1

Interactive WiFi Connectivity For Moving Vehicles

Interactive WiFi Connectivity For Moving Vehicles

Presented by Zhou YingguiPresented by Zhou Yinggui

2

Cellular networks tend to be expensiveCellular networks tend to be expensive

1

2

Background

the ubiquity of WiFithe ubiquity of WiFi

3

Background

Can WiFi deployments support common applications from moving vehicles

Can WiFi deployments support common applications from moving vehicles ??

Vi-FiVi-Fi

4

IntroductionIntroduction1

Outlines

Experimental platformExperimental platform2

Handoff strategiesHandoff strategies3

ViFi design & implementationViFi design & implementation4

EvaluationEvaluation5

ConclusionConclusion6

5

Introduction

ViFi is a protocol that minimizes disruptions to support interactive applications.

Hard Handoff --Clients communicate with only one basestation at any given time.

6

Hard handoffs are limited by

1.gray periods --- connectivity drops

sharply and unpredictably;

2.the difficulty of estimating the

continuously changing channel quality.

Hard handoff methods has frequent disruptions even when clients are close to WiFi basestations.

Introduction

7

Diversity--using multiple basestations simultaneously.

ViFi exploits diversity and opportunistic receptions to minimize disruptions.

Introduction

8

Experiment

Experiments are performed on two vehicular mobile network testbeds: VanLAN and DieselNet.

DieselNet--Downtown of Amherst, Massachusetts

trace-driven study

VanLAN--Microsoft campus in Washington.

live experiments.

9

VanLAN

•Eleven basestations (BSes)

•Two vehicles equipped with Atheros 5213 chipset, omnidirectional antennae ,GPS unit .

•The box covers 828×559 m2 area.in which at least one packet is received by vehicles from any BS.

•Not all pairs of BSes are within wireless range of one another.

•The vehicles shuttle around the campus ten times a day, with a speed of 40 Km/h.

Experiment

10

• The equipment of vehicles are the same with VanLAN• log all beacons heard from nearby BSes• traces from Channels 1 and 6• log more than 100,000 beacons in 3 days• Only analysis to BSes in the core of the town and to BSes that are visible on all three days

DieselNetExperiment

11

Handoff strategies

Using a trace-driven evaluation on VanLANEach BS and vehicle broadcasts a 500-byte packet at 1Mbps every 100 ms.

criterions:•aggregate performance--total number of packets delivered and the total time or distance…

•periods of uninterrupted connectivity--contiguous time intervals when the performance of an application is above a threshold.

12

hard handoff

1. RSSI:client associates to BSes with higher signal strength.

2. BRR:client associates to the BS with the highest strength.3. Sticky:client does not disassociate from the current

BS until connectivity is absent.4. History:client associates to the BS that has historically

provided the best average performance.5. BestBS:not practical, represents an upper bound of

hard handoff.

AllBSes:client opportunistically uses all BSes.Ideal method that represents an upper bound of any handoff protocol

diversity methods

Handoff strategies

13

Handoff strategiesAggregate Performance

pick BRR as representative

Average number of packets delivered per day in Van-LAN by various methods. Average number of packets delivered per day in Van-LAN by various methods.

14

adequate (uninterrupted) connectivity: at least 50% of the packets are receivedin a one-second interval.

black dot (gray periods) ： areas that connectivity drops sharply and unpredictably

Handoff strategiesUninterrupted Connectivity

go to 34go to 34

15

•connectivity is often undermined by gray periods even close to BSes.

•gray periods tend to be short-lived, and do not severely impact aggregate performance.

Handoff strategiesUninterrupted Connectivity

16

other definitions of adequate connectivity

Handoff strategiesUninterrupted Connectivity

17

The insight of losses

•upstream direction

losses are roughly independent across BSes and a packet sent by the vehicle is received by at least one BS with a high probability.

the conclusion has been shown previously by S. Biswas and R. Morris. ExOR: opportunistic multi-hop routing for wireless networks. In SIGCOMM, Aug. 2005.

Handoff strategies

18

•downstream direction

gray period:even when a vehicle is associated to a BS with a low average loss rate, it can lose many packets in a small time period,hurting interactive applications.

The insight of losses

Handoff strategies

19

most burst losses are path dependent rather than receiver dependent.

The insight of losses

Handoff strategies

•downstream direction

20

ViFi protocol

environment

•Diversity: A packet sent by a moving vehicle can often be heard by multiple BSes, and multiple BSes can often deliver packets to a moving vehicle.•Bandwidth-limited inter-BS communication

21

Motivated by AllBSes

imposes minimal additional load

does not increase per packet latency

can handle rapidlychanging sets of BSes.

ViFi protocol

22

•the vehicle designates one of the nearby BSes as the anchor(by BRR).

•The vehicle designates other nearby BSes as auxiliary.

•The vehicle embeds the identity of the current anchor and auxiliary into the beacons which are broadcasted periodically.

•The vehicle also embeds the identity of the previous anchor for salvaging.

Protocol overviewViFi protocol

23

1. src transmits the packet P.2. If dst receives P, it broadcasts an ACK.3. If an auxiliary overhears P, but within a small window has not heard an ACK, it probabilistically relays P.4. If dst receives relayed P and has not already sent an ACK, it broadcasts an ACK.5. If src does not receive an ACK within a retransmission interval, it retransmits P.

Protocol overviewViFi protocol

24

1. src transmits the packet P.2. If dst receives P, it broadcasts an ACK.3. If an auxiliary overhears P, but within a small window has not heard an ACK, it probabilistically relays P.4. If dst receives relayed P and has not already sent an ACK, it broadcasts an ACK.5. If src does not receive an ACK within a retransmission interval, it retransmits P.

Upstream packets are relayed on the inter-BS backplane and downstreampackets on the vehicle-BS channel

why relaying is better than a retransmission ?

Protocol overviewViFi protocol

25

Computing relaying probability

balance the trade-off between false negative(no diversity) and false positive (excessive load)

The guidelines of probability computation •G1: Account for relaying decisions made by other potentially relaying auxiliaries.•G2: Prefer auxiliaries with better connectivity to the destination.•G3: Limit the expected number of relayed transmissions.

ViFi protocol

26

The overall strategy is to compute relaying probabilities locallyand the expected number of packets is equal to 1.

11

K

i

iirc

Ci is the Bi’s probability that Bi has heard the packet but not an acknowledgment

ri is Bi’s relay times （ relay probability） which is less than one.

Computing relaying probabilityViFi protocol

27

dstBj

dstBi

p

p

r

r

j

i

,

,

Pab represents the probability that b correctly receives a transmission from a

Computing relaying probabilityViFi protocol

28

Ci = Psrc,Bi (1 − Psrc,dst Pdst,Bi )*

Ci is the Bi’s probability that Bi has heard the packet but not an acknowledgment(assumed independent)

Computing relaying probabilityViFi protocol

29

•Opportunistic receptions provide a low-overhead but unreliable means.

•With probabilistic relaying, each BS relays based on an independently computed relaying probability, which avoids the need for explicit coordination messages between BSes.

•The resulting protocol is lightweight , decentralized ,simple and works well.

ViFi protocol

30

Salvaging

•Sometimes a vehicle moves out of range before the anchor BS can deliver packets from the Internet.

•newly designated anchors salvage packets by contacting the previous anchor over the backplane.

•the old anchor transfers any unacknowledged packets within a certain time threshold(based on the minimum TCP retransmission timeout,3s).

ViFi protocol

31

current incoming reception probability

packet reception probability from them to other nodes(learn from other nodes)

Estimating packet reception probabilities using beacons

Pab:the number of beacons received in a given timeinterval divided by the number that must have been sent.

Beacons

ViFi protocol

32

close to ideal

two-fold compared to current methods

low false positive andfalse negative rates

Evaluation

link-layer

interactiveapplication

coordination mechanism

33

Link-layer performanceEvaluation

34

Link-layer performanceEvaluation

back to 14back to 14

35

most of ViFi’s gain is a result from diversity. Given thatonly 1.2% of the packets are salvaged, this benefit ofsalvaging is disproportionate（ 10% ） .

application performanceEvaluation

TCP

36

VoIP

VoIP is more challenging than TCP because quality is sensitive to both loss and delay.

The standard for evaluating a voice call is the Mean Opinion Score (MoS)

application performanceEvaluation

37

interruption:the MoS value drops below 2 for a three-second period.

ViFi average MoS is 3.4 compared with 3.0 of BRR.

VoIPapplication performance

Evaluation

38

Effectiveness of coordinationEvaluation

39

Conclusion

1.current WiFi is unsuitable for vehicular client

2.firstly studyed basestation diversity， then designed ViFi

3.key to its effectiveness is a decentralized probabilistic algorithm

4.excellent link-layer performance

5.doubled the number of successful TCP transfers and the length of disruption-free VoIP calls

40

Appendix

•Extent of diversity

•Efficiency of medium usage

•Limitations

•Deployment