QoS Routing for Multimedia Communication over Wireless Mobile Ad Hoc Networks: A Survey

Introduction

Delivering multimedia data over wireless MANETs has its own application domain, and also has its own challenges (Loo et al., 2012). Wireless MANET is a prominent solution in diverse emergency situations that require rescue operations, when disasters have destroyed the network infrastructure. There is no need to deploy any infrastructure to make MANET nodes to communicate with each other. The IEEE 802.11 wireless LAN can be implemented without any infrastructure or central controller (Crow et al., 1997). In ad hoc mode, all nodes participate in both data processing and routing task. The network also relies on the multi-hop type of routing for their data transmission. The concept of wireless MANET can give a new bloom to the multimedia industry because in MANETs we can provide the same without deploying any additional infrastructure. So, such networks can work in collaboration rendering services either without paying anything or paying a tiny share of revenue charged. In addition, we can define Quality of Service (QoS) models that will be used to provide the desired QoS efficiently (Nyambo et al., 2014). Existing high-speed networking and effective compression techniques utilize available bandwidth to serve a large number of connections, resulting in various multimedia applications. Basically, three types of multimedia services have been developed: accessing stored data, accessing live data, and accessing interactive data.

The diverse nature of networked multimedia applications causes various problems because each application has its own requirements (Rao et al., 2002). An application might demand better quality data and can tolerate some acceptable amount of delay, while another application may be more sensitive to delay by compromising the quality of data. Every data flow of packets from a source to a destination in the network needs certain and prerequisite resources in order desired QoS to each individual flow to be provided. Obviously, the transfer of continuous media data (e.g. video and audio) need much more bandwidth than the transfer of discrete media (i.e. text or images–i.e. lightweight data). Also, a decision must be made in order to allocate proper bandwidth to each flow available in the network. For example, Vaidya et al. (2005) have proposed a distributed fair scheduling scheme that ensures that all packets will get a proper bandwidth of the wireless channel.

Multimedia applications can be categorized as per their delay requirements (Rao et al., 2002). Non real-time applications (e.g. image transfer) have no delay requirements, while real-time applications have delay requirements, and can be classified into hard real-time and soft real-time applications. Hard real-time application has very strict delay requirements and failure in satisfying them might result in hazardous side effect. It includes various applications like nuclear-reactor control system, missile control system etc. Soft real-time application has some delay requirements, but that are somewhat less strict than hard real-time application, where failure in satisfying the delay requirements would not result in any hazardous side effects. Examples of soft real-time applications are video-conferencing and video on demand. There is also an extra categorization. Non-interactive applications (e.g. video-on-demand) do not need any type of feedback from the receiver in order to continue application function. There is only the requirement of handling one-way traffic. On the contrary, interactive applications (e.g. gaming and video streaming) require some kind of feedback or commands from the receiver, so that application can proceed further. An interactive application requires two-way traffic, each of which is very much delay sensitive. In interactive applications, one single path may lead to network congestion ultimately affecting the overall network throughput and especially QoS (Rao et al., 2002).