Motivated by the improved detection performance in multi-input multi-output (MIMO) radar systems, we apply the intelligent reflecting surface (IRS) in the phased radar. To deal with this, we consider a more practical radar antenna model and IRS element model, and the near-field channel model between phased radar and IRS is assumed. An active radar beamforming and passive IRS beamforming are proposed aiming at maximizing the signal-to-noise ratio. Meanwhile, the SNR gain by introducing IRS in phased radar is analyzed which shows that the IRS deployment has great impacts on the SNR improvement. Simulation results illustrate that the proposed method can achieve better SNR performance with proper IRS deployment. The IRS cannot be located far from the phased radar and the angle between the radar and the IRS should be carefully tuned, otherwise, the benefit is limited.
Frequency-modulated continuous-wave (FMCW) radar micro-Doppler signatures provide informative motion cues for small unmanned aerial vehicle (UAV) classification, but independent single-window predictions can vary across neighboring time windows. This study evaluates a compact scratch-trained depthwise convolutional neural network (CNN) with a deterministic three-channel observation interface and a parameter-free offline temporal logit fusion stage under a recording-disjoint six-class protocol. The classifier contains 52,918 trainable parameters; temporal fusion is applied only as final post-processing, without retraining or increasing model size. Across three recording-disjoint folds and three model seeds, the proposed three-channel single-window setting achieved macro-F1, compared with for a repeated-channel reference. Applying short-neighborhood fusion with further increased mean macro-F1 to . A seven-configuration per-channel ablation shows that no single channel accounts for this gain, and an expanded subset with varied modulation periods, frequency bands, and nominal ranges provides initial evidence on sensitivity to radar acquisition parameters. These results indicate that the proposed observation interface improves over simple channel repetition, while offline temporal aggregation provides a small additional gain for the compact classifier within the evaluated held-out-recording protocol.