30)
18
� Single point of failure at root node
I
(100)
H
(000)
K
(110)
L
(210)
J
(010)
M
(020)
N
(030)
O
(050)
Q
(0100)
P
(0210)
18MSc Final Exam - 7/21/2008
![Page 19: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/19.jpg)
Cross-links based routingA(0)
D(20)C(10)B(00) F(40) G(50)E(30)
19
� Routing through cross links
� Reduce burden on the root node
� Use hierarchical addresses
I
(100)
H
(000)
K
(110)
L
(210)
J
(010)
M
(020)
N
(030)
O
(050)
Q
(0100)
P
(0210)
19MSc Final Exam - 7/21/2008
![Page 20: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/20.jpg)
� Send message from U to K� Hierarchical routing - 5 hops� Cross links - 5 hops� Circular path - 4 hops
Circular path based routing
20
� Circular path� Connects clusters at the same
depth� Reduce workload on root node� Use hierarchical addresses
MSc Final Exam - 7/21/2008
![Page 21: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/21.jpg)
Circular path based routing – Analytical model
][
][
][][
dEr
E
rangeontransmissi
distancehopperenergyE
hopshopperenergyE
transmittoenergyEcostE
T ×=
×=
×=
=
� R – Radius of sensor field
� D – Radius of circular link
� r – Transmission range of a node
� r1 – distance to source node
� r2 – distance to destination node
� ET – energy to send a message
21
][
rrdwhere
dEr
+=
×=
θθθ ddrdrrrprrddE 212121 ),,(),,(][ ∫∫∫=
21MSc Final Exam - 7/21/2008
![Page 22: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/22.jpg)
Analytical model (cont.)
22MSc Final Exam - 7/21/2008
![Page 23: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/23.jpg)
Analytical model (cont.)
4
5
2
332
3
4][
R
D
R
DDRdE
πππ−+−=
4
4
2
2 592][
R
D
R
D
dD
dE
πππ−+−= 43
2
3
82
43
2
3
82
4
2
2
4
2
2
R
D
R
D P(C)
R
D
R
D P(R)
πππ
πππ
π
+−=
−+−
=
D/R0.0 0.2 0.4 0.6 0.8 1.0
Pro
ba
bili
ty
0.0
0.2
0.4
0.6
0.8
1.0
P(R)
P(C)
RD 509.0=∴
23
• P(R) – Probability of going through root node
• P(C) – Probability of going through circular path
![Page 24: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/24.jpg)
Routing - Performance analysis
14
16
18
Distance (m)
0 100 200 300 400 500
De
pth
0
2
4
6
8
10
12
24MSc Final Exam - 7/21/2008
PT = -20dBm, 5000 nodes
![Page 25: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/25.jpg)
Performance analysis (cont.)N
um
be
r o
f m
essa
ge
s
10000
15000
20000
25000
30000
35000
40000
Tree only
Tree + Cross-links
Tree + Circular path
Ene
rgy p
er
messag
e (
µJ)
2.5
2.6
2.7
2.8
2.9
Depth at which circular path is formed
0 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9
Num
be
r o
f m
essa
ges
15000
20000
25000
30000
35000
40000
25MSc Final Exam - 7/21/2008
� Cross links helps to deliver more messages
� Circular path delivers even more messages
PT = -20dBm
Transmission power (dBm)
-20 -18 -16 -14 -12 -10
Nu
mbe
r o
f m
essa
ge
s
0
5000
10000
5000 nodes, circular sensor field
Depth at which circular path is formed
0 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9
2.4
![Page 26: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/26.jpg)
Overview
� Motivation
� Cluster & cluster tree formation algorithm
� Simple Hierarchical Clustering (SHC)
� Hop-ahead Hierarchical Clustering (HHC)
� Routing� Routing
� Hierarchical addressing scheme
� Cross-links based routing
� Circular path based routing
� Forming Virtual Sensor Networks
� Secure backbone design
26MSc Final Exam - 7/21/2008
![Page 27: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/27.jpg)
Plume tracking
MSc Final Exam - 7/21/2008 27
![Page 28: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/28.jpg)
Forming VSNs
� Messages are guaranteed to meet at root node� Forms a virtual tree
� More efficient & reliable than rumor & ant routingMSc Final Exam - 7/21/2008 28
![Page 29: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/29.jpg)
Forming VSN –Virtual tree that connects VSN members
29MSc Final Exam - 7/21/2008
Single event region Three event regions
![Page 30: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/30.jpg)
VSN - Performance analysis
Num
ber
of
messag
es
6000
8000
10000
12000
14000
16000
18000
20000
Region 1
Region 1+2+3
distributed
Num
ber
of m
ultic
ast
messages
500
1000
1500
2000
2500
3000
3500
Region1
Region 1+2+3
Distributed
Routing scheme
Tree Only Tree+Cross-links Tree+Circular path
Nu
mber
of
messag
es
10000
15000
20000
25000
30000
35000
40000
45000
50000
Region 1
Region 1+2+3
Distributed
30
Transmission power (dBm)
-20 -18 -16 -14 -12 -10
2000
4000
Transmission power (dBm)
-20 -18 -16 -14 -12 -10
Num
ber
of m
ultic
ast
messages
0
500
PT = -20dBm
![Page 31: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/31.jpg)
Secure backbone design
� Can support
� dynamic key distribution
� secure upper layer functions
� Secure VSNs need dynamic key assignment
� GTC algorithm can be extended to build a secure backbone
Transmission power (dBm)
-20 -18 -16 -14 -12
Circu
lari
ty
40
45
50
55
60
65
70
75
80
85
HHC
HHC + RBMCD
HHC + DKR
HHC + RBMCD-Nei
HHC + DKR-Nei
Hexagonal
Number of compromized nodes
2 4 6 8 10
Num
ber
of
CH
s a
ffecte
d
0
10
20
30
40
50
60
70
80 HHC + RBMCD - Direct
HHC + RBMCD - Indirect
HHC + DKR - Direct
HHC + DKR - Indirect
� GTC algorithm can be extended to build a secure backbone
� No significant changes are required
� Retains most of its desirable characteristics
31MSc Final Exam - 7/21/2008
![Page 32: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/32.jpg)
Future directions
� Test performance of our algorithms on a rigorous
simulation platform such as TOSSIM
� Use actual data from a tank based testbed
� VSN management functions
� VSN dynamics
� e.g., migrating, disappearing, merging, & splitting plumes
� Identifying multiple networks
� Connecting them
� Routing algorithms that are not tied to the cluster tree
� Dynamic key distribution with & across VSNs
32MSc Final Exam - 7/21/2008
![Page 33: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/33.jpg)
Summary
� A mechanism to form VSNs� Connect nodes observing the same phenomenon
� Inter-VSN & intra-VSN communication
� A cluster & cluster tree formation algorithm� Hop-ahead Hierarchical Clustering (HHC)� Hop-ahead Hierarchical Clustering (HHC)
� More uniform & circular clusters
� Cluster tree with lower depth
� Properties are comparable with hexagonal packing
� Cluster tree based routing
� Hierarchical addressing scheme
� Cross-links & circular path based routing schemes increase network lifetime at least by a factor of 2
33MSc Final Exam - 7/21/2008
![Page 34: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/34.jpg)
Related publications
� Conference Papers� H. M. N. D. Bandara and A. P. Jayasumana, “An enhanced top-down cluster and
cluster tree formation algorithm for wireless sensor networks”, 2nd International
Conference on Industrial and Information Systems (ICIIS 2007), Sri Lanka, Aug.
2007.
� H. M. N. D. Bandara, A. P. Jayasumana, and I. Ray, “Key pre-distribution based
secure backbone design for wireless sensor networks”, 3rd Internationalsecure backbone design for wireless sensor networks”, 3 International
Workshop on Practical Issues in Building Sensor Network Applications
(SenseApp 08), Oct. 2008, to be published.
� Posters� H. M. N. D. Bandara, A. P. Jayasumana, T. H. Illangasekare, and Qi Han, “A
wireless sensor network based system for underground chemical plume
tracking,” CSU Ventures, Fort Collins, CO, Feb. 2008.
� First place - ISTeC Student Research Poster Contest
34MSc Final Exam - 7/21/2008
![Page 35: Top-Down Clustering Based Self-Organization of ...Self-Organization of Collaborative Wireless Sensor Networks M.Sc Final Exam Dilum Bandara Department of Electrical and Computer Engineering](https://reader030.fdocuments.net/reader030/viewer/2022040600/5e891bbc63760710b93420de/html5/thumbnails/35.jpg)
Questions ?
Thank You…Thank You…