Implementation and Evaluation of Mobility Models with OPNET Vortrag - - PowerPoint PPT Presentation

Lehrstuhl Netzarchitekturen und Netzdienste

Institut für Informatik Technische Universität München

Implementation and Evaluation of Mobility Models with OPNET

Vortrag zur Masterarbeit von Thomas Oberwallner 12.09.2012 Betreuer: Alexander Klein

Implementation and Evaluation of Mobility Models with OPNET

2

Implementation and Evaluation of Mobility Models with OPNET

3

Outline

I. Motivation II. Modules

 Import of OpenStreetMap (OSM) road maps  Routing  Mobility model  Statistics  Import/Export of GPS-Traces

III. Evaluation IV. Comparison of existing solutions

Implementation and Evaluation of Mobility Models with OPNET

4

Motivation

 Existing mobility framework in OPNET

• Mobility models: Random Waypoint, Random Direction,Random Walk,

Group Mobility

• Statistics: Spatial node distribution, node speed distribution, link duration,

transient phase

 VANET (Vehicular Ad Hoc Network) simulation

• Restricted area (roads)
• Interaction of nodes

 Goal

• Simulation of traffic on real maps
• Fast movement model of VANETs
• Import/Export Traces
• Comparison of routing protocols (AODV vs. OLSR vs. B.A.T.M.A.N.)

Implementation and Evaluation of Mobility Models with OPNET

5

Modules: Import of OpenStreetMap road maps

 File-Format: OSM

• XML-File contains 3 element types

 Node

• Point in map
• Contains latitude, longitude, id, (version, timestamp, userid, changeset)

 Way

• Contains references to nodes + additional data like highway type, speed

limit, name, surface, one way

 Relation

• Forms restrictions or areas
• Contains references to nodes and ways

 Important: Guessing of unknown data

Implementation and Evaluation of Mobility Models with OPNET

6

 Creation of routing graph by minimizing road graph

• Vertex in routing graph: intersection
• Edge: Connects two intersections

 Routing with Dijkstra algorithm  Weight of edges

• Distance for shortest path
• Time for fastest path

 Pending nodes stored in a Min-Heap

Modules: Routing

Implementation and Evaluation of Mobility Models with OPNET

7

 Creation of routing graph by minimizing road graph

• Vertex in routing graph: intersection
• Edge: Connects two intersections

 Routing with Dijkstra algorithm  Weight of edges

• Distance for shortest path
• Time for quickest path

 Pending nodes stored in a Min-Heap

Modules: Routing

Implementation and Evaluation of Mobility Models with OPNET

8

 Creation of routing graph by minimizing road graph

• Vertex in routing graph: intersection
• Edge: Connects two intersections

 Routing with Dijkstra algorithm  Weight of edges

• Distance for shortest path
• Time for fastest path

 Pending nodes stored in a Min-Heap

A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4 A B C D E 1 5 2 1 3 10 4

Modules: Routing

Implementation and Evaluation of Mobility Models with OPNET

9

Modules: Mobility model – Trip generation

 Routed-Geo-Waypoint

• Similar to Random Waypoint
• Random starting point
• Random destination

 “Levy-Flight”

• Random starting point
• Choosing destination with random distance

 Sightseeing

• Random starting point
• Routed-Geo-Waypoint route
• Return to starting point

Implementation and Evaluation of Mobility Models with OPNET

10

Modules: Mobility model – Speed

 Every car has its desired speed Normal Distribution (100, 10) Erlang-K (100,1) Beta (2,5)  Speed limits: Drivers obey speed limits depending on desired speed  Behaviour:

• If no car in front: speed = (desired speed / 100) * speed limit
• If slower car in front:
• If distance > safety-distance: speed = (desired speed / 100) * speed limit
• If distance = safety-distance: speed = speed of car in front
• If distance < safety-distance: speed = 0.95 * speed of car in front

7 2 . 5

• 7

7 . 5 : 7 7 . 5

• 8

2 . 5 : 8 2 . 5

• 8

7 . 5 : 8 7 . 5

• 9

2 . 5 : 9 2 . 5

• 9

7 . 5 : 9 7 . 5

• 1

2 . 5 : 1 2 . 5

• 1

7 . 5 : 1 7 . 5

• 1

1 2 . 5 : 1 1 2 . 5

• 1

1 7 . 5 : 1 1 7 . 5

• 1

2 2 . 5 : 1 2 2 . 5

• 1

2 7 . 5 : 0.05 0.1 0.15 0.2 .

• .

5 : . 1

• .

1 5 : . 2

• .

2 5 : . 3

• .

3 5 : . 4

• .

4 5 : . 5

• .

5 5 : . 6

• .

6 5 : . 7

• .

7 5 : 0.02 0.04 0.06 0.08 0.1 0.12 0.14

Implementation and Evaluation of Mobility Models with OPNET

11

Modules: Mobility model – Traffic Signals

 Position of traffic signal (TS) extracted from map  Timing:

• Every TS has a uniform distributed offset [0; 20]s
• Every TS has a uniform distributed duration [5; 15]s
• No yellow-phase

 Two roads with an angle closest

to 180° have a green signal at the same time

Implementation and Evaluation of Mobility Models with OPNET

12

Modules: Mobility model - Intersections

 First in, first out principle  Departure and destination of all cars which crossed the intersection

in the previous 2 seconds is stored

 Arriving car comes from direction A and drives in direction B  Waiting decision depends on the direction of previous cars:

• If no car crossed the intersection: ok
• If cars came from A: ok
• If cars went from B to A: ok
• Else: Wait 2 seconds

Implementation and Evaluation of Mobility Models with OPNET

13

Modules: Mobility Model – Driver Model/Overtaking

 Status: not implemented  Driver change after reaching destination  Characteristics of driver model:

• Desired speed
• Distance to car in front
• Overtaking possible

 Overtaking decision depends on:

• Driving out of town
• Desired speed >> Desired speed of car in front
• Distance to next intersection long enough
• No/Little opposing traffic

Implementation and Evaluation of Mobility Models with OPNET

14

Modules: Statistics

 Map metrics

• Type of roads
• Number of intersections
• Distance between intersections
• Speed limits
• Number of traffic signals

 Mobility metrics

• Node speed
• Node density
• Neighbour distance
• Speed ratio between neighbours
• Spatial dependence between neighbours
• Relative speed of neighbours
• Number of cars per section

 Network metrics

• Number of neighbours
• Number of network partitions

Implementation and Evaluation of Mobility Models with OPNET

15

Modules: Import/Export of GPS-Traces

 File Format: GPX  Wpt: Waypoint

• Attributes: Latitude, longitude
• Elements (optional): Elevation, timestamp

 Rte: Route

• Elements: Name, description, list of route points (wpt)

 Trk: Track

• Elements: Name, description, list of track segments

 Trkseg: Track segment

• Elements: List of waypoints

Implementation and Evaluation of Mobility Models with OPNET

16

Evaluation

 Impact of map

• Rural area, radial-concentric (z.B. Nördlingen), Manhattan grid (New York)

 Simulation parameters

• Number of cars
• Desired speed
• Overtaking disabled/enabled
• Range of signal

Implementation and Evaluation of Mobility Models with OPNET

17

Comparison of VANET simulators

VanetMobiSim SUMO GrooveNet Import OSM-Files No Not standalone No Mobility model Constant speed, fluid traffic motion,

• vertaking, traffic

lights Car-following, multilane roads, traffic lights Constant speed, car-following, traffic lights Trip generation Random or activity based Random Random, sightseeing Routing Dijkstra, slow Dijkstra, A*, fast Dijkstra, fast Statistics Node density Position dump, edge lane traffic, trip/route information Not aggregated Log file contains all events Not aggregated Import/Export Traces No/Yes (NS2, GloMoSim, QualNet, NET) No/Yes (unknown format) No/Yes (unknown format) Network Simulation No No Yes (hybrid simulation possible)

Implementation and Evaluation of Mobility Models with OPNET

18

Comparison of VANET simulators

Master-Thesis VanetMobiSim SUMO GrooveNet Import OSM-Files Yes No Not standalone No Mobility model Constant speed, car-following, traffic lights,

• vertaking,

driver-model Constant speed, fluid traffic motion,

• vertaking, traffic

lights Car-following, multilane roads, traffic lights Constant speed, car-following, traffic lights Trip generation Random, sightseeing, distance-based Random or activity based Random Random, sightseeing Routing Dijkstra or A*, fast Dijkstra, slow Dijkstra, A*, fast Dijkstra, fast Statistics Map metrics, Mobility metrics, network metrics Node density Position dump, edgelane traffic, trip/route information Not aggregated Logfile contains all events Not aggregated Import/Export Traces Yes/Yes GPX-Format No/Yes (NS2, GloMoSim, QualNet, NET) No/Yes unknown format No/Yes unknown format Network Simulation Yes (no hybrid simulation) No No Yes (hybrid simulation possible)

Implementation and Evaluation of Mobility Models with OPNET

19

Video

Implementation and Evaluation of Mobility Models with OPNET

20

Questions

Thank you for your time and attention. Questions?

Implementation and Evaluation of Mobility Models with OPNET

21

Timetable Task

May Jun

Jul Aug Sep Oct Nov

Study of related work / papers Familiarisation with OPNET Implementation in OPNET Modeler Simulation of different scenarios Evaluation of statistics Writing of Master Thesis