Yingjie Zhang

An offline, position-indexed channel-direction database can support location-aware unmanned aerial vehicle (UAV) beamforming, but sparse measurements leave intervals with different directional consistency. We propose Sampling-Adaptive Gauge-Projective Channel-Direction Reconstruction (SAGP-CDR). It treats normalized channel vectors that differ only by a common phase as the same array direction. Each anchor interval selects one of three reconstruction methods: anchor-calibrated line of sight (LOS), an aligned along-track chord, or a three-dimensional spatial affine fit. Four anchor-observable descriptors describe physical gap, endpoint agreement, spatial-neighbor agreement and directional turn. The LOS method additionally uses the station position and an assumed array geometry. Retrospective nested leave-one-flight-out evaluation uses five measured massive-MIMO UAV flights from one campaign, with 64 antennas and 100 subcarriers. The primary loss takes a median over query positions at each subcarrier, then a median across subcarriers. Across 450 matched masks, SAGP-CDR lowers mean primary direction loss by 30.25% relative to anchor-tuned aligned Gaussian interpolation. Reductions relative to the strongest fixed comparator are 14.24%, 18.84% and 3.27% at 2%, 5% and 10% retention. At 2%, loss is 0.1705 versus 0.2682 for three-dimensional nearest-anchor reuse. Relative to a global rate control calibrated on the primary metric with identical candidates, the aggregate reduction is 14.87%. These results support observable local selection for sparse channel-direction reconstruction within the evaluated campaign.

Accurate multi-step wind power forecasting is important for power-system scheduling and renewable energy integration, yet the strong nonstationarity and multi-scale structure of wind power remain challenging. This study evaluates a CEEMDAN-based empirical wavelet transform (EWT) and frequency-aligned gated recurrent unit (GRU) framework for direct six-step forecasting. Each monthly series is first partitioned chronologically into training, validation, and test sets at a 7:1:2 ratio; CEEMDAN and EWT are then applied independently within the fixed partitions used for comparative evaluation. The decomposition atoms are aggregated into fixed physical frequency bands, providing a common representation across partitions. The validation set selects the EWT band subset used to denoise the leading high-frequency representation, and independent single-channel GRUs with temporal attention forecast the retained frequency-aligned components. Component forecasts are summed and constrained to the physical operating range. On twelve monthly subsets of the T1 SCADA dataset, the block-wise framework achieves equal-month averages of 3.9846% NMAE, 236.8220 kW RMSE, and 143.4440 kW MAE, outperforming a persistence baseline and seven trained benchmarks. The paired one-sided Wilcoxon tests show significant improvements over both CEEMDAN-GRU and CEEMDAN-EWT-LSTM for all three metrics. A strict origin-wise implementation, AdaptiveEWT+SDG, recomputes features only from the history available at each forecast origin. Across six initializations, it achieves 6.0050% NMAE, 376.2803 kW mean-horizon RMSE, and 216.1808 kW MAE, versus 6.0149%, 376.9031 kW, and 216.5353 kW for Persistence, confirming the adaptive-frequency principle under strictly sequential information.