[Paper Review] Probability Based Adaptive Invoked Clustering Algorithm in MANETs

S. Rohini, K. Indumathi|arXiv (Cornell University)|Feb 9, 2011

Mobile Ad Hoc Networks5 references3 citations

TL;DR

This paper proposes a Probability Based Adaptive Invoked Weighted Clustering Algorithm (PAIWCA) for mobile ad hoc networks (MANETs) that enhances cluster stability by dynamically selecting cluster-heads based on node battery power and a probability-based metric. The algorithm reduces cluster re-election overhead and outperforms the Maximal Weighted Independent Set (MWIS) in network connectivity, throughput, and packet delivery ratio under varying network conditions.

ABSTRACT

A mobile ad hoc network (MANET), is a self-configuring network of mobile devices connected by wireless links. In order to achieve stable clusters, the cluster-heads maintaining the cluster should be stable with minimum overhead of cluster re-elections. In this paper we propose a Probability Based Adaptive Invoked Weighted Clustering Algorithm (PAIWCA) which can enhance the stability of the clusters by taking battery power of the nodes into considerations for the clustering formation and electing stable cluster-heads using cluster head probability of a node. In this simulation study a comparison was conducted to measure the performance of our algorithm with maximal weighted independent set (MWIS) in terms of the number of clusters formed, the connectivity of the network, dominant set updates,throughput of the overall network and packet delivery ratio. The result shows that our algorithm performs better than existing one and is also tunable to different kinds of network conditions.

Motivation & Objective

• To address instability in MANET clustering due to frequent cluster-head re-elections.
• To reduce network overhead caused by dynamic cluster reconfiguration in mobile environments.
• To enhance network performance by selecting stable cluster-heads based on energy efficiency and node reliability.
• To develop a tunable clustering algorithm adaptable to diverse network conditions.
• To improve key metrics such as throughput, packet delivery ratio, and network connectivity.

Proposed method

• The PAIWCA algorithm uses a probability-based mechanism to determine the likelihood of a node becoming a cluster-head, factoring in residual battery power.
• Nodes calculate their cluster-head probability using a weighted function that includes energy level and other stability metrics.
• The algorithm invokes clustering only when necessary, minimizing unnecessary reconfiguration and reducing signaling overhead.
• Cluster formation is based on a weighted selection process that prioritizes energy-efficient and stable nodes.
• The method dynamically updates the dominant set of cluster-heads to maintain network connectivity and reduce re-election frequency.
• Performance is evaluated against MWIS using metrics including number of clusters, connectivity, throughput, and packet delivery ratio.

Experimental results

Research questions

• RQ1 How does the proposed PAIWCA algorithm improve cluster stability compared to existing methods like MWIS?
• RQ2 To what extent does incorporating battery power into cluster-head selection reduce cluster re-election overhead?
• RQ3 How does the probability-based mechanism affect network connectivity and throughput in dynamic MANETs?
• RQ4 Can the algorithm be tuned to perform well across varying network densities and mobility patterns?
• RQ5 What is the impact of adaptive invocation on signaling overhead and overall network efficiency?

Key findings

• PAIWCA reduces cluster re-election frequency by leveraging energy-aware probability calculations, enhancing cluster stability.
• The algorithm achieves higher network throughput compared to the MWIS algorithm under varying mobility and density conditions.
• Packet delivery ratio is significantly improved due to more stable and energy-efficient cluster-head selection.
• Network connectivity remains consistently high across simulations, indicating robust cluster formation and maintenance.
• The performance of PAIWCA is tunable and adaptable to different network conditions, demonstrating flexibility.
• The algorithm outperforms MWIS in all evaluated metrics: number of clusters, dominant set updates, throughput, and delivery ratio.