LEPTON is a Lightweight Emulation PlatTform for Opportunistic Networking. It has been primarily designed to allow the developers of real opportunistic networking software (i.e., middleware and/or applications) to run their software systems with simulated mobility.
Unlike other emulation platforms, LEPTON does not require exotic networking equipment, and it does not even require deploying virtual machines on one or several hosts. A simple laptop or desktop workstation running Linux can easily support emulation-based experiments involving up to a couple hundred nodes, and larger experiments can be run on any cluster of Linux machines.
Being an emulator rather than a simulator, LEPTON is meant to drive the communication between full-featured instances of an opportunistic networking system (a.k.a. OppNet system), each instance determining the behavior of one system node (SN) during the simulation. To date only a few existing OppNet systems are openly distributed, but LEPTON can currently operate with several of these systems.
In the following we assume that LEPTON has been installed on your laptop or desktop workstation. If that is not the case yet, please refer to the installation procedure.
Start LEPTON on your host:
lepton.sh start
This command starts LEPTON with default configuration parameters. With
this configuration (which is defined in conf/lepton.conf) the
display window represents a 200 m. x 200 m. area, in which 20 nodes
are created automatically at startup. These nodes move according to
the Levy Walk mobility
pattern. A blue vertex between two nodes means that these nodes are
"within radio range", that is, the distance between them does not
exceed 30 meters.
In this example the nodes you observe are just blips on the display screen: they are otherwise non-existent, and have no set behavior. This is because LEPTON is just simulating the mobility of purely virtual nodes. Things will change when actual nodes (e.g., instances of an OppNet system) are running concurrently with LEPTON, which will then drive communications between these nodes.
Buttons in the display window allow you to change the speed at which time is running, and thus the speed at which the nodes are moving. Again this is possible because LEPTON is currently running in plain simulation mode. When running with instances of an OppNet system, the only possible speed will be "real time".
You can check if LEPTON is running at any time:
lepton.sh status
And the parameters used in the current execution can likewise be displayed:
lepton.sh info
At runtime LEPTON produces a number of file, which by default are
saved in /dev/shm/${USER}/lepton. Among these files is a
DGS
(Dynamic Graph Stream) file, which represents the evolution of the
dynamic graph you can see in the display window:
tail -f /dev/shm/${USER}/lepton/lepton.dgs
(Use Ctrl-C to terminate)
Now let us stop LEPTON:
lepton.sh stop
Let us make a copy of the DGS file we've just produced in /tmp, and let us start LEPTON again, but this time with an option specifying that it should replay that file:
cp /dev/shm/${USER}/lepton/lepton.dgs /tmp/myTuto.dgs
lepton.sh start in_dgs=/tmp/myTuto.dgs
This time LEPTON does not determine the mobility of each node according to the Levy Walk mobility pattern: it simply replays the movements recorded earlier in the DGS file. Radio contacts (visualized as vertices in the graph) are likewise determined by the DGS file.
Let us assume we would like to start LEPTON again, but this time with 40 nodes moving in the display window:
lepton.sh stop
lepton.sh start nodes=40
As you can see, any parameter defined in the default configuration file can be overwritten in the command line.
It may sometimes be useful to control the connectivity between mobile nodes, as this connectivity determines whether these nodes can communicate together or not. This can be obtained by setting LEPTON's mobility simulator to "manual mode":
lepton.sh stop
lepton.sh start nodes=5 manual=true
With the last command, only 5 nodes are created in the display window, and no mobility pattern is applied to move these nodes. LEPTON applies an "auto-layout" pattern, which means that it tries to distribute the nodes evenly in the window. Besides the distance between nodes in the window has no incidence on their being connected or not: you are the one who can decide which pairs of nodes should be connected or disconnected.
To connect two nodes together, first select the edge editing option
(button in the bottom-right side of the window). Then select a node,
and with a drag-and-drop action, connect this node to another node. A
blue vertex should appear to show that these nodes are now connected.
To disconnect two nodes, select the edge between these nodes and press the delete key on your keyboard.
This manual mode is mostly useful when each node on the screen corresponds to an instance of an OppNet system that is actually running in the background. In such a case the manual mode comes handly, as it makes it possible to run tests easily with a couple of nodes, before starting to run experiments involving dozens or hundreds of nodes.
Now let us stop LEPTON again, before proceeding with this tutorial.
lepton.sh stop
Examples of mobility and/or contact scenarios can be found in the
examples directory. After the first installation of LEPTON, this
directory is initially empty, but it can be populated by downloading
pre-defined scenarios.
The following command displays the list of available scenarios:
bin/util/load_example.sh
Let assume you would like to download the scenario entitled vannes_tohannic:
bin/util/load_example.sh vannes_tohannic
This scenario defines the mobility of 50 vehicles and 30 pedestrians in a residential area, in the city of Vannes (France), assuming a radio range of 50 meters. To run LEPTON with this scenario, just type:
lepton.sh start conf=examples/vannes_tohannic/lepton.conf
You will see that vehicles are depicted as (fast) red nodes, and pedestrians
as (slow) blue nodes. All the parameters required to run this
mobility scenario are defined in the files saved in
examples/vannes_tohannic.
Do not forget to stop DoDWAN when you are done watching blips moving on the screen:
lepton.sh stop
So far all you have done is using LEPTON as a mobility simulator: the nodes displayed on the screen are nothing but blips on a screen. Now if you have gone this far reading this tutorial, you are probably ready --and maybe eager-- to try running LEPTON wih "real nodes".
To date LEPTON can operate with a number of compatible OppNet systems.
You may now select one of these systems, and install the selected system on your host platform, together with the corresponding adapter.
Once both the OppNet system you have selected and its adapter have been installed, you are ready to start LEPTON again, but this time each mobile node displayed on the screen will correspond to a system node running in the background:
lepton.sh start oppnet_adapter=<path/to/oppnet/adapter.sh>
... where the last argument must of course indicate where is the adapter to the OppNet system you have installed.
Please refer to the documentation provided with each adapter for details about how to install and use this adapter.