Opportunistic Networks (ONs) are extensions of the legacy Mobile Ad-Hoc Networks (MANETs) concept. Thus, they are networks formed almost exclusively of mobile wireless devices devices (ranging from small wireless-capable sensors to smartphones and tablets) in which the presence of a continuous path between a sender and a destination is not assumed, since two nodes may never be connected to the network at the same time. An ON is highly dynamic, and the topology is thus extremely unstable and sometimes completely unpredictable.
The opportunistic networking paradigm is particularly suitable to those environments which are characterized by frequent and persistent partitions. In the field of wildlife tracking, for example, some kinds of sensor nodes are used to monitor wild species. In these cases it is not easy (nor possible sometimes) to have connectivity among a source sensor node and a destination data collector node. This happens because the animals to be monitored move freely and there is no possibility to control them in such a way to favor connectivity. Opportunistic networks may also be exploited to bridge the digital divide. In fact, they can support intermittent connectivity to the Internet for underdeveloped or isolated regions. This can be obtained by exploiting mobile nodes that collect information to upload to the Internet, as well as requests for Web pages or any kind of data that needs to be downloaded from the Internet. Both data and requests are uploaded to and downloaded from the Internet once the mobile data collector node reaches a location where connectivity is available.
Opportunistic message exchange
ONs help disseminate data and decongest currently existing backend protocols by using short-range communication over IEEE 802.11, Bluetooth, ZigBee, etc. They are also considered to be the solution that will offer vehicle-to-vehicle communication for future Vehicular Ad-Hoc Networks (VANETs). The composing nodes of an ON have no knowledge of the shape of the network when they join it. They are only aware of other nodes that they come in close proximity to, depending on the radius of their communication mechanism (e.g. Wi-Fi, Bluetooth, NFC, etc.). Thus, no assumptions are made regarding the existence of paths between nodes, since a network topology isn’t known by any node (and even if it would be known, it would much too dynamic to be used in routing mechanisms, since nodes are highly mobile and they hardly stay in the same place for long periods of time).
ONs are based on a paradigm entitled store-carry-and-forward. This implies that a node begins by storing some local data, which was either generated by itself, or was received from another node. Since ONs are characterized by a high degree of mobility, nodes move around a lot, thus carrying the stored data around the network. Finally, when the destination for a particular data item is encountered, a forwarding process is started in order to send the data to the destination. However, encountering a message’s destination is not the only situation where a data forward occurs. Opportunistic networks are based on the altruism of nodes: it is not enough for a node to see only to its interest, instead it must help other encountered nodes transfer their data (in exchange for them carrying the node’s messages as well). This mutual help is the key to ONs and helps ensure that the data is spread through the network as much as possible, thus increasing the probability of a message to reach its intended destination, and decreasing the time it takes to do so.