The DYMO Routing Protocol in VANET Scenarios: Christoph Sommer and Falko Dressler
The DYMO Routing Protocol in VANET Scenarios: Christoph Sommer and Falko Dressler
The DYMO Routing Protocol in VANET Scenarios: Christoph Sommer and Falko Dressler
treiber/MicroApplet
Time (s)
S
p
e
e
d
(
m
/
s
)
1500 2000 2500 3000 3500
0
5
1
0
1
5
2
0
2
5
3
0
3
5
28 vehicles/km/lane
4.2 vehicles/km/lane
Fig. 2. Speed samples of simulated cars at different trafc densities
highly unstable. A larger number of simulated vehicles forced
cars and trucks into a stop-and-go motion, reducing the cars
top speed to that of trucks. This stabilized links between
vehicles and reduced speed differences between vehicles and
roadside infrastructure, but caused large oscillations of local
node densities.
Realistic communication patterns of MANET nodes were
modeled using OMNeT++3.2p1 [16], a simulation environ-
ment free for non-commercial use, and its INET Frame-
work 20060330, a set of simulation modules released under
the GPL. OMNeT++ runs discrete, event-based simulations of
communicating nodes on a wide variety of platforms and is
getting increasingly popular in the communications commu-
nity. Scenarios in OMNeT++ are represented by a hierarchy
of reusable modules written in C++. Their relationships and
communication links are stored as Network Description (NED)
les and can be modeled graphically. Simulations are either
run interactively in a graphical environment or executed as
command-line applications. The INET Framework provides a
set of OMNeT++ modules that represent various layers of the
Internet protocol suite, e.g. the TCP, UDP, IPv4 and ARP
protocols. It also provides modules that allow the modeling
of spatial relations of mobile nodes and IEEE 802.11 trans-
missions between them.
IV. SIMULATION MODEL
The DYMO routing protocol was implemented as an
application-layer module of the INET Framework module set.
Following the specication [2], it employs UDP to communi-
cate with other instances of DYMO. Additionally, it uses two
helper modules to support DYMO operation on the network
layer. The complete protocol stack is shown in Figure 3. The
rst helper module is able to queue outbound packets before
routing in the network layer occurs, so that a route can be
set up by DYMO. The queue can then be signaled to release
buffered packets for a given destinationeither in order to
have them routed to the rst hop or to have them discarded
by the network layer because no route could be found. The
second helper module is installed as a hooking function in
Lay
DYMO
TanspLay
App1
DaaLinLay
App2
quu
h
Fig. 3. DYMO and support modules in the protocol stack
the inbound packet path. It noties DYMO of the arrival of
packets. This way, routing table entries can be refreshed and
route errors can be sent, respectively. DYMO and its helper
modules are assembled together with various components of
the INET Framework to form simulated MANET nodes.
Mobile nodes are represented by modules of type Car,
which perform DYMO along with TCP or UDP applications
that generate application specic trafc. Communication with
other nodes takes place via an IEEE 802.11 module. The road-
side infrastructure is provided by modules of type AccessPoint,
which execute DYMO only to route between the wireless and
the wired network, i.e. the Internet. Internet connectivity is
modeled by a node of type CSTMGateway that is also running
DYMO. It sends back delayed response messages to requests
via TCP or UDP, i.e. it simulates the application servers that
are used by the clients (the Cars).
For all communications, the complete network stack, includ-
ing ARP, was used and wireless modules were congured to
closely resemble IEEE 802.11b network cards transmitting at
11 Mbit/s with RTS/CTS disabled. The TCP protocol imple-
mentation follows the TCP Reno specication. Thus, results
can be readily compared with existing Linux implementations
of DYMO, e.g. NIST DYMO or DYMOUM. For the simulation
of radio wave propagation, a plain free-space model was
employed and the transmission ranges of all nodes adjusted
to a xed value of 180 m, a trade-off between varying real-
world measurements described in related work [17], [18]. All
simulation parameters used to parameterize the modules of the
INET Framework are summarized in Table I.
In order to ensure realistic application layer trafc, the fol-
lowing three different communication scenarios were modeled:
1) Vehicles polled trafc information from an Internet host.
At 5 minute intervals, starting at a random point in time
no more than 5 minutes from the start of a simulation,
a vehicle tried to send a 256 Byte UDP packet to the
gateway, which, upon reception of the packet, answered
with a 1024Byte response packet.
2) Mobile nodes checked a POP3 mailbox (using TCP)
for new messages, congured with a maximum segment
size of 1024 Byte and an advertised window size of
14 336Byte, to send eight 16 Byte commands, each
triggering a 32 Byte response. As in the rst case, the
mailbox check was repeated 5 minutes after sending the
Fig. 4. Simulated MANET scenario
TABLE I
INET FRAMEWORK MODULE PARAMETERS
Parameter Value
TCP.mss 1024 Byte
TCP.advertisedWindow 14 336 Byte
TCP.tcpAlgorithmClass TCPReno
ARP.retryTimeout 1 s
ARP.retryCount 3
ARP.cacheTimeout 100 s
mac.address auto
mac.bitrate 11 Mbit/s
mac.broadcastBackoff 31 slots
mac.maxQueueSize 14 Pckts
mac.rtsCts false
decider.bitrate 11 Mbit/s
decider.snirThreshold 4 dB
snrEval.bitrate 11 Mbit/s
snrEval.headerLength 192 bit
snrEval.snrThresholdLevel 3 dB
snrEval.thermalNoise 110 dB
snrEval.sensitivity 85 dB
snrEval.pathLossAlpha 1.9
snrEval.carrierFrequency 2.4 GHz
snrEval.transmitterPower 2 mW
channelcontrol.carrierFrequency 2.4 GHz
channelcontrol.pMax 2 mW
channelcontrol.sat 80 dBm
channelcontrol.alpha 1.9
rst command and the maximum session length limited
accordingly.
3) Vehicles requested RSS feeds from a web server (also
using TCP). This was represented by changing the
second cases parameters, so that nodes would only send
a single, 256 Byte request message and receive a single,
65 536Byte response message, with fragmentation and
reassembly taking place in lower layers.
The modeled nodes were then further combined to create
the MANET scenario shown in Figure 4, a simulated highway
with two lanes in each direction forming a 10 km long closed
ring with evenly spaced access points at distances of 2 km,
5 km or 10 km, depending on the scenario.
0
.
0
0
.
2
0
.
4
0
.
6
0
.
8
1
.
0
Node Density (vehicles/km/lane), AP Spacing
S
u
c
c
e
s
s
R
a
t
e
AP Spacing
a: 2 km
b: 5 km
c: 10 km
0.42 2.8 4.2 28
a b c a b c a b c a b c
Fig. 5. Success rate at different distances and
node densities; Scenario: UDP
0
.
0
0
.
2
0
.
4
0
.
6
0
.
8
1
.
0
Node Density (vehicles/km/lane), Traffic Pattern
S
u
c
c
e
s
s
R
a
t
e
Scenario
a: UDP
b: POP3
c: RSS
0.42 2.8 4.2 28
a b c a b c a b c a b c
Fig. 6. Success rate in different scenarios at
different node densities; AP distance: 5 km
0
.
0
0
.
2
0
.
4
0
.
6
0
.
8
1
.
0
Node Density (vehicles/km/lane), Traffic Pattern
D
Y
M
O
M
e
d
iu
m
U
s
a
g
e
Scenario
a: UDP
b: POP3
c: RSS
0.42 2.8 4.2 28
a b c a b c a b c a b c
Fig. 7. DYMO overhead in different scenarios
at different node densities; AP distance: 5 km
0
.
0
0
.
2
0
.
4
0
.
6
0
.
8
1
.
0
Node Density (vehicles/km/lane), Active Nodes
S
u
c
c
e
s
s
R
a
t
e
Active Nodes
a: 100%
b: 10%
2.8 4.2 28
a b a b a b
Fig. 8. Success rate improvement with passive
nodes; AP distance: 5 km, Scenario: UDP
0
1
0
0
0
2
0
0
0
3
0
0
0
4
0
0
0
5
0
0
0
Hops
D
is
t
a
n
c
e
(
m
)
5 10 15 20 25 30
Fig. 9. Hop counts of established bidirectional
links vs. distance bridged
0 0.1 0.5 1 2 5
0
.
0
0
.
2
0
.
4
0
.
6
0
.
8
1
.
0
Malus (hops)
S
u
c
c
e
s
s
R
a
t
e
Fig. 10. Improvement of success rate when
applying a hop count malus
V. PERFORMANCE ANALYSIS
Perceived performance of the VANET was estimated by
recording the overall success rate, i.e. the probability of
successful reception of a UDP information packet, the last
POP3 response, or a complete RSS feed, at the requesting
vehicle depending on the trafc pattern in use. Performance
was measured at four different node densities of 0.42, 2.8,
4.2, and 28 vehicles per kilometer and lane, corresponding
to the two chosen trafc densities and fractions of 10 % and
100 % DYMO-equipped vehicles, respectively. Three different
minimum route lifetimes of 1 s, 3 s, and 10 s were tried for
each of the simulated scenarios. A value of 1 s proved to
balance the amount of route request and route error messages
in the network best. Also, setting the DYMO Network Size
parameter to 50 hops instead of to the default 10 hops proved
to be benecial when access points were spaced more than
2 km apart and node densities did not exceed 28 vehicles per
kilometer and lane.
All results are shown as boxplots. For each data set, a box
is drawn from the rst quartile to the third quartile, and the
median is marked with a thick line. Additional whiskers extend
from the edges of the box towards the minimum and maximum
of the data set, but no further than 1.5 times the interquartile
range. Data points outside the range of box and whiskers are
considered outliers and drawn separately.
Figure 5 shows the overall probability of a simulated UDP
session being successfully completed for different node densi-
ties and access point distances. As can be seen, even low node
densities of 4.2 nodes per kilometer and lane, as well as sparse
access point deployment of one node per 5 km highway, suf-
ced to permit the exchange of UDP packets in approx. 50 %
of all tries. Results for other communication scenarios are
shown in Figure 6, which plots the overall probability of a
session being successfully completed for a xed access point
distance of 5 km. Due to the retry mechanisms offered by
the TCP protocol, POP3 sessions always had a signicantly
higher chance of being completed than plain UDP sessions,
even though completion of a POP3 session required the
exchange of more packets. Also visible is a rapidly decreasing
probability of sessions being successfully completed when
node densities increased to above 4.2 nodes per kilometer and
lane or when larger messages were to be delivered. While at
0.42 nodes per kilometer and lane, the probability of RSS
sessions completing was almost at par with that of POP3
sessions, at 2.8 nodes per kilometer and lane already only half
as many RSS sessions nished successfullyapprox. 20 %
compared to approx. 40 % POP3 sessions. Figures 7 and 8
illustrate a reason for this decrease. With a rising number of
communicating nodes, network trafc on the shared medium
was increasingly dedicated to DYMO packets until, at 28.0
vehicles per kilometer and lane, the MANET was almost
exclusively busy exchanging routing messages. Reducing the
number of actively participating nodes to 10 % signicantly
improved gureseven for node densities as low as 2.8
vehicles per kilometer and lane.
To estimate the impact of overload effects on the quality
of routes established by DYMO in the VANET, the relation
between the length of a route in number of hops and the
total distance bridged between vehicle and access point was
examined. As can be seen in Figure 9, the bridged distance
is closely related to the number of hops and it is increasing
linearly by approx. 150 m per hopnot much less than the
nodes communication range of 180 m.
In order to reduce the stress imposed on the network due
to constant link breakages and subsequent ooding of route
error and new route request messages, a promising mechanism
was implemented for estimating the potential route stability by
taking movement directions into account. When comparing
two routes to nd the shortest path, DYMO now added
a malus of 0.15.0 hops for each time a packet was sent
to a vehicle traveling in the opposite direction. Information
about a vehicles relative travel direction was assumed to be
estimable by the physical layer. Figure 10 shows the results
of this adaptation. Success rates of sessions could indeed
be signicantly improved by adding such a malus, but the
adaptation failed to produce the huge effects observed by
other groups [5] when completely ignoring oncoming trafc
for route selection.
VI. CONCLUSION AND FUTURE WORK
Evaluation of the feasibility and the expected quality of
VANETs operated with the routing protocol DYMO showed
that for small amounts of payload data to be transported,
ad hoc networks of vehicles and static highway infrastructure
can be successfully setup, maintained, and used with well-
known protocols from the Internet protocol suite alone. Even
low node densities and sparse access point deployment sufced
to support routine polling of information via an Internet
gateway, e.g. the checking of a POP3 mailbox. Larger amounts
of network trafc to be transported over the ad hoc network,
however, induced overload effects that noticeably destabilized
the VANET. Particularly at higher node densities, which
commonly occurred in micro-jams, the routing and transport
protocol behavior led to a drastic increase in network load.
When the network became congested and new connections
could not be established, simple retry mechanisms only fur-
thered congestion.
Simulation results therefore seem to encourage an adapta-
tion of the protocols in use, so problems perceived by lower
layers are reacted to in a sensible way and application re-
quirements are taken into account when the network becomes
overloaded. Cross-layer optimization might keep nodes from
using potentially unstable routes for low-priority messages in
favor of a reduction of network load. Also, the results of the
conducted simulations make a simple ooding of messages
through the VANET and the selection of routes without taking
node position and mobility into account, as proposed in the
current draft of DYMO, appear wasteful. An experimental
modication of DYMO, which penalized routes across the
lanes when assessing the quality of potential routes, proved
benecial, but failed to produce the predicted increase in
overall network quality that was claimed in related work.
REFERENCES
[1] L. Wischhof, A. Ebner, H. Rohling, M. Lott, and R. Halfmann, SOTIS
- a self-organizing trafc information system, in Proceedings of the
57th IEEE Vehicular Technology Conference (VTC 03 Spring), 2003.
[2] I. Chakeres and C. Perkins, Dynamic MANET On-Demand
(DYMO) Routing, Internet-Draft, draft-ietf-manet-dymo-06.txt,
October 2006. [Online]. Available: http://tools.ietf.org/wg/manet/
draft-ietf-manet-dymo/draft-ietf-manet-dymo-06.txt
[3] C. Perkins, E. Belding-Royer, and S. Das, Ad hoc On-Demand
Distance Vector (AODV) Routing, RFC 3561, July 2003. [Online].
Available: http://www.ietf.org/rfc/rfc3561.txt
[4] C. Perkins and E. Royer, Ad hoc On-Demand Distance Vector Routing,
in 2nd IEEE Workshop on Mobile Computing Systems and Applications,
New Orleans, LA, February 1999, pp. 90100.
[5] R. Baumann, Vehicular ad hoc networks (VANET) - engineering and
simulation of mobile ad hoc routing protocols for VANET on highways
and in cities, Masters thesis, ETH Z urich, 2004.
[6] C. Sommer, I. Dietrich, and F. Dressler, Realistic Simulation of
Network Protocols in VANET Scenarios, in 26th Annual IEEE Confer-
ence on Computer Communications (IEEE INFOCOM 2007): Mobile
Networking for Vehicular Environments (MOVE 2007), Poster Session.
Anchorage, Alaska, USA: IEEE, May 2007.
[7] T. Clausen, C. Dearlove, J. Dean, the OLSRv2 Design Team, and the
MANET Working Group, MANET Neighborhood Discovery Protocol
(NHDP), Internet-Draft, draft-ietf-manet-nhdp-00.txt, June 2006.
[Online]. Available: http://tools.ietf.org/pdf/draft-ietf-manet-nhdp-00.
pdf
[8] T. H. Clausen, C. M. Dearlove, J. W. Dean, and C. Adjih, Generalized
MANET Packet/Message Format, Internet-Draft, draft-ietf-manet-
packetbb-02.txt, July 2006. [Online]. Available: http://tools.ietf.org/pdf/
draft-ietf-manet-packetbb-02.pdf
[9] A. Mahajan, Urban mobility models for vehicular ad hoc networks,
Masters thesis, Department of Computer Science, Florida State Univer-
sity, 2006.
[10] A. K. Saha and D. B. Johnson, Modeling mobility for vehicular ad-
hoc networks, in Proceedings of the 1st ACM international workshop
on Vehicular ad hoc networks, 2004.
[11] C. Lochert, A. Barthels, A. Cervantes, M. Mauve, and M. Caliskan,
Multiple simulator interlinking environment for IVC, in Proceedings
of the 2nd ACM international workshop on Vehicular ad hoc networks,
2005.
[12] M. Treiber, A. Hennecke, and D. Helbing, Congested trafc states in
empirical observations and microscopic simulations, Physical Review
E, vol. 62, p. 1805, 2000.
[13] M. Treiber and D. Helbing, Realistische Mikrosimulation von Straen-
verkehr mit einem einfachen Modell, in ASIM 2002, Tagungsband 16.
Symposium Simulationstechnik, 2002.
[14] O. Kaumann, K. Froese, R. Chrobok, J. Wahle, L. Neubert, and
M. Schreckenberg, On-line simulation of the freeway network of
north rhine-westphalia, in Trafc and Granular Flow 99, D. Helbing,
H. Herrmann, M. Schreckenberg, and D. Wolf, Eds. Heidelberg:
Springer, 2000, pp. 351356.
[15] B. Tilch and D. Helbing, Evaluation of single vehicle data in depen-
dence of the vehicle-type, lane, and site, in Trafc and Granular Flow
99, D. Helbing, H. Herrmann, M. Schreckenberg, and D. Wolf, Eds.
Heidelberg: Springer, 2000.
[16] A. Varga, The OMNeT++ discrete event simulation system, in Pro-
ceedings of the European Simulation Multiconference (ESM2001), 2001.
[17] F. Hui and P. Mohapatra, Experimental characterization of multi-hop
communications in vehicular ad hoc network, in Proceedings of the
2nd ACM international workshop on Vehicular ad hoc networks, 2005.
[18] H. Wu, M. Palekar, R. Fujimoto, J. Lee, J. Ko, R. Guensler, and
M. Hunter, Vehicular networks in urban transportation systems, in
Proceedings of the 2005 national conference on Digital government
research, 2005.