A cross-layer fragmentation approach to video streaming over mobile ad-hoc network using BATMAN-Adv

The growing popularity of mobile devices equipped with wireless interfaces makes Mobile Ad-hoc Networks (MANET) an interesting architecture for the rapid deployment of related applications without relying on any infrastructure. This is a self-healing architecture that attracts this network in a challenging environment. Node mobility causes an ad-hoc network’s topology to be in a constant state of flux. Route changes and link failures are the most common problems in communication over ad hoc networks, causing an increase in delay, delay variation (jitter) and packet loss. In these networks, in the absence of a central scheduling node, the nodes must be always available to receive and forward packets even when they are not the origin or destination of transmission. When the MANET is dense, and the source is located more than one hop away from the destination, providing delivery ratios that meet real-time video streaming constraints is still an open issue. In this paper, we have used a cross-layered fragmentation approach to transmit high-resolution real-time multimedia data with minimal delay, high throughput and high fps despite limited bandwidth. Here image frames or video frames are segmented into smaller chunks and sent to the network as a payload of a network packet. This ensures that the data lost in our network is less compared to non-segmented frames.