on modeling low-power wireless protocols based on synchronous packet transmissions
DESCRIPTION
On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions . Marco Zimmerling , Federico Ferrari, Luca Mottola * , Lothar Thiele ETH Zurich * Politecnico di Milano and SICS Swedish ICT. TexPoint fonts used in EMF. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/1.jpg)
On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions Marco Zimmerling, Federico Ferrari, Luca Mottola*, Lothar Thiele
ETH Zurich*Politecnico di Milano and SICS Swedish ICT
![Page 2: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/2.jpg)
Accurate mathematical models of low-power wireless protocols are important
» Understanding (e.g., trade-offs, parameters) » Verification» System design (e.g., node placement, power
sources)» Prediction and adaptation at runtime
But traditional link-based multi-hop protocols are intricate and difficult to model (e.g., ZigBee)
Motivation
![Page 3: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/3.jpg)
Highly dynamic network topology due to:» Volatile low-power wireless links» Node mobility, failures
Traditional protocols maintain substantial network state that governs their operation (e.g., link qualities)
» Distributed across nodes» Concurrent, uncontrolled updates
Motivation
![Page 4: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/4.jpg)
Highly dynamic network topology due to:» Volatile low-power wireless links» Node mobility, failures
Traditional protocols maintain substantial network state that governs their operation (e.g., link qualities)
» Distributed across nodes» Concurrent, uncontrolled updates
Motivation
Modeling often stops at the link layer, where interactions are single-hop
insufficient!
![Page 5: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/5.jpg)
Example: End-to-End Delivery
Devise a model of the packet deliveryrate (PDR) observed at the sink
PDR = f()
![Page 6: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/6.jpg)
Example: End-to-End Delivery
A
B
A:B:
= network state
Media Access
Radio Driver
wake-uptimes
timePDR = f( , …)wake-up
times
one-hop
![Page 7: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/7.jpg)
Example: End-to-End Delivery
A
B
= network state
Tree Routing
Media Access
Radio Driver
linkqualitieswake-up
times0.9
0.7
PDR = f( , , …)wake-uptimes
linkqualities
one-hop
one-hop/end-to-end
![Page 8: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/8.jpg)
Example: End-to-End Delivery
A
B
= network state
Tree Routing
Media Access
Radio Driver
linkqualitieswake-up
times
PDR = f( , , …)wake-uptimes
linkqualities
one-hop
one-hop/end-to-end
![Page 9: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/9.jpg)
Example: End-to-End Delivery
A
B
= network state
Application
Reliable Transport
Tree Routing
Media Access
Radio Driver
queuesizeslink
qualitieswake-up
times
PDR = f( , , , …)wake-uptimes
linkqualities
queuesizes
one-hop
end-to-end
one-hop/end-to-end
![Page 10: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/10.jpg)
Example: End-to-End Delivery
A
B
= network state
Application
Reliable Transport
Tree Routing
Media Access
Radio Driver
queuesizeslink
qualitieswake-up
times
PDR = f( , , , …)wake-uptimes
linkqualities
queuesizes
one-hop
end-to-end
one-hop/end-to-end
Continuouslychanging
![Page 11: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/11.jpg)
Example: End-to-End Delivery
A
B
= network state
Application
Reliable Transport
Tree Routing
Media Access
Radio Driver
queuesizeslink
qualitieswake-up
times0.8
0.95
PDR = f( , , , …)wake-uptimes
linkqualities
queuesizes
one-hop
end-to-end
one-hop/end-to-end
Continuouslychanging
![Page 12: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/12.jpg)
Example: End-to-End Delivery
A
B
= network state
one-hop
end-to-end
one-hop/end-to-end
Application
Reliable Transport
Tree Routing
Media Access
Radio Driver
queuesizeslink
qualitieswake-up
times0.8
0.95
Protocols abstract the unreliable wireless channel as point-to-point links major
reason for complexity
Continuouslychanging
![Page 13: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/13.jpg)
Multiple nodes send simultaneously to the same receiver, as opposed to pairwise link-based transmissions (LT)
Synchronous Transmissions (ST)
RR
![Page 14: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/14.jpg)
Enabled by IEEE 802.15.4 physical-layer phenomena:
» Power (and delay) capture different/identical packets
» Constructive baseband interference identical packets
Several recent protocols exploit ST for various purposes:
» A-MAC [SenSys 10]: contention resolution» Glossy [IPSN 11]: network flooding » Splash [NSDI 13]: code updates» CAOS [SenSys 13]: all-to-all communication
Synchronous Transmissions (ST)
![Page 15: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/15.jpg)
Enabled by IEEE 802.15.4 physical-layer phenomena:
» Power (and delay) capture different/identical packets
» Constructive baseband interference identical packets
Several recent protocols exploit ST for various purposes:
» A-MAC [SenSys 10]: contention resolution» Glossy [IPSN 11]: network flooding » Splash [NSDI 13]: code updates» CAOS [SenSys 13]: all-to-all communication
Synchronous Transmissions (ST)
ST enable efficient and reliable multi-hop protocols with very little network
state
![Page 16: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/16.jpg)
Open Question
Do ST also enable simple yet accurate modeling?
![Page 17: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/17.jpg)
Modeling Trade-Off Space
Accuracyof Models
Modeling Scope
Tree Routing
Radio Driver
High
Low
Low High
![Page 18: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/18.jpg)
Modeling Trade-Off Space
Park et al.[IPSN 10]
Buettner et al.[SenSys 06]
Zimmerling et al.[IPSN 12]
Langendoen & Meier[TOSN 10]
Polastre et al.[SenSys 04]
Challen et al.[MobiSys 10]
Tree Routing
Radio Driver
2-7 % error
Accuracyof Models
Modeling Scope
High
Low
Low High
Linklayer
![Page 19: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/19.jpg)
Modeling Trade-Off Space
Park et al.[IPSN 10]
Buettner et al.[SenSys 06]
Zimmerling et al.[IPSN 12]
Langendoen & Meier[TOSN 10]
Polastre et al.[SenSys 04]
Challen et al.[MobiSys 10] Bruneo et al.
[PE 12] Gelenbe et al.[TOSN 07]Guenther et al.[UKPEW 12]
Tree Routing
Radio Driver
2-7 % error
Accuracyof Models
Modeling Scope
High
Low
Low High
Linklayer
Fullstack
![Page 20: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/20.jpg)
Modeling Trade-Off Space
Park et al.[IPSN 10]
Buettner et al.[SenSys 06]
Zimmerling et al.[IPSN 12]
Langendoen & Meier[TOSN 10]
Polastre et al.[SenSys 04]
Challen et al.[MobiSys 10]
Our work
Tree Routing
Radio Driver
2-7 % error 0.25 % error
Bruneo et al.[PE 12] Gelenbe et al.
[TOSN 07]Guenther et al.[UKPEW 12]
Accuracyof Models
Modeling Scope
High
Low
Low High
Linklayer
Fullstack
![Page 21: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/21.jpg)
Validity of the Bernoulli assumption to ST vs LTModeling an ST-based multi-hop protocolModel validation through real-world experiments
Outline
![Page 22: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/22.jpg)
Validity of the Bernoulli assumption to ST vs LTModeling an ST-based multi-hop protocolModel validation through real-world experiments
Outline
![Page 23: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/23.jpg)
Intended receiver of a sequence of packets observes a sequence of i.i.d. Bernoulli trials
» Success (packet received) with probability p» Failure (packet lost) with probability 1 – p
Empirical studies showed that this assumption is not always valid to LT (esp. when packets are close in time)
Bernoulli Assumption
S R
S
SReceptions/losses areindependent over time
![Page 24: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/24.jpg)
Intended receiver of a sequence of packets observes a sequence of i.i.d. Bernoulli trials
» Success (packet received) with probability p» Failure (packet lost) with probability 1 – p
Empirical studies have shown that this assumption is not always valid to LT
Bernoulli Assumption
S R
S
S
What about ST?
Receptions/losses areindependent over time
![Page 25: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/25.jpg)
Glossy [IPSN 11] leverages ST for efficient and reliable network flooding
Methodology
![Page 26: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/26.jpg)
Glossy [IPSN 11] leverages ST for efficient and reliable network flooding
Methodology
![Page 27: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/27.jpg)
Glossy [IPSN 11] leverages ST for efficient and reliable network flooding
Methodology
![Page 28: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/28.jpg)
Glossy [IPSN 11] leverages ST for efficient and reliable network flooding
Methodology
![Page 29: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/29.jpg)
Glossy [IPSN 11] leverages ST for efficient and reliable network flooding
Methodology
![Page 30: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/30.jpg)
Methodology139-node testbed at NUS
20-byte packets at 20 ms IPI
![Page 31: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/31.jpg)
Methodology139-node testbed at NUS
20-byte packets at 20 ms IPI
ST-Type: how protocols perceive Glossy» 70 equally distributed nodes» 50,000 Glossy floods each» Remaining 138 nodes log
![Page 32: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/32.jpg)
Methodology139-node testbed at NUS
20-byte packets at 20 ms IPI
ST-Type: how protocols perceive Glossy» 70 equally distributed nodes» 50,000 Glossy floods each» Remaining 138 nodes log
LT-Type: how protocols perceive LT» All 139 nodes» 50,000 broadcasts each» Nodes within radio range log
![Page 33: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/33.jpg)
Methodology139-node testbed at NUS
20-byte packets at 20 ms IPI
ST-Type: how protocols perceive Glossy» 70 equally distributed nodes» 50,000 Glossy floods each» Remaining 138 nodes log
LT-Type: how protocols perceive LT» All 139 nodes» 50,000 broadcasts each» Nodes within radio range log
Tx power: 0 dBm (max) and -15 dBm (lowest possible)
![Page 34: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/34.jpg)
Methodology139-node testbed at NUS
20-byte packets at 20 ms IPI
ST-Type: how protocols perceive Glossy» 70 equally distributed nodes» 50,000 Glossy floods each» Remaining 138 nodes log
LT-Type: how protocols perceive LT» All 139 nodes» 50,000 broadcasts each» Nodes within radio range log
Tx power: 0 dBm (max) and -15 dBm (lowest possible)Collected >1,200,000,000 packet reception events
![Page 35: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/35.jpg)
1) Represent every trace as a discrete-time binary time series {xi}n of length n = 50,000:
010011100111110011101010…
2) Keep only weakly stationary time series, using two empirical tests for non-constant mean/variance
3) Check the validity of the Bernoulli assumption, using a statistical test based on the sample autocorrelation
Trace Analysis
![Page 36: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/36.jpg)
Bernoulli Results
20 200 10000
10
20
30
40
Inter-packet interval (milliseconds)
Trac
es fo
r w
hich
the
Bern
oulli
ass
umpt
ion
does
no
t ho
ld (%
)
1010111110111101100110111111101111101110110001111111001111011101011111000111110100111101011001100110
200 1000
![Page 37: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/37.jpg)
Bernoulli Results
20 200 10000
10
20
30
40ST-Type 0 dBmST-Type -15 dBmLT-Type 0 dBmLT-Type -15 dBm
Inter-packet interval (milliseconds)
Trac
es fo
r w
hich
the
Bern
oulli
ass
umpt
ion
does
no
t ho
ld (%
)
0.5 0.1 0.03
200 1000
![Page 38: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/38.jpg)
Bernoulli Results
20 200 10000
10
20
30
40ST-Type 0 dBmST-Type -15 dBmLT-Type 0 dBmLT-Type -15 dBm
Inter-packet interval (milliseconds)
Trac
es fo
r w
hich
the
Bern
oulli
ass
umpt
ion
does
no
t ho
ld (%
)40.0
14.5
2.60.5 0.1 0.03
200 1000
![Page 39: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/39.jpg)
Bernoulli Results
20 200 10000
10
20
30
40ST-Type 0 dBmST-Type -15 dBmLT-Type 0 dBmLT-Type -15 dBm
Inter-packet interval (milliseconds)
Trac
es fo
r w
hich
the
Bern
oulli
ass
umpt
ion
does
no
t ho
ld (%
)
8.7
34.6
4.0
14.1
0.3 2.50.5
40.0
0.1
14.5
0.032.6
200 1000
![Page 40: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/40.jpg)
Bernoulli Results
20 200 10000
10
20
30
40ST-Type 0 dBmST-Type -15 dBmLT-Type 0 dBmLT-Type -15 dBm
Inter-packet interval (milliseconds)
Trac
es fo
r w
hich
the
Bern
oulli
ass
umpt
ion
does
no
t ho
ld (%
)
8.7
34.6
4.0
14.1
0.3 2.50.5
40.0
0.1
14.5
0.032.6
The Bernoulli assumption is highly valid to ST and significantly more valid
to ST than to LT
![Page 41: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/41.jpg)
Validity of the Bernoulli assumption to ST vs LTModeling an ST-based multi-hop protocolModel validation through real-world experiments
Outline
![Page 42: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/42.jpg)
Low-Power Wireless Bus (LWB)LWB [SenSys 12] uses only ST for communication
» Turns a multi-hop network into a “virtual” single-hop network, similar to a shared bus
Centralized scheduling» A controller node orchestrates all
communication
LWB
controllermulti-hop network Glossy floods shared bus
![Page 43: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/43.jpg)
Low-Power Wireless Bus (LWB)LWB [SenSys 12] uses only ST for communication
» Turns a multi-hop network into a “virtual” single-hop network, similar to a shared bus
Centralized scheduling» A controller node orchestrates all
communication
LWB
controllermulti-hop network Glossy floods shared bus
Does the validity of the Bernoulli assumption to ST help devise a simple yet accurate
energy model of LWB?
![Page 44: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/44.jpg)
A single event − the reception of schedules from the controller − drives the operation of all LWB nodes
Schedule-Driven Operation
Communication rounds
schedule data data …data
![Page 45: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/45.jpg)
Finite-State Machine
= schedule received= schedule missed
![Page 46: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/46.jpg)
Finite-State Machine
= schedule received= schedule missed
The Bernoulli assumption allows us to characterize schedule receptions through
a single parameter p
![Page 47: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/47.jpg)
Discrete-Time Markov Chain
p
1-p
pp
p
p
p p
p
p
1-p
1-p
1-p1-p
1-p
1-p1-p
1-p1-p1-p
p
p
= schedule received, with probability p= schedule missed, with probability 1-p
p
![Page 48: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/48.jpg)
0 0.2 0.4 0.6 0.8 10
0.5
1
Probability of receiving a schedule p
Stat
iona
rydi
stri
buti
on
Stationary distribution of the DTMC gives the frequency of visits to each state in the long run for a given p
Combined with the well-defined cost of each state, we get the long-term expected energy cost of a LWB node
Energy Model
![Page 49: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/49.jpg)
Validity of the Bernoulli assumption to ST vs LTModeling an ST-based multi-hop protocolModel validation through real-world experiments
Outline
![Page 50: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/50.jpg)
30-node testbed at ETH Zurich
15-byte payload at 6 sec IPI
Tx power: o dBm (max)
Estimate several model parameters at runtime, such as:
Measure energy consumption using established software-based methods
Model Validation
p = number of received schedulesnumber of expected schedules
![Page 51: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/51.jpg)
Energy Validation Results
0 5 10 15 200
50100150200250300350400450
Discarded schedule and data packets (%)
Radi
o on
-tim
e pe
r ro
und
(mse
c)
reality
![Page 52: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/52.jpg)
Energy Validation Results
0 5 10 15 200
50100150200250300350400450 Mea-
sured
Discarded schedule and data packets (%)
Radi
o on
-tim
e pe
r ro
und
(mse
c)
reality
![Page 53: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/53.jpg)
Energy Validation Results
0 5 10 15 200
50100150200250300350400450 Mea-
sured
Discarded schedule and data packets (%)
Radi
o on
-tim
e pe
r ro
und
(mse
c)average model error: 0.25%
reality artificial
![Page 54: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/54.jpg)
Modeling Trade-Off Space
Tree Routing
Radio Driver
Application
LWB
Glossy
Radio DriverPark et al.[IPSN 10]
Buettner et al.[SenSys 06]
Zimmerling et al.[IPSN 12]
Langendoen & Meier[TOSN 10]
Polastre et al.[SenSys 04]
Challen et al.[MobiSys 10]
Our work
2-7 % error 0.25 % error
Bruneo et al.[PE 12] Gelenbe et al.
[TOSN 07]Guenther et al.[UKPEW 12]
Accuracyof Models
Modeling Scope
High
Low
Low High
Linklayer
Fullstack
![Page 55: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/55.jpg)
The Bernoulli assumption is highly valid to ST but often illegitimate to LT
ST enable simple yet accurate modeling of a complete state-of-the-art low-power wireless networking stack
Conclusions
![Page 56: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/56.jpg)
0 200 400 600 800 1000Time (seconds)
0.2
0
.4
0.
6
0.8
1Pa
cket
rec
epti
on r
ate
Weakly stationaryNon-stationaryLinear fit
Formal tests (e.g., KPSS, ADF) often fail in practice
Empirically declare a trace as non-stationary if» PRR changes by 0.015 or more over the entire
trace, or» PRR drops/rises by >0.05 within 40 seconds
(2,000 packets)
Test for Weak Stationarity
![Page 57: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/57.jpg)
Type Tx power Total Non-stationary
Weakly stationary
ST-Type 0 dBm 9660 47 9613ST-Type -15 dBm 9660 256 9404LT-Type 0 dBm 4189 1418 2771LT-Type -15 dBm 1777 588 1189
Trace Statistics
![Page 58: On Modeling Low-Power Wireless Protocols Based On Synchronous Packet Transmissions](https://reader035.vdocuments.mx/reader035/viewer/2022062315/56816376550346895dd45325/html5/thumbnails/58.jpg)
0 5 10 15 20Lag
-0.0
5
0
0.0
5
0.
1Sa
mpl
e au
toco
rrel
atio
n
Trace 1Trace 2Bounds
Let {xi}n a realization of an i.i.d. sequence {Xi}∞ of random variables with finite variance
» For large n, about 95% of the sample autocorrelation values should lie within the confidence bounds ±1.96/n1/2
» We consider the Bernoulli assumption valid for a given trace if the above holds already at lag 1
Validating Bernoulli