[Paper Review] A Novel Adaptive Routing through Fitness Function Estimation Technique with Multiple QoS Parameters Compliance

T. R. Gopalakrishnan Nair, Kavitha Sooda|arXiv (Cornell University)|Jul 11, 2011

Mobile Ad Hoc Networks3 references3 citations

TL;DR

This paper proposes an adaptive routing algorithm that uses a fitness function estimation technique to optimize paths based on multiple QoS parameters (e.g., delay, jitter, bandwidth). By integrating decision attributes to prevent routing loops and the count-to-infinity problem, the method outperforms classical distance vector routing in topologies of 1–64 nodes, achieving efficient and optimal path selection with improved stability and QoS compliance.

ABSTRACT

The paper presents a method which shows a significant improvement in discovering the path over the distance vector protocol. The proposed method is a multi-parameter QoS along with the fitness function which shows that it overcomes the limitation of DV like routing loops by spanning tree approach, count-to-infinity problem by decision attribute. The input considered is a topology satisfying the QoS parameters of size 1 to 64 nodes and it was shown that an optimal path selection was obtained efficiently over the classical distance vector algorithm

Motivation & Objective

• To address the limitations of classical distance vector routing protocols, such as routing loops and the count-to-infinity problem.
• To enable optimal path selection in dynamic networks while satisfying multiple QoS constraints.
• To improve routing efficiency and stability in topologies ranging from 1 to 64 nodes.
• To integrate multiple QoS parameters (e.g., delay, jitter, bandwidth) into a unified fitness function for path evaluation.
• To develop a scalable and adaptive routing mechanism that outperforms traditional DV algorithms in performance and reliability.

Proposed method

• The method employs a fitness function that evaluates candidate paths based on multiple QoS parameters, including delay, jitter, and bandwidth.
• A decision attribute mechanism is introduced to detect and prevent routing loops and the count-to-infinity problem.
• The algorithm uses a topology-aware approach, analyzing network structures of 1 to 64 nodes to compute optimal paths.
• Path selection is guided by the fitness function, which dynamically ranks routes based on QoS compliance.
• The approach is implemented as an enhancement to the distance vector protocol, maintaining its simplicity while adding adaptive intelligence.
• The fitness function is estimated iteratively using network state information, enabling real-time adaptation to changing conditions.

Experimental results

Research questions

• RQ1 How can routing loops and the count-to-infinity problem be effectively mitigated in distance vector protocols?
• RQ2 To what extent can a fitness function based on multiple QoS parameters improve path selection in dynamic networks?
• RQ3 Can the proposed method achieve optimal path selection in small to medium-sized topologies (1–64 nodes) with better stability than classical DV?
• RQ4 How does the integration of decision attributes enhance the reliability and convergence of routing decisions?
• RQ5 What is the performance gain of the proposed method over traditional distance vector routing in terms of QoS compliance and path efficiency?

Key findings

• The proposed method successfully prevents routing loops and the count-to-infinity problem through the use of decision attributes.
• Optimal path selection was achieved efficiently in network topologies of 1 to 64 nodes, demonstrating scalability.
• The fitness function-based approach enabled effective compliance with multiple QoS parameters, including delay, jitter, and bandwidth.
• The method showed significant improvement in path discovery compared to classical distance vector routing.
• The integration of QoS-aware fitness estimation enhanced routing stability and performance in dynamic network environments.
• The results confirm that the proposed technique is a viable and effective enhancement to traditional distance vector protocols.