Per-Class Precision, Recall, and F1-Score for a Dual-Static Passive Radar System

Hongmin Wang published performance metrics for a joint spatiotemporal-geometry framework in passive radar on 2026-06-02. The dataset contains per-class precision, recall, and F1-score results from 500 Monte Carlo trials per SNR point across five evaluation seeds. The proposed method achieved a mean classification accuracy of 93.7% and a mean localization error of 1.15 km at -6 dB SNR.

Use Cases

• Benchmarking classification algorithms based on reported precision, recall, and F1-scores.
• Analyzing the performance of joint optimization methods under varying SNR conditions mentioned in the description.
• Comparing localization error reduction against geometry-only baseline methods described in the study.

Strengths

• Performance metrics are derived from 500 Monte Carlo trials per SNR point, providing a statistical basis.
• Results include specific numerical outcomes: a mean classification accuracy of 93.7% and a 28.1% reduction in localization error.
• Data is associated with a published research methodology involving a joint spatiotemporal-geometry framework.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 5.5 KB, indicating limited scope, likely containing summary statistics rather than raw trial data.

Provenance

Source

figshare, author Hongmin Wang

Collection Method

Likely contains aggregated results from Monte Carlo simulations of a passive radar system using DVB-T broadcast signals at 650 MHz.

Freshness

Last updated 2026-06-02 17:36:00; freshness should be verified.