Upload
vantu
View
217
Download
2
Embed Size (px)
Citation preview
Wireless Sensor Networks for SmartMetering
Andreas Weigel, Christian Renner, Volker Turau, Holger Ernst
EnergyCon 2014 - IEEE International Energy ConferenceDubrovnik, Crotia, 13-16 May, 2014
May 14th, 2014
TUHHTUHHInstitute of TelematicsInstitute of TelematicsHamburg University of TechnologyHamburg University of Technology
IntroductionIntroduction
Introduction� What? Smart Metering / Advanced Metering Infrastructure.� How? Wireless sensor network.
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 11
IntroductionIntroduction
Introduction� What? Smart Metering / Advanced Metering Infrastructure.� How? Wireless sensor network.
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 11
IntroductionIntroduction
Alternatives
� PLC: straightforward, viable option around electricity meters⇒ not feasible in some regions
� Customer DSL/Broadband: fast, reliable, high bandwidth⇒ legal implications; may necesssitate additional WiFi
� Cellular: Simple deployment, no additional infrastructure⇒ additional communication cost, communication problems in
cellars� Wireless sensor networks
⇒ license free bands, no communication cost⇒ low cost, low power consumption; but also: lossy
environment, low bandwidth
⇒ attractive alternative
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 22
IntroductionIntroduction
Alternatives
� PLC: straightforward, viable option around electricity meters⇒ not feasible in some regions
� Customer DSL/Broadband: fast, reliable, high bandwidth⇒ legal implications; may necesssitate additional WiFi
� Cellular: Simple deployment, no additional infrastructure⇒ additional communication cost, communication problems in
cellars� Wireless sensor networks
⇒ license free bands, no communication cost⇒ low cost, low power consumption; but also: lossy
environment, low bandwidth⇒ attractive alternative
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 22
RequirementsRequirements
Use Cases / Latency Requirements
Meter → DC� Collect energy consumption, load profiles: 1 h to 1 day
DC → Meter� Regular distribution of price tables: 1 h (individual) 24 h
(whole network)� Ad-hoc commands (e.g., (dis)connection): 10 min to 1 h
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 33
Protocol StackProtocol Stack
Software Architecture
NwkLayer(Forwarding, Association) LQE
OMNeT++MiXiM
LBT-basedlink layer
Dispatch
MacAbstractionLayer
NetworkManager
Otap
Queue
ReliableBlockTrRbtSenderRbtSenderRbtSenderRbtSender
RbtReceiverRbtReceiverRbtReceiver
ReliableBlockTrRbtSenderRbtSenderRbtSenderRbtSender
RbtReceiverRbtReceiverRbtReceiver
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 44
Protocol StackProtocol Stack
Network Layer
Different routing for different directions of communication:
Meters → DC (many-to-one)� Tree constructed from periodic beacons� Expected number of transmissions metric to determine next
hop
DC → Meters (one-to-many)� Restricted re-broadcasting based on subtree tables
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 55
EvaluationEvaluation
Experiment Setup
� Hardware: LPC1763 (CortexM3) + CC1120 transceiver� Experiment:
� 1 DC, 63 meters, 3 floors of large office building� 5 blocks of 1 kB, 50 blocks of 75 B, both directions� 5 runs per setup
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 66
EvaluationEvaluation
Evaluation - Collection and Distribution
0 1 2 3 4 5 60
6
12
18
24
30
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
many-to-one: 5× 1000byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
6
12
18
24
30
Avg. number of hopsA
vg.
Late
ncy
perB
lock
[s]
one-to-many: 5× 1000byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
1
2
3
4
5
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
many-to-one: 50× 75byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
1
2
3
4
5
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
one-to-many: 50× 75byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 77
EvaluationEvaluation
Evaluation - Collection and Distribution
0 1 2 3 4 5 60
6
12
18
24
30
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
many-to-one: 5× 1000byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
6
12
18
24
30
Avg. number of hopsA
vg.
Late
ncy
perB
lock
[s]
one-to-many: 5× 1000byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
1
2
3
4
5
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
many-to-one: 50× 75byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
1
2
3
4
5
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
one-to-many: 50× 75byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
⇒ Latency Requirements fulfilled
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 77
ConclusionConclusion
Conclusion
� Latency requirements could be fulfilled
� Wireless sensor networks: possible solution for AMI
� Unacknowledged broadcasts cause performance problems
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 88
Wireless Sensor Networks for SmartMetering
Andreas Weigel, Christian Renner, Volker Turau, Holger Ernst
EnergyCon 2014 - IEEE International Energy ConferenceDubrovnik, Crotia, 13-16 May, 2014
May 14th, 2014
Andreas WeigelResearch Assistant
Phone +49 / (0)40 428 78 3746
e-Mail [email protected]
http://www.ti5.tuhh.de/staff/weigel
TUHHTUHHInstitute of TelematicsInstitute of TelematicsHamburg University of TechnologyHamburg University of Technology
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 99
Association Results
� Duration of association for different groups of nodes:� All 64 nodes� 7 co-located nodes� 6 spatially scattered nodes
0 25 50 75 100 125 150 175 200 225 250 275 300 3250.0
0.2
0.4
0.6
0.8
1.0
Duration of association (s)
Port
ion
ofas
soci
ated
node
s
scattered groupco-located groupwhole network
Andreas Weigel Wireless Sensor Networks for Smart MeteringAndreas Weigel Wireless Sensor Networks for Smart Metering 1010
Evaluation - Collection and Distribution (1)
0 1 2 3 4 5 60
6
12
18
24
30
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
many-to-one: 5× 1000byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
6
12
18
24
30
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
one-to-many: 5× 1000byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
1
2
3
4
5
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
many-to-one: 50× 75byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
1
2
3
4
5
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
one-to-many: 50× 75byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
Evaluation - Collection and Distribution (2)
0 1 2 3 4 5 60
6
12
18
24
30
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
many-to-one: 5× 1000byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
6
12
18
24
30
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
one-to-many: 5× 1000byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
1
2
3
4
5
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
many-to-one: 50× 75byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io
0 1 2 3 4 5 60
1
2
3
4
5
Avg. number of hops
Avg
.La
tenc
ype
rBlo
ck[s
]
one-to-many: 50× 75byte
0.0
0.2
0.4
0.6
0.8
1.0
Blo
ckR
ecep
tion
Rat
io