Ratul Mahajan Jitu Padhye Sharad Agarwal Brian Zill Connectivity onboard vehicles Increasingly common Provided by many public transit agencies And by corporations Riders love the facility ID: 810420
Download The PPT/PDF document "High-performance vehicular connectivity ..." 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
High-performance vehicular connectivity with opportunistic erasure coding
Ratul Mahajan
Jitu Padhye
Sharad Agarwal
Brian Zill
Slide2Connectivity on-board vehicles
Increasingly commonProvided by many public transit agenciesAnd by corporationsRiders love the facilityBoosts ridership
Slide3But performance can be poor
Expectation setting by service operators:“there can be lapses in the backhaul coverage or system congestion” “
cancel a failed download and re-try in approximately 5 minutes”
Slide4Vehicular connectivity uses WWAN links
WiFi
WWAN link (e.g., 3G, EVDO,
WiMax
VanProxy
Slide5Vehicular WWAN connectivity is lossy
Slide6Methods to mask losses
Retransmissions (ARQ)unsuitable for high delay paths
Erasure codingexisting methods are capacity-oblivious
P1
P2
Sender
Receiver
P1+P2
P1
P1+P2
Slide7Opportunistic erasure coding (OEC): A new erasure coding method
Use all spare capacity for redundancyChallenge: highly
bursty traffic
Slide8OEC: Transmission strategy
Send erasure coded packets iff the bottleneck queue is emptyData packets are sent right awayProperties:Dynamically adjusts coding redundancy to match “instantaneous” spare capacity
Delays data packets by at most one packet
Slide9OEC: Encoding strategy
Conventional codes are not appropriateNeed redundancy level to be known in advance
Greedy encoding: each coded packet maximizes the amount of new information at the receiverXOR
of packets
P1
Sndr
Rcvr
P1
P2
P2
P10
P10
P1+……+P10
. . . . . . .
. . . . . .
Slide10OEC: Encoding strategy
Conventional codes are not appropriateNeed redundancy level to be known in advance
Greedy encoding: each coded packet maximizes the amount of new information at the receiverXOR
of packets
P1
Sndr
Rcvr
P1
P2
P10
P10
P4
. . . . . . .
. .
.
Slide11OEC properties
Greedily maximizes goodput with each packet transmission (coded or data)Retains this property even when traffic is striped across multiple pathsCombine with delay-based path selection
Slide12PluriBus: OEC for moving vehicles
OEC
OEC needs
Fraction of received packets
Queue length
Least-delay path
PluriBus
estimates
Path loss rate
Path capacity
Propagation
delay diff.
VanProxy
L
anProxy
Slide13On aggressive use of spare capacity
Paths are not busy all the time in practiceWWAN charges are likely a small fraction of operating cost for transit operatorsMedia access protocol isolates users from each other
Slide14Evaluation
Deployment on two buses plying on MS campusTwo WWAN links on each: EVDO and WiMaxReal conditionsTrace-driven workloadEmulationRepeatability and controlled conditions
Allows consideration of different environments
Slide15PluriBus improves performance by 4x
[Results based on deployment]
Slide16PluriBus improves performance even
when load increases multifold
[Results based on deployment]
Slide17PluriBus outperforms other loss recovery methods
[Results based on emulation]
Slide18Other results in the paper
Loss rate estimation error is lowThe impact of any inaccuracy on OEC is minimalPath delay estimation error is lowImportant to account for queue build upFraction of coded packets reduces with load
67% 35% when load is increased 8x
Slide19Summary
OEC is a new erasure coding method to mask losses while using all spare capacityOpportunistic transmissionsGreedy encodingIts application to the vehicular context reduces connection completion time by 4x
Slide20Loss recovery performance under realistic conditions
Slide21Path capacity of WiMax
Downlink
Uplink
Slide22OEC: Encoding strategy
Conventional codes are not appropriateNeed redundancy level to be known in advance
Greedy encoding: each coded packet maximizes the amount of new information at the receiverXOR
pkts; = fraction
of data pkts at rcvr
P1
Sndr
Rcvr
P1
P2
P2
P10
P10
P1+……+P10
. . . . . . .
. . . . . .
P1
P1
P2
P10
P10
P4
. . . . . . .
. .
.