D-ODMRP: a destination-driven on-demand multicast routing protocol for mobile ad hoc networks

Yong‐Bin Yan, Tian Kang, Kaiqi Huang, B. Zhang, Jun ZhengIET 2012 · DOI

TL;DR

This paper introduces D-ODMRP, a novel multicast routing protocol for mobile ad hoc networks (MANETs) that significantly improves forwarding efficiency. By prioritizing paths that pass through existing multicast group members, D-ODMRP reduces the number of forwarding nodes and associated overhead, outperforming the established ODMRP protocol in simulations.

What Problem Does This Paper Address?

Mobile ad hoc networks (MANETs) are dynamic, infrastructure-less wireless networks where nodes constantly move, making efficient communication challenging. Multicast routing, which allows a single source to send data to multiple destinations simultaneously, is crucial for many MANET applications. However, existing multicast routing protocols often struggle with high forwarding overhead and fragility due to the network's dynamic nature. The paper specifically targets improving the forwarding efficiency of multicast routing in MANETs, aiming to reduce the number of nodes involved in forwarding data packets without sacrificing delivery reliability.

Methods

The core innovation of D-ODMRP lies in its "destination-driven" strategy, integrated into the on-demand routing approach of the existing ODMRP protocol. Instead of simply finding the shortest path to a destination, D-ODMRP encourages multicast paths that traverse other existing multicast group members. This is achieved by introducing a "deferring time" for Join Query packets at intermediate nodes. The longer the path from the last encountered group member, the longer the deferral. This mechanism prioritizes Join Queries that represent a "cheaper" path in terms of extra hops to existing group members, effectively encouraging the formation of more compact forwarding structures.

Key aspects of the protocol design include:

  • Join Query Phase: Nodes receiving Join Queries calculate a deferring time (JQDelay) based on the "ExtraHop" distance (hops from the last met group member). This delay is designed to favor paths with lower ExtraHop values.
  • Duplicate Join Query Handling: Intermediate nodes can process duplicate Join Queries and select the one with the least ExtraHop, updating their internal state.
  • Join Reply Phase: Similar to ODMRP, Join Replies are sent back along the learned reverse path to establish the forwarding structure.
  • Data Forwarding Phase: Forwarders broadcast received multicast data packets, with a small random delay to mitigate collisions.
  • Loop Avoidance: A simple rule is implemented: a node cannot accept a Join Query that has already passed through itself in the current routing discovery round, preventing loops.

Datasets

Not discussed in the paper. The evaluation relies on simulations.

Key Insights and Ideas

  • The "Destination-Driven" Advantage: The core insight is that by intelligently guiding the path discovery to leverage existing group members, the overall forwarding structure can be made more efficient. This is a departure from purely source-centric or shortest-path approaches.
  • Deferring Time as a Cost-Aware Mechanism: The introduction of a calculated deferring time for Join Queries is a clever way to implicitly encode cost (in terms of extra hops) into the path selection process without requiring complex calculations or global network state.
  • Trade-off Between Efficiency and Robustness: The paper highlights that while D-ODMRP improves efficiency, ODMRP's inherent redundancy in forwarders makes it more resilient to network dynamics at higher mobility speeds, albeit at the cost of efficiency. This suggests a tunable trade-off depending on network requirements.
  • Generalizability of the Strategy: The authors suggest that the destination-driven strategy is not limited to ODMRP and can be applied to other multicast routing protocols, indicating its potential broad impact.

Key Results

Simulations using GloMoSim with 50 randomly moving nodes demonstrated that D-ODMRP significantly improves forwarding efficiency compared to ODMRP.

  • Forwarding Overhead Reduction: D-ODMRP reduced normalized forwarding overhead by up to 15% at higher mobility speeds, up to 23% with larger multicast group sizes, and up to 21% with more multicast groups.
  • Packet Delivery Ratio: D-ODMRP achieved a similar packet delivery ratio to ODMRP, indicating that the efficiency gains did not come at the expense of reliability.
  • Mobility Impact: At low mobility speeds, D-ODMRP performed comparably to ODMRP. As mobility increased, the difference in forwarding overhead became more pronounced, with D-ODMRP showing better efficiency.

Limitations and Future Work

  • Loop Avoidance Simplicity: While the loop avoidance mechanism is simple, its effectiveness in all complex dynamic scenarios might warrant further investigation.
  • Performance at High Mobility: While D-ODMRP shows efficiency gains, ODMRP's higher redundancy offers better robustness at very high mobility speeds, suggesting a potential area for hybrid approaches.
  • Parameter Tuning: The optimal values for parameters like T (deferring time base) and Cmax (maximum deferral constant) might be network-dependent and require careful tuning.
  • Future Work: The authors suggest that the destination-driven strategy can be applied to other existing multicast routing protocols, implying potential for further research in this direction. They also mention the possibility of introducing this strategy into other protocols like MAODV and PatchODMRP.

Summary updated

Abstract

This article proposes a destination-driven on-demand multicast routing protocol (D-ODMRP) to improve the multicast forwarding efficiency in mobile ad hoc networks (MANETs). In D-ODMRP, the path from the multicast source to a multicast destination tends to use those paths passing through another multicast destination. If such multiple paths are available, the one leading to the least extra cost is preferred. This destination-driven strategy is introduced into the on-demand construction process of a multicast forwarding structure in a popular multicast protocol ODMRP. Simulation results show that D-ODMRP can significantly improve the forwarding efficiency as compared with ODMRP. Moreover, the destination-driven strategy can also be introduced into other existing multicast routing protocols for MANETs.
BibTeX
@inproceedings{yan2012d,
  author = {Yong‐Bin Yan and Tian Kang and Kaiqi Huang and B. Zhang and Jun Zheng},
  title = {{D-ODMRP: a destination-driven on-demand multicast routing protocol for mobile <i>ad hoc</i> networks}},
  booktitle = {IET},
  year = {2012},
  doi = {10.1049/iet-com.2011.0033},
}
Figure