Wireless ad hoc network

An ad -hoc network (Latin ad hoc, literally " for this moment made ​​" ) is a radio network that connects two or more devices to a meshed network. Networks that build self-employed and configure, also called mobile ad hoc networks (english mobile ad hoc network, MANet ) or mesh networks ( engl. mesh [ mɛʃ ] " mesh" or " grid ").

Ad -hoc networks connect mobile devices ( nodes ) such as cellular phones, personal digital assistants and laptops without any fixed infrastructure such as wireless access points. Data is passed from node to node until they reach their destinations, thus increasing the data load is distributed more advantageous than in networks with a central point of contact. Scarce resources such as computation time, energy and data rate demand an effective cooperation between the network nodes. Special routing method to ensure that the network constantly adapt as they move, add or node failure.

  • 4.1 Topology-based routing methods 4.1.1 Proactive process
  • 4.1.2 Reactive Process
  • 4.1.3 Hybrid Methods

Operation

Direct link

The individual nodes see and communicate with each other.

Indirect connection

The individual nodes are sometimes so far apart that they can not communicate directly with each other. Here, the intermediate nodes forward the data. In this manner may result in a nearly arbitrarily large, self-managing network.

Pros and Cons of a meshed wireless network

Advantages:

  • Secure version of a network;
  • If one terminal is by redirecting the data communication still possible;
  • Highly efficient;
  • Good load distribution;
  • Low network costs;
  • No central administration

Cons:

  • Comparatively complex routing required;
  • Store a routing table in each terminal;
  • Each terminal acting as a router and is therefore often active;
  • The terminal is to remain switched as possible;
  • Higher levels of consumption in the terminal

For details, see article Meshed network.

Routing process in general

In order to provide effective data forwarding in a mobile ad hoc network, special routing protocols are used. Their task is to determine a path from source to destination node. Depending on the metric used, this should be as short as possible, for example, or lightly loaded regions of the network use. Additional requirements for the protocols are as small as possible routing tables, which must be constantly updated when nodes disappear, move or appear new. The time and the number of messages that are required to find a route should be minimized.

Because of the special conditions in a mobile ad -hoc network are usually used in the Internet routing algorithms can not be used. The main reasons for this are:

  • Nodes have no prior knowledge about the topology of the network, they have to explore yourself;
  • No central instances for storing routing information;
  • Mobility of the nodes and the associated constant topology change;
  • Changing metrics of the transmission lines, for example, by interference;
  • Limited resources of the nodes ( eg, system performance, power consumption).

Routing protocols

There are more than 70 competing designs for the routing of packets by a mobile ad-hoc or mesh network. A uniform classification of protocols does not exist, but a classification into reactive and proactive protocols in the literature widely different, such as paths found and exchanged on the network. Alternative classifications distinguish whether the geodetic position is used, or how many receivers are addressed.

Topology-based routing methods

The topology- based routing methods do not require geodetic information about the positions of the nodes of the mobile ad hoc network. You only require logical information about the neighborhood relations of the nodes, ie which nodes have a direct connection or can contact through one or more intermediate nodes ( hop ). This neighboring node can communicate with each other. The topological information is often obtained by sending so-called HELLO packets. Depending on the time of building the topology database is proactive or reactive routing. An example of a protocol of this class is the Neighbourhood Discovery Protocol ( NHDP ), the elements of the Optimized Link State Routing Protocol ( OLSR ) was used.

Proactive process

Proactive routing method to determine the paths to be used between two nodes already before they are required for transmission of user data. If user data is then sent, so no need to wait for the determination of the path to the destination node. Is disadvantageous for it, however, that these procedures also send traffic from user data without a lot of control packets to determine paths that are probably not needed later. An example of a protocol of this class is the Optimized Link State Routing Protocol ( OLSR ).

Reactive methods

In contrast to the proactive method to determine reactive routing method for mobile ad -hoc networks the required paths between two nodes only if user data to be transmitted. It follows that the first data packet of a compound can only be transmitted with a delay, there must first be waiting for the completion of the route determination. Thus we need only control packets are sent when user data is sent and this is necessary for route determination. This is positively reflected in the energy consumption of the nodes. The protocol Ad hoc On-demand Distance Vector ( AODV ) is an example of a protocol this category.

Hybrid Methods

Hybrid method combining proactive and reactive routing method. Here, the target is to be achieved to combine the advantages of both approaches in a new routing protocol. For example, a proactive method can be used in a locally restricted area, while for more distant destinations, a reactive process is employed. This reduces the load on the network by the control packets, which would be shipped in a purely proactive method over the entire network. Nevertheless stand for local targets immediately paths are available, would be waiting for their determination as in purely reactive methods without. Zone Routing Protocol ( ZRP ) is a routing protocol that implements this approach.

Position -based routing methods

Position -based routing methods use geodetic information about the exact locations of the nodes. This information is obtained, for example, via the GPS receiver. Based on this local information, it is the shortest or otherwise best path between source and destination nodes determine. An example of a position-based routing protocol is LAR.

Commercial importance

In 2004, the theme was still academically and existed almost exclusively a number of pilot projects. The acquisition of MeshNetworks, a leading provider in the field of mobile ad -hoc networks, Motorola showed by the end of 2004 that the industry is expected here with fast-growing markets.

In Road Casting a mobile ad hoc network is used to distribute a customized radio program, but still is this software only a prototype.

In large-scale networking of specially designed student laptops will be done automatically via a mesh network within the non-profit project 100 dollar laptop. For this, the Massachusetts Institute of Technology, a novel protocol is installed on the mobile computers.

Metrics (networking) Ad hoc On-Demand Distance Vector Routing OLPC XO-1 Wireless community network
28387
de