Robust dissimilarity measure for Network Localization

TL;DR

This paper introduces a robust dissimilarity measure for network localization that effectively handles outliers and Gaussian noise, using a convex underestimator based on the Huber penalty for efficient polynomial-time optimization.

Contribution

It proposes a novel robust dissimilarity model with a convex underestimator for network localization, improving performance under noisy and outlier conditions.

Findings

Outperforms traditional methods in the presence of outliers

Effective under high-power noise and Gaussian noise

Convex underestimator enables polynomial-time optimization

Abstract

In practice, network applications have to deal with failing nodes, malicious attacks, or, somehow, nodes facing highly corrupted data --- generally classified as outliers. This calls for robust, uncomplicated, and efficient methods. We propose a dissimilarity model for network localization which is robust to high-power noise, but also discriminative in the presence of regular gaussian noise. We capitalize on the known properties of the M-estimator Huber penalty function to obtain a robust, but nonconvex, problem, and devise a convex underestimator, tight in the function terms, that can be minimized in polynomial time. Simulations show the performance advantage of using this dissimilarity model in the presence of outliers and under regular gaussian noise.