A new adaptive broadcasting approach for mobile ad hoc networks

Mobile ad hoc networks (MANETs) have been gaining tremendous attention owing to the advances in wireless technologies accompanied by many applications and implementations. However, there are still a number of issues in MANETs which require further investigations and efficient solutions. Out of these issues, broadcasting in MANETs has been a major problem for both industry and the research community. The broadcast communication is usually required to disseminate a message to all the nodes of a network. This operation is highly required in MANETs to distribute necessary information and ensure efficient control and coordination over the network nodes. However, broadcasting in MANETs is usually susceptible to several challenging communication issues, including, flooding, packets contentions and collisions, i.e., these problems all together are called the Broadcast Storm Problem (BSP). Despite a number of suggested solutions for BSP, the probabilistic scheme is considered the most promising solution due to its simplicity and suitability for MANETs. Under the umbrella of this scheme, many dynamic probabilistic broadcasting algorithms have been proposed in the literature to solve the BSP. However, most of them are not suitable for many applications including those real life scenarios as there are many limitations such as the probability of rebroadcasting operation and thresholds rebroadcasting permission, which is caused by collecting local neighbourhoods' connectivity by broadcasting HELLO packets. In an attempt to enhance and promote the quality of the probabilistic scheme, this paper proposes a new probabilistic approach to overcome these limitations. Our proposed approach is augmented with a well-know ad hoc routing protocols including Ad hoc On demand Distance Victor protocol (AODV). We have conducted intensive simulation experiments under different operating condition. The simulation results show that our proposed approach outperforms its counterparts including the well known blind flooding, fixed probabilistic and traditional dynamic probabilistic approaches.