Scalable Multicasting through Hexagonal Zone Based Structure Over Mobile Adhoc Networks

Abstract

Mobile Ad-hoc NETwork (MANET) is a multi-hop autonomous network formed exclusively with a collection of mobile nodes without any centralized infrastructure. A multicast routing protocol is essential to enable efficient group communication for critical applications. Recently, several stateless multicast routing protocols have evolved to achieve scalability. However, overhead in group membership management and zone construction under a highly dynamic network topology are the key issues in scalability. Hence, this paper proposes a Location Aware Multicasting Protocol (LAMP) to improve the scalability of multicast routing with limited overhead. The proposed LAMP encompasses three mechanisms such as minimum length multicast tree construction, zone based greedy multicast forwarding and mobility adaptive tree maintenance. The LAMP employs minimum length based multicast tree construction and creates the hexagonal zone-based structure. This type of structure significantly increases the coverage and the number of nodes associated with the leader of each zone. The large coverage reduces the communication overhead as well as the propagation delay considerably while performing multicast routing and tree maintenance. The hexagonal zone based greedy multicast forwarding scheme decides the multicast tree branches and strategically selects the greedy forwarder for a group of receivers to reduce the overall path length. The mobility adaptive tree maintenance optimizes the performance of multicast routing by adjusting the tree structure over concurrently changing topology. The zone members facilitate the ultimate forwarding decision by enabling the up-to-date positions. The experimental results demonstrate that the LAMP has a high throughput and significantly lower routing overhead regardless of network size, group size, and node speed. From the results, it is observed that the proposed LAMP protocol improves the throughput by 7.3% and reduces the delay by 23 % compared to the EGMP.