Accelerating the Kamada-Kawai algorithm for boundary detection in a mobile ad hoc network

TL;DR

This paper introduces an improved Kamada-Kawai algorithm that accelerates boundary detection in large, non-convex mobile ad hoc networks by employing novel heuristics, resulting in faster processing and accurate boundary identification.

Contribution

The paper presents a novel extension to the Kamada-Kawai algorithm that enhances scalability and accuracy for boundary detection in complex mobile ad hoc networks.

Findings

Significantly reduced processing time.

Maintained acceptable boundary detection accuracy.

Effective in large, non-convex networks.

Abstract

Force-directed algorithms such as the Kamada-Kawai algorithm have shown promising results for solving the boundary detection problem in a mobile ad hoc network. However, the classical Kamada-Kawai algorithm does not scale well when it is used in networks with large numbers of nodes. It also produces poor results in non-convex networks. To address these problems, this paper proposes an improved version of the Kamada-Kawai algorithm. The proposed extension includes novel heuristics and algorithms that achieve a faster energy level reduction. Our experimental results show that the improved algorithm can significantly shorten the processing time and detect boundary nodes with an acceptable level of accuracy.