The effect of transmission variance on observer placement for source-localization.

Area of Science:

• Epidemiology
• Network Science
• Data Science

Background:

• Source localization of epidemics is critical for intervention.
• Sparse and noisy data present significant challenges in epidemic source detection.
• Observer placement significantly impacts the accuracy of epidemic source identification.

Purpose of the Study:

• To develop optimal observer placement strategies for epidemic source localization.
• To investigate the influence of network topology and transmission delay variance on observer placement.
• To propose algorithms for observer placement in both low- and high-variance transmission delay regimes.

Main Methods:

• Analyzing network topology and node-to-node transmission delay variance.
• Developing observer placement algorithms tailored to different variance regimes.
• Simulating and validating proposed methods on real-world network data.

Main Results:

• Optimal observer placement is dependent on both network topology and transmission delay variance.
• In low-variance regimes, placement based on network distances is sufficient.
• In high-variance regimes, observers must be placed closer together to mitigate noise and ensure informative data.

Conclusions:

• Novel observer placement strategies improve epidemic source localization accuracy.
• The proposed methods outperform state-of-the-art strategies in both low- and high-variance transmission delay scenarios.
• Effective observer placement is crucial for accurate epidemic source detection in complex networks.