Uichin Lee, Joon-Sang Park
Eyal Amir, Mario Gerla
Network Research Lab,
Computer Science Dept., UCLA



Emerging VANET applications

Safety driving (e.g., TrafficView)

Content distribution (e.g.,
CarTorrent/AdTorrent)

Vehicular sensors (e.g., MobEyes)
What about commerce “on wheels”?



Examples

A mobile user wants to sell “iPod Mini, 4G”

A road side store wants to advertise a special offer
How to form a “virtual” market place using wireless communications among mobile users as well as pedestrians (including roadside stores)?






FleaNet architecture
FleaNet protocol design
Feasibility analysis
Simulation
Conclusions



-- System Components
Vehicle-to-vehicle communications
Vehicle-to-infrastructure ( ad-station ) communications
Vehicle-to-adstation communications
Inter-vehicle communications
Private Adstation
* Roadside stores (e.g., a gas station)

-- Query Formats and Management


Users express their interests using formatted queries
 eBay-like category is provided

E.g., Consumer Electronics/Mp3 Player/Apple iPod
Query management

Query storage using a light weight DB (e.g., Berkeley
DB )

Spatial/temporal queries

Process an incoming query to find matched queries
(i.e., exact or approximate match)

E.g. Query(buy an iPod)  Query(sell an iPod)




FleaNet building blocks

Query dissemination

Distributed query processing

Transaction notification

Seller and buyer are notified

This requires routing in the VANET
VANET challenges

Large scale, dense, and highly mobile
Goal: designing “efficient, scalable, and non-interfering protocols” for VANETs



Query dissemination exploiting vehicle mobility
Query “originator” periodically advertises its query to 1-hop neighbors

Vehicles “carry” received queries w/o further relaying
Yellow Car w/ Q
1
Q
2
Q
1
Q
2
Red Car w/ Q
2
Q
1


Received query is processed to find a match of interests

Eg. Q
1
– buy iPod / Q
M
– sell iPod / Q
2
– buy Car
Q
M
Q
2
(1) Find a matching query for Q
2
 No match found
Q M Cyan car w/ Q
M
(2) Send a match notification msg to the originator of query Q
M
Q2
Q
M
 Q
1
(1) Find a matching query for Q
M
 Found query Q
1
Q
M
Q
1 Red car w/ Q
2
& carries Q
1


After seeing a match, use Last Encounter
Routing (LER) to notify seller/buyer

Forward a packet to the node with more “recent” encounter
Q
M
LocalMatch
Q
M
 Q
1
Cyan car Green car
Q
1
T-1s
Yellow car
RESP
TRX
REQ
Blue car
Q
1
T-5s
Red car
Q
1
T
Originator of Q
1
Q
1
T-10s Current Time: T
Encounter timestamp




Restricted mobility patterns are harmful to opportunistic data dissemination
However, latency can be greatly improved by the popularity of queries
Popularity distribution of 16,862 posting (make+model) in the vehicle ad section of Craigslist (Mar. 2006)
Items (log)





Assume that only the query originator can
“periodically” advertise a query to its neighbors
We are interested in link load
Load depends only on average number of neighbors and advertisement period (not on network size)
Example:

Parameter setting : R=250m, 1500B packet size,
BW=11Mbps



N=1,000 nodes in 2,400m x 2,400m (i.e., 90 nodes within one’s communication range)
Advertisement period: 2 seconds
Worst case link utilization: < 4%






Ns-2 network simulator
802.11b - 2Mbps, 250M radio range
Two-ray ground reflection model
“Track” mobility model

Vehicles move in the
2400mx2400m Westwood area in the vicinity of the UCLA campus
Metric

Average latency: time to find a matched query of interest
Westwood area, 2400mx2400m


Impact of density and speed
450
400
350
300
250
200
150
100
50
0
5 10 15 20
Average Speed (m/s)
N=100
N=200
N=300
25


Impact of query popularity


Popularity: the fraction of users with the same interest
For a single buyer, increase the number of sellers (e.g.,
N=200/0.1 = 20 sellers)
40
30
20
10
0
70
60
50
N=100/V=5
N=100/V=25
N=300/V=5
N=300/V=25
0.05
0.1
0.15
Popularity
0.2
0.25


Impact of ad-station location


Given N=100, fix each node in its initial location, and set it as a
“stationary” ad-station (as a buyer) measure the average latency to the remaining 99 mobile nodes
(run 99 times, by taking turns as a seller: 1 buyer  1 seller)
500
450 N=100/V=25m/s
400
350
300
250
200
150
100
50
0
1 11 21 31 41 51 61 71 81 91 avg. stationary avg. mobile




Proposed a virtual market concept in VANETs:

A mix of mobile and stationary users carry out buy/sell transactions (or any other matching of common interests) using vehicular networks
Mobility-assisted query dissemination and resolution
(scalable and non-interfering)



Node density/speed are closely related to the performance
Popularity of a query greatly improves the performance
Location of an ad-station is important to the performance
Future work



Query aggregation to improve the performance

Unpopular queries/queries from ad-stations
How to enforce cooperativeness of users?
Security: false query injection and spamming?