on multicore wireless sensor networks Nicoletta Triolo Francesco Baldini Susanna Pelagatti Stefano Chessa University of Pisa Italy Background wireless sensor networks Sink Internet ID: 797315
Download The PPT/PDF document "Structured parallel programming" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Structured parallel programming on multi-core wireless sensor networks
Nicoletta Triolo, Francesco Baldini, Susanna Pelagatti, Stefano Chessa University of Pisa, Italy
Slide2Background: wireless sensor networks
Sink
Internet,
Satellite
Networks,
etc
..
User
Slide3Wireless sensor networks (WSN)
Main tasks of a sensor:Sense(pre)-process
Communicate
Models
of
computation
Centralized (at the sink
, a sensor can make some pre-processing)Distributed
Slide4Wireless sensor network platforms
platforms for conventional (single-core) sensors:8 bits/8MHz microcontrollers (MicaZ, T-Mote)
128 KB flash memory
8 KB RAM
IEEE 802.15.4
TinyOS
+ NesC
Slide5Trends in WSN
Architectural level:Application level:
Example of Application Domains: Multimedia WSN and VSN
Slide6Trends in WSNParallel architectures of the single sensor
nodeDistributed computations throughout the WSNNeed for high-level abstractions to support
Parallel
Distributed programming
Slide7Example of multi-core WSN node
Raspberry PI 2 Model BARM Cortex-A7 CPU900 MHz quad-core1GB RAMLinux-Like + C/C++/Java + Wi-Fi IEEE 802.11
Slide8Wireless/Visual Sensor Networks (W/VSN)
Number of networked devices Each device:Microsystem with processor/memoryCameraWireless/wired network interfaceConstrained resourcesProcessing, communicationsEnergy Often used for tracking applications
Mix of video processing, distributed communications
Slide9Tracking with W/VSN
A number of cameras cooperatively track a mobile target
Detection when a target is in the Field of View (
FoV
) of a camera
Each camera computes location information of the target
All location information from different cameras are fused together
Improve localization accuracyAlert other cameras in advance
Slide10Example of a tracking application (I)
A generic node ci runs an infinite
loop
.
In a
generic iteration
k:1. Acquisition phase:Acquires an image from own camera: s
k
Slide11Example of a tracking application (
II)2. Exchange phase:ci receives the output mi
k
from
its logical
neighbors in n(ci )where each Nj in n(c
i) shares (part) of the FOV with ci
Slide12Example of an application: tracking (III)
3. Computation phase: xk+1 = f (x
k
,
m
1
k, m2k , … , mi
k, sk)f is the aggregation function
xk+1 is the output of tracking (estimated position of the target) at step k+1mik is the output of
neighbor Ni at step ksk is the local image acquired at
step k
Slide13Example of an application: tracking (IV)
4. Transmission phase:Broadcasts xk+1 to its
logical
neighbors
Slide14Iterative Neighbor Stencil (INS) skeleton
Stencil-like computationcomputation on matrix data structureFits common patterns of tracking apps in W/VSN
Local image acquisition
Local processing
Exchange processed data with physical/logical neighbors (cameras with intersection FOVs)
Slide15Skeletons: programming abstractions
efficient, portable, reusable and parametric
Slide16Modeling tracking applications with INSReal time execution
τ : max. latency of each roundAt next round data of this round are outdatedρ : max. fraction of packets lost in each round per nodePacket loss affects the quality of trackingNon-functional requirementMaximize network lifetime by reducing cameras duty cycle.
Slide17Modeling tracking applications with INS
Implications on the underlying MAC layerDetermine a communication patternAffects packet loss and latencyAffects energy consumptionTwo MAC protocols: MACAW and T-MACTwo extremes: MACAW keeps radio always onT-MAC schedules off-periods for the radio Tuning of MACAW and T-MAC for INS
Slide18T-MAC and MACAW parameters
T-MAC Preamble sampling basedFs: frame sizeTa: length of
active
time
Vl
:
contention intervalMACAWCSMA/CA, exponential backoff
, radio always onInitial backoff length
Slide19SimulationsCastalia simulatorCC2420 wireless radio (IEEE 802.15.4)
4,7,10 nodes in a single hop network100 iterations of the INS skeletonRound of 0.5 sec.Camera processing time of 30 msec.
τ=470 msec. (max communication Latency)
ρ=0.1
(max packet loss)
Slide20Results
T-MACFrame size (Fs)vs round latency (τ
)
Slide21Results
T-MACFrame size (Fs)vspacket received in time
(r
)
Slide22Results
T-MACpacket received in time (r)vsContention Interval (
Vl
)
Slide23Results
T-MACEnergy consumption for communications *vsTimeout (Ta)
*
of the camera
that
spends more
Slide24Results
T-MACEnergy consumption for communicationsvsFrame size (Fs
)
Active time
Ta
=10ms
Slide25Results
Energy consumption with T-MAC and MACAWT-MAC configured according to the previous experiments
Slide26ConclusionsINS Skeleton
fits well processing & communication patterns of tracking applications of W/VSN Knowledge of
communication
pattern
allows
for fine
configuration of MAC paramsetersto achieve energy efficiency
to meet requirements on latency and packet loss
Slide27Future worksAnalyse
the tracking accuracy w.r.t. energy behaviour of INSAnalyse the
behaviour
of INS in
multihop
W/VSN
cameras with intersecting FOV may be far in the communication
topologyExtend this study to other patterns (skeletons
) for WSN