Clustered Multicast Routing for Mobile Ad Hoc Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional multicast routing protocols in ad hoc networks face scalability issues and high overhead due to the need for extensive control message exchange, especially in large networks with dynamic topologies and limited resources.
Innovation Solution
A multicast routing method that partitions the network into clusters, where each cluster autonomously decides data packet forwarding and maintains a distance vector routing table to minimize control message exchange and reliance on core nodes, allowing data packets to hop between clusters without a complex shared tree or mesh structure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If tree-based multicast routing protocols are used, then the number of datagrams to be forwarded is reduced, but the protocol heavily depends on the integrity of the core node and may lead to severe multicast routing failure if a core node is deficient
Solution Approach 1:
The patent segments the network into multiple clusters with independent cluster heads, eliminating the single point of failure (core node) in tree-based protocols. Each cluster autonomously manages its own routing, distributing the routing function across multiple nodes rather than relying on a central core node, thus improving reliability while maintaining efficient data forwarding.
2Reliability
If mesh-based multicast protocols are used, then robustness is improved through multiple paths, but the number of redundant paths increases the complexity of routing
Solution Approach 1:
The patent divides the network into clusters where each cluster head maintains a simplified routing table with only essential entries. This segmentation reduces the complexity at each node while maintaining overall network robustness through inter-cluster routing paths.
Solution Approach 2:
Each cluster head operates with local autonomy, maintaining only the routing information necessary for its own cluster and immediate neighbors. This local quality approach reduces the routing complexity at each node while preserving robustness through distributed decision-making across cluster boundaries.
3Device complexity
If conventional flat multicast routing protocols are used, then simplicity is maintained, but scalability deteriorates as the network size increases
Solution Approach 1:
The patent segments the network into clusters with dedicated cluster heads that manage routing locally. This segmentation enables the network to scale by adding more clusters without increasing the complexity at individual nodes, as each cluster head only manages its local cluster and neighboring clusters.
Solution Approach 2:
The patent introduces a hierarchical dimension by electing cluster heads that operate at a higher level of abstraction. This dimensional change allows the network to scale from flat to hierarchical structure, improving scalability while maintaining manageable complexity through the new organizational layer.
4Productivity
If clustered routing protocols are used to improve scalability, then control message overhead increases due to extensive exchange of control messages
Solution Approach 1:
The patent implements self-service by allowing each cluster head to autonomously maintain its own routing table and make forwarding decisions without requiring extensive control message exchange. Cluster heads independently manage their clusters and only exchange necessary routing information with neighboring cluster heads, significantly reducing control message overhead while maintaining scalability.
Solution Approach 2:
The patent extracts the routing function from centralized control and assigns it to individual cluster heads. This extraction eliminates the need for extensive control message exchange across the entire network, as each cluster head operates independently with locally maintained routing information, reducing control message overhead while preserving scalability.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
The invention relates to a method for routing a multicast data packet in an ad hoc network, said network being partitioned into clusters of nodes, each cluster having a cluster head, comprising: - a step (220, 620) of discovering the topology of the network by said cluster heads so that each cluster head has a description (330-335,730) of the network topology; - a step (230, 630) of exchanging multicast membership information among said cluster heads so that each cluster head has a multicast membership table (SM) indicating, for each cluster, which multicast group(s) is(are) represented therein; - said multicast data packet being routed through the network by a plurality of cluster heads, each cluster head independently deciding on the next cluster head(s) to forward the packet to, if any, on the basis of the multicast group (m) the data packet is destined to, said multicast membership table and said description of the network topology.