Reactive routing protocols, such as ad hoc on demand distance vector routing (AODV) or dynamic source routing (DSR), only create a link between two nodes when there is data to be transmitted. This prevents the huge amount of overhead to maintain unused routes. The obvious problem with this is that the multimedia content can not be transmitted until the link is set up. In the case of a large network, this could take on the order of seconds. Another problem is route request flooding. Flooding occurs when a node is looking for a destination, but has no knowledge of where the destination node is. It has to broadcast route requests in order to eventually find a path to the destination. These flooded route request packets could potentially disrupt ongoing communications, causing packet loss and distortion.
When multimedia data (i.e. audio, video) is transmitted through a wireless network, the end user has the ability to observe the quality of the network through the quality of the transmission. In other words, if the quality of a streaming video or audio link is poor, the end user will be able to perceive the low quality link in the form of distortion. Though this is unavoidable in some cases, using this distortion as a basis for designing and selecting networking protocols could help reduce the occurrence of such distortion and help increase the experience for the end user. The delay and reliability of a link have to be enforced.
This can be done using hybrid routing schemes for general networks, or by modeling the routing protocol based on the specific requirements of the ad hoc network. For instance, in an office environment, proactive networks would make sense since there is little motion and generally many different transmissions from each node. In contrast, a vehicle network would need a much more adaptive scheme to adapt to the constantly changing links. By intelligently choosing the protocol that both fits the situation and requirements of the end users, the overall quality of the transmitted content can be increased, resulting in a better perceived quality for the end user.