Study of target detection and tracking using phased array radar
He, Xin · DR-NTU (Nanyang Technological University) · 2025
This dissertation investigates target detection, direction-of-arrival (DOA) estimation, and target tracking in phased-array radar systems. A uniform linear array (ULA) with sixteen elements operating at 10 GHz is used in the simulations. Four algorithms are examined: CA-CFAR detection, MVDR beamforming, and the subspace-based DOA estimators MUSIC and ESPRIT. Monte Carlo tests are performed to study how the signal-to-noise ratio (SNR) and the number of snapshots influence detection performance and angular estimation accuracy. The results show that MUSIC can reach sub-degree precision (about 0.8 RMSE) when the SNR is above 9 dB, performing roughly 25 percent better than ESPRIT. The MVDR beamformer forms a narrow main lobe of about 6 degrees and provides more than 30 dB side-lobe suppression when moderate diagonal loading is applied. When CA-CFAR detections are combined with a constant-velocity Kalman filter, the angular variance is reduced by nearly 40 percent. This creates a complete pipeline that covers detection, localization, and tracking. These findings support an integrated processing framework for phased-array radar and offer practical guidance for parameter selection under moderate-SNR conditions.