The videos available below illustrate how LEPTON can be used to run experiments involving different kinds of mobility scenarios.
This video shows a communication scenario involving pedestrians moving in a two-level subway station. The mobility of pedestrians is here based on the kth/walkers dataset, which is available in the CRAWDAD database. This dataset has itself been produced with the Legion Studio simulator, using a predefined scenario. This scenario models a train platform connected via escalators to the upper entry-level. Pedestrians can arrive from several entry points of the subway station, or when trains arrive at the platforms. They can likewise leave the station through an exit point, or by boarding a train. The video covers an experiment that lasted about 11 minutes in real time, but it is played at fast speed and lasts actually less than 3 minutes. Pedestrians, represented by nodes in an opportunistic network, are depicted as dots, initially colored in orange. Radio communication between these nodes is simulated, with a transmission range of 10 meters. A grey edge between two nodes appears when these two nodes are close enough to communicate. The edge turns yellow when a transmission actually occurs.
The OppNet system used during this experiment is DoDWAN, that supports content-based communication through the publish-subscribe model. A very simple application scenario is considered here: a few seconds after the beginning of the experiment, a node (that appears in the middle-top of the area, depicted as a square) publishes a message that then starts disseminating epidemically in the network. All the other nodes turn red as soon as they receive a copy of this message.
It can be observed in this video that epidemic dissemination makes it possible for a message to continue disseminating in a given area (in that case a subway station) long after the original sender of this message has left the area. In fact the message somehow remains in the "collective memory" of all mobile devices, and it would only disappear from the subway station if that station was completely deserted.
This video shows a communication scenario involving pedestrians moving in the district of Östermalm (downtown Stockholm). The mobility of these pedestrians is based on the kth/walkers dataset, which is available in the CRAWDAD database. This dataset has itself been produced with the Legion Studio simulator in order to illustrate a typical network with churn (in which nodes keep entering and leaving the network over time). In this video the transmission range between two pedestrians is assumed to be 50 meters.
This video shows a communication scenario involving 40 vehicles moving in a residential area.The map on which vehicles are moving has been extracted automatically from OpenStreetMap. Vehicles are depicted as red dots, and a blue edge appears between two dots when they are within transmission range (200 meters in that case). An edge turns magenta when two vehicles have "discovered" each other, that is, when a connection is established between them. SInce neighbor discovery is here based on each vehicle broadcasting a beacon every 10 seconds, it sometimes takes a few seconds for two neighbor vehicles to discover one another.
After about 10 seconds in the simulation, a vehicle appears that is circled in green. This vehicle is carrying a message that is going to disseminate "epidemically" to all other vehicles in the neighborhood. As soon as a vehicle has received a copy of the message (and therefore become a "carrier" for that message), the corresponding dot turns green. In that particular scenario it takes about 60 seconds for the message to reach all vehicles roaming the area.
This video has been produced using LEPTON combined with DoDWAN nodes.
This video illustrates the mobility of buses in the city of Vannes (France). The map on which these buses are moving has been extracted automatically from OpenStreetMap, and the mobility of each bus is derived from the bus company's timetables. The color of each bus depends on the line it is running. A short edge appears between two buses when they are within transmission range (edges are almost invisible at that scale since we assume a radio range of 200 meters, and because the video is accelerated).
This video illustrates the mobility of 15 persons in an office space (which bears a striking resemblance with the place where the members of team CASA earn a living!). People are shown moving from one room to another, and sometimes staying in a room for a while (to work a little, maybe...). A grey edge appears between two persons when they are within transmission range from one another. In that case the range is assumed to be 5 meters, which corresponds approximately to what can be observed with Bluetooth LE (Low Energy) devices.
This video illustrates how mobile nodes can count themselves in a distributed way. In this particular case all nodes implement a GO-Counter (Grow Only Counter), that is, a particular kind of Conflict-Free Replicated Datatype (CRDT) that constitutes a distributed counter. As soon as a node is created, it increments its own replica of the counter, so its own presence in the network is accounted for. Afterwards each contact between two nodes is an opportunity for these nodes to synchronize their replicas. Eventually all replicas converge to the same value (in that case, 50), which is simply the number of nodes in the network.